- Email: [email protected]
Efflux-mediated heavy metal resistance in prokaryotes
FEMS Microbiology Reviews 27 (2003) 313^339
www.fems-microbiology.org
E¥ux-mediated heavy metal resistance in prokaryotes Dietrich H. Nies
Institute of Microbiology, Molecular Microbiology, Martin-Luther-University Halle-Wittenberg, Kurt-Mothes-Str. 3, 06099 Halle/Saale, Germany Received 3 October 2002; received in revised form 14 January 2003; accepted 31 March 2003 First published online 30 April 2003
Abstract What makes a heavy metal resistant bacterium heavy metal resistant? The mechanisms of action, physiological functions, and distribution of metal-exporting proteins are outlined, namely: CBA efflux pumps driven by proteins of the resistance^nodulation^cell division superfamily, P-type ATPases, cation diffusion facilitator and chromate proteins, NreB- and CnrT-like resistance factors. The complement of efflux systems of 63 sequenced prokaryotes was compared with that of the heavy metal resistant bacterium Ralstonia metallidurans. This comparison shows that heavy metal resistance is the result of multiple layers of resistance systems with overlapping substrate specificities, but unique functions. Some of these systems are widespread and serve in the basic defense of the cell against superfluous heavy metals, but some are highly specialized and occur only in a few bacteria. Possession of the latter systems makes a bacterium heavy metal resistant. 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : CDF; RND ; Cadmium; Cobalt; Zinc ; Nickel; Chromate ; Ralstonia metallidurans
Contents 1. 2.
3.
4.
Introduction: what kind of heavy metal resistance will be described . . . . . . . . . . . . . . . . . The ¢rst layer of heavy metal resistance: members of the resistance^nodulation^cell division (RND) protein family export super£uous cations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. CzcCBA-mediated e¥ux is the outer shell of Co2þ , Zn2þ , Cd2þ and Ni2þ resistance in R. metallidurans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Localization and speciation of the cations e¥uxed by CzcCBA . . . . . . . . . . . . . . . . . 2.3. The Cus system from E. coli: is there direct export from the periplasm? . . . . . . . . . . 2.4. Export of organic substrates by HAE-RNDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Biochemistry of RND proteins: towards a model of function for these proton-driven antiporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. MFPs and OMFs enhance the function of RND proteins . . . . . . . . . . . . . . . . . . . . . 2.7. Distribution of RND proteins in the kingdoms of life . . . . . . . . . . . . . . . . . . . . . . . . The second layer of metal resistance : cation di¡usion facilitators (CDF family) serve as secondary cation ¢lters in bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Physiological function and substrate speci¢city of CDF proteins . . . . . . . . . . . . . . . . 3.2. Biochemistry of CDF proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Distribution of CDF proteins in the living world . . . . . . . . . . . . . . . . . . . . . . . . . . . . The third layer of heavy metal resistance: P-type ATPases are the basic defense against heavy metal cations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Families of CPx-type ATPases and their substrates . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Structure and function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Distribution of CPx-type ATPases in prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . .
314 314 315 316 316 317 318 319 322 324 324 325 325 326 328 328 328
* Tel. : +49 (345) 5526352; Fax: +49 (345) 5527010. E-mail address : [email protected] (D.H. Nies). 0168-6445 / 03 / $22.00 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-6445(03)00048-2
FEMSRE 781 2-6-03
Cyaan Magenta Geel Zwart
314
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
5.
........ ........ ........ ........ ........ many sys........
330
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
6. 7.
Other heavy metal export systems: CHR protein family, NreB, CnrT . . . . . . . . 5.1. CHR proteins detoxify chromate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. NreB-like proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. CnrT-like proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Heavy metal resistance in prokaryotes revisited . . . . . . . . . . . . . . . . . . . . . . . . Concluding remarks: heavy metal resistance is the product of the interaction of tems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction : what kind of heavy metal resistance will be described Metal resistance determinants were initially found on bacterial plasmids [1^13]. However, after the genome sequences of several bacteria became available, homologues to metal resistance determinants popped up in many bacteria (see below). If the possession of a heavy metal resistance determinant makes a bacterium metal resistant, which bacteria remain sensitive? Since Escherichia coli serves as model bacterium for many purposes, metal resistance of a given bacterium could be compared to metal resistance of E. coli [14]. This comparison was done with the well-characterized heavy metal resistant bacterium Ralstonia metallidurans strain CH34 (previously Alcaligenes eutrophus or Ralstonia sp. [15]). It originates from a zinc decantation tank and related bacteria have been found in many metal-contaminated environments around the world [16,17]. The minimal inhibitory concentrations (MICs) of the biologically important heavy metals [18] for R. metallidurans strain CH34 and for E. coli strain K38 were compared in Trisbu¡ered mineral salts medium [19]. Growth of neither bacterium was inhibited by up to 50 mM MoO23 4 , 23 3þ WO4 and, due to its low solubility, Fe . Both species had a MIC of 2 mM UO23 2 . E. coli was more resistant to 33 and SbO3 VO3 3 , AsO4 2 than R. metallidurans; however, R. metallidurans was 2.5-fold more resistant to CrO23 4 than E. coli. Interaction of microbes with arsenic has recently been reviewed [20] and will not be elucidated here. Thus, of these seven biologically important oxyanionforming heavy metals [18], only chromate resistance will be discussed. The ¢rst part of this review will concentrate on the heavy metal cations. The MIC values of both bacteria to biologically important heavy metal cations (with the exception of iron) were plotted against the solubility constants of the respective metal sul¢de complexes ([21], Fig. 1). This was done because heavy metal cations that escape metal e¥ux systems may cause toxicity by interaction with cellular thiol compounds such as glutathione. The solubility of the sul¢de complexes may therefore serve as a measure of the a⁄nity of a metal to thiol compounds and for the severity of damage resulting from this interaction. The di¡erent de-
FEMSRE 781 2-6-03
329 329 329 329 330
grees of metal resistance roughly followed the solubility constants (regression coe⁄cients s 80%) ; so, this interaction may indeed be part of metal toxicity. Metal resistance of R. metallidurans was in general higher than metal resistance of E. coli (Fig. 1). Especially resistance of R. metallidurans to the cations of cobalt, zinc, nickel, and copper was very high, cadmium resistance was similar to that of E. coli. This review will focus on the description of all resistance systems that are responsible for this remarkable heavy metal resistance of R. metallidurans and it will give an overview of the distribution of these heavy metal resistance determinants in other organisms.
2. The ¢rst layer of heavy metal resistance : members of the resistance^nodulation^cell division (RND) protein family export super£uous cations The RND protein family was ¢rst described as a related group of bacterial transport proteins involved in heavy metal resistance (R. metallidurans), nodulation (Mesorhizobium loti) and cell division (E. coli) [22]. This family has grown into a huge superfamily that includes seven protein families that can be found in all major kingdoms of life [23]. In bacteria and archaea members of this superfamily are involved in transport of heavy metals, hydrophobic compounds, amphiphiles, nodulation factors, and, with SecDF, in protein export. The members of the RND superfamily in eukaryotes transport sterols or serve as receptors [24^45]. This section deals with the heavy metal e¥ux family (HME-RND [23]) of the RND superfamily that has received the number TC 2.A.6.1.1 in Milton Saier’s functional-phylogenetic classi¢cation system for transmembrane solute transporters [46]. RND proteins involved in export of hydrophobic and amphiphilic compounds (HAE-RND) will be included, since data obtained for these systems allow the development of models for the function of RND proteins and RND-driven e¥ux complexes. Therefore, the expression ‘RND protein’ will be used exclusively hereafter for HME-RND and HAERND proteins, but not for eukaryotic proteins or bacterial proteins involved in protein export. Bacterial RND proteins like company. Adjacent to a
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 Mn2+
Zn2+ Co2+ Ni2+
Cd2+
Ag+ Hg2+
Cu2+
Pb2+
log (MIC, [mM])
10
1
0.1
0.01 20
30
40
50
- log (KMeS) Fig. 1. Heavy metal toxicity depends on the a⁄nity of a cation to sulfur. The logarithm of the MIC of heavy metal cations for E. coli (open circles) and R. metallidurans (closed circles) was plotted against the logarithm of the solubility for the respective metal sul¢de as a measure of the a⁄nity to sulfur [22]. The positions of the cations are indicated with arrows at the top. An exponential curve ¢t was performed for both data sets that yielded regression coe⁄cients for E. coli (thin line) of 81.1% and for R. metallidurans (thick line) of 86.7%.
gene that encodes a RND protein is ^ in most cases ^ a second gene that encodes a member of the membrane fusion protein family MFP [22]. This family has also been designated periplasmic e¥ux protein family [47] or, more recently, periplasmic adapter protein family [48]. This review will use the oldest scienti¢c name, that is MFP. Besides MFPs, many RND proteins cooperate with yet a third protein that belongs to the family of outer membrane factors (OMF) [47,49]. These three proteins form an e¥ux protein complex that may export its substrate from the cytoplasm, the cytoplasmic membrane or the periplasm across the outer membrane directly to the outside [50^53]. To distinguish this type of transport system from ATP-driven ABC transport systems, the RND-driven export systems will be referred to as CBA e¥ux systems or CBA transporters. 2.1. CzcCBA-mediated e¥ux is the outer shell of Co2+ , Zn 2+, Cd 2+ and Ni 2+ resistance in R. metallidurans The ¢rst identi¢ed member of the RND family was the CzcA protein from R. metallidurans followed by the CnrA protein from the same bacterium [22]. This L-proteobacterium contains two megaplasmids involved in heavy metal resistance [17,19]. The presence of the larger of both plasmids, pMOL30, increased the MICs to Zn2þ 50-fold, to
FEMSRE 781 2-6-03
315
Co2þ 33-fold and to Cd2þ seven-fold [19], while the second plasmid, pMOL28, mediated resistance to Ni2þ and a 16fold increase in cobalt resistance. The plasmid pMOL30-mediated resistance to Co2þ , Zn2þ and Cd2þ was designated Czc and the structural gene region of the czc resistance determinant of pMOL30 was cloned [54] and sequenced [55,56]. This structural gene region contains the genes for the OMF CzcC, the MFP CzcB and the CzcA protein of the RND family. The three genes form an operon czcCBA that is transcribed tricistronically [57] and is £anked by a multitude of genes involved in metal-dependent regulation of czcCBA expression (C. GroMe, A. Anton, T. Ho¡mann, S. Franke, G. Schleuder and D.H. Nies, unpublished) [58^ 60]. The czcCBA structural gene region was expressed in a plasmid-free R. metallidurans strain under the control of a constitutive promoter on a low copy number vector. The presence of the czcCBA genes in this bacterial strain mediated a metal resistance level similar to that reached by the complete megaplasmid pMOL30 [54,55]. Czc-mediated resistance to Co2þ , Zn2þ and Cd2þ was based on a diminished cellular accumulation of the three cations resulting from cation e¥ux [61] driven by the proton motive force [62]. The cobalt^nickel resistance determinant cnr on plasmid pMOL28 is also based on cation e¥ux [61,63]. The resistance determinant is composed of a cnrCBA structural region [64] that is preceded by the regulatory gene region cnrYXH [65,66]. Similar to czcCBA, cnrCBA encodes the OMF CnrC, the MFP CnrB and the RND protein CnrA. Another cobalt^nickel resistance determinant, ncc, was cloned from R. metallidurans strain 31A (previously Achromobacter xylosoxidans subsp. xylosoxidans) and further characterized [67]. Similar to cnr, ncc is composed of a regulatory gene region nccYXH followed by the structural region nccCBA, again encoding a putative outer membrane protein (NccC), a MFP (NccB), and an RND protein (NccA). In contrast to Cnr, Ncc mediates resistance to cadmium in addition to cobalt and nickel resistance [67]. Czc alone seemed to be responsible for the increase in Co2þ , Zn2þ and Cd2þ resistance that was assigned to plasmid pMOL30. However, the genomic sequence of R. metallidurans (preliminary sequence obtained from DOE/ JGI, http://www.jgi.doe.gov/JGI_microbial/html/ralstonia/ ralston_homepage.html) indicated the presence of a ncc determinant very similar to ncc from R. metallidurans strain 31A in the genome of strain CH34. The ncc determinant from R. metallidurans was assigned to plasmid pMOL30 by a polymerase chain reaction and homologous recombination experiment (S. Franke and D.H. Nies, unpublished). Thus, Ncc from R. metallidurans strain CH34 may contribute to the cobalt, zinc and cadmium resistance mediated by plasmid pMOL30. The exclusive assignment of this resistance to Czc could be an experimental artifact, e.g. resulting from the higher copy number of the vector
Cyaan Magenta Geel Zwart
316
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
plasmid compared to the native plasmid pMOL30. However, this possibility was excluded when the czcA gene was deleted from pMOL30. The resulting pMOL30(vczcA)containing mutant strain was even more sensitive to cobalt, zinc and cadmium than the plasmid pMOL30-free strain [68]. Since polar e¡ects were carefully avoided when the vczcA deletion was constructed, this indicates clearly the essential function of Czc as the outermost shell of cobalt, zinc, and cadmium resistance in R. metallidurans. In agreement with this fact, no regulatory genes were found upstream of nccCBA on plasmid pMOL30 and no expression of this operon could be measured (G. Rehbein, C. Grosse and D.H. Nies, unpublished). Thus, ncc seems to be a silent determinant in R. metallidurans CH34. 2.2. Localization and speciation of the cations e¥uxed by CzcCBA Why does the CzcCBA e¥ux system mediate such a huge increase in metal resistance? Maybe, this system diminishes not only the cytoplasmic concentration of heavy metal cations but additionally (or even exclusively?) the periplasmic concentration. In this case, the cations could be removed before they had the opportunity to enter the cell. Moreover, the CzcCBA e¥ux system could mediate further export of the cation that had been removed from the cytoplasm by other e¥ux systems. Such export of heavy metal cations from the periplasm may happen indirectly, that is by the combined action of an uptake and a CBA transenvelope e¥ux system, but periplasmic cations could also be directly funneled into a CBA transenvelope e¥ux system. Yet, both mechanisms may contribute to the physiological function of CzcCBA and other CBA transport systems. As shown with puri¢ed protein that has been reconstituted into proteoliposomes, CzcA should be able to accept heavy metal cations from the cytoplasm [69]. Since transport kinetics of the reconstituted CzcA protein and the CzcCBA complex as characterized in inside-out vesicles are similar [62], the complete CzcCBA complex should also be able to export cations from the cytoplasm. E¥ux of cadmium by CzcCBA ¢ts into a quantitative physiological model describing cadmium homeostasis [68]. This favors the possibility of export from the cytoplasm over export directly from the periplasm. On the other hand, the K50 values of CzcA and the CzcCBA complex for Co2þ , Zn2þ and Cd2þ are in the lower mM range [62,69]. The trace elements Co2þ and Zn2þ are transported with a sigmoidal substrate saturation function (with a Hill coe⁄cient of 2). This keeps the e¥ux velocity of the CzcCBA complex slow when cytoplasmic concentrations are below the K50 value, but allows rapid export at higher concentrations. A cytoplasmic £ow equilibrium model predicts the apparent cytoplasmic concentration of Co2þ to be 8.4 mM and of Zn2þ to be below 6.6 mM [68]. These values are in contrast to the situation
FEMSRE 781 2-6-03
described for E. coli by Outten and O’Halloran [70] ; in this bacterium, the apparent concentration of Zn2þ is kept at 0.2 mM. Moreover, the authors predict by analysis of the interaction of Zn2þ with regulatory proteins a free cellular zinc concentration in the fM range. If R. metallidurans is not completely di¡erent from E. coli, the concentration of free zinc in the cell will not allow e⁄cient export of zinc from the cytoplasm by CzcCBA in the cellular model of Outten and O’Halloran. As shown in Fig. 1, heavy metal cations di¡er in their a⁄nity for sulfur. This a⁄nity relates to the toxicity of the individual heavy metal cation. Gram-negative bacteria contain an intracellular glutathione pool of about 10 mM [71,72], other bacteria harbor di¡erent thiol compounds [73]. Glutathione complexes of extremely sulfurloving heavy metal cations have complex binding constants of 1042 (Hg2þ ^bis-glutathionate complex [74]) and 1039 (Cuþ ^glutathionate complex [75]). Copper should never occur ‘free’ in a cell [76], but the complex binding constants for Cd2þ , Zn2þ , Co2þ , and Ni2þ should be below the copper value. To explain the fM concentrations of ‘free’ zinc observed by the regulatory proteins in E. coli [70] at an apparent cytoplasmic concentration of 0.2 mM Zn2þ and 10 mM glutathione, a binding constant of at least 1013 is required, which is a moderate assumption of this value. Thus, in the cells of Gram-negative bacteria cellular zinc that is not needed in the active centers of enzymes may be present exclusively in glutathionate complexes once the thermodynamic equilibrium has been reached. However, CzcA was not able to transport 65 Zn2þ into proteoliposomes when glutathione was present [68]. Formation of a zinc^mono-glutathionate complex in vitro has a rate constant of 3.9U103 M31 s31 [77]. This would lead to cytoplasmic concentrations of ‘free’ Zn2þ in the WM range in a £ow equilibrium between uptake and a theoretical £ux into a glutathione-bound pool (calculation not shown). This concentration is 1010 -fold higher than the fM concentration of ‘free’ zinc predicted for E. coli [70], but still 1000-fold lower than the K50 value of CzcA for zinc [69]. If the in vivo rate of the formation of zinc^ mono-glutathionate complexes is two or three orders of magnitude lower than the rate determined in vitro [77], CzcCBA will be able to detoxify cytoplasmic Zn2þ . Otherwise, this transport system relies on other transporters that export Zn2þ into the periplasm, meaning that in this case CzcCBA-mediated e¥ux is a direct periplasmic detoxi¢cation process. But, is there any evidence for a direct export of metal cations out of the periplasm ? To investigate this question, a copper/silver-transporting system from E. coli will be discussed. 2.3. The Cus system from E. coli : is there direct export from the periplasm? Copper and silver have a much higher a⁄nity towards
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
sulfur than does zinc. The half-life of Cuþ and Agþ as ‘free’ ions in the cytoplasm is thus probably much shorter than the half-life of ‘free’ Zn2þ , and no ‘free’ copper should occur in a bacterial cell [76]. Thus, evidence for direct periplasmic detoxi¢cation may be more easily obtained for silver and copper than for zinc or cobalt. A plasmid-encoded copper resistance determinant for E. coli has been known for a long time [9,59,78^86]. However, copper resistance factors encoded by the chromosome of this bacterium were described only recently. A copper-exporting P-type ATPase of E. coli, CopA [87,88], transports the heavy metal cation from the cytoplasm into the periplasm like other soft metal-transporting P-type or CPx-type ATPases [89], without any evidence for further transport from the periplasm to the outside. The only HME-RND of E. coli is the CusA protein. Its gene is embedded in the structural gene operon cusCFBA on the bacterial chromosome that encodes CusA, the MFP CusB, the OMF CusC and a small additional protein, CusF [90]. Upstream of cusCFBA and divergently transcribed on the other DNA strand are two regulatory genes that encode the factors of a two-component regulatory system CusS (sensor/histidine kinase) and CusR (response regulator). Both are involved in control of cellular copper homeostasis [91] and of cusCFBA expression [90]. The proteins of the Cus system mediate resistance to silver [90,92]. Resistance to copper could not be assigned to Cus in the ¢rst study [90], not even in a mutant background with the copA gene for the copper-detoxifying Ptype ATPase disrupted. However, copper was a better inducer of cus expression than silver [90], indicating involvement of Cus in copper resistance. Outten et al. found a clear contribution of Cus to copper resistance under anaerobic conditions [93] while copper detoxi¢cation under aerobic conditions was performed by CopA and the periplasmic protein CueO. Under anaerobic conditions periplasmic copper ions control copper homeostasis via CusRS regulation, while under aerobic conditions cytoplasmic copper exerts control via the regulator CueR [93]. Finally, Grass and Rensing [94] demonstrated that an essential function of Cus for copper resistance is easily visible in mutant strains devoid of CueO. This protein is a multi-copper oxidase located in the periplasm [95] explaining why CopA and CueO work together in detoxifying copper, but only under aerobic conditions [93]. CopA exports Cuþ from the cytoplasm into the periplasm [87]. If not prevented by further transport, Cuþ may react with periplasmic components to form Cu2þ and would therefore reduce and impair these components. CueO could reoxidize and repair these periplasmic components again. Electrons can be transferred by the copper atoms of CueO [95] to molecular oxygen leaving a repaired periplasmic component and water. Since Agþ cannot be oxidized to Ag2þ so easily, Agþ should not be able to damage periplasmic components by oxidation (but it may damage
FEMSRE 781 2-6-03
317
them by binding). Thus, CueO cannot repair silver damage and silver resistance mediated by Cus was visible in the presence of CueO [90] In the absence of CueO or in the absence of molecular oxygen [93,94], however, the Cus system is required to maintain copper homeostasis. Cus complements a missing repair or protection process in the periplasm. Recent work on the plasmid-encoded copper resistance system Pco discusses the importance of oxidizing periplasmic Cuþ to Cu2þ to protect the cell [83,96]. Periplasmic proteins, the MFP CusB and the OMF CusC, were required for full copper resistance [97]. Substitutions of methionine residues in the periplasmic portion of the CusA protein resulted in loss of copper resistance demonstrating their functional importance [97]. The major di¡erence in the organization of the czcCBA and the cusCFBA structural gene regions is the presence of cusF in the cus determinant. Similar genes only occur in putative or identi¢ed silver/copper resistance determinants. The small 10-kDa protein CusF is a periplasmic protein that binds one copper per polypeptide and may function as a copper usher that facilitates access of periplasmic copper towards the CusCBA e¥ux pump [97]. Therefore, the main function of Cus may be the direct export of Cuþ and Agþ from the periplasm to the outside [93,94]. This is necessary because both monovalent cations should never occur ‘free’ in the cytoplasm owing to their high a⁄nity to sulfur. More evidence for possible direct periplasmic access of substrates into CBA e¥ux systems comes from the HAE-RND proteins that transport organic substances [98,99]. 2.4. Export of organic substrates by HAE-RNDs Because of their eminent medical importance [99], many HAE-RND proteins have recently been characterized. These RND proteins are involved in the detoxi¢cation of organic substances, including antibiotics, in pathogenic or non-pathogenic bacteria. Resistance to antibiotics mediated by RND proteins was responsible for half of the multiple drug resistant Pseudomonas aeruginosa strains that emerged in vivo in a hospital [100], showing the importance of RND proteins for the medical problem of antibiotic resistance. This pathogenic bacterium contains at least 13 RND transport systems (see below). One is a HME-RND protein and exports heavy metals [101], the other 12 are probably HAE-RND proteins. Best characterized of these proteins are MexB, MexD and MexF. Together with their respective MFPs (MexA, MexC, MexE) and OMFs (OprM, OprJ, OprN), these systems detoxify a variety of organic substances. Unfortunately, nomenclature of the HAE-RNDs did not follow the older nomenclature of the HME-RNDs, which may lead to confusion. MexB is the CzcA-related RND protein while MexA is the CzcB-related MFP. The speci¢city of the MexB-driven system mainly in-
Cyaan Magenta Geel Zwart
318
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
cludes L-lactam antibiotics, L-lactamase inhibitors and organic solvents [102^104], while the MexD-driven system expels cepheme compounds [105]. Chloramphenicol, £uoroquinolones and imipenem are best exported by the MexF-driven exporter [106]. Recently, the substrate speci¢cities of two more Mex systems were investigated, MexJK and MexGHI [107,108], and the functions of some systems were characterized in multiple deletion mutant strains [109]. E. coli contains six HAE-RND proteins and the Cus system as the sole HME-RND protein. The di¡erent substrate speci¢cities of all six HAE-RNDs have been characterized [110,111], but most data are available for the AcrAB-TolC system [98,112^115] that protects the bacterium against a variety of toxic substances especially during slow growth [116]. The AcrD-driven system from E. coli is an aminoglycoside e¥ux pump [117]. Examples of characterized HAE-RND-driven systems from other bacteria are AcrAB from Haemophilus in£uenzae [118], MtrCDE from Neisseria gonorrhoeae [119,120], SrpABC and TtgDEF from Pseudomonas putida [121,122], CmeABC from Campylobacter jejuni [123,124], AdeABC from Acinetobacter baumannii [125], SmeDEF from Stenotrophomonas maltophilia [126], IfeAB and AmeABC from Agrobacterium tumefaciens [127,128], NorMI from Brucella melitense [129], XepCAB from Porphyromonas gingivalis [130], and a CBA e¥ux system from Serratia marcescens [131]. The substrate speci¢city of an RND-driven export system can include many organic substances with di¡erent chemical structures. Common to all these components is only a hydrophobic ability or at least a hydrophobic side chain. Examples are various L-lactam antibiotics [132^ 134], L-lactamase inhibitors [135], other hydrophobic antibiotics and toxic substances [117,129,130,136,137], organic solvents [104,112,114,115,121,122,138], bile salts [113], indole [139], and even homoserine lactone compounds involved in quorum sensing [127,140,141]. A title announcing resistance to vanadate mediated by a RND-driven system [108], however, has to be considered with caution, since the e¡ect on vanadium resistance is via the wellknown action of an RND-driven system on quorum sensing. 2.5. Biochemistry of RND proteins: towards a model of function for these proton-driven antiporters Contribution to resistance to a multitude of toxic substances has been assigned to RND proteins, but biochemical and genetic analysis has only reached a few members. The RND protein CzcA that expels Co2þ , Zn2þ , and Cd2þ was puri¢ed and reconstituted actively into ammonium chloride-charged proteoliposomes [69]. The result demonstrated that CzcA is a proton^cation antiporter depending solely on the ZWvpH portion of the proton motive force and not on its v8 portion [69]. This was in agreement with previous data obtained with the CzcCBA complex that
FEMSRE 781 2-6-03
Fig. 2. Signature sequences of selected RND proteins. Representative RND proteins of the HAE family (AcrB from E. coli), HME group 1 (CzcA from R. metallidurans), HME group 2 (CnrA from R. metallidurans), HME group 3 (two putative proteins from R. metallidurans denoted as ‘p’contig number/gene number, and two proteins from Helicobacter pylori and Aquifex aeolicus), HME group 4 (SilA from Salmonella enterica pv. typhimurium), and HME group 5 (all7631 from Nostoc) were analyzed to demonstrate the relation of the groups to each other. A short sequence alignment of the transmembrane helix IV is shown, which contains the signature motifs of these proteins. The red box indicates the ‘DFG’ occurring in HME1, HME2 and a few HME3 proteins. The green box above the red one is the ‘AI/VG’ at the same position in the other protein groups. The blue box frames the ‘DD’ motif of HAE and HME5 proteins, the gray one below the ‘DSS’ at the same position that is a characteristic of HME group 3. Finally, the purple arrows mark unique and indicative polar and acidic residues that occur at this position only in HME5 proteins.
was heterologously expressed in E. coli and characterized in inside-out vesicles [62]. Not many transport data are available for other RND proteins. Only AcrB has also been puri¢ed and reconstituted into proteoliposomes [51]. Depending on the presence of vpH, the protein catalyzed the extrusion of £uorescent phospholipids. Known substrates of AcrAB, such as bile acids, erythromycin, and cloxacillin, inhibited this activity. Action of various drugs in the presence of an inverted pH gradient (inside acidic) suggested that AcrB is a proton antiporter [51] like CzcA. Thus, members of both RND families, HME and HAE, are proton^substrate antiporters. With reporter gene fusions, the topology of several RND proteins was investigated. These proteins include CzcA [69], MexB [142], and MexD [143]. RND proteins are huge proteins of more than 1000 amino acid residues (aa) in size. They are composed of two related halves that are probably the result of an ancient gene duplication and fusion [144], each containing a large periplasmic loop of about 300 aa in size that is £anked by 12 transmembrane K-helices, TMH I to TMH XII. TMH IV contains amino acid residues that are conserved in most RND proteins [23]. HME-RND proteins involved in the export of divalent heavy metal cations (CzcA, CnrA) exhibit the consensus sequence DFGX3 DGAX3 VEN in this TMH (Fig. 2). Exporters for monovalent cations like CusA and the silver transporter SilA [145] have consensus sequences similar to AVGX3 DAAX3 IEN (Fig. 2) and do not contain the ‘DFG’ motif. The HAE-RND proteins such as AcrB and MexB that export organic substances do not have the ¢rst aspartate residue in the DFG region, but a double
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
319
DD in the middle (characteristic DD motif, Fig. 2). These conserved sequences allow the prediction for the speci¢city of many RND proteins with merely a glance. Both aspartate residues and the glutamate residue in TMH IV of CzcA were essential for function [69]. The same was true for the single aspartate and the glutamate residue of CusA [97]. For MexF, one of the two aspartate residues in the DD motif and the glutamate residue were also essential [146]. Thus, the presence of charged amino acid residues in TMH IV of RND proteins is required for the proton^substrate antiport. Both aspartate residues of CzcA are probably involved in proton translocation [69]. As in other proteins [147^150], the aspartate residues may therefore be part of a charge relay system that forms the path of the protons across the cytoplasmic membrane. These data led to a preliminary model of how CzcA and
other RND proteins might function [69] (Fig. 3). This simple model for the e¥ux of cytoplasmic cations by an RND export system can easily be modi¢ed to ¢t other postulated functions for RND proteins (Fig. 3). It is not relevant whether the substrates access the periplasmic substrate binding sites from the cytoplasm (as in the case of Zn2þ ?), or from the periplasm (as in the case of Cuþ ?), or from inside the cytoplasmic membrane (maybe in the case of hydrophobic substrates for HAE-RND proteins). In all cases, a protonation/deprotonation cycle couples the process of substrate binding and release to the exergonic proton import reaction. Moreover, protonation may easily change the substrate binding site between negatively charged and neutral (HME-RND proteins) or hydrophilic and hydrophobic (HAE-RND proteins). The substrates bound initially to some substrate binding site of the RND protein are consecutively forwarded to the other two components of the CBA e¥ux system for further transport. A lot of insight into the possible mechanisms of action has come from the recently available three-dimensional structure of the AcrB protein obtained after crystallization [151,152]. Three AcrB protomers are organized as a homotrimer in the shape of a jelly¢sh. Each protomer is composed of a transmembrane region 5 nm thick, composed of the previously identi¢ed 12 transmembrane spans, and a 7-nm protruding headpiece containing a central cavity at the bottom. The cavity has three vestibules at the side of the headpiece that may lead into the periplasm. This structure implies that substrates may be translocated from the cell interior through the transmembrane region and/or from the periplasm through the vestibules and are collected in the central cavity [151], before they are forwarded to the other two components of the CBA e¥ux complex.
Fig. 3. Proposed cellular function and mechanism of RND proteins. This model is based on the structure of AcrB, TolC and published biochemical data (see text for details). The RND protein (red) is shown as a trimer with a periplasmic substrate binding site. The MFP protein is shown in green, the OMF in yellow. Action by RND proteins is driven by protons (black dots) that were previously expelled by the respiratory chain (1). In the case of HME1 and HME2 proteins, divalent heavy metal cations (blue, purple and gray circles, M2þ ) migrate through the cation channel of the protein and are bound to a periplasmic substrate binding site (orange box, 2a). Alternatively (blue arrows), this site may also be accessible for periplasmic cations (2b), especially in the case of Cuþ and HME4 proteins. HAE transporter of organic substrates may bind these substrates from the hydrophobic environment of the cytoplasmic membrane (2c). A⁄nity of the hypothetical periplasmic binding site is decreased by protonation with periplasmic protons (3) and this energy is used to accumulate the substrates into the OMF tube or a microenvironment formed by the RND protein (4). The high concentration in this tube allows easy di¡usion of the substrates from the tube to the outside (5). For energetic coupling, the hypothetical periplasmic substrate binding site is deprotonated by import of the protons to the cytoplasm (6). This complex reaction allows transport of substrates across the complete envelope of a Gram-negative bacterium that is composed of outer membrane (OM), cell wall (CW), periplasmic space (PP) and cytoplasmic membrane (CPM).
2.6. MFPs and OMFs enhance the function of RND proteins
FEMSRE 781 2-6-03
Most RND proteins are encoded in operons that additionally contain the gene for a MFP. Many of these operons harbor a third gene, encoding an OMF [50]. The position of this OMF-encoding gene varies from operon to operon. It may be located at the 5P end of operons like czc, at the 3P end of the operon as in most mex operons, it could be not even part of the operon as in the case of tolC [48]. Cross-linking experiments with the CzcCBA e¥ux complex displayed a CzcB multimer [153]. This protein migrates in polyacrylamide gels at a much lower apparent molecular mass than predicted from the derived amino acid sequence. This unusual migration behavior may be based on the extensive coil-coil structure proposed for MFPs [52]. Depending on which size for the CzcB monomer is used in the calculation, the CzcB multimer is a trimer or a tetramer of the MFP [153]. For the MFP
Cyaan Magenta Geel Zwart
320
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
Table 1 Distribution of members of heavy metal-exporting protein families in the genomes of 64 prokaryotesa RND proteinsb
CDF proteinsc
P-type ATPasesd
Other transport systemsd
0 0 0 0
0 1: 2a 1: 2a 2: 2a, 3aii
2: 2: 1: 1:
0 0 0 0
0 0 0
2: 3f, 3f 1: 3f 2: 3f, 3f
3: Cu9, Cu12, ¡ 4: Cu2, Cu14, Zn1, ¡ 6: Cu9, Cu12, Zn1, ¡
0 0 0
0 0 0 0 0 0
1: 1: 1: 2: 0 0
2: 0 1: 1: 3: 2:
Zn7 Cu9 Cu1, KdpB, ¡ Cu1, KdpB
CHR4 0 0 0 0 0
4: M3b, M4, A5¡ 1: A5 1: A5
2: 2g, 3ai 2: 2c, 3e 2: 2h, 3c
3: Cu1, Cu12, ¡ 1: Cu13 3: Cu3, Zn2, KdpB
0 0 CHR1
6: M5, A4, A5¡ 8: M5, M5, A1, A4, A5¡
1: 3c 3: 1b, 2e, 3ai
9: Cu5, Cu10, Zn7, Zn9, ¡ 13: Cu2, Cu5, Cu5, Cu10, Zn4, Zn9, Me1, Me1, ¡
CHR3, NreB CHR3, NreB
1: A2 1: A2 0 0 1: A2 0 0
4: 2: 3: 3: 3: 1: 1:
CHR4 CHR1, CHR4 0 0 0 CHR4 0
Mycoplasma genitalium Mycoplasma pneumoniae Ureaplasma urealyticum Mycoplasma pulmonis Streptococcus pneumoniae Streptococcus pyogenes Lactococcus lactis High G+C Mycobacterium tuberculosis
0 0 0 0 0 0 0
0 0 0 0 2: 2c, 3e 2: 2c, 3e 2: 2c, 3e
4: Cu6, Zn7, Zn10, ¡ 4: Cu6, Zn5, Zn10, ¡ 6: Cu6, Zn10, KdpG, ¡ 7: Cu6, Zn7, Zn10, KdpG, ¡ 2: Cu6, KdpG 5: Cu9, Zn7, ¡ 13: Cu9, Cu12, Zn7, Zn10, Zn12, ¡ 1: MgtA 1: MgtA 0 1: MgtA 4: Cu2, Me1, ¡ 3: Cu9, Zn10, ¡ 9: Cu9, Cu12, Zn8, Zn12, ¡
0
1: 3aii
0
Mycobacterium leprae Proteobacteria K subdivision Caulobacter crescentus Brucella melitensis Mesorhizobium loti Agrobacterium tumefaciens Sinorhizobium meliloti
0
1: 1b
12: Cu2, Cu2, Cu2, Zn9, Zn9, Zn12, Zn12, ¡ 4: Cu2, Cu2, Zn12, ¡
7: M1, M2, A3, A5¡ 7: A5¡ 10: A5¡ 10: A3, A5¡ 11: A3, A5¡
2: 2: 2: 2: 2:
Rickettsia prowazekii Rickettsia conorii Magnetococcus sp. L subdivision Neisseria meningitidis Ralstonia solanacearum
1: A5 4: A5¡ 14: M4, A1, A4, A5¡
1: 3b 1: 3b 4: 1a, 3f, 3f, 3f
3: Cu14, Zn9, ¡ 5: Cu3, Cu14, Zn2, Zn9, ¡ 7: Cu4, Cu14, Zn2, Zn12, ¡ 5: Cu3, Cu3, Cu14, Zn2, ¡ 9: Cu3, Cu3, Cu4, Cu14, Cu14, Zn2, Zn12, KdpG, ¡ 0 0 2: Cu7, Cu14
1: A5 15: M1, M3a, M3b, M4, A3, A5¡
0 2: 1a, 3f
2: Cu8, Cu14 4: Cu7, Cu14, Zn6, ¡
Prokaryotic species Archaea Crenarchaeota Aeropyrum pernix Sulfolobus solfataricus Sulfolobus tokodaii Pyrobaculum aerophilum Euryarchaeota Archaeoglobus fulgidus Halobacterium sp. Methanobacterium thermoautotrophicum Methanocaldococcus jannaschii Pyrococcus horikoshii Pyrococcus abyssi Pyrococcus furiosus Thermoplasma acidophilum Thermoplasma volcanium Bacteria Early-branching Aquifex aeolicus Thermotoga maritima Deinococcus radiodurans Cyanobacteria Synechocystis sp. Nostoc sp. Gram-positive bacteria Low G+C Bacillus subtilis Bacillus halodurans Listeria innocua Listeria monocytogenes Staphylococcus aureus Clostridium perfringens Enterococcus faecalis
FEMSRE 781 2-6-03
3f 2c 3f 2c, 3f
2g, 3e, 3f, 3f 2g, 3c, 2g, 3c, 2f, 2h, 3d 3d
3e, 3f 3e 3e 3e
2e, 3b 2h, 3c 2g, 3c 1a, 3c 1a, 3c
Cu1, Me0 Cu1, Cu16 Cu1 Cu1
MgtA, ¡
Cyaan Magenta Geel Zwart
0 0 0 0 0 0 0
0
CHR5 0 CHR5, CnrT CHR1 CHR5, NreB 0 0 NreB
CHR2, CHR5, CnrT
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
321
Table 1 (Continued). Prokaryotic species
RND proteinsb
CDF proteinsc
P-type ATPasesd
Other transport systemsd
Ralstonia metallidurans
20: 2UM1, 2UM2, 3UM3a, 3UM3b, 2UM4, A5¡
3: 1a, 2h, 3f
10: Zn6, Zn6, Zn6, Cu4, Cu7, KdpG, ¡
CHR1, 2UCHR2, CHR5, CnrT
2: A5¡ 3: M3b, M4, A5
2: 2c, 3b 0
3: Cu15, Cu15, KdpG 3: Cu8, Cu14, Zn7
0 0
7: M4, A5¡ 6: A5¡ 4: A5¡ 7: A1, A5¡ 0 1: A5 1: A5 13: M1, A5¡ 7: A1, A1, A5¡ 5: A1, A5¡
2: 2: 2: 2: 0 0 2: 3: 1: 1:
4: 6: 5: 5: 0 1: 1: 7: 3: 0
0 0 0 CnrT 0 0 0 CHR1 CHR1 0
2: M5, A4
2: 2b, 3f
1: MgtA
0
0
3: 2b, 3ai, 3aii
0
0
0 0 0
0 0 0
1: Zn11 1: Zn11 1: Zn11
0 0 0
1: A2 0
0 0
0 1: Cu11
CHR4 0
O subdivision Campylobacter jejuni Helicobacter pylori Q subdivision Escherichia coli Salmonella typhimurium Salmonella enterica Yersinia pestis Buchnera sp. Haemophilus in£uenzae Pasteurella multocida Pseudomonas aeruginosa Vibrio cholerae Xylella fastidiosa Flexible Bacteria CFB group Porphyromonas gingivalis Green sulfur bacteria Chlorobium tepidum Chlamydiales Chlamydophila pneumoniae Chlamydia trachomatis Chlamydia muridarum Spirochaetales Borrelia burgdorferi Treponema pallidum
2g, 2g, 2g, 2g,
3b 3b 3b 3b
2g, 3b 1a, 2h, 3b 3b 2h
Cu8, Cu3, Cu8, Cu8,
Zn3, Cu8, Zn3, Zn3,
KdpG, MgtA Zn3, KdpG, ¡ ¡ ¡
Cu8 Cu8 Cu7, Cu14, Zn6, Zn12, ¡ Cu8, Cu14, Zn3
a The comprehensive microbial resource at http://www.tigr.org and the R. metallidurans sequencing homepage of DOE/JGI at http://www.jgi.doe.gov/ JGI_microbial/html/ralstonia/ralston_homepage.html was used. A BLAST search [280^282] was performed against the ‘hook’ proteins CzcA from R. metallidurans (RND family), CzcD from R. metallidurans (CDF family), CadA from S. aureus (CPx type ATPases), ChrA from R. metallidurans (CHR family), NreB from R. metallidurans strain 31A and CnrT from R. metallidurans. Sequences with an E value smaller than 0.01 were collected and analyzed in two ways. Firstly, the occurrence of speci¢c signatures was used to pre-group the proteins into putative functional clusters. Secondly, assignment of a sequence to a group of proteins was cross-checked by multiple alignments performed with BioTechnix3d (http://www.biotechnix3d.com). For RND proteins, sequences with low similarity to CzcA and a signature of a RND protein (‘DD motif’) were not further analyzed. Likewise, analysis of a single genome was not further pursued when a magnesium- or potassium-transporting P-type ATPase was encountered. ‘¡’ indicates the presence of such non-characterized RND- and P-type proteins that were non-RND and non-CPx-type proteins, respectively. For RND-, CDF-, and P-type proteins, the total number of the respective proteins encoded by the particular genome is listed. After a colon, the association of the single proteins to a group within the protein family is given (see below). For CHR proteins, the number of these proteins is given and the assignment to the ¢ve groups of these proteins (similarity tree not shown). CnrT and NreB proteins were only mentioned. A zero (0) indicates that the organism does not contain any protein from the respective protein group or family. b For RND proteins, ‘M’ refers to the respective cluster of the HME-RND proteins, ‘A’ to the respective cluster of the HAE-RND proteins (Table 2). c Likewise, the cluster assignments of the speci¢c CDF proteins (Table 3) is indicated. d The data for the P-type ATPases (Cu, Cu-CPx-type ATPase ; Zn, Zn-CPx-type ATPase. MeO, MgtA and KdpG are other P-type proteins, the number after ‘Cu’ or ‘Zn’ indicates the assignment to a group within the respective subfamily and a possible ¢ne di¡erentiation of function) and the other protein groups are not shown.
AcrA, a trimeric form has also been determined in crosslinking experiments [53]. In many CBA protein complexes, loss of the MFP or the RND protein abolishes the respective resistance [55,90,94,97,154,155]. In contrast, loss of the OMF usually has a relatively moderate in£uence [55,97]. Since many Gram-negative bacteria contain a variety of OMF-encoding genes, another OMF may take over and complement for the loss of the native OMF. TolC is known to interact with a variety of transenvelope e¥ux systems that are driven by RND proteins or even ABC export systems [48]. The CnrC protein could be substituted partially in
FEMSRE 781 2-6-03
arti¢cially mixed operons with CzcC or NccC (G. Grass and D.H. Nies, unpublished). Chimeric MexCBA e¥ux pumps were also functional [156^158]. Thus, the OMF part of a CBA e¥ux complex is not involved in the substrate speci¢city, although its overproduction might lead to increased resistance to antibiotics [159]. TolC is the ¢rst OMF for which the structure is known at high resolution [160]. A TolC homotrimer forms a tubelike channel with a length of 14 nm and an outer diameter of 3.5 nm. This tube is hollow, allowing substrates to pass. The tube is su⁄cient in length to span the entire periplasmic space and the outer membrane, connecting the peri-
Cyaan Magenta Geel Zwart
322
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
plasmic domains of the RND protein with the outside. Insertion in the outer membrane and the ability to form a channel has been determined for OprM and TolC [48,161,162]. OprM formed a channel for small molecules, but was only able to allow transport of antibiotics after protease treatment [163]. This may indicate that the iris-like structure at the periplasmic end of the OMF tube [160] could serve as a gate, controlling access of the substrates from the RND protein. The top of the headpiece of AcrB opens like a funnel, where TolC might directly dock into AcrB [151]. This funnel was connected to the central cavity of AcrB by a pore that may allow substrate transport from the central cavity to TolC. All these pores and cavities in RND and OMF proteins may serve as microenvironments explaining a common mechanism of action of CBA e¥ux systems : an active accumulation of a substrate into a tube or cavity may yield a high substrate concentration inside this microenvironment allowing a rapid discharge of the substrate to the outside by di¡usion (Fig. 3). MFPs interact with their RND proteins and are attached to the cytoplasmic membrane by a hydrophobic residue, some by a hydrophobic amino-terminus [53,164]. MFPs contain a £exible coil-coil structure that may allow a £exible contact between the RND pump and the OMF tube [52]. A recent structure proposal based on liquid layer crystallization [165] describes AcrA as a doughnut-like structure with an inner diameter of 3 nm. Andersen et al. have suggested that MFPs may serve as adapter proteins between RND or other export pumps and OMF tubes [48]. MFPs probably do not fuse the cytoplasmic and the outer membrane directly, but they attach the OMFs to the pump proteins (Fig. 3). Thus, OMF and MFP proteins have a rather static function during CBA-mediated transenvelope e¥ux. Re-
cent ¢ndings indicate that the large periplasmic loops (LPL) of HAE-RND proteins have a function in substrate recognition. When the LPLs of AcrD were exchanged with the corresponding LPLs from AcrB, the chimeric protein displayed the substrate speci¢city of AcrB rather than AcrD [166]. Similar results were obtained when chimeric AcrB^MexB proteins were constructed [167]. Unfortunately, chimeric CzcA^CnrA proteins were not stable (K. Helbig, A. Legatzki and D.H. Nies, unpublished) and similar experiments could not be performed for these HMERNDs. Methionine residues in the LPL of CusA were essential for CusA function as mentioned above [97]. Similarly, essential amino acids probably involved in substrate recognition were identi¢ed in both LPLs of MexD [168]. Moreover, the LPLs of MexF were also involved in correct assembly of this CBA e¥ux pump [169], assigning a second essential function to the headpiece of RND proteins. Taken together, the RND proteins are indeed the central and most important component of CBA e¥ux systems: they mediate the active part of the transport process, determine the substrate speci¢city and are involved in the assembly of the transenvelope protein complex. Thus, distribution of RND proteins in the kingdoms of life may give some clues about the distribution and frequency of RND-mediated metal resistance. 2.7. Distribution of RND proteins in the kingdoms of life The genomes of 63 prokaryotes from both kingdoms (Archaea and Bacteria, tigrblast.tigr.org/cmr-blast, [170]) were compared to the genome of R. metallidurans (http://www.jgi.doe.gov/JGI_microbial/html/ralstonia/ ralston_homepage.html, Table 1). All RND proteins with
Table 2 Groups of heavy metal-transporting RND proteins Name HME proteins HME1 (M1) HME2 (M2) HME3a (M3a) HME3b (M3b) HME4 (M4) HME5 (M5) HAE proteins HAE1 (A1) HAE2 (A2) HAE3 (A3) HAE4 (A4) HAE5 (A5)
Signature motifa
Predicted substrate
Examplesb
DFG-DGA-VEN DFG-DGA-VEN GFG-D(G,S,A)(S,A)-(V,M)EN (A,g)(I,L)G-D(G,A,s)-VEN A(I,V)G-DA(A,s)-(V,I)(E,d)N AIG-DDX-(M,V)EN
Zn2þ , Co2þ , Cd2þ Ni2þ , Co2þ Divalent cations? Monovalent cations ? Cuþ , Agþ Ni2þ ?
CzcA, CztA, HelA CnrA, NccA RSA1040 HP0969 SilA, CusA All7631
X(V,L)G-D(N,G)A-X(D,E)N XXG-D(D,X)(S,a)-X(D,E)(N,s) XXG-DDA-XEN XXG-D(D,S,N)(S,A)-XE(N,t) AIG-DDA-XEN
Unknown Unknown Unknown Unknown Organic molecules
Alr1656 YerP NolG Alr5294 AcrB
a
For better orientation, the amino acid residues corresponding to G404 , D408 , E415 , and N416 of CzcA are shown in bold. The dash corresponds to three mostly hydrophobic amino acid residues. Amino acids shown in parentheses are alternatives for the same position. Amino acid names in lower-case letters indicate rare choices, X is any amino acid. b Representative sequences are from R. metallidurans strains CH34 (CzcA, CnrA) and 31A (NccA), Ralstonia solanacearum (RSA1040), Pseudomonas £uorescens (CztA), Legionella pneumophila (HelA), Helicobacter pylori (HP0969), Salmonella enterica pv. typhimurium (SilA), E. coli (CusA, AcrB), Nostoc sp. (All7631, Alr1656, Alr5294), Bacillus subtilis (YerP), and S. meliloti (NolG).
FEMSRE 781 2-6-03
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
su⁄cient homology to CzcA (E value 6 0.01 in the E value of the BLAST score) were selected and compared to CzcA and the 50 protein sequences in the databases (1 034 995 sequences) most similar to CzcA. Multiple alignments yielded homology trees (data not shown) that were used to group and subgroup the RND proteins (Table 2). Moreover, the occurrence of the Czc-, Cus-, and Mex-speci¢c consensus motifs DFG/D/EN, AIA/D/EN and DD, respectively (see above), was used to con¢rm the group assignments (Table 2). HME- or HAE-RND proteins were found in low frequency in early-branching bacteria (Table 1), but Aquifex aeolicus contains members of both RND protein families. RND proteins are very rare in £exible bacteria. Since spirochetes contain members of the unusual HAE3 family [23], this observation may indicate a limitation of the method used. Gram-positive bacteria are the third bacterial group with a low frequency of RND proteins (Table 1). Given the current hypothesis about the function of RND proteins as transenvelope e¥ux pumps, a CBA ef£ux complex should not function in a Gram-positive cell wall. However, CzcA alone mediates a small degree of resistance [171], thus, RND proteins in Gram-positive bacteria might function as single-subunit e¥ux systems. All RND proteins found in Gram-positive bacteria grouped in the HAE2 cluster (identical with the HAE2 protein family [23]) that appeared only in these bacteria and in Borrelia burgdorferi. This indicates that the mechanism of action of the RND proteins in Gram-positives might be di¡erent from that in most Gram-negative bacteria. Cyanobacteria and proteobacteria contained many RND proteins from both families (Table 1). Exceptions are the symbiotic bacterium Buchnera with no RND protein and three pathogenic bacteria with only one RND (Rickettsia prowazekii, H. in£uenzae, Pasteurella multocida). Most RND proteins were found in R. metallidurans strain CH34 (20 putative or candidate genes, some of them not complete), followed by its phytopathogenic relative Ralstonia solanacearum (15 genes), Magnetococcus (14 genes) and P. aeruginosa (13 genes). Thus, R. metallidurans contains more RND proteins than every other bacterium in the analyzed group. Most of the many RND proteins in cyanobacteria and proteobacteria belong to the HAE family. When bacteria contain only a few RND proteins, these are HAE-RND proteins in most cases. Examples of bacteria that exclusively harbored HAE-RND proteins are H. in£uenzae (with one RND protein that has already been characterized [118]), R. prowazekii, C. jejuni, P. multocida, Thermotoga maritima, Deinococcus radiodurans, B. burgdorferi, and Gram-positive bacteria. When bacteria contain putative HME-RND proteins at all, most of these are putative copper/silver exporters of the HME4 group (Tables 1 and 2). Examples are A. aeolicus (two genes, HME3b and HME4), Magnetococcus (HME4), Helicobacter pylori (HME3b and HME4) and
FEMSRE 781 2-6-03
323
E. coli (HME4 = CusA). A second group of bacteria harbored members of the highly unusual HME5 cluster, the cyanobacteria Synechocystis (one HME5 gene), Nostoc (two HME5 genes), Legionella pneumophila, and P. gingivalis (one HME5 gene). The HME5 cluster contains members of the HME-RND protein family by homology. These proteins exhibit the ‘DD’ motif in the TMH IV that is characteristic of HAE proteins, but contain some conserved polar residues in their TMH IV unique to this group (Fig. 2). There is only indirect evidence that these HME5 proteins might transport heavy metals, especially Ni2þ . A putative HME5 protein (sequence slr0794, http://www.kazusa.or.jp/cyano/Synechocystis) is part of an operon designated nrsBACD [172]. Expression of this operon is controlled by nickel via a two-component regulatory system NrsRS ( = RppAB) encoded upstream of nrsBACD on the other DNA strand. The genes in the nrsBACD operon refer to the genomic designations slr0793 (MFP), slr0794 (HME5), slr0795 (hypothetical protein), and slr0796 (putative NreB protein). It has been shown for NreB-like proteins that they are able to exert nickel resistance alone ([173], see below), so the nickel resistance assigned to the nrsBACD may be mediated by nrsD. It is not clear if nrsC is an OMF and if a complete CBA e¥ux complex can be encoded by this operon or parts of it. Therefore, con¢rmation of the HME5 proteins as nickel-exporting RND proteins has to await further experimental evidence. A third group of bacteria that contain putative HMERND proteins may harbor proteins of the HME1 cluster highly related to CzcA. In the analyzed genomes and the current databases, 10 HME1 sequences were found. These were proteins from R. metallidurans (CzcA from di¡erent strains [55,56]), P. aeruginosa (CzrA [101]), Pseudomonas £uorescens (CztA), a fragment from R. metallidurans (gene 848 on contig 456), R. solanacearum (Rsp 0493), Caulobacter (CC2390), two proteins from either pathovar of Xanthomonas campestris and HelA from L. pneumophila. Only these eight bacteria can be expected to exhibit Co2þ / Zn2þ /Cd2þ resistance based on RND-driven transenvelope e¥ux. The Cnr-related HME2 group is much smaller and contains only four members, two of them from R. metallidurans strain CH34 (CnrA and NccA), one from strain 31A (NccA [67]) and one from Caulobacter (CC2724). These three bacteria should exhibit RND-based nickel/cobalt/ cadmium resistance. Group HME4 contains 10 sequences of putative copper/ silver transport systems. These sequences came from E. coli (CusA [90,94,97]), Salmonella enterica pv. typhimurium (SilA [145]), two sequences from R. metallidurans and one sequence from R. solanacearum, Pseudomonas syringae, H. pylori, L. pneumophila, A. aeolicus and Magnetococcus sp. CusA has been introduced above and by Grass and Rensing [88], SilA by Silver [92]. The H. pylori copper export system has also been characterized [174]. Because
Cyaan Magenta Geel Zwart
324
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
of the similarity of the other eight sequences to CusA and SilA, these bacteria can be expected to detoxify copper and/or silver in a way that has been shown for CusA and SilA. The remaining HME3 cluster contains all sequences not closely associated with HME1, HME2 or HME4, but with a unique DSS motif in TMH IV (Fig. 2). There are two subgroups in this cluster (Table 2). Subgroup HME3a exhibits the DFG consensus motif and the proteins in this subgroup may transport divalent heavy metal cations. The ¢ve members of HME3a come from R. solanacearum, Xanthomonas axonopolis pv. citri, and three sequences from R. metallidurans strain CH34. The other subgroup, HME3b, shows mostly the AIG motif characteristic of transporters of monovalent cations. The ¢ve proteins of this cluster come from R. solanacearum, H. pylori, A. aeolicus and again three sequences from R. metallidurans. The sequence from A. aeolicus is unusual since it contains a GIG motif instead of AIG, which gives this putative protein (aq_469) an ancient touch. The relationship of RND proteins leads to a scenario of how heavy metal-transporting HME proteins may have evolved: the earliest branch of HME proteins are the HME5 proteins that still contain the DD motif of RND proteins, and which are much more diverse and widely distributed than HME proteins. Have HME proteins originated from HAE proteins ? The other four groups, HME1 to HME4, are currently found only in proteobacteria and A. aeolicus. This could indicate that the divalent heavy metal transport systems originated from an ancient monovalent transport system such as aq_469 from A. aeolicus during evolution of the proteobacteria.
3. The second layer of metal resistance: cation di¡usion facilitators (CDF family) serve as secondary cation ¢lters in bacteria CDF proteins form a protein family of metal transporters (TC 2.A.4.1.1^2 [46]) occurring in all three domains of life [175]. Primary substrates of CDF proteins are Zn2þ , but also Co2þ , Ni2þ , Cd2þ , and Fe2þ . CDF-mediated transport is driven by a concentration gradient, a chemiosmotic gradient, v8, vpH or a potassium gradient [176^ 178] (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished). First, the physiological function, biochemistry and substrate speci¢cities of the CDF protein family will be analyzed, before the distribution of CDF proteins in sequenced bacterial genomes will be compared to that of R. metallidurans strain CH34. 3.1. Physiological function and substrate speci¢city of CDF proteins All CDF proteins described so far in bacteria are involved in resistance to Zn2þ and other heavy metal cati-
FEMSRE 781 2-6-03
ons. The archetype of the family, CzcD from R. metallidurans strain CH34, was ¢rst described as a regulator of expression of the CzcCBA high resistance system [60], but CzcD is also able to mediate a small degree of Zn2þ /Co2þ / Cd2þ resistance in the absence of the high resistance CzcCBA system [179]. The presence of CzcD decreases the cytoplasmic concentration of these metals [179] which serve as inducers for czc expression (C. GroMe, A. Anton, T. Ho¡mann, S. Franke, G. Schleuder and D.H. Nies, unpublished). Heterologously expressed in E. coli, CzcD was driven by the proton motive force and e¥uxes Zn2þ and Cd2þ , but not Ni2þ and Co2þ (. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished). Cross-linking experiments indicated that CzcD is probably a dimer in vivo. Although CDF-encoding genes are found on the chromosomes of many sequenced organisms, the function of the proteins has only been characterized in a few bacteria. In Bacillus subtilis, the czcD gene is in an operon together with the gene for a TrkA dehydrogenase [180] and mutant studies show some involvement in cobalt resistance [177]. CzcD from B. subtilis was driven by the proton motive force as well as by a potassium gradient, and the exchange seemed to be electroneutral. Substrates for this CzcD protein were Zn2þ , Co2þ and Cd2þ [176]. Two groups cloned from Staphylococcus aureus a small determinant containing a CDF-encoding gene and a regulator. The CDF protein was named ZntA (now RzcB [181]) or CzrB [182]. Both proteins are nearly the same size (325 or 326 aa) and di¡er only in nine positions; they are essentially identical. CzcD from S. aureus is involved in zinc and cobalt resistance. A CDF protein from Thermus thermophilus was shown to mediate zinc and cadmium, but not cobalt resistance [183]. Recently, one of the two CDF proteins from E. coli was characterized, ZitB (the ybgR gene product), which is also a zinc resistance system [184]. This protein diminishes the cellular accumulation of Zn2þ and is driven by a potassium gradient in addition to the proton motive force [185]. E. coli contains another putative CDF-encoding gene, the enigmatic yiiP. Expression of this protein is induced by zinc, deletion of the gene does not lead to an increase in zinc resistance, but overexpression has no e¡ect [184]. The function of this protein is therefore unknown. The bacterial CDF proteins mentioned so far are mainly Zn2þ /Co2þ /Cd2þ -exporting cytoplasmic membrane proteins. Magnetotactic bacteria are able to sense the magnetic ¢eld of the Earth with the help of speci¢c intracellular organelles, the magnetosomes, which contain the mineral magnetite (Fe3 O4 ) as ¢eld-sensing material. MamB was described as a CDF protein located in the magnetosome membrane [186]. Its function is probably to transport iron into this organelle. Magnetosome-like magnetite was discussed as a marker for remnants of ancient life forms from the planet Mars [187,188]. That would increase the distribution of CDF proteins to anoth-
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
er planet! However, a magnetic ¢eld is the prerequisite for the occurrence of magnetotactic life forms. Since the magnetic ¢eld of Mars was only 1/30 as strong as the ¢eld of our Earth when the putative Martian fossils were liberated by a giant impact [189] and this is rarely su⁄cient to drive any magnetotactic behavior, we probably have to wait for better evidence for life forms and CDF proteins from other planets. Usually, bacteria contain one or a few CDF-encoding genes. In eukaryotes, the number of CDF-encoding genes increases in parallel with evolution from yeasts to animals and plants. Saccharomyces cerevisiae contains ¢ve CDFencoding genes [190]. ZRC1 was initially described in the same year as CzcD and this protein mediates resistance to zinc and cadmium [191]. Another yeast CDF, COT1, mediates cobalt resistance [192], but ZRC1 and COT1 can complement each other [193]. Both can even substitute for CzcD in its transport and regulatory function in R. metallidurans [179]. ZRC1 and COT1 are located in the vacuolar membrane and transport the cations Zn2þ , Cd2þ and Co2þ into this compartment for detoxi¢cation and possible future re-use [194^196]. This explains why both proteins are also needed to survive zinc starvation [194,197,198]. Moreover, since the cytoplasm is not detoxi¢ed in ZRC1 knock-out strains, the cell reacts with an increased glutathione level [199]. Similar to the function of CzcD from R. metallidurans, the postulated regulatory function of ZRC1 is therefore an indirect consequence of its transport function. A third S. cerevisiae protein, MSC2 (YDR205W), is also involved in zinc transport [200]. The respective deletion strain, however, accumulated excess zinc in the nucleus-enriched membrane fractions. Therefore, a function of MSC2 in zinc homeostasis of the nucleus was suggested [200]. The other two CDF proteins from S. cerevisiae are not involved in Zn2þ /Cd2þ /Co2þ transport, but these two proteins, MFT1 and MFT2, are probably Fe2þ transport systems located in the mitochondrial membrane [201]. That adds to the evidence that CDF proteins may transport Fe2þ . The ZRC1-related CDF protein Zhf from the ¢ssion yeast Schizosaccharomyces pombe is located in the endoplasmic reticulum and the nuclear envelope like MSC2 [202]. Deletion of zhf leads to increased zinc and cobalt sensitivity, but to increased cadmium and nickel resistance. The cytoplasmic content of zinc, cobalt and cadmium was diminished and detoxi¢cation of cadmium by phytochelatin synthesis was not increased [202]. Thus, this protein may be responsible for supply of the endoplasmic reticulum and the nucleus with zinc and/or cobalt. Involuntary transport of cadmium and nickel into the same compartment by Zhf could mediate increased toxicity of these cations. The currently known CDF proteins in higher animals are all zinc transport systems. ZnT1 is located in the cytoplasmic membrane and seems to be a zinc e¥ux system
FEMSRE 781 2-6-03
325
[203]. In contrast, ZnT2 transports Zn2þ into vesicles and is therefore comparable in function to ZRC1 and COT1 [204]. ZnT3 also transports Zn2þ into vesicles, but exclusively into synaptic vesicles [205]. The protein is most abundant in the zinc-enriched mossy ¢bers that project from the dentate granule cells to hilar and CA3 pyramidal neurons in the brain. Thus, ZnT3 seems to be responsible for the ability of these neurons to release zinc upon excitation [206]. Age-dependent low expression of ZnT3 leads to de¢ciencies of Zn2þ in synaptic vesicles of the mossy ¢ber pathway and induces glutamatergic excitotoxicity in the hippocampal neurons and the deterioration of learning and memory [207]. The fourth known CDF protein from higher animals, ZnT4, is similar to ZnT3 with respect to zinc transport and abundant expression in the brain, but this protein is also expressed in the mammary epithelia and responsible for excretion of zinc into mother’s milk [208]. The CDF family in mammals kept growing with ZnT5 that transports zinc into secretory granules in pancreatic L cells [209] and also plays an important role in maturation of osteoblasts and in maintenance of the cells involved in the cardiac conduction system [210]. ZnT5 proteins from mouse and man are twice the size of other CDF proteins. Since dimerization has been indicated for CzcD (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) and ZnT4 [211], gene duplication followed by gene fusion could have been responsible for the occurrence of ZnT5 and other large CDF proteins. Another large CDF protein from mammals was named ZTL1 (ZnT-like protein). It is located in the enterocyte apical membrane, e.g. in kidney [212]. Comparison of the 765-aa ZnT5 protein with the 523-aa ZTL1 proteins displays a 495-aa middle part that di¡ers in only two amino acid residues (99.6% identity). This middle part starts at M172 of ZnT5 that equals M1 of ZTL1, and extends to K666 of ZnT5. The 100-aa ZnT5 and 28-aa ZTL1 carboxy-termini are completely di¡erent. The genomic sequence encoding the ZTL1 protein consists of 11 exons separated by large introns with the ¢rst ones encoding possible alternative splicing [212]. The ZnT5 protein has a predicted size of 84 kDa, but its size in vivo was determined by Western blot analysis to be 55 kDa [209], which is similar to the predicted size of ZTL1 of 57.3 kDa. Thus, the original gene may encode at least two proteins by splicing of the RNA and post-translational processing of a pre-protein, and the two proteins may di¡er in function and localization. Mammals contain at least two more CDF proteins, ZnT6 and ZnT7. The protein ZnT7 is currently only putative (AY094606) awaiting publication of more data. ZnT6 may transport cytoplasmic zinc into the Golgi apparatus and its expression is regulated by zinc in kidney cells [213]. All these results suggest that intracellular zinc homeostasis in mammals is mediated by a variety of ZnTlike proteins that di¡er in intracellular localization and
Cyaan Magenta Geel Zwart
326
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
organ-speci¢c expression. These proteins transport zinc from the cytoplasm into another compartment, to the outside (ZnT1, ZnT4) like bacterial transport systems, or into vesicles (ZnT2, ZnT3, ZnT5) like ZRC1 and COT1, or into the Golgi apparatus (ZnT6) like Zhf. However, experiments with oocytes expressing ZTL1 indicate that CDF proteins may also be involved in uptake into the cell [212]. The ¢rst CDF protein from plants was ZAT [214]. Arabidopsis thaliana plants overexpressing ZAT (AtCDF1) exhibited enhanced Zn2þ resistance and strongly increased Zn2þ content in the roots under high Zn2þ exposure. Antisense mRNA-producing plants were viable, with a wildtype level of Zn2þ resistance and content [214]. The ZAT protein was puri¢ed and functioned as a zinc uptake system in several microbial model systems [178]. ZAT did not transport other heavy metal cations with the exception of a slow transport of cadmium. A. thaliana contains at least seven more CDF proteins [215]. As in mammals, a variety of CDF proteins may also be responsible for zinc homeostasis in plants. A ZAT-like gene was also cloned from the heavy metal hyperaccumulator plant Thlaspi caerulescens [216]. From the related plant T. goesingense, a CDF-encoding gene with interesting properties was cloned. Transcription of the MTP1 gene from this plant leads to a spliced and an unspliced mRNA. Both mRNAs encode proteins which di¡er within a histidine-rich putative metal binding domain and in their substrate speci¢city [217]. The protein translated from the unspliced mRNA confers high level tolerance to Cd2þ , Co2þ and Zn2þ , whereas the product of the spliced mRNA confers nickel resistance [217]. This is reminiscent of the ZnT5/ZTL1 situation in animals and of nickel as a possible substrate of Zhf. 3.2. Biochemistry of CDF proteins With transport of Ni2þ by the T. goesingense protein, the substrate spectrum of CDF proteins includes Fe2þ , Co2þ , Ni2þ , Zn2þ and Cd2þ . The ionic radii of these cations are 76 pm, 74 pm, 72 pm, 74 pm and 97 pm, respectively [21]. Of the other divalent heavy metal cations of biological importance, Mn2þ (80 pm) may also fall in the size range of possible CDF substrates. In contrast, Mg2þ (65 pm) and Cu2þ (69 pm) may be too small and Ca2þ (99 pm) and Hg2þ (110 pm) too big to serve as CDF substrates in general. Additionally, a⁄nity for a substrate binding site should also di¡er between substrate cations. The cations of the ¢rst transition period usually have a higher a⁄nity for free electron pairs of oxygen and nitrogen atoms, whereas the cations of the second and third transition periods prefer the ‘soft’ electrons of sulfur atoms. The majority of the CDF substrates are divalent metal cations of the ¢rst transition period with sizes of 74 Z 2 pm diameter with Cd2þ being a piggy-back rider due to its chemical similarity to Zn2þ .
FEMSRE 781 2-6-03
Most CDF proteins have a size of 300^400 aa. CzcD from R. metallidurans exhibits six transmembrane spans [179] and this can also be assumed as the rule for most other CDF proteins. CDF proteins contain many histidine residues that are located at the amino-terminus, the carboxy-terminus, or between transmembrane helices IV and V, or at all three locations. Histidine residues of CzcD bound two to three Zn2þ per CzcD polypeptide (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) and were also essential for transport of zinc by ZitB from E. coli [185]. Although the histidine-rich carboxy-termini of CzcD and ZitB were not essential for transport (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) [185], they could have a function as regulators of transport activity of these proteins. The CzcDs from B. subtilis and R. metallidurans and ZitB from E. coli are e¥ux pumps driven by the proton motive force or a potassium gradient (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) [176,177,180,185] and function of ZRC1 from S. cerevisiae also required a proton gradient [196]. In eukaryotic cells, however, not all membranes harbor a proton, sodium or potassium gradient. Therefore, some CDF proteins may function as simple di¡usion facilitators in these organisms. Cation transport by the ZAT protein from A. thaliana was not stimulated in proteoliposomes by a proton gradient. Moreover, ZAT functioned as an uptake system in several microbial cells [178]. Therefore, cation transport by CDF proteins could be either an antiport reaction against a chemiosmotic gradient or a facilitated di¡usion driven by the concentration gradient of the respective cation. 3.3. Distribution of CDF proteins in the living world In a similar manner to the RND proteins, the distribution of CDF proteins in sequenced bacterial genomes was analyzed (Table 1). The proteins in these genomes and in current databases were compared in multiple alignments, and used to group and subgroup these proteins (Table 3). Three groups of CDF proteins could be established. Group 1 contained two subgroups, 1a and 1b. Cluster 1b is composed of the large proteins ZnT5, ZTL1 and MSC2, together with three other sequences. This assignment may be an artefact of their size and the multiple cluster algorithm used. Cluster 1a, on the other hand, consists of six sequences of putative proteins from proteobacteria signi¢cantly related to each other. This might indicate a common mechanism of function. The remaining CDF proteins fall into two groups, each containing several subgroups. Group 2 contains most of the known zinc-transporting CDF proteins in bacteria (CzcDs, ZitB), eukaryotes and archaea (Table 3). Cluster 2d occurs exclusively in eukaryotes and also contains the nickel-transporting plant protein that is highly related to ZAT from A. thaliana [215]. Group 2a, which branches o¡ ¢rst, is found in Crenarchaeota (Table 3). This might in-
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
dicate that the mainly zinc-transporting CDF proteins may have originated very early in evolution. Group 3 is assembled around YiiP from E. coli. No data are available for proteins in this group with the exception of MamB, the probable Fe2þ -transporting CDF protein from magnetosome membranes of magnetotactic bacteria [186]. In aerobic environments, iron occurs mainly as the insoluble Fe3þ , however, in anaerobic or micro-aerobic environments, a considerable concentration of Fe2þ may be present. Fe2þ can also be a substrate of CorA-like magnesium uptake systems [198]. Similar to Zn2þ , a second-step ¢ltering of Fe2þ against Mg2þ could be required, which yields a speculative initial idea for the function of the group 3 CDF proteins. When the distribution of CDF proteins is analyzed (Table 1), most prokaryotes contain two putative CDF proteins, one from group 2 and one from group 3. Of the 64 genomes analyzed, 24 contained two putative CDF proteins, 15 one, 16 none, seven three and only two contained four. Prokaryotes with two putative CDF proteins always harbored one member of group 3 plus a putative CDF from group 2 (75%), group 1 (12.5%), or a second protein from group 3 (also 12.5%). Thus, the general out¢t of a bacterium is one group 3 and one group 2 CDF protein. The prokaryotes with three or four CDF proteins always contained a group 3 protein, exclusively or in addition to a group 2 or a group 1 protein. This could indicate that the gene for a group 3 CDF protein is less likely to have been sacri¢ced during evolution than that of a group 2 zinctransporting CDF protein. If group 3 proteins are indeed
327
iron transporters, this may mirror the higher biological importance of iron in comparison to zinc.
4. The third layer of heavy metal resistance : P-type ATPases are the basic defense against heavy metal cations P-type ATPases constitute a superfamily of transport proteins that are driven by ATP hydrolysis [218]. Members of this family occur in all three kingdoms of life. Substrates are inorganic cations such as Hþ , Naþ , Kþ , Mg2þ , Ca2þ , Cuþ , Agþ , Zn2þ and Cd2þ [144]. An individual P-type ATPase can either import its substrate from the outside or from the periplasm to the cytoplasm (importing P-type ATPase) or export it from the cytoplasm to the outside/periplasm (exporting P-type ATPase). With respect to heavy metal homeostasis, P-type ATPases may be important for two reasons. Firstly, import systems for macroelements such as Mg2þ may also import heavy metals [219]. Secondly, exporting P-type ATPases may detoxify heavy metal cations by e¥ux. Members of the huge P-type ATPase superfamily that transport heavy metal cations carry a conserved proline residue, preceded and/or followed by a cysteine residue. This motif is essential for the function of at least one member of this family [220]. Work of the last few years on these ‘CPx-type’ or soft metal-transporting ATPases has been reviewed before [89] and will be partially covered in other reviews in this issue. A survey of heavy metal
Table 3 Groups of CDF proteins Name CDF1 1a 1b CDF2 2a 2b 2c 2d 2e 2f 2g 2h CDF3 3a 3b 3c 3d 3e 3f
Occurrence
Predicted substratea
Examplesb
Proteobacteria (K, L, Q) Bacteria, yeasts, mammals
Zn2þ ? Zn2þ
PA1297 MSC2, ZnT5, ZTL1, ZnT7
Crenarchaeota Euryarchaeota, £exible bacteria, Gram-positive bacteria Bacteria and thermophilic Euryarchaeota Multicellular eukaryotes Proteo- and cyanobacteria Gram-positive bacteria Bacteria Bacteria
Zn2þ ???? Zn2þ ??? Zn2þ ?? Zn2þ , (Ni2þ ) Zn2þ ? Zn2þ ? Zn2þ Zn2þ , Co2þ , Cd2þ
PAE2715 Spr1672 PH0896 ZAT, ZnT2 and 3 All7610 SA0155 ZitB, CzcD_Bacsu CzcD_Ralme
Bacteria Proteobacteria (K, N) Bacteria Gram-positive bacteria (low G+C) Thermotoga and Gram-positive bacteria (low G+C) Prokaryotes
unknown unknown unknown unknown unknown Fe2þ ?
Aq_2073 YiiP DR1236 EF0859 YeaB, YbdO MamB
a
The number of question marks after the predicted substrates indicates the probability of this prediction ; the more question marks, the lower the probability for the putative substrate. b Representative sequences are from P. aeruginosa (PA1297), S. cerevisiae (MSC2), Pyrobaculum aerophilum (PAE2715), Streptococcus pneumoniae (Spr1672), Pyrococcus horikoshii (PH0896), A. thaliana (ZAT), mammals (ZnT2, ZnT3, ZnT5, ZnT7, ZTL1), Nostoc (All7610), S. aureus (SA0155), E. coli (ZitB, YiiP), B. subtilis (CzcD_Bacsu, YeaB, YbdO), R. metallidurans (CzcD_Ralme), A. aeolicus (Aq_2073), D. radiodurans (DR1236), Enterococcus faecalis (EF0859), Magnetococcus gryphiswaldensis (MamB).
FEMSRE 781 2-6-03
Cyaan Magenta Geel Zwart
328
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
cation-transporting CPx-type ATPases in this section will therefore concentrate on data that are important in understanding the interconnections between CPx-type ATPases and members of the other two metal e¥ux protein families, CDF and RND. 4.1. Families of CPx-type ATPases and their substrates One family of CPx-type ATPases is involved in export of Cuþ (but not Cu2þ [220]) and Agþ (Cu-CPx-type ATPases [88,89]. Members of this group were characterized in a variety of organisms, in the archaeon Archaeoglobus fulgidus [221], the cyanobacterium Synechocystis [222], the Gram-positive bacteria B. subtilis [223], Enterococcus faecium [224], Enterococcus hirae [225^227], Streptococcus mutans [228], Lactobacillus sakei [229], the proteobacteria Rhizobium leguminosarum, Sinorhizobium meliloti [230], P. putida [231] and E. coli [87]. Cu-CPx-type ATPases also occur in lower eukaryotes as Dictyostelium [232], Cryptosporidium parvum [233] and Candida albicans [234]. Seven of the 45 genes from A. thaliana that may encode Ptype ATPases are putative CPx-type ATPases [235]. One of these proteins may transport copper and is required for ethylene signaling in the plant [236]. In mammals including human, miss-expression or mutant proteins of CuCPx-type ATPases were assigned to the copper homeostasis defects Menkes’ and Wilson’s diseases [237,238]. This demonstrates a broad distribution of Cu-CPx-type ATPases in the living world. A second family transports Zn2þ , Cd2þ and Pb2þ (ZnCPx-type ATPases), again, mainly from the cytoplasm to the outside or to the periplasm. The very ¢rst member of the CPx-type ATPases described was CadA, a cadmium resistance-mediating protein encoded on a plasmid from S. aureus [13,239]. Examples from other bacteria such as B. subtilis [240], Listeria monocytogenes [241], P. putida [242], S. maltophilia [243], E. coli [244^248], and the yeast S. cerevisiae [249] followed, but the distribution of ZnCPx-ATPases seems to be not as broad as the distribution of the Cu-CPx-type ATPases. The substrate speci¢city of CPx-type ATPases is Cuþ / Agþ or Zn2þ /Cd2þ /Pb2þ . A CPx-type ATPase from the cyanobacterium Oscillatoria brevis uses heavy metal cations from either group as substrates [250]. Overexpression of CadA from H. pylori has some e¡ect on cobalt resistance, so Co2þ may also be a substrate of Zn-CPx-type ATPases [251], as has been suggested before for a P-type ATPase from Synechocystis [252]. There is, however, no clear evidence for the transport of Ni2þ by a CPx-type ATPase. A publication claiming the cloning of a P-type ATPase [253] involved in nickel resistance deals with a possible nickel uptake protein from a completely di¡erent protein family (NiCoT) not known to be involved in metal resistance, but in metal uptake [254^256]. The direction of transport of CPx-type ATPases in bacteria is mostly export to the outside or the periplasm, but
FEMSRE 781 2-6-03
the other direction of transport also seems to occur. The ¢rst examples of Cu-CPx-ATPases came from the Grampositive bacterium E. hirae. Two Cu-CPx-type ATPases were found, CopA involved in copper uptake and CopB in copper resistance by e¥ux [257^261]. These are reviewed in [283]. Physiological evidence indicates that two CPx-type ATPases from the cyanobacterium Synechocystis, CtaA and PacS, may be involved in copper uptake [284]. They facilitate the synthesis of the copper-dependent electron transfer factor plastocyanin [222], although disruption of pacS decreased copper tolerance. Disruption of ctaA diminished the amount of cell-bound copper [222], clearly indicating a function of CtaA in copper uptake. There is also evidence from B. subtilis that the ZnCPx-type ATPase ZosA is a zinc uptake system [262]. It seems to be easy for evolution to change an e¥ux system into an uptake system when this is required by the speci¢c environment the cell struggles to survive in. 4.2. Structure and function The membrane topology of CPx-type ATPases was analyzed exemplarily for the ¢rst described member, CadA from S. aureus [263]. The protein contains eight transmembrane segments with the conserved, eponymous CPC motif located in the sixth of these segments. The ATP binding domain is situated in a large cytoplasmic domain that follows this transmembrane domain. The amino-terminal 109 aa carry a CXXC motif that is also present in most members of the CPx-ATPases [263]. Although heavy metal cations are able to bind to the amino-terminal domain [264], this domain is not essential for transport [248] and a regulatory function has been discussed [241]. When the HME-RND proteins were discussed above, the problems in detoxifying thiol-bound heavy metal cations were mentioned. The complex binding constant for Cuþ ^glutathionate complexes is 1039 [75]. The energy to remove the Cuþ cation from this complex is vGPo = 3RT ln1039 = 3226 kJ mol31 at 30‡C. Thus, to detoxify heavy metal cations bound to thiols, paired cysteine residues in the substrate binding or substrate funneling site are required and additionally a lot of energy. CPx-type ATPases exhibit both features and it has already been demonstrated that the substrates in vivo are likely metal^thiolate complexes rather than the ‘free’ metals [247]. 4.3. Distribution of CPx-type ATPases in prokaryotes If CPx-type ATPases clean out the thiol-bound heavy metal pool, they should occur even more frequently in the living world than CDF or RND proteins. In a similar manner to the approach described above for the other two protein families, the distribution of putative CPxtype ATPase in prokaryotic genomes was analyzed. A number of 16 groups of Cu-CPx-type ATPases and 12 groups of Zn-CPx-type ATPases were found by multiple
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
alignments (data not shown). Members of the other families of P-type ATPases such as MgtA-like magnesium transporters and KdpG-like potassium transporters were only considered as examples. Generally, the KdpG-like proteins were more related to the CPx-type ATPases than the MgtA-like proteins (data not shown). Starting from the 64 prokaryotes analyzed, only eight species did not contain a gene for a putative P-type ATPase at all (Table 1). Five species contained at least one P-type ATPase, but none of the CPx-type. Most (29 species) had CPx-type ATPases of both families, the Cu-CPxtype and the Zn-CPx-type ATPases. If prokaryotes harbored only a member of one of these two families, this was mostly a member of the Cu-CPx-type ATPases (18 species) and in just four cases only a member of the Zn-CPx-type ATPases. Thus, nearly half of the species analyzed contained members of both CPx families, which indicates the importance of the resistance mediated by these proteins. As with the RND proteins, export of copper/silver was more important than e¥ux of zinc/cadmium/lead, mirroring the di¡erent toxicities of these cations. R. metallidurans contains genes or candidate genes for two Cu-CPx-type ATPases, three Zn-CPx-like ATPases and ¢ve other P-type ATPases giving a total number of 10 (Table 1). One of the Zn-CPx-type ATPases, PbrA, is part of a plasmid-mediated lead resistance determinant in R. metallidurans [265], the other two, ZntA and CadA, are involved in resistance to Zn2þ and Cd2þ [68]. This complement of CPx-type ATPases is not unusually high for a bacterium; cyanobacteria, mycobacteria or Rhizobium species exhibited similar numbers (Table 1).
5. Other heavy metal export systems: CHR protein family, NreB, CnrT Before the data about RND, CDF, and CPx proteins are assembled into a picture that describes heavy metal detoxi¢cation in prokaryotic cells, three small groups of transporters will be introduced. All three are linked to resistance determinants that may allow a bacterium to survive in the speci¢c nickel-, cobalt- and chromate-rich serpentine environment [266]. ChrA-like proteins of the CHR protein family [267] detoxify the oxyanion chromate, while NreB and CnrT are nickel e¥ux systems.
329
mode of their action [267]. CHR proteins are probably chromate e¥ux pumps [268,269] driven by the chemiosmotic gradient. Since the export of an anion is in the direction of the electric ¢eld of the proton motive force, v8 alone is su⁄cient to drive chromate e¥ux. The CHR proteins of 64 prokaryotic genomes were compared by multiple alignments and grouped into ¢ve categories that may mirror di¡erences in the physiological or biochemical modes of action. Only a quarter of the bacterial genomes contained at least one putative CHR member (Table 1). Two CHR-encoding genes were found in R. solanacearum and Bacillus halodurans, four in R. metallidurans. Again, R. metallidurans contained the highest number of members of a protein family involved in heavy metal export. Of the 21 putative CHR proteins analyzed, involvement in chromate e¥ux has only been shown for two of the four proteins from R. metallidurans [267,270^273], the ChrA protein from P. aeruginosa [1,268,269,274], and Synechococcus, a cyanobacterium roughly related to Synechocystis [275]. 5.2. NreB-like proteins In three bacterial species (Synechocystis sp., Nostoc sp., S. meliloti), the presence of a CHR protein was accompanied by the presence of a NreB-like protein in the genome (Table 1), in Magnetococcus it appeared without a CHR protein. The NreB protein has been characterized in R. metallidurans strain 31A [173]. This bacterium contains two distinct nickel resistance determinants on its native megaplasmids pTOM8 and pTOM9: ncc [67], which was mentioned above, and nreB [173]. Expression of the nreB gene was speci¢cally induced by nickel and conferred nickel resistance in both strain 31A and E. coli. Cells expressing nreB showed decreased accumulation of Ni2þ , suggesting that NreB mediates nickel e¥ux. The NreB-like proteins have a size of about 400 aa and contain probably 12 transmembrane spans. One putative protein from the enterobacterium Hafnia alvei is much smaller and may be the result of a sequencing error. The other six NreB-like proteins found in current databases all exhibit a characteristic, histidine-rich carboxy-terminus that resembles a natural His tag. 5.3. CnrT-like proteins
5.1. CHR proteins detoxify chromate The CHR family contains proteins that are bound to the cytoplasmic membrane by 10 transmembrane spans [267]. The protein probably originated by gene duplication followed by gene fusion. Several bacteria contain operons that comprise two successive genes encoding the aminoand the carboxy-terminal parts, respectively, of a fulllength CHR protein. This led to the hypothesis that both halves of the full-length proteins may di¡er in the
FEMSRE 781 2-6-03
Only three proteins in the analyzed genomes and the current databases were similar to CnrT from R. metallidurans. These were putative proteins from Yersinia pestis, R. solanacearum and M. loti (Table 1). The cnrT gene is located directly downstream of the cnr cobalt^nickel resistance determinant in plasmid pMOL28 of R. metallidurans and encodes a small degree of nickel resistance by itself (G. Grass, S. Ku«hnemund, B. Fricke, K. Sutter, G. KrauM and D.H. Nies, unpublished) [65]. CnrT-like proteins ex-
Cyaan Magenta Geel Zwart
330
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
hibit 12 or 14 transmembrane spans and, like NreB proteins, a histidine-rich carboxy-terminus. CnrT- and NreB-like proteins do not exhibit any potential ATP binding sites and are thus probably driven by the proton motive force, like CDF proteins. Moreover, all three groups of proteins contain cytoplasmic loops of the polypeptide chain rich in putative metal binding regions and export their substrates across the cytoplasmic membrane. Therefore, CnrT- and NreB-like proteins may act as secondary ¢lter proteins for Ni2þ similar to CDF proteins that may act as secondary ¢lter proteins for Zn2þ , Co2þ and Cd2þ .
6. Heavy metal resistance in prokaryotes revisited Which modes of heavy metal e¥ux do most bacteria have? Export of zinc and cadmium is accomplished at least by CDF2, Zn-CPx and HME1-RND proteins. Of the 64 prokaryotic genomes analyzed, there are approximately equal numbers of species harboring no system (17 species), only one CDF2 protein (13 species), only one ZnCPx (15 species), or a CDF2 plus a Zn-CPx protein (13 species). Only a minority of prokaryotes analyzed have all three proteins (three species), an ATPase plus a HME1RND protein (two species), a CDF2 plus a HME1-RND protein (one species) or exclusively an RND protein (no species). Thus, the presence of a HME1-RND protein is the exception and marks high level resistance to zinc, cobalt and cadmium, while possession of a CDF2 and/or a Zn-CPx-type ATPase is the general case and therefore part of the cellular homeostasis system. Moreover, since no CBA transport system occurs without the presence of a helper protein that imports the cations into the periplasm, this shows that periplasmic access to the CBA pump may also be important for resistance to this group of heavy metal cations. Bacteria relying exclusively on a CDF2 protein for their zinc homeostasis are more adapted to an anaerobic or micro-aerobic environment than to higher oxygen concentrations. The need to remove metals from the thiol pool by a CPx-type ATPase may be smaller when the additional oxidation stress by molecular oxygen is absent. As far as we know today, CPx-type ATPases or CDF proteins detoxify nickel only in rare cases. RND proteins known to detoxify nickel and cobalt (HME2 group) or candidates for this process (HME5 group) are rare and mark species as adapted to nickel-rich environments. Moreover, NreB- and CnrT-like nickel transport systems occur in only a few species. Nickel is used seldom as a cofactor [18] and if needed, is imported into the cell by chemiosmotically driven slow and speci¢c uptake systems of the NiCoT family [46,254^256,276^278] or ABC transport systems [279]. The concentration of nickel in seawater and its toxicity to E. coli is comparable to that of zinc, but this comparison has to be viewed in the light of two
FEMSRE 781 2-6-03
known zinc-exporting systems in E. coli (Table 1) and no known nickel e¥ux system. The toxicity of nickel seems to be a smaller problem than the toxicity of zinc or, especially, copper. Copper is mainly detoxi¢ed by a Cu-CPx-type ATPase. Of the analyzed 64 genomes, 41 contained the gene for a putative Cu-CPx-ATPase, 17 no such protein, and six an additional HME4-RND protein. The presence of such a RND protein may mark a demand for a supplementary export of the periplasmic copper, maybe under anaerobic conditions [93]. The presence of an RND-driven CBA pump is again an exception, indicating increased heavy metal resistance. However, the highest degree of copper resistance is mediated by plasmid-bound copper resistance determinants [9,79^81,83,84] that are mentioned in the article by Grass and Rensing [88]. Seventeen species did not contain any known zinc export system and the same number no known copper e¥ux system (Table 1). The two groups overlap, but not completely ; nine species contained neither zinc- nor copperexporting systems. Four of them, the pathogenic spirochete B. burgdorferi, two species of Rickettsia and the methanogenic archaeon M. jannaschii, contained at least a CDF3 and/or a CHR protein indicating some skills in export of heavy metals: the unknown CDF3 substrate and/or chromate. The remaining ¢ve species without known competence to detoxify heavy metals are the cellwall-free pathogenic Mycoplasma and Ureaplasma species and the aphid symbiont Buchnera. In their pathogenic or symbiotic way of life, these bacteria are probably not confronted with heavy metal stress, rendering metal homeostasis genes super£uous.
7. Concluding remarks: heavy metal resistance is the product of the interaction of many systems We may now be able to answer the question we started with: what makes a heavy metal resistant bacterium like R. metallidurans heavy metal resistant? Firstly, there is a multitude of resistance systems. When R. metallidurans is compared to R. solanacearum, an increase in the numbers of the CPx-type, CDF and RND proteins is observed (Table 1). These additional genes are the product either of horizontal gene transfer or of a duplication of genes already present in the common Ralstonia ancestor. If the duplicated genes in R. metallidurans were closely related to a respective gene in R. solanacearum, this would indicate a possible gene duplication that had happened during the evolution of the genus R. metallidurans. If this is not the case and if the genes are additionally harbored by a plasmid, this would point towards horizontal gene transfer. The number of CPx-type ATPases increases from R. solanacearum to R. metallidurans from one to three Zn-CPx proteins, while the number of Cu-CPx-ATPases is constant at two. The four Zn-CPx proteins from both
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
Ralstonia species belong to the same group of proteins, Zn6, which indicates an origin by gene multiplication. CHR proteins duplicated from two to four. R. metallidurans contains one CDF protein more than R. solanacearum. Both species harbor a CDF1 and a CDF3 protein, the additional protein in R. metallidurans is the CDF2 protein CzcD that is encoded by plasmid pMOL30. The number of HME-RND proteins increased from four in R. solanacearum to an outstanding number of 12 in R. metallidurans. When the various clusters of HME proteins are compared, the number of HME1 and HME4 proteins is duplicated, the number of two HME3 proteins in R. solanacearum is even tripled to six proteins in R. metallidurans. In all cases, predicted proteins from R. metallidurans and R. solanacearum group closely with each other indicating an origin from a common ancestral gene. The remaining two HME2 proteins, CnrA and NccA, which occur only in R. metallidurans and not in R. solanacearum, are both encoded on plasmids that may have been acquired by horizontal gene transfer. Secondly, there is di¡erentiation of the function of metal resistance systems after the duplication event has occurred. The three Zn-CPx-type ATPases of R. metallidurans became lead-, cadmium- and zinc-speci¢c e¥ux pumps. Thirdly, there is combination of the specialized proteins with others to form resistance systems with increased e⁄ciency. Of the two determinants that encode a HME2 pump, ncc on plasmid pMOL30 of strain CH34 became inactivated while cnr combined with the cnrT gene to form an e⁄cient nickel/cobalt pump. The HME3 proteins that may detoxify copper and silver combined with other copper resistance systems. The additional CDF protein in R. metallidurans, CzcD, became attached to the Czc system and is involved in regulation of expression and provides additionally a minor degree of metal resistance. Other genes also became part of the Czc system that is intriguing and unique in its complexity. The three steps, (1) gene multiplication by duplication and horizontal transfer, (2) di¡erentiation of function and (3) combination of genes into highly e⁄cient operons, created the outstanding heavy metal resistance of R. metallidurans.
Acknowledgements This work was funded by the Deutsche Forschungsgemeinschaft, the state Sachsen-Anhalt and by the Fonds der Chemischen Industrie. I am grateful to Chris Rensing, Gregor Grass and Cornelia GroMe for their comments on this review. The R. metallidurans CH34 web site of DOE/ JGI (http://www.jgi.doe.gov/JGI_microbial/html/ralstonia/ ralston_homepage.html) and the comprehensive microbial resource http://www.tigr.org/tigr-scripts/CMR2/CMR HomePage.spl of The Institute of Genomic Research (TIGR) were used intensively to compile Table 1. I would also like to thank all the colleagues, mentors and students
FEMSRE 781 2-6-03
331
in my lab who in the last 20 years have contributed to the progress of the CH34 story.
References [1] Ohtake, H., Cervantes, C. and Silver, S. (1987) Decreased chromate uptake in Pseudomonas £uorescens carrying a chromate resistance plasmid. J. Bacteriol. 169, 3853^3856. [2] Silver, S. and Phung, L.T. (1996) Bacterial heavy metal resistance: New surprises. Annu. Rev. Microbiol. 50, 753^789. [3] Broer, S., Ji, G., Broer, A. and Silver, S. (1993) Arsenic e¥ux governed by the arsenic resistance determinant of Staphylococcus aureus plasmid pI258. J. Bacteriol. 175, 3480^3485. [4] Brown, N.L., Misra, T.K., Winnie, J.N., Schmidt, A., Sei¡, M. and Silver, S. (1986) The nucleotide sequence of the mercuric resistance operons of plasmid R100 and transposon Tn501: further evidence for mer genes which enhance the activity of the mercuric ion detoxi¢cation system. Mol. Gen. Genet. 202, 143^151. [5] Endo, G. and Silver, S. (1995) CadC, the transcriptional regulatory protein of the cadmium resistance system of Staphylococcus aureus plasmid pI258. J. Bacteriol. 177, 4437^4441. [6] Silver, S., Budd, K., Leahy, K.M., Shaw, W.V., Hammond, D., Novick, R.P., Willsky, G.R., Malamy, M.H. and Rosenberg, H. (1981) Inducible plasmid-determined resistance to arsenate, arsenite, and antimony(III) in Escherichia coli and Staphylococcus aureus. J. Bacteriol. 146, 983^996. [7] Silver, S., Misra, T.K. and Laddaga, R.A. (1989) DNA sequence analysis of bacterial toxic heavy metal resistances. Biol. Trace Elem. Res. 21, 145^163. [8] Silver, S. (1992) Plasmid-determined metal resistance mechanisms : range and overview. Plasmid 27, 1^3. [9] Silver, S., Lee, B.T.O., Brown, N.L. and Cooksey, D.A. (1993) Bacterial plasmid resistances to copper, cadmium, and zinc. In: Chemistry of Copper and Zinc Triads (Welch, A.J. and Chapman, S.K., Eds.), pp. 38^53. The Royal Society of Chemistry, London. [10] Silver, S. and Ji, G. (1994) Newer systems for bacterial resistances to toxic heavy metals. Environ. Health Perspect. 102, 107^113. [11] Silver, S. and Walderhaug, M. (1995) Bacterial plasmid-mediated resistances to mercury, cadmium and copper. In: Toxicology of Metals. Biochemical Aspects (Goyer, R.A. and Cherian, M.G., Eds.), pp. 435^458. Springer Verlag, Berlin. [12] Summers, A.O. and Silver, S. (1972) Mercury resistance in a plasmidbearing strain of Escherichia coli. J. Bacteriol. 112, 1228^1236. [13] Nucifora, G., Chu, L., Misra, T.K. and Silver, S. (1989) Cadmium resistance from Staphylococcus aureus plasmid pI258 cadA gene results from a cadmium-e¥ux ATPase. Proc. Natl. Acad. Sci. USA 86, 3544^3548. [14] Nies, D.H. (2000) Heavy metal resistant bacteria as extremophiles: molecular physiology and biotechnological use of Ralstonia spec. CH34. Extremophiles 4, 77^82. [15] Goris, J., De Vos, P., Coenye, T., Hoste, B., Janssens, D., Brim, H., Diels, L., Mergeay, M., Kersters, K. and Vandamme, P. (2001) Classi¢cation of metal-resistant bacteria from industrial biotopes as Ralstonia campinensis sp. nov., Ralstonia metallidurans sp. nov. and Ralstonia basilensis Steinle et al., 1998 emend. Int. J. Syst. Evol. Microbiol. 51, 1773^1782. [16] Diels, L. and Mergeay, M. (1990) DNA probe-mediated detection of resistant bacteria from soils highy polluted by heavy metals. Appl. Environ. Microbiol. 56, 1485^1491. [17] Mergeay, M., Monchy, S., Vallaeys, T., Auquier, V., Bentomane, A., Bertin, P., Taghavi, S., Dunn, J., van der Lelie, D. and Wattiez, R. (2003) Ralstonia metallidurans, a bacterium speci¢cally adapted to toxic metals: towards a catalogue of metal-responsive genes. FEMS Microbiol. Rev. 27, 385^410.
Cyaan Magenta Geel Zwart
332
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[18] Nies, D.H. (1999) Microbial heavy metal resistance. Appl. Microbiol. Biotechnol. 51, 730^750. [19] Mergeay, M., Nies, D., Schlegel, H.G., Gerits, J., Charles, P. and van Gijsegem, F. (1985) Alcaligenes eutrophus CH34 is a facultative chemolithotroph with plasmid-bound resistance to heavy metals. J. Bacteriol. 162, 328^334. [20] Mukhopadhyay, R., Rosen, B.P., Pung, L.T. and Silver, S. (2002) Microbial arsenic: from geocycles to genes and enzymes. FEMS Microbiol. Rev. 26, 311^325. [21] Weast, R.C. (1984) CRC Handbook of Chemistry and Physics, 64th edn. CRC Press, Boca Raton, FL. [22] Saier Jr., M.H., Tam, R., Reizer, A. and Reizer, J. (1994) Two novel families of bacterial membrane proteins concerned with nodulation, cell division and transport. Mol. Microbiol. 11, 841^847. [23] Tseng, T.-T., Gratwick, K.S., Kollman, J., Park, D., Nies, D.H., Go¡eau, A. and Saier, M.H.J. (1999) The RND superfamily : an ancient, ubiquitous and diverse family that includes human disease and development proteins. J. Mol. Microbiol. Biotechnol. 1, 107^ 125. [24] Brown, D.E., Thrall, M.A., Walkley, S.U., Wurzelmann, S., Wenger, D.A., Allison, R.W. and Just, C.A. (1996) Metabolic abnormalities in feline Niemann-Pick type C heterozygotes. J. Inherit. Metab. Dis. 19, 319^330. [25] Campo, J.V., Stowe, R., Slomka, G., Byler, D. and Gracious, B. (1998) Psychosis as a presentation of physical disease in adolescence: a case of Niemann-Pick disease, type C. Dev. Med. Child. Neurol. 40, 126^129. [26] Carstea, E.D., Morris, J.A., Coleman, K.G., Loftus, S.K., Zhang, D., Cummings, C., Gu, J., Rosenfeld, M.A., Pavan, W.J., Krizman, D.B., Nagle, J., Polymeropoulos, M.H., Sturley, S.L., Ioannou, Y.A., Higgins, M.E., Comly, M., Cooney, A., Brown, A., Kaneski, C.R., Blanchette Mackie, E.J., Dwyer, N.K., Neufeld, E.B., Chang, T.Y., Liscum, L. and Tagle, D.A. (1997) Niemann-Pick C1 disease gene: homology to mediators of cholesterol homeostasis. Science 277, 228^231. [27] Erickson, R.P., Aviles, R.A., Zhang, J., Kozloski, M.A., Garver, W.S. and Heidenreich, R.A. (1997) High-resolution mapping of the spm (Niemann-Pick type C) locus on mouse chromosome 18. Mamm. Genome 8, 355^356. [28] Garver, W.S., Erickson, R.P., Wilson, J.M., Colton, T.L., Hossain, G.S., Kozloski, M.A. and Heidenreich, R.A. (1997) Altered expression of caveolin-1 and increased cholesterol in detergent insoluble membrane fractions from liver in mice with Niemann-Pick disease type C. Biochim. Biophys. Acta 1361, 272^280. [29] Garver, W.S., Hsu, S.C., Erickson, R.P., Greer, W.L., Byers, D.M. and Heidenreich, R.A. (1997) Increased expression of caveolin-1 in heterozygous Niemann-Pick type II human ¢broblasts. Biochem. Biophys. Res. Commun. 236, 189^193. [30] Gillan, T.L., Byers, D.M., Riddell, D.C., Neumann, P.E. and Greer, W.L. (1997) Limiting the Niemann-Pick type C critical region to a 1-cM interval. Clin. Invest. Med. 20, 339^343. [31] Goodrum, J.F. and Pentchev, P.G. (1997) Cholesterol reutilization during myelination of regenerating PNS axons is impaired in Niemann-Pick disease type C mice. J. Neurosci. Res. 49, 389^392. [32] Greer, W.L., Riddell, D.C., Byers, D.M., Welch, J.P., Girouard, G.S., Sparrow, S.M., Gillan, T.L. and Neumann, P.E. (1997) Linkage of Niemann-Pick disease type D to the same region of human chromosome 18 as Niemann-Pick disease type C. Am. J. Hum. Genet. 61, 139^142. [33] Greer, W.L., Riddell, D.C., Gillan, T.L., Girouard, G.S., Sparrow, S.M., Byers, D.M., Dobson, M.J. and Neumann, P.E. (1998) The Nova Scotia (type D) form of Niemann-Pick disease is caused by a G3097CT transversion in NPC1. Am. J. Hum. Genet. 63, 52^54. [34] Gu, J.Z., Carstea, E.D., Cummings, C., Morris, J.A., Loftus, S.K., Zhang, D., Coleman, K.G., Cooney, A.M., Comly, M.E., Fandino, L., Ro¡, C., Tagle, D.A., Pavan, W.J., Pentchev, P.G. and Rosenfeld, M.A. (1997) Substantial narrowing of the Niemann-Pick C
FEMSRE 781 2-6-03
[35]
[36] [37] [38]
[39] [40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
candidate interval by yeast arti¢cial chromosome complementation. Proc. Natl. Acad. Sci. USA 94, 7378^7383. Kovesi, T.A., Lee, J., Shuckett, B., Clarke, J.T., Callahon, J.W. and Phillips, M.J. (1996) Pulmonary in¢ltration in Niemann-Pick disease type C. J. Inherit. Metab. Dis. 19, 792^793. Lange, Y. and Steck, T.L. (1998) Four cholesterol-sensing proteins. Curr. Opin. Struct. Biol. 8, 435^439. Liscum, L. and Klansek, J.J. (1998) Niemann-Pick disease type C. Curr. Opin. Lipidol. 9, 131^135. Osborne, T.F. and Rosenfeld, J.M. (1998) Related membrane domains in proteins of sterol sensing and cell signaling provide a glimpse of treasures still buried within the dynamic realm of intracellular metabolic regulation. Curr. Opin. Lipidol. 9, 137^140. Schi¡mann, R. (1996) Niemann-Pick disease type C. From bench to bedside. J. Am. Med. assoc. 276, 561^564. Schedin, S., Sindelar, P.J., Pentchev, P., Brunk, U. and Dallner, G. (1997) Peroxisomal impairment in Niemann-Pick type C disease. J. Biol. Chem. 272, 6245^6251. Suresh, S., Yan, Z., Patel, R.C., Patel, Y.C. and Patel, S.C. (1998) Cellular cholesterol storage in the Niemann-Pick disease type C mouse is associated with increased expression and defective processing of apolipoprotein D. J. Neurochem. 70, 242^251. Schofer, O., Mischo, B., Puschel, W., Harzer, K. and Vanier, M.T. (1998) Early-lethal pulmonary form of Niemann-Pick type C disease belonging to a second, rare genetic complementation group. Eur. J. Pediatr. 157, 45^49. Sequeira, J.S., Vellodi, A., Vanier, M.T. and Clayton, P.T. (1998) Niemann-Pick disease type C and defective peroxisomal beta-oxidation of branched-chain substrates. J. Inherit. Metab. Dis. 21, 149^ 154. Steinberg, S.J., Mondal, D. and Fensom, A.H. (1996) Co-cultivation of Niemann-Pick disease type C ¢broblasts belonging to complementation groups alpha and beta stimulates LDL-derived cholesterol esteri¢cation. J. Inherit. Metab. Dis. 19, 769^774. Lange, Y., Ye, J. and Steck, T.L. (1998) Circulation of cholesterol between lysosomes and the plasma membrane. J. Biol. Chem. 273, 18915^18922. Saier, M.H.J. (2000) A functional-phylogenetic classi¢cation system for transmembrane solute transporters. Microbiol. Mol. Biol. Rev. 64, 354^411. Johnson, J.M. and Church, G.M. (1999) Alignment and structure prediction of divergent protein families: periplasmic and outer membrane proteins of bacterial e¥ux pumps. J. Mol. Biol. 287, 695^715. Andersen, C., Hughes, C. and Koronakis, V. (2001) Protein export and drug e¥ux through bacterial channel-tunnels. Curr. Opin. Cell Biol. 13, 412^416. Paulsen, I.T., Park, J.H., Choi, P.S. and Saier, M.H.J. (1997) A family of Gram-negative bacterial outer membrane factors that function in the export of proteins, carbohydrates, drugs and heavy metals from Gram-negative bacteria. FEMS Microbiol. Lett. 156, 1^8. Zgurskaya, H.I. and Nikaido, H. (2000) Multidrug resistance mechanisms: drug e¥ux across two membranes. Mol. Microbiol. 37, 219^ 225. Zgurskaya, H.I. and Nikaido, H. (1999) Bypassing the periplasm : Reconstitution of the AcrAB multidrug e¥ux pump of Escherichia coli. Proc. Natl. Acad. Sci. USA 96, 7190^7195. Zgurskaya, H.I. and Nikaido, H. (1999) AcrA is a highly asymmetric protein capable of spanning the periplasm. J. Mol. Biol. 285, 409^ 420. Zgurskaya, H.I. and Nikaido, H. (2000) Cross-linked complex between oligomeric periplasmic lipoprotein AcrA and the inner-membrane-associated multidrug e¥ux pump AcrB from Escherichia coli. J. Bacteriol. 182, 4264^4267. Nies, D., Mergeay, M., Friedrich, B. and Schlegel, H.G. (1987) Cloning of plasmid genes encoding resistance to cadmium, zinc, and cobalt in Alcaligenes eutrophus CH34. J. Bacteriol. 169, 4865^4868. Nies, D.H., Nies, A., Chu, L. and Silver, S. (1989) Expression and
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71] [72]
[73] [74]
[75]
nucleotide sequence of a plasmid-determined divalent cation e¥ux system from Alcaligenes eutrophus. Proc. Natl. Acad. Sci. USA 86, 7351^7355. Kunito, T., Kusano, T., Oyaizu, H., Senoo, K., Kanazawa, S. and Matsumoto, S. (1996) Cloning and sequence analysis of czc genes in Alcaligenes sp. strain CT14. Biosci. Biotechnol. Biochem. 60, 699^ 704. van der Lelie, D., Schwuchow, T., Schwidetzky, U., Wuertz, S., Baeyens, W., Mergeay, M. and Nies, D.H. (1997) Two component regulatory system involved in transcriptional control of heavy metal homoeostasis in Alcaligenes eutrophus. Mol. Microbiol. 23, 493^ 503. GroMe, C., Grass, G., Anton, A., Franke, S., Navarrete Santos, A., Lawley, B., Brown, N.L. and Nies, D.H. (1999) Transcriptional organization of the czc heavy metal homoeostasis determinant from Alcaligenes eutrophus. J. Bacteriol. 181, 2385^2393. Nies, D.H. and Brown, N.L. (1998) Two-component systems in the regulation of heavy metal resistance. In: Metal Ions in Gene Regulation (Silver, S. and Walden, W., Eds.), pp. 77^103. Chapman and Hall, London. Nies, D.H. (1992) CzcR and CzcD, gene products a¡ecting regulation of resistance to cobalt, zinc and cadmium (czc system) in Alcaligenes eutrophus. J. Bacteriol. 174, 8102^8110. Nies, D.H. and Silver, S. (1989) Plasmid-determined inducible e¥ux is responsible for resistance to cadmium, zinc, and cobalt in Alcaligenes eutrophus. J. Bacteriol. 171, 896^900. Nies, D.H. (1995) The cobalt, zinc, and cadmium e¥ux system CzcABC from Alcaligenes eutrophus functions as a cation-protonantiporter in Escherichia coli. J. Bacteriol. 177, 2707^2712. Sensfuss, C. and Schlegel, H.G. (1988) Plasmid pMOL28-encoded resistance to nickel is due to speci¢c e¥ux. FEMS Microbiol. Lett. 55, 295^298. Liesegang, H., Lemke, K., Siddiqui, R.A. and Schlegel, H.-G. (1993) Characterization of the inducible nickel and cobalt resistance determinant cnr from pMOL28 of Alcaligenes eutrophus CH34. J. Bacteriol. 175, 767^778. Grass, G., GroMe, C. and Nies, D.H. (2000) Regulation of the cnr cobalt/nickel resistance determinant from Ralstonia sp. CH34. J. Bacteriol. 182, 1390^1398. Tibazarwa, C., Wuertz, S., Mergeay, M., Wyns, L. and van der Lelie, D. (2000) Regulation of the cnr cobalt and nickel resistance determinant of Ralstonia eutropha (Alcaligenes eutrophus) CH34. J. Bacteriol. 182, 1399^1409. Schmidt, T. and Schlegel, H.G. (1994) Combined nickel-cobalt-cadmium resistance encoded by the ncc locus of Alcaligenes xylosoxidans 31A. J. Bacteriol. 176, 7045^7054. Legatzki, A., Franke, S., Lucke, S., Ho¡mann, T., Anton, A., Neumann, D. and Nies, D.H. (2003) First step towards a quantitative model describing Czc-mediated heavy metal resistance in Ralstonia metallidurans. Biodegradation (in press). Goldberg, M., Pribyl, T., Juhnke, S. and Nies, D.H. (1999) Energetics and topology of CzcA, a cation/proton antiporter of the RND protein family. J. Biol. Chem. 274, 26065^26070. Outten, C.E. and O’Halloran, T.V. (2001) Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 292, 2488^2492. Fahey, R.C., Wc, B., Adams, W.B. and Worsham, M.B. (1978) Occurrence of glutathione in bacteria. J. Bacteriol. 133, 1126^1129. McLaggan, D., Logan, T.M., Lynn, D.G. and W, E. (1990) Involvement of gamma-glutamyl peptides in osmoadaptation of Escherichia coli. J. Bacteriol. 172, 3631^3636. Fahey, R.C. (2001) Novel thiols of prokaryotes. Annu. Rev. Microbiol. 55, 333^356. Oram, P.D., Fang, X., Fernando, Q., Letkeman, P. and Letkeman, D. (1996) The formation constants of mercury(II)-glutathione complexes. Chem. Res. Toxicol. 9, 709^712. Osterberg, R., Ligaarden, R. and Persson, D. (1979) Copper(I) com-
FEMSRE 781 2-6-03
[76]
[77]
[78]
[79] [80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89] [90]
[91]
[92]
[93]
[94]
333
plexes of penicillamine and glutathione. J. Inorg. Biochem. 10, 341^ 355. Rae, T.D., Schmidt, P.J., Pufahl, R.A., Culotta, V.C. and O’ Halloran, T.V. (1999) Undetectable intracellular free copper: The requirement of a copper chaperone for superoxide dismutase. Science 284, 805^808. Dominey, L.A. and Kustin, K. (1983) Kinetics and mechanism of Zn(II) complexation with reduced glutathione. J. Inorg. Biochem. 18, 153^160. Brown, N.L., Camakaris, J., Lee, B.T., Williams, T., Morby, A.P., Parkhill, J. and Rouch, D.A. (1991) Bacterial resistances to mercury and copper. J. Cell. Biochem. 46, 106^114. Brown, N.L., Rouch, D.A. and Lee, B.T.O. (1992) Copper resistance determinants in bacteria. Plasmid 27, 41^51. Brown, N.L., Lee, B.T.O. and Silver, S. (1993) Bacterial transport of and resistance to copper. In: Metal Ions in Biological Systems, Vol. 30 (Sigel, H. and Sigel, A., Eds.), pp. 405^430. Marcel Dekker, New York. Brown, N.L., Barrett, S.R., Camakaris, J., Lee, B.T. and Rouch, D.A. (1995) Molecular genetics and transport analysis of the copper-resistance determinant (pco) from Escherichia coli plasmid pRJ1004. Mol. Microbiol. 17, 1153^1166. Bryson, J.W., O’Halloran, T.V., Rouch, D.A., Brown, N.L., Camakaris, J. and Lee, B.T.O. (1993) Chemical and genetic studies of copper resistance in E. coli. In: Bioinorganic Chemistry of Copper (Karlin, K.D. and Tylekar, Z., Eds.), pp. 101^109. Chapman and Hall, New York. Lee, S.M., Grass, G., Rensing, C., Barrett, S.R., Yates, C.J., Stoyanov, J.V. and Brown, N.L. (2002) The Pco proteins are involved in periplasmic copper handling in Escherichia coli. Biochem. Biophys. Res. Commun. 295, 616^620. Lee, B.T.O., Brown, N.L., Rogers, S., Bergemann, A., Camakaris, J. and Rouch, D.A. (1990) Bacterial response to copper in the environment: Escherichia coli as a model system. In: Metal Speciation in the Environment (Gucer, S. and Broeckaert, J.A.C., Eds.). Springer-Verlag, New York. Rouch, D.A. and Brown, N.L. (1997) Copper-inducible transcriptional regulation at two promoters in the Escherichia coli copper resistance determinant pco. Microbiology 141, 1191^1202. Williams, J.R., Morgan, A., Rouch, D.A., Brown, N.L. and Lee, B.T.O. (1993) Copper resistant enteric bacteria from United Kingdom and Australian piggeries. Appl. Environ. Microbiol. 59, 2531^ 2537. Rensing, C., Fan, B., Sharma, R., Mitra, B. and Rosen, B.P. (2000) CopA: An Escherichia coli Cu(I)-translocating P-type ATPase. Proc. Natl. Acad. Sci. USA 97, 652^656. Rensing, C. and Grass, G. (2003) Escherichia coli mechanisms of copper homeostasis in a changing environment. FEMS Microbiol. Rev. 27, 197^213. Rensing, C., Ghosh, M. and Rosen, B.P. (1999) Families of softmetal-ion-transporting ATPases. J. Bacteriol. 181, 5891^5897. Franke, S., Grass, G. and Nies, D.H. (2001) The product of the ybdE gene of the Escherichia coli chromosome is involved in detoxi¢cation of silver ions. Microbiology 147, 965^972. Munson, G.P., Lam, D.L., Outten, F.W. and O’Halloran, T.V. (2000) Identi¢cation of a copper-responsive two-component system on the chromosome of Escherichia coli K-12. J. Bacteriol. 182, 5864^ 5871. Silver, S. (2003) Bacterial silver resistance: Molecular biology and uses and misuses of silver compounds. FEMS Microbiol. Rev. 27, 341^354. Outten, F.W., Hu¡man, D.L., Hale, J.A. and O’Halloran, T.V. (2001) The independent cue and cus systems confer copper tolerance during aerobic and anaerobic growth in Escherichia coli. J. Biol. Chem. 276, 30670^30677. Grass, G. and Rensing, C. (2001) Genes involved in copper homeostasis in Escherichia coli. J. Bacteriol. 183, 2145^2147.
Cyaan Magenta Geel Zwart
334
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[95] Grass, G. and Rensing, C. (2001) CueO is a multi-copper oxidase that confers copper tolerance in Escherichia coli. Biochem. Biophys. Res. Commun. 286, 902^908. [96] Hu¡man, D.L., Huyett, J., Outten, F.W., Doan, P.E., Finney, L.A., Ho¡man, B.M. and O’Halloran, T.V. (2002) Spectroscopy of Cu(II)-PcoC and the multicopper oxidase function of PcoA, two essential components of Escherichia coli pco copper resistance operon. Biochemistry 41, 10046^10055. [97] Franke, S., Grass, G., Rensing, C. and Nies, D.H. (2003) Molecular analysis of the copper-transporting CusCFBA e¥ux system from Escherichia coli. J. Bacteriol. (in press). [98] Nikaido, H. (1998) Multiple antibiotic resistance and e¥ux. Curr. Opin. Microbiol. 1, 516^523. [99] Kohler, T., Pechere, J.C. and Plesiat, P. (1999) Bacterial antibiotic e¥ux systems of medical importance. Cell. Mol. Life Sci. 56, 771^ 778. [100] Ziha Zari¢, I., Llanes, C., Kohler, T., Pechere, J.C. and Plesiat, P. (1999) In vivo emergence of multidrug-resistant mutants of Pseudomonas aeruginosa overexpressing the active e¥ux system MexAMexB-OprM. Antimicrob. Agents Chemother. 43, 287^291. [101] Hassan, M.E.T., van der Lelie, D., Springael, D., Ro«mling, U., Ahmed, N. and Mergeay, M. (1999) Identi¢cation of a gene cluster, czr, involved in cadmium and zinc resistance in Pseudomonas aeruginosa. Gene 238, 417^425. [102] Srikumar, R., Kon, T., Gotoh, N. and Poole, K. (1998) Expression of Pseudomonas aeruginosa multidrug e¥ux pumps MexA-MexBOprM and MexC-MexD-OprJ in a multidrug-sensitive Escherichia coli strain. Antimicrob. Agents. Chemother. 42, 65^71. [103] Li, X.Z. and Poole, K. (1999) Organic solvent-tolerant mutants of Pseudomonas aeruginosa display multiple antibiotic resistance. Can. J. Microbiol. 45, 18^22. [104] Li, X.Z., Zhang, L. and Poole, K. (1998) Role of the multidrug e¥ux systems of Pseudomonas aeruginosa in organic solvent tolerance. J. Bacteriol. 180, 2987^2991. [105] Gotoh, N., Tsujimoto, H., Poole, K., Yamagishi, J.I. and Nishino, T. (1995) The outer membrane protein OprM of Pseudomonas aeruginosa is encoded by oprK of the mexA-mexB-oprK multidrug resistance operon. Antimicrob. Agents. Chemother. 39, 2567^ 2569. [106] Maseda, H., Yoneyama, H. and Nakae, T. (2000) Assignment of the substrate-selective subunits of the MexEF-OprN multidrug e¥ux pump of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 44, 658^664. [107] Chuanchuen, R., Narasaki, C.T. and Schweizer, H.P. (2002) The MexJK e¥ux pump of Pseudomonas aeruginosa requires OprM for antibiotic e¥ux but not for e¥ux of triclosan. J. Bacteriol. 184, 5036^5044. [108] Aendekerk, S., Ghysels, B., Cornelis, C. and Baysse, C. (2002) Characterization of a new e¥ux pump, MexGHI-OpmD, from Pseudomonas aeruginosa that confers resistance to vanadium. Microbiology 148, 2371^2381. [109] Morita, Y., Komori, Y., Mima, T., Kuroda, T., Mizushima, T. and Tsuchiya, T. (2001) Construction of a series of mutants lacking all of the four major mex operons for multidrug e¥ux pumps or possessing each one of the operons from Pseudomonas aeruginosa PAO1: MexCD-OprJ is an inducible pump. FEMS Microbiol. Lett. 202, 139^143. [110] Nishino, K. and Yamaguchi, A. (2001) Analysis of a complete library of putative drug transporter genes in Escherichia coli. J. Bacteriol. 183, 5803^5812. [111] Sulavik, M.C., Houseweart, C., Cramer, C., Jiwani, N., Murgolo, N., Greene, J., DiDomenico, B., Shaw, K.J., Miller, G.H., Hare, R. and Shimer, G. (2001) Antibiotic susceptibility pro¢les of Escherichia coli strains lacking multidrug e¥ux pump genes. Antimicrob. Agents Chemother. 45, 1126^1136. [112] White, D.G., Goldman, J.D., Demple, B. and Levy, S.B. (1997) Role of the acrAB locus inorganic solvent tolerance mediated by
FEMSRE 781 2-6-03
[113] [114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
[126]
[127]
[128]
[129]
[130]
[131]
expression of marA, soxS, or robA in Escherichia coli. J. Bacteriol. 1997, 6122^6126. Thanassi, D.G., Cheng, L.W. and Nikaido, H. (1997) Active e¥ux of bile salts by Escherichia coli. J. Bacteriol. 179, 2512^2518. Aono, R. (1998) Improvement of organic solvent tolerance level of Escherichia coli by overexpression of stress-responsive genes. Extremophiles 2, 239^248. Aono, R., Tsukagoshi, N. and Yamamoto, M. (1998) Involvement of outer membrane protein TolC, a possible member of the mar-sox regulon, in maintenance and improvement of organic solvent tolerance of Escherichia coli K-12. J. Bacteriol. 180, 938^944. Rand, J.D., Danby, S.G., Greenway, D.L.A. and England, R.R. (2002) Increased expression of the multidrug e¥ux genes acrAB occurs during slow growth of Escherichia coli. FEMS Microbiol. Lett. 207, 91^95. Rosenberg, E.Y., Ma, D. and Nikaido, H. (2000) AcrD of Escherichia coli is an aminoglycoside e¥ux pump. J. Bacteriol. 182, 1754^ 1756. Sa¤nchez, L., Pan, W., Vin‹as, M. and Nikaido, H. (1998) The acrAB homolog of Haemophilus in£uenzae codes for a functional multidrug e¥ux pump. J. Bacteriol. 179, 6855^6857. Hagman, K.E., Lucas, C.E., Balthazar, J.T., Snyder, L., Nilles, M., Judd, R.C. and Shafer, W.M. (1997) The MtrD protein of Neisseria gonorrhoeae is a member of the resistance/nodulation/division protein family constituting part of an e¥ux system. Microbiology 7, 2117^2125. Veal, W.L., Nicholas, R.A. and Shafer, W.M. (2002) Overexpression of the MtrC-MtrD-MtrE e¥ux pump due to an mtrR mutation is required for chromosomally mediated penicillin resistance in Neisseria gonorrhoeae. J. Bacteriol. 184, 5619^5624. Ramos, J.L., Duque, E., Godoy, P. and Segura, A. (1998) E¥ux pumps involved in toluene tolerance in Pseudomonas putida DOTT1E. J. Bacteriol. 180, 3323^3329. Mosqueda, G. and Ramos, J.L. (2000) A set of genes encoding a second toluene e¥ux system in Pseudomonas. J. Bacteriol. 182, 937^ 943. Lin, J., Michel, L.O. and Zhang, Q.J. (2002) CmeABC functions as a multidrug e¥ux system in Campylobacter jejuni. Antimicrob. Agents Chemother. 46, 2124^2131. Pumbwe, L. and Piddock, L.J.V. (2002) Identi¢cation and molecular characterisation of CmeB, a Campylobacter jejuni multidrug e¥ux pump. FEMS Microbiol. Lett. 206, 185^189. Magnet, S., Courvalin, P. and Lambert, T. (2001) Resistance-nodulation-cell division-type e¥ux pump involved in aminoglycoside resistance in Acinetobacter baumannii strain BM4454. Antimicrob. Agents Chemother. 45, 3375^3380. Alonso, A. and Martinez, J.L. (2000) Cloning and characterization of SmeDEF, a novel multidrug e¥ux pump from Stenotrophomonas maltophilia. Antimicrob. Agents Chemother. 44, 3079^3086. Palumbo, J.D., Kado, C.I. and Phillips, D.A. (1998) An iso£avonoid-inducible e¥ux pump in Agrobacterium tumefaciens is involved in competitive colonization of roots. J. Bacteriol. 180, 3107^ 3113. Peng, W.T. and Nester, E.W. (2001) Characterization of a putative RND-type e¥ux system in Agrobacterium tumefaciens. Gene 270, 245^252. Braibant, M., Guilloteau, L. and Zygmunt, M.S. (2002) Functional characterization of Brucella melitensis NorMI, an e¥ux pump belonging to the multidrug and toxic compound extrusion family. Antimicrob. Agents Chemother. 46, 3050^3053. Ikeda, T. and Yoshimura, F. (2002) A resistance-nodulation-cell division family xenobiotic e¥ux pump in an obligate anaerobe, Porphyromonas gingivalis. Antimicrob. Agents Chemother. 46, 3257^ 3260. Kumar, A. and Worobec, E.A. (2002) Fluoroquinolone resistance of Serratia marcescens : involvement of a proton gradient-dependent e¥ux pump. J. Antimicrob. Chemother. 50, 593^596.
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 [132] Nikaido, H., Basina, M., Nguyen, V. and Rosenberg, E.Y. (1998) Multidrug e¥ux pump AcrAB of Salmonella typhimurium excretes only those L-lactam antibiotics containing lipophilic side chains. J. Bacteriol. 180, 4682^4692. [133] Gotoh, N., Tsujimoto, H., Tsuda, M., Okamoto, K., Nomura, A., Wada, T., Nakahashi, M. and Nishino, T. (1998) Characterization of the MexC-MexD-OprJ multidrug e¥ux system in delta mexAmexB-oprM mutants of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 42, 1938^1943. [134] Lomovskaya, O., Lee, A., Hoshino, K., Ishida, H., Mistry, A., Warren, M.S., Boyer, E., Chamberland, S. and Lee, V.J. (1999) Use of a genetic approach to evaluate the consequences of inhibition of e¥ux pumps in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 43, 1340^1346. [135] Li, X.Z., Zhang, L., Srikumar, R. and Poole, K. (1998) beta-lactamase inhibitors are substrates for the multidrug e¥ux pumps of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 42, 399^ 403. [136] Oethinger, M., Kern, W.V., Jellen-Ritter, A.S., McMurry, L.M. and Levy, S.B. (2000) Ine¡ectiveness of topoisomerase mutations in mediating clinically signi¢cant £uoroquinolone resistance in Escherichia coli in the absence of the AcrAB e¥ux pump. Antimicrob. Agents Chemother. 44, 10^13. [137] Potrykus, J., Baranska, S. and Wegrzyn, G. (2002) Inactivation of the acrA gene is partially responsible for chloramphenicol sensitivity of Escherichia coli CM2555 strain expressing the chloramphenicol acetyltransferase gene. Microb. Drug Resist.-Mech. Epidemiol. Dis. 8, 179^185. [138] Ramos, J.L., Duque, E., Gallegos, M.T., Godoy, P., Ramos-Gonzalez, M.I., Rojas, A., Teran, W. and Segura, A. (2002) Mechanisms of solvent tolerance in gram-negative bacteria. Annu. Rev. Microbiol. 56, 743^768. [139] Kawamura-Sato, K., Shibayama, K., Horii, T., Iimuma, Y., Arakawa, Y. and Ohta, M. (1999) Role of multiple e¥ux pumps in Escherichia coli in indole expulsion. FEMS Microbiol. Lett. 179, 345^352. [140] Evans, K. and Poole, K. (1999) The MexA-MexB-OprM multidrug e¥ux system of Pseudomonas aeruginosa is growth-phase regulated. FEMS Microbiol. Lett. 173, 35^39. [141] Pearson, J.P., Van Delden, C. and Iglewski, B.H. (1999) Active e¥ux and di¡usion are involved in transport of Pseudomonas aeruginosa cell-to-cell signals. J. Bacteriol. 181, 1203^1210. [142] Guan, L., Ehrmann, M., Yoneyama, H. and Nakae, T. (1999) Membrane topology of the xenobiotic-exporting subunit, MexB, of the MexA, B-OprM extrusion pump in Pseudomonas aeruginosa. J. Biol. Chem. 274, 10517^10522. [143] Gotoh, N., Kusumi, T., Tsujimoto, H., Wada, T. and Nishino, T. (1999) Topological analysis of an RND family transporter, MexD of Pseudomonas aeruginosa. FEBS Lett. 458, 32^36. [144] Saier, M.H.J. (1994) Computer-aided analyses of transport protein sequences: gleaning evidence concerning function, structure, biogenesis, and evolution. Microbiol. Rev. 58, 71^93. [145] Gupta, A., Matsui, K., Lo, J.F. and Silver, S. (1999) Molecular basis for resistance to silver in Salmonella. Nat. Med. 5, 183^188. [146] Aires, J.R., Pechere, J.C., Van Delden, C. and Kohler, T. (2002) Amino acid residues essential for function of the MexF e¥ux pump protein of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 46, 2169^2173. [147] Kaback, H.R. (1988) Site-directed mutagenesis and ion-gradient driven active transport: on the path of the proton. Annu. Rev. Physiol. 50, 243^256. [148] Kaback, H.R., Voss, J. and Wu, J. (1997) Helix packing in polytopic membrane proteins: the lactose permease of Escherichia coli. Curr. Opin. Struct. Biol. 7, 537^542. [149] Varela, M.F. and Wilson, T.H. (1996) Molecular biology of the lactose carrier of Escherichia coli. Biochim. Biophys. Acta 276, 21^34.
FEMSRE 781 2-6-03
335
[150] Richter, H.-T., Brown, L.S., Needleman, R. and Lanyi, J.K. (1996) A linkage of the pKa’s of asp-85 and glu-204 forms part of the reprotonation switch of bacteriorhodopsin. Biochemistry 35, 4054^ 4062. [151] Murakami, S., Nakashima, R., Yamashita, R. and Yamaguchi, A. (2002) Crystal structure of bacterial multidrug e¥ux transporter AcrB. Nature 419, 587^593. [152] Pos, K.M. and Diederichs, K. (2002) Puri¢cation, crystallization and preliminary di¡raction studies of AcrB, the inner-membrane multi-drug e¥ux protein. Acta Crystallogr. D 58, 1865^1867. [153] Pribyl, T. (2001) Topologie des CzcCBA-E¥ux-Komplexes aus Ralstonia metallidurans CH34. Thesis, Martin-Luther-Universita«t, Halle-Wittenberg. [154] Lacroix, F.J.C., Cloeckaert, A., Grepinet, O., Pinault, C., Popo¡, M.Y. and Waxin, H. (1996) Salmonella typhimurium AcrB-like gene ^ Identi¢cation and role in resistance to biliary salts and detergents and in murine infection. FEMS Microbiol. Lett. 135, 161^167. [155] Ma, D., Cook, D.N., Alberti, M., Pon, N.G., Kikaido, H. and Hearst, J.E. (1993) Molecular cloning and characterization of acrA and acrE genes of Escherichia coli. J. Bacteriol. 175, 6299^ 6313. [156] Srikumar, R., Li, X.Z. and Poole, K. (1997) Inner membrane e¥ux components are responsible for beta-lactam speci¢city of multidrug e¥ux pumps in Pseudomonas aeruginosa. J. Bacteriol. 179, 7875^ 7881. [157] Yoneyama, H., Ocaktan, A., Gotoh, N., Nishino, T. and Nakae, T. (1998) Subunit swapping in the Mex-extrusion pumps in Pseudomonas aeruginosa. Biochem. Biophys. Res. Commun. 244, 898^902. [158] Murata, T., Kuwagaki, M., Shin, T., Gotoh, N. and Nishino, T. (2002) The substrate speci¢city of tripartite e¥ux systems of Pseudomonas aeruginosa is determined by the RND component. Biochem. Biophys. Res. Commun. 299, 247^251. [159] Cavallo, J.D., Plesiat, P., Couetdic, G. and Leblanc, F. (1997) J. Antimicrob. Chemother. 50, 1039^1043. [160] Koronakis, V., Shar¡, A., Koronakis, E., Luisi, B. and Hughes, C. (2000) Crystal structure of the bacterial membrane protein TolC central to multidrug e¥ux and protein export. Nature 405, 914^ 919. [161] Wong, K.K.Y. and Hancock, R.E.W. (2000) Insertion mutagenesis and membrane topology model of the Pseudomonas aeruginosa outer membrane protein OprM. J. Bacteriol. 182, 2402^2410. [162] Andersen, C., Koronakis, E., Hughes, C. and Koronakis, V. (2002) An aspartate ring at the TolC tunnel entrance determines ion selectivity and presents a target for blocking by large cations. Mol. Microbiol. 44, 1131^1139. [163] Yoshihara, E., Maseda, H. and Saito, K. (2002) The outer membrane component of the multidrug e¥ux pump from Pseudomonas aeruginosa may be a gated channel. Eur. J. Biochem. 269, 4738^ 4745. [164] Kawabe, T., Fujihira, E. and Yamaguchi, A. (2000) Molecular construction of a multidrug exporter system, AcrAB: Molecular interaction between AcrA and AcrB, and cleavage of the N-terminal signal sequence of AcrA. J. Biochem. 128, 195^200. [165] Avila-Sakar, A.J., Misaghi, S., Wilson-Kubalek, E.M., Downing, K.H., Zgurskaya, H., Nikaido, H. and Nogales, E. (2001) Lipidlayer crystallization and preliminary three-dimensional structural analysis of AcrA, the periplasmic component of a bacterial multidrug e¥ux pump. J. Struct. Biol. 136, 81^88. [166] Elkins, C.A. and Nikaido, H. (2002) Substrate speci¢city of the RND-type multidrug e¥ux pumps AcrB and AcrD of Escherichia coli is determined predominately by two large periplasmic loops. J. Bacteriol. 184, 6490^6498. [167] Tikhonova, E.B., Wang, Q.J. and Zgurskaya, H.I. (2002) Chimeric analysis of the multicomponent multidrug e¥ux transporters from gram-negative bacteria. J. Bacteriol. 184, 6499^6507. [168] Mao, W.M., Warren, M.S., Black, D.S., Satou, T., Murata, T., Nishino, T., Gotoh, N. and Lomovskaya, O. (2002) On the mech-
Cyaan Magenta Geel Zwart
336
[169]
[170]
[171]
[172]
[173]
[174]
[175]
[176]
[177]
[178]
[179]
[180]
[181]
[182]
[183]
[184]
[185]
[186]
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 anism of substrate speci¢city by resistance nodulation division (RND)-type multidrug resistance pumps: the large periplasmic loops of MexD from Pseudomonas aeruginosa are involved in substrate recognition. Mol. Microbiol. 46, 889^901. Maseda, H., Kitao, M., Eda, S., Yoshihara, E. and Nakae, T. (2002) A novel assembly process of the multicomponent xenobiotic e¥ux pump in Pseudomonas aeruginosa. Mol. Microbiol. 46, 677^686. Peterson, J.D., Umayam, L.A., Dickinson, T., Hickey, E.K. and White, O. (2001) The comprehensive microbial resource. Nucleic Acids Res. 29, 123^125. Rensing, C., Pribyl, T. and Nies, D.H. (1997) New functions for the three subunits of the CzcCBA cation-proton-antiporter. J. Bacteriol. 179, 6871^6879. Lopez-Maury, L., Garcia-Dominguez, M., Florencio, F.J. and Reyes, J.C. (2002) A two-component signal transduction system involved in nickel sensing in the cyanobacterium Synechocystis sp. PCC 6803. Mol. Microbiol. 43, 247^256. Grass, G., Fan, B., Rosen, B.P., Lemke, K., Schlegel, H.G. and Rensing, C. (2001) NreB from Achromobacter xylosoxidans 31A is a nickel-induced transporter conferring nickel resistance. J. Bacteriol. 183, 2803^2807. Waidner, B., Melchers, K., Ivanov, I., Loferer, H., Bensch, K.W., Kist, M. and Bereswill, S. (2002) Identi¢cation by RNA pro¢ling and mutational analysis of the novel copper resistance determinants CrdA (HP1326), CrdB (HP1327), and CzcB (HP1328) in Helicobacter pylori. J. Bacteriol. 184, 6700^6708. Paulsen, I.T. and Saier Jr., M.H. (1997) A novel family of ubiquitous heavy metal ion transport proteins. J. Membr. Biol. 156, 99^ 103. Gu¡anti, A.A., Wei, Y., Rood, S.V. and Krulwich, T.A. (2002) An antiport mechanism for a member of the cation di¡usion facilitator family: divalent cations e¥ux in exchange for Kþ and Hþ . Mol. Microbiol. 45, 145^153. Wang, W., Gu¡anti, A.A., Wei, Y., Ito, M. and Krulwich, T.A. (2000) Two types of Bacillus subtilis tetA(L) deletion strains reveal the physiological importance of TetA(L) in Kþ acquisition as well as in Naþ , alkali, and tetracycline resistance. J. Bacteriol. 182, 2088^ 2095. Bloss, T., Clemens, S. and Nies, D.H. (2002) Characterisation of the ZAT1p zinc transporter from Arabidopsis thaliana in microbial model organisms and reconstituted proteoliposomes. Planta 214, 783^791. Anton, A., GroMe, C., ReiMman, J., Pribyl, T. and Nies, D.H. (1999) CzcD is a heavy metal ion transporter involved in regulation of heavy metal resistance in Ralstonia sp. strain CH34. J. Bacteriol. 181, 6876^6881. Sturr, M.G., Ablooglu, A.J. and Krulwich, T.A. (1997) A Bacillus subtilis locus encoding several gene products a¡ecting transport of cation. Gene 188, 91^94. Xiong, A.M. and Jayaswal, R.K. (1998) Molecular characterization of a chromosomal determinant conferring resistance to zinc and cobalt ions in Staphylococcus aureus. J. Bacteriol. 180, 4024^4029. Kuroda, M., Hayashi, H. and Ohta, T. (1999) Chromosome-determined zinc-responsible operon czr in Staphylococcus aureus strain 912. Microbiol. Immunol. 43, 115^125. Spada, S., Pembroke, J.T. and Wall, J.G. (2002) Isolation of a novel Thermus thermophilus metal e¥ux protein that improves Escherichia coli growth under stress conditions. Extremophiles 6, 301^308. Grass, G., Fan, B., Rosen, B.P., Franke, S., Nies, D.H. and Rensing, C. (2001) ZitB (YbgR), a member of the cation di¡usion facilitator family, is an additional zinc transporter in Escherichia coli. J. Bacteriol. 183, 4664^4667. Lee, S.M., Grass, G., Haney, C.J., Fan, B., Rosen, B.P., Anton, A., Nies, D.H. and Rensing, C. (2002) Functional analysis of the Escherichia coli zinc transporter ZitB. FEMS Microbiol. Lett. 215, 273^278. Gru«nberg, K., Wawer, C., Tebo, B.M. and Schu«ler, D. (2001) A
FEMSRE 781 2-6-03
[187]
[188]
[189]
[190]
[191]
[192]
[193]
[194]
[195]
[196]
[197]
[198]
[199]
[200]
[201]
[202]
[203]
[204]
large gene cluster encoding several magnetosome proteins is conserved in di¡erent species of magnetotactic bacteria. Appl. Environ. Microbiol. 67, 4573^4582. Buseck, P.R., Dunin-Borkowski, R.E., Devouard, B., Frankel, R.B., McCartney, M.R., Midgley, P.A., Po¤sfai, M. and Weyland, M. (2001) Magnetite morphology and life on Mars. Proc. Natl. Acad. Sci. USA 98, 13490^13495. Thomas-Keprta, K.L., Clemett, S.J., Bazylinski, D.A., Kirschvink, J.L., McKay, D.S., Wentworth, S.J., Vali, H., Gibson Jr., J.E.K. and Romanek, C.S. (2002) Magnetofossils from ancient Mars: a robust biosignature in the Martian meteorite ALH84001. Appl. Environ. Microbiol. 68, 3663^3672. Luhmann, J.G. and Russell, C.T. (1997) Mars: magnetic ¢eld and magnetosphere. In: Encyclopedia of Planetary Sciences (Shirley, J.H. and Fainbridge, R.W., Eds.), pp. 454^456. Chapman and Hall, New York. Paulsen, I.T., Sliwinski, M.K., Nelissen, B., Go¡eau, A. and Saier, M.H. (1998) Uni¢ed inventory of established and putative transporters encoded within the complete genome of Saccharomyces cerevisiae. FEBS Lett. 430, 116^125. Kamizomo, A., Nishizawa, M., Teranishi, A., Murata, K. and Kimura, A. (1989) Identi¢cation of a gene conferring resistance to zinc and cadmium ions in the yeast Saccharomyces cerevisiae. Mol. Gen. Genet. 219, 161^167. Conklin, D.S., McMaster, J.A., Culbertson, M.R. and Kung, C. (1992) COT1, a gene involved in cobalt accumulation in Saccharomyces cerevisiae. Mol. Cell. Biol. 12, 3678^3688. Conklin, D.S., Culbertson, M.R. and Kung, C. (1994) Interactions between gene products involved in divalent cation transport in Sacchromyces cerevisiae. Mol. Gen. Genet. 244, 303^311. MacDiarmid, C.W., Gaither, L.A. and Eide, D. (2000) Zinc transporters that regulate vacuolar zinc storage in Saccharomyces cerevisiae. EMBO J. 19, 2845^2855. Miyabe, S., Izawa, S. and Inoue, Y. (2001) The Zrc1 is involved in zinc transport system between vacuole and cytosol in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 282, 79^83. MacDiarmid, C.W., Milanick, M.A. and Eide, D.J. (2002) Biochemical properties of vacuolar zinc transport systems of Saccharomyces cerevisiae. J. Biol. Chem. 277, 39187^39194. Miyabe, S., Izawa, S. and Inoue, Y. (2000) Expression of ZRC1 coding for suppressor of zinc toxicity is induced by zinc-starvation stress in Zap1-dependent fashion in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 276, 879^884. MacDiarmid, C.W. and Gardner, R.C. (1998) Overexpression of the Saccharomyces cerevisiae magnesium transport system confers resistance to aluminum ion. J. Biol. Chem. 273, 1727^1732. Kobayashi, S., Miyabe, S., Izawa, S., Inoue, Y. and Kimura, A. (1996) Correlation of the Osr/Zrc1 gene product and the intracellular glutathione levels in Saccharomyces cerevisiae. Biotechnol. Appl. Biochem. 23, 3^6. Li, L. and Kaplan, J. (2001) The yeast gene MSC2, a member of the cation di¡usion facilitator family, a¡ects the cellular distribution of zinc. J. Biol. Chem. 276, 5036^5043. Li, L. and Kaplan, J. (1997) Characterization of two homologous yeast genes that encode mitochondrial iron transporters. J. Biol. Chem. 272, 28485^28493. Clemens, S., Bloss, T., Vess, C., Neumann, D., Nies, D.H. and zur Nieden, U. (2002) A transporter in the endoplasmic reticulum/nuclear envelope of Schizosaccharomyces pombe cells di¡erentially a¡ects transition metal tolerance. J. Biol. Chem. 277, 18215^ 18221. Palmiter, R.D. and Findley, S.D. (1995) Cloning and functional characterization of a mammalian zinc transporter that confers resistance to zinc. EMBO J. 14, 639^649. Palmiter, R.D., Cole, T.B. and Findley, S.D. (1996) ZnT-2, a mammalian protein that confers resistance to zinc by facilitating vesicular sequestration. EMBO J. 15, 1784^1791.
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 [205] Palmiter, R.D., Cole, T.B., Quaife, C.J. and Findley, S.D. (1996) ZnT3, a putative transporter of zinc into synaptic vesicles. Proc. Natl. Acad. Sci. USA 93, 14934^14939. [206] Wenzel, H.J., Cole, T.B., Born, D.E., Schwartzkroin, P.A. and Palmiter, R.D. (1997) Ultrastructral localization of zinc transporter-3 (ZnT-3) to synaptic vesicle membranes within mossy ¢ber boutons in the hippocampus of mouse and monkey. Proc. Natl. Acad. Sci. USA 94, 12676^12681. [207] Saito, T., Takahashi, K., Nakagawa, N., Hosokawa, T., Kurasaki, M., Yamanoshita, O., Yamamoto, Y., Sasaki, H., Nagashima, K. and Fujita, H. (2000) De¢ciencies of hippocampal Zn and ZnT3 accelerate brain aging of rat. Biochem. Biophys. Res. Commun. 279, 505^511. [208] Michalczyk, A.A., Allen, J., Blomeley, R.C. and Ackland, M.L. (2002) Constitutive expression of hZnT4 zinc transporter in human breast epithelial cells. Biochem. J. 364, 105^113. [209] Kambe, T., Narita, H., Yamaguchi-Iwai, Y., Hirose, J., Amano, T., Sugiura, N., Sasaki, R., Mori, K., Iwanaga, T. and Nagao, M. (2002) Cloning and characterization of a novel mammalian zinc transporter, zinc transporter 5, abundantly expressed in pancreatic beta cells. J. Biol. Chem. 277, 19049^19055. [210] Inoue, K., Matsuda, K., Itoh, M., Kawaguchi, H., Tomoike, H., Aoyagi, T., Nagai, R., Hori, M., Nakamura, Y. and Tanaka, T. (2002) Osteopenia and male-speci¢c sudden cardiac death in mice lacking a zinc transporter gene, Znt5. Hum. Mol. Genet. 11, 1775^ 1784. [211] Murgia, C., Vespignani, I., Cerase, J., Nobili, F. and Perozzi, G. (1999) Cloning, expression, and vesicular localization of zinc transporter Dri 27/ZnT4 in intestinal tissue and cells. Am. J. Physiol. Gastrointest. Liver Physiol. 277, G1231^G1239. [212] Cragg, R.A., Christie, G.R., Phillips, S.R., Russi, R.M., Kury, S., Mathers, J.C., Taylor, P.M. and Ford, D. (2002) A novel zinc-regulated human zinc transporter, hZTL1, is localized to the enterocyte apical membrane. J. Biol. Chem. 277, 22789^22797. [213] Huang, L.P., Kirschke, C.P. and Gitschier, J. (2002) Functional characterization of a novel mammalian zinc transporter, ZnT6. J. Biol. Chem. 277, 26389^26395. [214] van der Zaal, B.J., Neuteboom, L.W., Pinas, J.E., Chardonnens, A.N., Schat, H., Verkleij, J.A.C. and Hooykaas, P.J.J. (1999) Overexpression of a novel Arabidopsis gene related to putative zinc-transporter genes from animals can lead to enhanced zinc resistance and accumulation. Plant Physiol. 119, 1047^1055. [215] Ma«ser, P., Thomine, S., Schroeder, J.I., Ward, J.M., Hirschi, K., Sze, H., Talke, I.N., Amtmann, A., Maathuis, F.J.M., Sanders, D., Harper, J.F., Tchieu, J., Gribskov, M., Persans, M.W., Salt, D.E., Kim, S.A. and Guerinot, M.L. (2001) Phylogenetic relationships within cation transporter families of Arabidopsis. Plant Physiol. 126, 1646^1667. [216] Assuncao, A.G.L., Martins, P.D., De Folter, S., Vooijs, R., Schat, H. and Aarts, M.G.M. (2001) Elevated expression of metal transporter genes in three accessions of the metal hyperaccumulator Thlaspi caerulescens. Plant Cell Environ. 24, 217^226. [217] Persans, M.W., Nieman, K. and Salt, D.E. (2001) Functional activity and role of cation-e¥ux family members in Ni hyperaccumulation in Thlaspi goesingense. Proc. Natl. Acad. Sci. USA 98, 9995^ 10000. [218] Fagan, M.J. and Saier Jr., M.H. (1994) P-type ATPases of eukaryotes and bacteria : sequence comparisons and construction of phylogenetic trees. J. Mol. Evol. 38, 57^99. [219] Snavely, M.D., Florer, J.B., Miller, C.G. and Maguire, M.E. (1989) Magnesium transport in Salmonella typhimurium: 28 Mg2þ transport by CorA, MgtA, and MgtB systems. J. Bacteriol. 171, 4761^ 4766. [220] Fan, B. and Rosen, B.P. (2002) Biochemical characterization of CopA, the Escherichia coli Cu(I)-translocating P-type ATPase. J. Biol. Chem. 277, 46987^46992. [221] Mandal, A.K., Cheung, W.D. and Arguello, J.M. (2002) Character-
FEMSRE 781 2-6-03
[222]
[223]
[224]
[225]
[226]
[227]
[228]
[229]
[230]
[231]
[232]
[233]
[234]
[235]
[236]
[237]
[238] [239]
337
ization of a thermophilic P-type Agþ /Cuþ -ATPase from the extremophile Archaeoglobus fulgidus. J. Biol. Chem. 277, 7201^7208. Tottey, S., Rich, P.R., Rondet, S.A.M. and Robinson, N.J. (2001) Two Menkes-type ATPases supply copper for photosynthesis in Synechocystis PCC 6803. J. Biol. Chem. 276, 19999^20004. Banci, L., Bertini, L., Cio¢-Ba¡oni, S., D’Onofrio, M., Gonnelli, L., Marhuenda-Egea, F.C. and Ruiz-Duenas, F.J. (2002) Solution structure of the N-terminal domain of a potential copper-translocating P-type ATPase from Bacillus subtilis in the apo and Cu(I) loaded states. J. Mol. Biol. 317, 415^429. Hasman, H. and Aarestrup, F.M. (2002) tcrB, a gene conferring transferable copper resistance in Enterococcus faecium : occurrence, transferability, and linkage to macrolide and glycopeptide resistance. Antimicrob. Agents Chemother. 46, 1410^1416. Bissig, K.D., Wunderli-Ye, H., Duda, P.W. and Solioz, M. (2001) Structure-function analysis of puri¢ed Enterococcus hirae CopB copper ATPase : e¡ect of Menkes/Wilson disease mutation homologues. Biochem. J. 357, 217^223. Bissig, K.D., Voegelin, T.C. and Solioz, M. (2001) Tetrathiomolybdate inhibition of the Enterococcus hirae CopB copper ATPase. FEBS Lett. 507, 367^370. Wunderli-Ye, H. and Solioz, M. (2001) Puri¢cation and functional analysis of the copper ATPase CopA of Enterococcus hirae. Biochem. Biophys. Res. Commun. 280, 713^719. Vats, N. and Lee, S.F. (2001) Characterization of a copper-transport operon, copYAZ, from Streptococcus mutans. Microbiology 147, 653^662. Stentz, R., Loizel, C., Malleret, C. and Zagorec, M. (2000) Development of genetic tools for Lactobacillus sakei: Disruption of the betagalactosidase gene and use of lacZ as a reporter gene to study regulation of the putative copper ATPase, AtkB. Appl. Environ. Microbiol. 66, 4272^4278. Reeve, W.G., Tiwari, R.P., Kale, N.B., Dilworth, M.J. and Glenn, A.R. (2002) ActP controls copper homeostasis in Rhizobium leguminosarum bv. viciae and Sinorhizobium meliloti preventing low pHinduced copper toxicity. Mol. Microbiol. 43, 981^991. Adaikkalam, V. and Swarup, S. (2002) Molecular characterization of an operon, cueAR, encoding a putative P1-type ATPase and a MerR-type regulatory protein involved in copper homeostasis in Pseudomonas putida. Microbiology (UK) 148, 2857^2867. Burlando, B., Evangelisti, V., Dondero, F., Pons, G., Camakaris, J. and Viarengo, A. (2002) Occurrence of Cu-ATPase in Dictyostelium : possible role in resistance to copper. Biochem. Biophys. Res. Commun. 291, 476^483. LaGier, M.J., Zhu, G. and Keithly, J.S. (2001) Characterization of a heavy metal ATPase from the apicomplexan Cryptosporidium parvum. Gene 266, 25^34. Riggle, P.J. and Kumamoto, C.A. (2000) Role of a Candida albicans P1-type ATPase in resistance to copper and silver ion toxicity. J. Bacteriol. 182, 4899^4905. Axelsen, K.B. and Palmgren, M.G. (2001) Inventory of the superfamily of P-type ion pumps in Arabidopsis. Plant Physiol. 126, 696^ 706. Hirayama, T., Kieber, J.J., Hirayama, N., Kogan, M., Guzman, P., Nourizadeh, S., Alonso, J.M., Dailey, W.P., Dancis, A. and Ecker, J.R. (1999) Responsive-to-antagonist1, a Menkes/Wilson disease-related copper transporter, is required for ethylene signaling in Arabidopsis. Cell 97, 383^393. Larin, D., Mekios, C., Das, K., Ross, B., Yang, A.S. and Gilliam, T.C. (1999) Characterization of the interaction between the Wilson and Menkes disease proteins and the cytoplasmic copper chaperone, HAH1p. J. Biol. Chem. 274, 28497^28504. Butler, P., McIntyre, N. and Mistry, P.K. (2001) Molecular diagnosis of Wilson disease. Mol. Genet. Metab. 72, 223^230. Udo, E.E., Jacob, L.E. and Mathew, B. (2000) A cadmium resistance plasmid, pXU5, in Staphylococcus aureus, strain ATCC25923. FEMS Microbiol. Lett. 189, 79^80.
Cyaan Magenta Geel Zwart
338
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[240] Solovieva, I.M. and Entian, K.D. (2002) Investigation of the yvgW Bacillus subtilis chromosomal gene involved in Cd2þ ion resistance. FEMS Microbiol. Lett. 208, 105^109. [241] Bal, N., Mintz, E., Guillain, F. and Catty, P. (2001) A possible regulatory role for the metal-binding domain of CadA, the Listeria monocytogenes Cd2þ -ATPase. FEBS Lett. 506, 249^252. [242] Lee, S.W., Glickmann, E. and Cooksey, D.A. (2001) Chromosomal locus for cadmium resistance in Pseudomonas putida consisting of a cadmium-transporting ATPase and a MerR family response regulator. Appl. Environ. Microbiol. 67, 1437^1444. [243] Alonso, A., Sanchez, P. and Martinez, J.L. (2000) Stenotrophomonas maltophilia D457R contains a cluster of genes from gram-positive bacteria involved in antibiotic and heavy metal resistance. Antimicrob. Agents Chemother. 44, 1778^1782. [244] Beard, S.J., Hashim, R., MembrilloHernandez, J., Hughes, M.N. and Poole, R.K. (1997) Zinc(II) tolerance in Escherichia coli K12: evidence that the zntA gene (o732) encodes a cation transport ATPase. Mol. Microbiol. 25, 883^891. [245] Rensing, C., Mitra, B. and Rosen, B.P. (1997) The zntA gene of Escherichia coli encodes a Zn(II)-translocating P-type ATPase. Proc. Natl. Acad. Sci. USA 94, 14326^14331. [246] Rensing, C., Sun, Y., Mitra, B. and Rosen, B.P. (1998) Pb(II)-translocating P-type ATPases. J. Biol. Chem. 273, 32614^32617. [247] Sharma, R., Rensing, C., Rosen, B.P. and Mitra, B. (2000) The ATP hydrolytic activity of puri¢ed ZntA, a Pb(II)/Cd(II)/Zn(II)-translocating ATPase from Escherichia coli. J. Biol. Chem. 275, 3873^3878. [248] Mitra, B. and Sharma, R. (2001) The cysteine-rich amino-terminal domain of ZntA, a Pb(II)/Zn(II)/Cd(II)-translocating ATPase from Escherichia coli, is not essential for its function. Biochemistry 40, 7694^7699. [249] Shiraishi, E., Inouhe, M., Joho, M. and Tohoyama, H. (2000) The cadmium-resistant gene, CAD2, which is a mutated putative coppertransporter gene (PCA1), controls the intracellular cadmium-level in the yeast S. cerevisiae. Curr. Genet. 37, 79^86. [250] Tong, L., Nakashima, S., Shibasaka, M., Katsuhara, M. and Kasamo, K. (2002) A novel histidine-rich CPx-ATPase from the ¢lamentous cyanobacterium Oscillatoria brevis related to multiple-heavymetal cotolerance. J. Bacteriol. 184, 5027^5035. [251] Herrmann, L., Schwan, D., Garner, R., Mobley, H.L.T., Haas, R., Scha«fer, K.P. and Melchers, K. (1999) Helicobacter pylori cadA encodes an essential Cd(II)-Zn(II)-Co(II) resistance factor in£uencing urease activity. Mol. Microbiol. 33, 524^536. [252] Rutherford, J.C., Cavet, J.S. and Robinson, N.J. (1999) Cobalt-dependent transcriptional switching by a dual-e¡ector MerR-like protein regulates a cobalt-exporting variant CPx-type ATPase. J. Biol. Chem. 274, 25827^25832. [253] Amoroso, M.J., Schubert, D., Mitscherlich, P., Schumann, P. and Kothe, E. (2000) Evidence for high a⁄nity nickel transporter genes in heavy metal resistant Streptomyces spec. J. Basic Microbiol. 40, 295^301. [254] Degen, O. and Eitinger, T. (2002) Substrate speci¢city of nickel/ cobalt permeases: Insights from mutants altered in transmembrane domains I and II. J. Bacteriol. 184, 3569^3577. [255] Eitinger, T. and Friedrich, B. (1994) A topological model for the high-a⁄nity nickel transporter of Alcaligenes eutrophus. Mol. Microbiol. 12, 1025^1032. [256] Eitinger, T., Wolfram, L., Degen, O. and Anthon, C. (1997) A Ni2þ binding motif is the basis of high a⁄nity transport of the Alcaligenes eutrophus nickel permease. J. Biol. Chem. 272, 17139^17144. [257] Odermatt, A., Suter, H., Krapf, R. and Solioz, M. (1992) An ATPase operon involved in copper resistance by Enterococcus hirae. Ann. NY Acad. Sci. 671, 484^486. [258] Odermatt, A., Suter, H., Krapf, R. and Solioz, M. (1993) Primary structure of two P-type ATPases involved in copper homeostasis in Enterococcus hirae. J. Biol. Chem. 268, 12775^12779. [259] Odermatt, A., Krapf, R. and Solioz, M. (1994) Induction of the putative copper ATPases, CopA and CopB, of Enterococcus hirae
FEMSRE 781 2-6-03
[260]
[261]
[262]
[263]
[264]
[265]
[266] [267]
[268]
[269]
[270]
[271]
[272]
[273]
[274]
[275]
[276]
[277]
[278]
[279]
by Agþ and Cu2þ , and Agþ extrusion by CopB. Biochem. Biophys. Res. Commun. 202, 44^48. Odermatt, A. and Solioz, M. (1995) Two trans-acting metalloregulatory proteins controlling expression of the copper-ATPases of Enterococcus hirae. J. Biol. Chem. 270, 4349^4354. Lu, Z.H., Dameron, C.T. and Solioz, M. (2003) The Enterococcus hirae paradigm of copper homeostasis: Copper chaperone turnover, interactions, and transactions. Biometals 16, 137^143. Gaballa, A. and Helmann, J.D. (2002) A peroxide-induced zinc uptake system plays an important role in protection against oxidative stress in Bacillus subtilis. Mol. Microbiol. 45, 997^1005. Tsai, K.J., Lin, Y.F., Wong, M.D., Yang, H.H.C., Fu, H.L. and Rosen, B.P. (2002) Membrane topology of the pl258 CadA Cd(II)/ Pb(II)/Zn(II)-translocating P-type ATPase. J. Bioenerg. Biomembr. 34, 147^156. DiDonato, M., Zhang, J.Y., Que, L. and Sarkar, B. (2002) Zinc binding to the NH2 -terminal domain of the Wilson disease copper-transporting ATPase ^ Implications for in vivo metal ion-mediated regulation of ATPase activity. J. Biol. Chem. 277, 13409^ 13414. Borremans, B., Hobman, J.L., Provoost, A., Brown, N.L. and Van der Lelie, D. (2001) Cloning and functional analysis of the pbr lead resistance determinant of Ralstonia metallidurans CH34. J. Bacteriol. 183, 5651^5658. Baker, A.J.M. (1987) Metal tolerance in plants. New Phytol. 106, 93^111. Nies, D.H., Koch, S., Wachi, S., Peitzsch, N. and Saier, M.H.J. (1998) CHR, a novel family of prokaryotic proton motive forcedriven transporters probably containing chromate/sulfate transporters. J. Bacteriol. 180, 5799^5802. Alvarez, A.H., Moreno-Sanchez, R. and Cervantes, C. (1999) Chromate e¥ux by means of the ChrA chromate resistance protein from Pseudomonas aeruginosa. J. Bacteriol. 181, 7398^7400. Pimentel, B.E., Moreno-Sanchez, R. and Cervantes, C. (2002) E¥ux of chromate by Pseudomonas aeruginosa cells expressing the ChrA protein. FEMS Microbiol. Lett. 212, 249^254. Nies, A., Nies, D.H. and Silver, S. (1990) Nucleotide sequence and expression of a plasmid-encoded chromate resistance determinant from Alcaligenes eutrophus. J. Biol. Chem. 265, 5648^5653. Nies, A., Nies, D.H. and Silver, S. (1989) Cloning and expression of plasmid genes encoding resistances to chromate and cobalt in Alcaligenes eutrophus. J. Bacteriol. 171, 5065^5070. Peitzsch, N., Eberz, G. and Nies, D.H. (1998) Alcaligenes eutrophus as a bacterial chromate sensor. Appl. Environ. Microbiol. 64, 453^ 458. Juhnke, S., Peitzsch, N., Hu«bener, N., GroMe, C. and Nies, D.H. (2002) New genes involved in chromate resistance in Ralstonia metallidurans strain CH34. Arch. Microbiol. 179, 15^25. Cervantes, C., Ohtake, H., Chu, L., Misra, T.K. and Silver, S. (1990) Cloning, nucleotide sequence, and expression of the chromate resistance determinant of Pseudomonas aeruginosa plasmid pUM505. J. Bacteriol. 172, 287^291. Nicholson, M.L. and Laudenbach, D.E. (1995) Genes encoded on a cyanobacterial plasmid are transcriptionally regulated by sulfur availability and cysR. J. Bacteriol. 177, 2143^2150. Lenz, O., Schwartz, E., Dernedde, J., Eitinger, T. and Friedrich, B. (1994) The Alcaligenes eutrophus H16 hoxX gene participates in hydrogenase regulation. J. Bacteriol. 176, 4385^4393. Mobley, H.L.T., Garner, R.M. and Bauerfeind, P. (1995) Helicobacter pylori nickel-transport gene NixA ^ Synthesis of catalytically active urease in Escherichia coli independent of growth conditions. Mol. Microbiol. 16, 97^109. Fulkerson, J.F., Garner, R.M. and Mobley, H.L.T. (1998) Conserved residues and motifs in the NixA protein of Helicobacter pylori are critical for the high a⁄nity transport of nickel ions. J. Biol. Chem. 273, 235^241. Depina, K., Navarro, C., Mcwalter, L., Boxer, D.H., Price, N.C.
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 and Kelly, S.M. (1995) Puri¢cation and characterization of the periplasmic nickel-binding protein NikA of Escherichia coli. Eur. J. Biochem. 227, 857^865. [280] Altschul, S.F., Madden, T.L., Scha«¡er, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSIBLAST : a new generation of protein database search programs. Nucleic Acids Res. 25, 3389^3402. [281] Altschul, K., Altschul, S. and Altschul, S.F. (1990) Methods for assessing the statistical signi¢cance of molecular sequence features by using general scoring schemes. Proc. Natl. Acad. Sci. USA 87, 2264^2268.
FEMSRE 781 2-6-03
339
[282] Altschul, K., Altschul, S. and Altschul, S.F. (1993) Applications and statistics for multiple high-scoring segments in molecular sequences. Proc. Natl. Acad. Sci. USA 90, 5873^5877. [283] Solioz, M. and Stoyanov, J.V. (2003) Copper homeostasis in Enterococcus hirae. FEMS Microbiol. Rev. 27, 183^195. [284] Cavet, J.S., Borrelly, G.P.M. and Robinson, N.J. (2003) Zn, Cu and Co in cyanobacteria : selective control of metal availability. FEMS Microbiol. Rev. 27, 165^181.
Cyaan Magenta Geel Zwart
www.fems-microbiology.org
E¥ux-mediated heavy metal resistance in prokaryotes Dietrich H. Nies
Institute of Microbiology, Molecular Microbiology, Martin-Luther-University Halle-Wittenberg, Kurt-Mothes-Str. 3, 06099 Halle/Saale, Germany Received 3 October 2002; received in revised form 14 January 2003; accepted 31 March 2003 First published online 30 April 2003
Abstract What makes a heavy metal resistant bacterium heavy metal resistant? The mechanisms of action, physiological functions, and distribution of metal-exporting proteins are outlined, namely: CBA efflux pumps driven by proteins of the resistance^nodulation^cell division superfamily, P-type ATPases, cation diffusion facilitator and chromate proteins, NreB- and CnrT-like resistance factors. The complement of efflux systems of 63 sequenced prokaryotes was compared with that of the heavy metal resistant bacterium Ralstonia metallidurans. This comparison shows that heavy metal resistance is the result of multiple layers of resistance systems with overlapping substrate specificities, but unique functions. Some of these systems are widespread and serve in the basic defense of the cell against superfluous heavy metals, but some are highly specialized and occur only in a few bacteria. Possession of the latter systems makes a bacterium heavy metal resistant. 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : CDF; RND ; Cadmium; Cobalt; Zinc ; Nickel; Chromate ; Ralstonia metallidurans
Contents 1. 2.
3.
4.
Introduction: what kind of heavy metal resistance will be described . . . . . . . . . . . . . . . . . The ¢rst layer of heavy metal resistance: members of the resistance^nodulation^cell division (RND) protein family export super£uous cations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. CzcCBA-mediated e¥ux is the outer shell of Co2þ , Zn2þ , Cd2þ and Ni2þ resistance in R. metallidurans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Localization and speciation of the cations e¥uxed by CzcCBA . . . . . . . . . . . . . . . . . 2.3. The Cus system from E. coli: is there direct export from the periplasm? . . . . . . . . . . 2.4. Export of organic substrates by HAE-RNDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Biochemistry of RND proteins: towards a model of function for these proton-driven antiporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. MFPs and OMFs enhance the function of RND proteins . . . . . . . . . . . . . . . . . . . . . 2.7. Distribution of RND proteins in the kingdoms of life . . . . . . . . . . . . . . . . . . . . . . . . The second layer of metal resistance : cation di¡usion facilitators (CDF family) serve as secondary cation ¢lters in bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Physiological function and substrate speci¢city of CDF proteins . . . . . . . . . . . . . . . . 3.2. Biochemistry of CDF proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Distribution of CDF proteins in the living world . . . . . . . . . . . . . . . . . . . . . . . . . . . . The third layer of heavy metal resistance: P-type ATPases are the basic defense against heavy metal cations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Families of CPx-type ATPases and their substrates . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Structure and function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Distribution of CPx-type ATPases in prokaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . .
314 314 315 316 316 317 318 319 322 324 324 325 325 326 328 328 328
* Tel. : +49 (345) 5526352; Fax: +49 (345) 5527010. E-mail address : [email protected] (D.H. Nies). 0168-6445 / 03 / $22.00 8 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-6445(03)00048-2
FEMSRE 781 2-6-03
Cyaan Magenta Geel Zwart
314
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
5.
........ ........ ........ ........ ........ many sys........
330
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
331
6. 7.
Other heavy metal export systems: CHR protein family, NreB, CnrT . . . . . . . . 5.1. CHR proteins detoxify chromate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. NreB-like proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. CnrT-like proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Heavy metal resistance in prokaryotes revisited . . . . . . . . . . . . . . . . . . . . . . . . Concluding remarks: heavy metal resistance is the product of the interaction of tems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction : what kind of heavy metal resistance will be described Metal resistance determinants were initially found on bacterial plasmids [1^13]. However, after the genome sequences of several bacteria became available, homologues to metal resistance determinants popped up in many bacteria (see below). If the possession of a heavy metal resistance determinant makes a bacterium metal resistant, which bacteria remain sensitive? Since Escherichia coli serves as model bacterium for many purposes, metal resistance of a given bacterium could be compared to metal resistance of E. coli [14]. This comparison was done with the well-characterized heavy metal resistant bacterium Ralstonia metallidurans strain CH34 (previously Alcaligenes eutrophus or Ralstonia sp. [15]). It originates from a zinc decantation tank and related bacteria have been found in many metal-contaminated environments around the world [16,17]. The minimal inhibitory concentrations (MICs) of the biologically important heavy metals [18] for R. metallidurans strain CH34 and for E. coli strain K38 were compared in Trisbu¡ered mineral salts medium [19]. Growth of neither bacterium was inhibited by up to 50 mM MoO23 4 , 23 3þ WO4 and, due to its low solubility, Fe . Both species had a MIC of 2 mM UO23 2 . E. coli was more resistant to 33 and SbO3 VO3 3 , AsO4 2 than R. metallidurans; however, R. metallidurans was 2.5-fold more resistant to CrO23 4 than E. coli. Interaction of microbes with arsenic has recently been reviewed [20] and will not be elucidated here. Thus, of these seven biologically important oxyanionforming heavy metals [18], only chromate resistance will be discussed. The ¢rst part of this review will concentrate on the heavy metal cations. The MIC values of both bacteria to biologically important heavy metal cations (with the exception of iron) were plotted against the solubility constants of the respective metal sul¢de complexes ([21], Fig. 1). This was done because heavy metal cations that escape metal e¥ux systems may cause toxicity by interaction with cellular thiol compounds such as glutathione. The solubility of the sul¢de complexes may therefore serve as a measure of the a⁄nity of a metal to thiol compounds and for the severity of damage resulting from this interaction. The di¡erent de-
FEMSRE 781 2-6-03
329 329 329 329 330
grees of metal resistance roughly followed the solubility constants (regression coe⁄cients s 80%) ; so, this interaction may indeed be part of metal toxicity. Metal resistance of R. metallidurans was in general higher than metal resistance of E. coli (Fig. 1). Especially resistance of R. metallidurans to the cations of cobalt, zinc, nickel, and copper was very high, cadmium resistance was similar to that of E. coli. This review will focus on the description of all resistance systems that are responsible for this remarkable heavy metal resistance of R. metallidurans and it will give an overview of the distribution of these heavy metal resistance determinants in other organisms.
2. The ¢rst layer of heavy metal resistance : members of the resistance^nodulation^cell division (RND) protein family export super£uous cations The RND protein family was ¢rst described as a related group of bacterial transport proteins involved in heavy metal resistance (R. metallidurans), nodulation (Mesorhizobium loti) and cell division (E. coli) [22]. This family has grown into a huge superfamily that includes seven protein families that can be found in all major kingdoms of life [23]. In bacteria and archaea members of this superfamily are involved in transport of heavy metals, hydrophobic compounds, amphiphiles, nodulation factors, and, with SecDF, in protein export. The members of the RND superfamily in eukaryotes transport sterols or serve as receptors [24^45]. This section deals with the heavy metal e¥ux family (HME-RND [23]) of the RND superfamily that has received the number TC 2.A.6.1.1 in Milton Saier’s functional-phylogenetic classi¢cation system for transmembrane solute transporters [46]. RND proteins involved in export of hydrophobic and amphiphilic compounds (HAE-RND) will be included, since data obtained for these systems allow the development of models for the function of RND proteins and RND-driven e¥ux complexes. Therefore, the expression ‘RND protein’ will be used exclusively hereafter for HME-RND and HAERND proteins, but not for eukaryotic proteins or bacterial proteins involved in protein export. Bacterial RND proteins like company. Adjacent to a
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 Mn2+
Zn2+ Co2+ Ni2+
Cd2+
Ag+ Hg2+
Cu2+
Pb2+
log (MIC, [mM])
10
1
0.1
0.01 20
30
40
50
- log (KMeS) Fig. 1. Heavy metal toxicity depends on the a⁄nity of a cation to sulfur. The logarithm of the MIC of heavy metal cations for E. coli (open circles) and R. metallidurans (closed circles) was plotted against the logarithm of the solubility for the respective metal sul¢de as a measure of the a⁄nity to sulfur [22]. The positions of the cations are indicated with arrows at the top. An exponential curve ¢t was performed for both data sets that yielded regression coe⁄cients for E. coli (thin line) of 81.1% and for R. metallidurans (thick line) of 86.7%.
gene that encodes a RND protein is ^ in most cases ^ a second gene that encodes a member of the membrane fusion protein family MFP [22]. This family has also been designated periplasmic e¥ux protein family [47] or, more recently, periplasmic adapter protein family [48]. This review will use the oldest scienti¢c name, that is MFP. Besides MFPs, many RND proteins cooperate with yet a third protein that belongs to the family of outer membrane factors (OMF) [47,49]. These three proteins form an e¥ux protein complex that may export its substrate from the cytoplasm, the cytoplasmic membrane or the periplasm across the outer membrane directly to the outside [50^53]. To distinguish this type of transport system from ATP-driven ABC transport systems, the RND-driven export systems will be referred to as CBA e¥ux systems or CBA transporters. 2.1. CzcCBA-mediated e¥ux is the outer shell of Co2+ , Zn 2+, Cd 2+ and Ni 2+ resistance in R. metallidurans The ¢rst identi¢ed member of the RND family was the CzcA protein from R. metallidurans followed by the CnrA protein from the same bacterium [22]. This L-proteobacterium contains two megaplasmids involved in heavy metal resistance [17,19]. The presence of the larger of both plasmids, pMOL30, increased the MICs to Zn2þ 50-fold, to
FEMSRE 781 2-6-03
315
Co2þ 33-fold and to Cd2þ seven-fold [19], while the second plasmid, pMOL28, mediated resistance to Ni2þ and a 16fold increase in cobalt resistance. The plasmid pMOL30-mediated resistance to Co2þ , Zn2þ and Cd2þ was designated Czc and the structural gene region of the czc resistance determinant of pMOL30 was cloned [54] and sequenced [55,56]. This structural gene region contains the genes for the OMF CzcC, the MFP CzcB and the CzcA protein of the RND family. The three genes form an operon czcCBA that is transcribed tricistronically [57] and is £anked by a multitude of genes involved in metal-dependent regulation of czcCBA expression (C. GroMe, A. Anton, T. Ho¡mann, S. Franke, G. Schleuder and D.H. Nies, unpublished) [58^ 60]. The czcCBA structural gene region was expressed in a plasmid-free R. metallidurans strain under the control of a constitutive promoter on a low copy number vector. The presence of the czcCBA genes in this bacterial strain mediated a metal resistance level similar to that reached by the complete megaplasmid pMOL30 [54,55]. Czc-mediated resistance to Co2þ , Zn2þ and Cd2þ was based on a diminished cellular accumulation of the three cations resulting from cation e¥ux [61] driven by the proton motive force [62]. The cobalt^nickel resistance determinant cnr on plasmid pMOL28 is also based on cation e¥ux [61,63]. The resistance determinant is composed of a cnrCBA structural region [64] that is preceded by the regulatory gene region cnrYXH [65,66]. Similar to czcCBA, cnrCBA encodes the OMF CnrC, the MFP CnrB and the RND protein CnrA. Another cobalt^nickel resistance determinant, ncc, was cloned from R. metallidurans strain 31A (previously Achromobacter xylosoxidans subsp. xylosoxidans) and further characterized [67]. Similar to cnr, ncc is composed of a regulatory gene region nccYXH followed by the structural region nccCBA, again encoding a putative outer membrane protein (NccC), a MFP (NccB), and an RND protein (NccA). In contrast to Cnr, Ncc mediates resistance to cadmium in addition to cobalt and nickel resistance [67]. Czc alone seemed to be responsible for the increase in Co2þ , Zn2þ and Cd2þ resistance that was assigned to plasmid pMOL30. However, the genomic sequence of R. metallidurans (preliminary sequence obtained from DOE/ JGI, http://www.jgi.doe.gov/JGI_microbial/html/ralstonia/ ralston_homepage.html) indicated the presence of a ncc determinant very similar to ncc from R. metallidurans strain 31A in the genome of strain CH34. The ncc determinant from R. metallidurans was assigned to plasmid pMOL30 by a polymerase chain reaction and homologous recombination experiment (S. Franke and D.H. Nies, unpublished). Thus, Ncc from R. metallidurans strain CH34 may contribute to the cobalt, zinc and cadmium resistance mediated by plasmid pMOL30. The exclusive assignment of this resistance to Czc could be an experimental artifact, e.g. resulting from the higher copy number of the vector
Cyaan Magenta Geel Zwart
316
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
plasmid compared to the native plasmid pMOL30. However, this possibility was excluded when the czcA gene was deleted from pMOL30. The resulting pMOL30(vczcA)containing mutant strain was even more sensitive to cobalt, zinc and cadmium than the plasmid pMOL30-free strain [68]. Since polar e¡ects were carefully avoided when the vczcA deletion was constructed, this indicates clearly the essential function of Czc as the outermost shell of cobalt, zinc, and cadmium resistance in R. metallidurans. In agreement with this fact, no regulatory genes were found upstream of nccCBA on plasmid pMOL30 and no expression of this operon could be measured (G. Rehbein, C. Grosse and D.H. Nies, unpublished). Thus, ncc seems to be a silent determinant in R. metallidurans CH34. 2.2. Localization and speciation of the cations e¥uxed by CzcCBA Why does the CzcCBA e¥ux system mediate such a huge increase in metal resistance? Maybe, this system diminishes not only the cytoplasmic concentration of heavy metal cations but additionally (or even exclusively?) the periplasmic concentration. In this case, the cations could be removed before they had the opportunity to enter the cell. Moreover, the CzcCBA e¥ux system could mediate further export of the cation that had been removed from the cytoplasm by other e¥ux systems. Such export of heavy metal cations from the periplasm may happen indirectly, that is by the combined action of an uptake and a CBA transenvelope e¥ux system, but periplasmic cations could also be directly funneled into a CBA transenvelope e¥ux system. Yet, both mechanisms may contribute to the physiological function of CzcCBA and other CBA transport systems. As shown with puri¢ed protein that has been reconstituted into proteoliposomes, CzcA should be able to accept heavy metal cations from the cytoplasm [69]. Since transport kinetics of the reconstituted CzcA protein and the CzcCBA complex as characterized in inside-out vesicles are similar [62], the complete CzcCBA complex should also be able to export cations from the cytoplasm. E¥ux of cadmium by CzcCBA ¢ts into a quantitative physiological model describing cadmium homeostasis [68]. This favors the possibility of export from the cytoplasm over export directly from the periplasm. On the other hand, the K50 values of CzcA and the CzcCBA complex for Co2þ , Zn2þ and Cd2þ are in the lower mM range [62,69]. The trace elements Co2þ and Zn2þ are transported with a sigmoidal substrate saturation function (with a Hill coe⁄cient of 2). This keeps the e¥ux velocity of the CzcCBA complex slow when cytoplasmic concentrations are below the K50 value, but allows rapid export at higher concentrations. A cytoplasmic £ow equilibrium model predicts the apparent cytoplasmic concentration of Co2þ to be 8.4 mM and of Zn2þ to be below 6.6 mM [68]. These values are in contrast to the situation
FEMSRE 781 2-6-03
described for E. coli by Outten and O’Halloran [70] ; in this bacterium, the apparent concentration of Zn2þ is kept at 0.2 mM. Moreover, the authors predict by analysis of the interaction of Zn2þ with regulatory proteins a free cellular zinc concentration in the fM range. If R. metallidurans is not completely di¡erent from E. coli, the concentration of free zinc in the cell will not allow e⁄cient export of zinc from the cytoplasm by CzcCBA in the cellular model of Outten and O’Halloran. As shown in Fig. 1, heavy metal cations di¡er in their a⁄nity for sulfur. This a⁄nity relates to the toxicity of the individual heavy metal cation. Gram-negative bacteria contain an intracellular glutathione pool of about 10 mM [71,72], other bacteria harbor di¡erent thiol compounds [73]. Glutathione complexes of extremely sulfurloving heavy metal cations have complex binding constants of 1042 (Hg2þ ^bis-glutathionate complex [74]) and 1039 (Cuþ ^glutathionate complex [75]). Copper should never occur ‘free’ in a cell [76], but the complex binding constants for Cd2þ , Zn2þ , Co2þ , and Ni2þ should be below the copper value. To explain the fM concentrations of ‘free’ zinc observed by the regulatory proteins in E. coli [70] at an apparent cytoplasmic concentration of 0.2 mM Zn2þ and 10 mM glutathione, a binding constant of at least 1013 is required, which is a moderate assumption of this value. Thus, in the cells of Gram-negative bacteria cellular zinc that is not needed in the active centers of enzymes may be present exclusively in glutathionate complexes once the thermodynamic equilibrium has been reached. However, CzcA was not able to transport 65 Zn2þ into proteoliposomes when glutathione was present [68]. Formation of a zinc^mono-glutathionate complex in vitro has a rate constant of 3.9U103 M31 s31 [77]. This would lead to cytoplasmic concentrations of ‘free’ Zn2þ in the WM range in a £ow equilibrium between uptake and a theoretical £ux into a glutathione-bound pool (calculation not shown). This concentration is 1010 -fold higher than the fM concentration of ‘free’ zinc predicted for E. coli [70], but still 1000-fold lower than the K50 value of CzcA for zinc [69]. If the in vivo rate of the formation of zinc^ mono-glutathionate complexes is two or three orders of magnitude lower than the rate determined in vitro [77], CzcCBA will be able to detoxify cytoplasmic Zn2þ . Otherwise, this transport system relies on other transporters that export Zn2þ into the periplasm, meaning that in this case CzcCBA-mediated e¥ux is a direct periplasmic detoxi¢cation process. But, is there any evidence for a direct export of metal cations out of the periplasm ? To investigate this question, a copper/silver-transporting system from E. coli will be discussed. 2.3. The Cus system from E. coli : is there direct export from the periplasm? Copper and silver have a much higher a⁄nity towards
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
sulfur than does zinc. The half-life of Cuþ and Agþ as ‘free’ ions in the cytoplasm is thus probably much shorter than the half-life of ‘free’ Zn2þ , and no ‘free’ copper should occur in a bacterial cell [76]. Thus, evidence for direct periplasmic detoxi¢cation may be more easily obtained for silver and copper than for zinc or cobalt. A plasmid-encoded copper resistance determinant for E. coli has been known for a long time [9,59,78^86]. However, copper resistance factors encoded by the chromosome of this bacterium were described only recently. A copper-exporting P-type ATPase of E. coli, CopA [87,88], transports the heavy metal cation from the cytoplasm into the periplasm like other soft metal-transporting P-type or CPx-type ATPases [89], without any evidence for further transport from the periplasm to the outside. The only HME-RND of E. coli is the CusA protein. Its gene is embedded in the structural gene operon cusCFBA on the bacterial chromosome that encodes CusA, the MFP CusB, the OMF CusC and a small additional protein, CusF [90]. Upstream of cusCFBA and divergently transcribed on the other DNA strand are two regulatory genes that encode the factors of a two-component regulatory system CusS (sensor/histidine kinase) and CusR (response regulator). Both are involved in control of cellular copper homeostasis [91] and of cusCFBA expression [90]. The proteins of the Cus system mediate resistance to silver [90,92]. Resistance to copper could not be assigned to Cus in the ¢rst study [90], not even in a mutant background with the copA gene for the copper-detoxifying Ptype ATPase disrupted. However, copper was a better inducer of cus expression than silver [90], indicating involvement of Cus in copper resistance. Outten et al. found a clear contribution of Cus to copper resistance under anaerobic conditions [93] while copper detoxi¢cation under aerobic conditions was performed by CopA and the periplasmic protein CueO. Under anaerobic conditions periplasmic copper ions control copper homeostasis via CusRS regulation, while under aerobic conditions cytoplasmic copper exerts control via the regulator CueR [93]. Finally, Grass and Rensing [94] demonstrated that an essential function of Cus for copper resistance is easily visible in mutant strains devoid of CueO. This protein is a multi-copper oxidase located in the periplasm [95] explaining why CopA and CueO work together in detoxifying copper, but only under aerobic conditions [93]. CopA exports Cuþ from the cytoplasm into the periplasm [87]. If not prevented by further transport, Cuþ may react with periplasmic components to form Cu2þ and would therefore reduce and impair these components. CueO could reoxidize and repair these periplasmic components again. Electrons can be transferred by the copper atoms of CueO [95] to molecular oxygen leaving a repaired periplasmic component and water. Since Agþ cannot be oxidized to Ag2þ so easily, Agþ should not be able to damage periplasmic components by oxidation (but it may damage
FEMSRE 781 2-6-03
317
them by binding). Thus, CueO cannot repair silver damage and silver resistance mediated by Cus was visible in the presence of CueO [90] In the absence of CueO or in the absence of molecular oxygen [93,94], however, the Cus system is required to maintain copper homeostasis. Cus complements a missing repair or protection process in the periplasm. Recent work on the plasmid-encoded copper resistance system Pco discusses the importance of oxidizing periplasmic Cuþ to Cu2þ to protect the cell [83,96]. Periplasmic proteins, the MFP CusB and the OMF CusC, were required for full copper resistance [97]. Substitutions of methionine residues in the periplasmic portion of the CusA protein resulted in loss of copper resistance demonstrating their functional importance [97]. The major di¡erence in the organization of the czcCBA and the cusCFBA structural gene regions is the presence of cusF in the cus determinant. Similar genes only occur in putative or identi¢ed silver/copper resistance determinants. The small 10-kDa protein CusF is a periplasmic protein that binds one copper per polypeptide and may function as a copper usher that facilitates access of periplasmic copper towards the CusCBA e¥ux pump [97]. Therefore, the main function of Cus may be the direct export of Cuþ and Agþ from the periplasm to the outside [93,94]. This is necessary because both monovalent cations should never occur ‘free’ in the cytoplasm owing to their high a⁄nity to sulfur. More evidence for possible direct periplasmic access of substrates into CBA e¥ux systems comes from the HAE-RND proteins that transport organic substances [98,99]. 2.4. Export of organic substrates by HAE-RNDs Because of their eminent medical importance [99], many HAE-RND proteins have recently been characterized. These RND proteins are involved in the detoxi¢cation of organic substances, including antibiotics, in pathogenic or non-pathogenic bacteria. Resistance to antibiotics mediated by RND proteins was responsible for half of the multiple drug resistant Pseudomonas aeruginosa strains that emerged in vivo in a hospital [100], showing the importance of RND proteins for the medical problem of antibiotic resistance. This pathogenic bacterium contains at least 13 RND transport systems (see below). One is a HME-RND protein and exports heavy metals [101], the other 12 are probably HAE-RND proteins. Best characterized of these proteins are MexB, MexD and MexF. Together with their respective MFPs (MexA, MexC, MexE) and OMFs (OprM, OprJ, OprN), these systems detoxify a variety of organic substances. Unfortunately, nomenclature of the HAE-RNDs did not follow the older nomenclature of the HME-RNDs, which may lead to confusion. MexB is the CzcA-related RND protein while MexA is the CzcB-related MFP. The speci¢city of the MexB-driven system mainly in-
Cyaan Magenta Geel Zwart
318
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
cludes L-lactam antibiotics, L-lactamase inhibitors and organic solvents [102^104], while the MexD-driven system expels cepheme compounds [105]. Chloramphenicol, £uoroquinolones and imipenem are best exported by the MexF-driven exporter [106]. Recently, the substrate speci¢cities of two more Mex systems were investigated, MexJK and MexGHI [107,108], and the functions of some systems were characterized in multiple deletion mutant strains [109]. E. coli contains six HAE-RND proteins and the Cus system as the sole HME-RND protein. The di¡erent substrate speci¢cities of all six HAE-RNDs have been characterized [110,111], but most data are available for the AcrAB-TolC system [98,112^115] that protects the bacterium against a variety of toxic substances especially during slow growth [116]. The AcrD-driven system from E. coli is an aminoglycoside e¥ux pump [117]. Examples of characterized HAE-RND-driven systems from other bacteria are AcrAB from Haemophilus in£uenzae [118], MtrCDE from Neisseria gonorrhoeae [119,120], SrpABC and TtgDEF from Pseudomonas putida [121,122], CmeABC from Campylobacter jejuni [123,124], AdeABC from Acinetobacter baumannii [125], SmeDEF from Stenotrophomonas maltophilia [126], IfeAB and AmeABC from Agrobacterium tumefaciens [127,128], NorMI from Brucella melitense [129], XepCAB from Porphyromonas gingivalis [130], and a CBA e¥ux system from Serratia marcescens [131]. The substrate speci¢city of an RND-driven export system can include many organic substances with di¡erent chemical structures. Common to all these components is only a hydrophobic ability or at least a hydrophobic side chain. Examples are various L-lactam antibiotics [132^ 134], L-lactamase inhibitors [135], other hydrophobic antibiotics and toxic substances [117,129,130,136,137], organic solvents [104,112,114,115,121,122,138], bile salts [113], indole [139], and even homoserine lactone compounds involved in quorum sensing [127,140,141]. A title announcing resistance to vanadate mediated by a RND-driven system [108], however, has to be considered with caution, since the e¡ect on vanadium resistance is via the wellknown action of an RND-driven system on quorum sensing. 2.5. Biochemistry of RND proteins: towards a model of function for these proton-driven antiporters Contribution to resistance to a multitude of toxic substances has been assigned to RND proteins, but biochemical and genetic analysis has only reached a few members. The RND protein CzcA that expels Co2þ , Zn2þ , and Cd2þ was puri¢ed and reconstituted actively into ammonium chloride-charged proteoliposomes [69]. The result demonstrated that CzcA is a proton^cation antiporter depending solely on the ZWvpH portion of the proton motive force and not on its v8 portion [69]. This was in agreement with previous data obtained with the CzcCBA complex that
FEMSRE 781 2-6-03
Fig. 2. Signature sequences of selected RND proteins. Representative RND proteins of the HAE family (AcrB from E. coli), HME group 1 (CzcA from R. metallidurans), HME group 2 (CnrA from R. metallidurans), HME group 3 (two putative proteins from R. metallidurans denoted as ‘p’contig number/gene number, and two proteins from Helicobacter pylori and Aquifex aeolicus), HME group 4 (SilA from Salmonella enterica pv. typhimurium), and HME group 5 (all7631 from Nostoc) were analyzed to demonstrate the relation of the groups to each other. A short sequence alignment of the transmembrane helix IV is shown, which contains the signature motifs of these proteins. The red box indicates the ‘DFG’ occurring in HME1, HME2 and a few HME3 proteins. The green box above the red one is the ‘AI/VG’ at the same position in the other protein groups. The blue box frames the ‘DD’ motif of HAE and HME5 proteins, the gray one below the ‘DSS’ at the same position that is a characteristic of HME group 3. Finally, the purple arrows mark unique and indicative polar and acidic residues that occur at this position only in HME5 proteins.
was heterologously expressed in E. coli and characterized in inside-out vesicles [62]. Not many transport data are available for other RND proteins. Only AcrB has also been puri¢ed and reconstituted into proteoliposomes [51]. Depending on the presence of vpH, the protein catalyzed the extrusion of £uorescent phospholipids. Known substrates of AcrAB, such as bile acids, erythromycin, and cloxacillin, inhibited this activity. Action of various drugs in the presence of an inverted pH gradient (inside acidic) suggested that AcrB is a proton antiporter [51] like CzcA. Thus, members of both RND families, HME and HAE, are proton^substrate antiporters. With reporter gene fusions, the topology of several RND proteins was investigated. These proteins include CzcA [69], MexB [142], and MexD [143]. RND proteins are huge proteins of more than 1000 amino acid residues (aa) in size. They are composed of two related halves that are probably the result of an ancient gene duplication and fusion [144], each containing a large periplasmic loop of about 300 aa in size that is £anked by 12 transmembrane K-helices, TMH I to TMH XII. TMH IV contains amino acid residues that are conserved in most RND proteins [23]. HME-RND proteins involved in the export of divalent heavy metal cations (CzcA, CnrA) exhibit the consensus sequence DFGX3 DGAX3 VEN in this TMH (Fig. 2). Exporters for monovalent cations like CusA and the silver transporter SilA [145] have consensus sequences similar to AVGX3 DAAX3 IEN (Fig. 2) and do not contain the ‘DFG’ motif. The HAE-RND proteins such as AcrB and MexB that export organic substances do not have the ¢rst aspartate residue in the DFG region, but a double
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
319
DD in the middle (characteristic DD motif, Fig. 2). These conserved sequences allow the prediction for the speci¢city of many RND proteins with merely a glance. Both aspartate residues and the glutamate residue in TMH IV of CzcA were essential for function [69]. The same was true for the single aspartate and the glutamate residue of CusA [97]. For MexF, one of the two aspartate residues in the DD motif and the glutamate residue were also essential [146]. Thus, the presence of charged amino acid residues in TMH IV of RND proteins is required for the proton^substrate antiport. Both aspartate residues of CzcA are probably involved in proton translocation [69]. As in other proteins [147^150], the aspartate residues may therefore be part of a charge relay system that forms the path of the protons across the cytoplasmic membrane. These data led to a preliminary model of how CzcA and
other RND proteins might function [69] (Fig. 3). This simple model for the e¥ux of cytoplasmic cations by an RND export system can easily be modi¢ed to ¢t other postulated functions for RND proteins (Fig. 3). It is not relevant whether the substrates access the periplasmic substrate binding sites from the cytoplasm (as in the case of Zn2þ ?), or from the periplasm (as in the case of Cuþ ?), or from inside the cytoplasmic membrane (maybe in the case of hydrophobic substrates for HAE-RND proteins). In all cases, a protonation/deprotonation cycle couples the process of substrate binding and release to the exergonic proton import reaction. Moreover, protonation may easily change the substrate binding site between negatively charged and neutral (HME-RND proteins) or hydrophilic and hydrophobic (HAE-RND proteins). The substrates bound initially to some substrate binding site of the RND protein are consecutively forwarded to the other two components of the CBA e¥ux system for further transport. A lot of insight into the possible mechanisms of action has come from the recently available three-dimensional structure of the AcrB protein obtained after crystallization [151,152]. Three AcrB protomers are organized as a homotrimer in the shape of a jelly¢sh. Each protomer is composed of a transmembrane region 5 nm thick, composed of the previously identi¢ed 12 transmembrane spans, and a 7-nm protruding headpiece containing a central cavity at the bottom. The cavity has three vestibules at the side of the headpiece that may lead into the periplasm. This structure implies that substrates may be translocated from the cell interior through the transmembrane region and/or from the periplasm through the vestibules and are collected in the central cavity [151], before they are forwarded to the other two components of the CBA e¥ux complex.
Fig. 3. Proposed cellular function and mechanism of RND proteins. This model is based on the structure of AcrB, TolC and published biochemical data (see text for details). The RND protein (red) is shown as a trimer with a periplasmic substrate binding site. The MFP protein is shown in green, the OMF in yellow. Action by RND proteins is driven by protons (black dots) that were previously expelled by the respiratory chain (1). In the case of HME1 and HME2 proteins, divalent heavy metal cations (blue, purple and gray circles, M2þ ) migrate through the cation channel of the protein and are bound to a periplasmic substrate binding site (orange box, 2a). Alternatively (blue arrows), this site may also be accessible for periplasmic cations (2b), especially in the case of Cuþ and HME4 proteins. HAE transporter of organic substrates may bind these substrates from the hydrophobic environment of the cytoplasmic membrane (2c). A⁄nity of the hypothetical periplasmic binding site is decreased by protonation with periplasmic protons (3) and this energy is used to accumulate the substrates into the OMF tube or a microenvironment formed by the RND protein (4). The high concentration in this tube allows easy di¡usion of the substrates from the tube to the outside (5). For energetic coupling, the hypothetical periplasmic substrate binding site is deprotonated by import of the protons to the cytoplasm (6). This complex reaction allows transport of substrates across the complete envelope of a Gram-negative bacterium that is composed of outer membrane (OM), cell wall (CW), periplasmic space (PP) and cytoplasmic membrane (CPM).
2.6. MFPs and OMFs enhance the function of RND proteins
FEMSRE 781 2-6-03
Most RND proteins are encoded in operons that additionally contain the gene for a MFP. Many of these operons harbor a third gene, encoding an OMF [50]. The position of this OMF-encoding gene varies from operon to operon. It may be located at the 5P end of operons like czc, at the 3P end of the operon as in most mex operons, it could be not even part of the operon as in the case of tolC [48]. Cross-linking experiments with the CzcCBA e¥ux complex displayed a CzcB multimer [153]. This protein migrates in polyacrylamide gels at a much lower apparent molecular mass than predicted from the derived amino acid sequence. This unusual migration behavior may be based on the extensive coil-coil structure proposed for MFPs [52]. Depending on which size for the CzcB monomer is used in the calculation, the CzcB multimer is a trimer or a tetramer of the MFP [153]. For the MFP
Cyaan Magenta Geel Zwart
320
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
Table 1 Distribution of members of heavy metal-exporting protein families in the genomes of 64 prokaryotesa RND proteinsb
CDF proteinsc
P-type ATPasesd
Other transport systemsd
0 0 0 0
0 1: 2a 1: 2a 2: 2a, 3aii
2: 2: 1: 1:
0 0 0 0
0 0 0
2: 3f, 3f 1: 3f 2: 3f, 3f
3: Cu9, Cu12, ¡ 4: Cu2, Cu14, Zn1, ¡ 6: Cu9, Cu12, Zn1, ¡
0 0 0
0 0 0 0 0 0
1: 1: 1: 2: 0 0
2: 0 1: 1: 3: 2:
Zn7 Cu9 Cu1, KdpB, ¡ Cu1, KdpB
CHR4 0 0 0 0 0
4: M3b, M4, A5¡ 1: A5 1: A5
2: 2g, 3ai 2: 2c, 3e 2: 2h, 3c
3: Cu1, Cu12, ¡ 1: Cu13 3: Cu3, Zn2, KdpB
0 0 CHR1
6: M5, A4, A5¡ 8: M5, M5, A1, A4, A5¡
1: 3c 3: 1b, 2e, 3ai
9: Cu5, Cu10, Zn7, Zn9, ¡ 13: Cu2, Cu5, Cu5, Cu10, Zn4, Zn9, Me1, Me1, ¡
CHR3, NreB CHR3, NreB
1: A2 1: A2 0 0 1: A2 0 0
4: 2: 3: 3: 3: 1: 1:
CHR4 CHR1, CHR4 0 0 0 CHR4 0
Mycoplasma genitalium Mycoplasma pneumoniae Ureaplasma urealyticum Mycoplasma pulmonis Streptococcus pneumoniae Streptococcus pyogenes Lactococcus lactis High G+C Mycobacterium tuberculosis
0 0 0 0 0 0 0
0 0 0 0 2: 2c, 3e 2: 2c, 3e 2: 2c, 3e
4: Cu6, Zn7, Zn10, ¡ 4: Cu6, Zn5, Zn10, ¡ 6: Cu6, Zn10, KdpG, ¡ 7: Cu6, Zn7, Zn10, KdpG, ¡ 2: Cu6, KdpG 5: Cu9, Zn7, ¡ 13: Cu9, Cu12, Zn7, Zn10, Zn12, ¡ 1: MgtA 1: MgtA 0 1: MgtA 4: Cu2, Me1, ¡ 3: Cu9, Zn10, ¡ 9: Cu9, Cu12, Zn8, Zn12, ¡
0
1: 3aii
0
Mycobacterium leprae Proteobacteria K subdivision Caulobacter crescentus Brucella melitensis Mesorhizobium loti Agrobacterium tumefaciens Sinorhizobium meliloti
0
1: 1b
12: Cu2, Cu2, Cu2, Zn9, Zn9, Zn12, Zn12, ¡ 4: Cu2, Cu2, Zn12, ¡
7: M1, M2, A3, A5¡ 7: A5¡ 10: A5¡ 10: A3, A5¡ 11: A3, A5¡
2: 2: 2: 2: 2:
Rickettsia prowazekii Rickettsia conorii Magnetococcus sp. L subdivision Neisseria meningitidis Ralstonia solanacearum
1: A5 4: A5¡ 14: M4, A1, A4, A5¡
1: 3b 1: 3b 4: 1a, 3f, 3f, 3f
3: Cu14, Zn9, ¡ 5: Cu3, Cu14, Zn2, Zn9, ¡ 7: Cu4, Cu14, Zn2, Zn12, ¡ 5: Cu3, Cu3, Cu14, Zn2, ¡ 9: Cu3, Cu3, Cu4, Cu14, Cu14, Zn2, Zn12, KdpG, ¡ 0 0 2: Cu7, Cu14
1: A5 15: M1, M3a, M3b, M4, A3, A5¡
0 2: 1a, 3f
2: Cu8, Cu14 4: Cu7, Cu14, Zn6, ¡
Prokaryotic species Archaea Crenarchaeota Aeropyrum pernix Sulfolobus solfataricus Sulfolobus tokodaii Pyrobaculum aerophilum Euryarchaeota Archaeoglobus fulgidus Halobacterium sp. Methanobacterium thermoautotrophicum Methanocaldococcus jannaschii Pyrococcus horikoshii Pyrococcus abyssi Pyrococcus furiosus Thermoplasma acidophilum Thermoplasma volcanium Bacteria Early-branching Aquifex aeolicus Thermotoga maritima Deinococcus radiodurans Cyanobacteria Synechocystis sp. Nostoc sp. Gram-positive bacteria Low G+C Bacillus subtilis Bacillus halodurans Listeria innocua Listeria monocytogenes Staphylococcus aureus Clostridium perfringens Enterococcus faecalis
FEMSRE 781 2-6-03
3f 2c 3f 2c, 3f
2g, 3e, 3f, 3f 2g, 3c, 2g, 3c, 2f, 2h, 3d 3d
3e, 3f 3e 3e 3e
2e, 3b 2h, 3c 2g, 3c 1a, 3c 1a, 3c
Cu1, Me0 Cu1, Cu16 Cu1 Cu1
MgtA, ¡
Cyaan Magenta Geel Zwart
0 0 0 0 0 0 0
0
CHR5 0 CHR5, CnrT CHR1 CHR5, NreB 0 0 NreB
CHR2, CHR5, CnrT
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
321
Table 1 (Continued). Prokaryotic species
RND proteinsb
CDF proteinsc
P-type ATPasesd
Other transport systemsd
Ralstonia metallidurans
20: 2UM1, 2UM2, 3UM3a, 3UM3b, 2UM4, A5¡
3: 1a, 2h, 3f
10: Zn6, Zn6, Zn6, Cu4, Cu7, KdpG, ¡
CHR1, 2UCHR2, CHR5, CnrT
2: A5¡ 3: M3b, M4, A5
2: 2c, 3b 0
3: Cu15, Cu15, KdpG 3: Cu8, Cu14, Zn7
0 0
7: M4, A5¡ 6: A5¡ 4: A5¡ 7: A1, A5¡ 0 1: A5 1: A5 13: M1, A5¡ 7: A1, A1, A5¡ 5: A1, A5¡
2: 2: 2: 2: 0 0 2: 3: 1: 1:
4: 6: 5: 5: 0 1: 1: 7: 3: 0
0 0 0 CnrT 0 0 0 CHR1 CHR1 0
2: M5, A4
2: 2b, 3f
1: MgtA
0
0
3: 2b, 3ai, 3aii
0
0
0 0 0
0 0 0
1: Zn11 1: Zn11 1: Zn11
0 0 0
1: A2 0
0 0
0 1: Cu11
CHR4 0
O subdivision Campylobacter jejuni Helicobacter pylori Q subdivision Escherichia coli Salmonella typhimurium Salmonella enterica Yersinia pestis Buchnera sp. Haemophilus in£uenzae Pasteurella multocida Pseudomonas aeruginosa Vibrio cholerae Xylella fastidiosa Flexible Bacteria CFB group Porphyromonas gingivalis Green sulfur bacteria Chlorobium tepidum Chlamydiales Chlamydophila pneumoniae Chlamydia trachomatis Chlamydia muridarum Spirochaetales Borrelia burgdorferi Treponema pallidum
2g, 2g, 2g, 2g,
3b 3b 3b 3b
2g, 3b 1a, 2h, 3b 3b 2h
Cu8, Cu3, Cu8, Cu8,
Zn3, Cu8, Zn3, Zn3,
KdpG, MgtA Zn3, KdpG, ¡ ¡ ¡
Cu8 Cu8 Cu7, Cu14, Zn6, Zn12, ¡ Cu8, Cu14, Zn3
a The comprehensive microbial resource at http://www.tigr.org and the R. metallidurans sequencing homepage of DOE/JGI at http://www.jgi.doe.gov/ JGI_microbial/html/ralstonia/ralston_homepage.html was used. A BLAST search [280^282] was performed against the ‘hook’ proteins CzcA from R. metallidurans (RND family), CzcD from R. metallidurans (CDF family), CadA from S. aureus (CPx type ATPases), ChrA from R. metallidurans (CHR family), NreB from R. metallidurans strain 31A and CnrT from R. metallidurans. Sequences with an E value smaller than 0.01 were collected and analyzed in two ways. Firstly, the occurrence of speci¢c signatures was used to pre-group the proteins into putative functional clusters. Secondly, assignment of a sequence to a group of proteins was cross-checked by multiple alignments performed with BioTechnix3d (http://www.biotechnix3d.com). For RND proteins, sequences with low similarity to CzcA and a signature of a RND protein (‘DD motif’) were not further analyzed. Likewise, analysis of a single genome was not further pursued when a magnesium- or potassium-transporting P-type ATPase was encountered. ‘¡’ indicates the presence of such non-characterized RND- and P-type proteins that were non-RND and non-CPx-type proteins, respectively. For RND-, CDF-, and P-type proteins, the total number of the respective proteins encoded by the particular genome is listed. After a colon, the association of the single proteins to a group within the protein family is given (see below). For CHR proteins, the number of these proteins is given and the assignment to the ¢ve groups of these proteins (similarity tree not shown). CnrT and NreB proteins were only mentioned. A zero (0) indicates that the organism does not contain any protein from the respective protein group or family. b For RND proteins, ‘M’ refers to the respective cluster of the HME-RND proteins, ‘A’ to the respective cluster of the HAE-RND proteins (Table 2). c Likewise, the cluster assignments of the speci¢c CDF proteins (Table 3) is indicated. d The data for the P-type ATPases (Cu, Cu-CPx-type ATPase ; Zn, Zn-CPx-type ATPase. MeO, MgtA and KdpG are other P-type proteins, the number after ‘Cu’ or ‘Zn’ indicates the assignment to a group within the respective subfamily and a possible ¢ne di¡erentiation of function) and the other protein groups are not shown.
AcrA, a trimeric form has also been determined in crosslinking experiments [53]. In many CBA protein complexes, loss of the MFP or the RND protein abolishes the respective resistance [55,90,94,97,154,155]. In contrast, loss of the OMF usually has a relatively moderate in£uence [55,97]. Since many Gram-negative bacteria contain a variety of OMF-encoding genes, another OMF may take over and complement for the loss of the native OMF. TolC is known to interact with a variety of transenvelope e¥ux systems that are driven by RND proteins or even ABC export systems [48]. The CnrC protein could be substituted partially in
FEMSRE 781 2-6-03
arti¢cially mixed operons with CzcC or NccC (G. Grass and D.H. Nies, unpublished). Chimeric MexCBA e¥ux pumps were also functional [156^158]. Thus, the OMF part of a CBA e¥ux complex is not involved in the substrate speci¢city, although its overproduction might lead to increased resistance to antibiotics [159]. TolC is the ¢rst OMF for which the structure is known at high resolution [160]. A TolC homotrimer forms a tubelike channel with a length of 14 nm and an outer diameter of 3.5 nm. This tube is hollow, allowing substrates to pass. The tube is su⁄cient in length to span the entire periplasmic space and the outer membrane, connecting the peri-
Cyaan Magenta Geel Zwart
322
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
plasmic domains of the RND protein with the outside. Insertion in the outer membrane and the ability to form a channel has been determined for OprM and TolC [48,161,162]. OprM formed a channel for small molecules, but was only able to allow transport of antibiotics after protease treatment [163]. This may indicate that the iris-like structure at the periplasmic end of the OMF tube [160] could serve as a gate, controlling access of the substrates from the RND protein. The top of the headpiece of AcrB opens like a funnel, where TolC might directly dock into AcrB [151]. This funnel was connected to the central cavity of AcrB by a pore that may allow substrate transport from the central cavity to TolC. All these pores and cavities in RND and OMF proteins may serve as microenvironments explaining a common mechanism of action of CBA e¥ux systems : an active accumulation of a substrate into a tube or cavity may yield a high substrate concentration inside this microenvironment allowing a rapid discharge of the substrate to the outside by di¡usion (Fig. 3). MFPs interact with their RND proteins and are attached to the cytoplasmic membrane by a hydrophobic residue, some by a hydrophobic amino-terminus [53,164]. MFPs contain a £exible coil-coil structure that may allow a £exible contact between the RND pump and the OMF tube [52]. A recent structure proposal based on liquid layer crystallization [165] describes AcrA as a doughnut-like structure with an inner diameter of 3 nm. Andersen et al. have suggested that MFPs may serve as adapter proteins between RND or other export pumps and OMF tubes [48]. MFPs probably do not fuse the cytoplasmic and the outer membrane directly, but they attach the OMFs to the pump proteins (Fig. 3). Thus, OMF and MFP proteins have a rather static function during CBA-mediated transenvelope e¥ux. Re-
cent ¢ndings indicate that the large periplasmic loops (LPL) of HAE-RND proteins have a function in substrate recognition. When the LPLs of AcrD were exchanged with the corresponding LPLs from AcrB, the chimeric protein displayed the substrate speci¢city of AcrB rather than AcrD [166]. Similar results were obtained when chimeric AcrB^MexB proteins were constructed [167]. Unfortunately, chimeric CzcA^CnrA proteins were not stable (K. Helbig, A. Legatzki and D.H. Nies, unpublished) and similar experiments could not be performed for these HMERNDs. Methionine residues in the LPL of CusA were essential for CusA function as mentioned above [97]. Similarly, essential amino acids probably involved in substrate recognition were identi¢ed in both LPLs of MexD [168]. Moreover, the LPLs of MexF were also involved in correct assembly of this CBA e¥ux pump [169], assigning a second essential function to the headpiece of RND proteins. Taken together, the RND proteins are indeed the central and most important component of CBA e¥ux systems: they mediate the active part of the transport process, determine the substrate speci¢city and are involved in the assembly of the transenvelope protein complex. Thus, distribution of RND proteins in the kingdoms of life may give some clues about the distribution and frequency of RND-mediated metal resistance. 2.7. Distribution of RND proteins in the kingdoms of life The genomes of 63 prokaryotes from both kingdoms (Archaea and Bacteria, tigrblast.tigr.org/cmr-blast, [170]) were compared to the genome of R. metallidurans (http://www.jgi.doe.gov/JGI_microbial/html/ralstonia/ ralston_homepage.html, Table 1). All RND proteins with
Table 2 Groups of heavy metal-transporting RND proteins Name HME proteins HME1 (M1) HME2 (M2) HME3a (M3a) HME3b (M3b) HME4 (M4) HME5 (M5) HAE proteins HAE1 (A1) HAE2 (A2) HAE3 (A3) HAE4 (A4) HAE5 (A5)
Signature motifa
Predicted substrate
Examplesb
DFG-DGA-VEN DFG-DGA-VEN GFG-D(G,S,A)(S,A)-(V,M)EN (A,g)(I,L)G-D(G,A,s)-VEN A(I,V)G-DA(A,s)-(V,I)(E,d)N AIG-DDX-(M,V)EN
Zn2þ , Co2þ , Cd2þ Ni2þ , Co2þ Divalent cations? Monovalent cations ? Cuþ , Agþ Ni2þ ?
CzcA, CztA, HelA CnrA, NccA RSA1040 HP0969 SilA, CusA All7631
X(V,L)G-D(N,G)A-X(D,E)N XXG-D(D,X)(S,a)-X(D,E)(N,s) XXG-DDA-XEN XXG-D(D,S,N)(S,A)-XE(N,t) AIG-DDA-XEN
Unknown Unknown Unknown Unknown Organic molecules
Alr1656 YerP NolG Alr5294 AcrB
a
For better orientation, the amino acid residues corresponding to G404 , D408 , E415 , and N416 of CzcA are shown in bold. The dash corresponds to three mostly hydrophobic amino acid residues. Amino acids shown in parentheses are alternatives for the same position. Amino acid names in lower-case letters indicate rare choices, X is any amino acid. b Representative sequences are from R. metallidurans strains CH34 (CzcA, CnrA) and 31A (NccA), Ralstonia solanacearum (RSA1040), Pseudomonas £uorescens (CztA), Legionella pneumophila (HelA), Helicobacter pylori (HP0969), Salmonella enterica pv. typhimurium (SilA), E. coli (CusA, AcrB), Nostoc sp. (All7631, Alr1656, Alr5294), Bacillus subtilis (YerP), and S. meliloti (NolG).
FEMSRE 781 2-6-03
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
su⁄cient homology to CzcA (E value 6 0.01 in the E value of the BLAST score) were selected and compared to CzcA and the 50 protein sequences in the databases (1 034 995 sequences) most similar to CzcA. Multiple alignments yielded homology trees (data not shown) that were used to group and subgroup the RND proteins (Table 2). Moreover, the occurrence of the Czc-, Cus-, and Mex-speci¢c consensus motifs DFG/D/EN, AIA/D/EN and DD, respectively (see above), was used to con¢rm the group assignments (Table 2). HME- or HAE-RND proteins were found in low frequency in early-branching bacteria (Table 1), but Aquifex aeolicus contains members of both RND protein families. RND proteins are very rare in £exible bacteria. Since spirochetes contain members of the unusual HAE3 family [23], this observation may indicate a limitation of the method used. Gram-positive bacteria are the third bacterial group with a low frequency of RND proteins (Table 1). Given the current hypothesis about the function of RND proteins as transenvelope e¥ux pumps, a CBA ef£ux complex should not function in a Gram-positive cell wall. However, CzcA alone mediates a small degree of resistance [171], thus, RND proteins in Gram-positive bacteria might function as single-subunit e¥ux systems. All RND proteins found in Gram-positive bacteria grouped in the HAE2 cluster (identical with the HAE2 protein family [23]) that appeared only in these bacteria and in Borrelia burgdorferi. This indicates that the mechanism of action of the RND proteins in Gram-positives might be di¡erent from that in most Gram-negative bacteria. Cyanobacteria and proteobacteria contained many RND proteins from both families (Table 1). Exceptions are the symbiotic bacterium Buchnera with no RND protein and three pathogenic bacteria with only one RND (Rickettsia prowazekii, H. in£uenzae, Pasteurella multocida). Most RND proteins were found in R. metallidurans strain CH34 (20 putative or candidate genes, some of them not complete), followed by its phytopathogenic relative Ralstonia solanacearum (15 genes), Magnetococcus (14 genes) and P. aeruginosa (13 genes). Thus, R. metallidurans contains more RND proteins than every other bacterium in the analyzed group. Most of the many RND proteins in cyanobacteria and proteobacteria belong to the HAE family. When bacteria contain only a few RND proteins, these are HAE-RND proteins in most cases. Examples of bacteria that exclusively harbored HAE-RND proteins are H. in£uenzae (with one RND protein that has already been characterized [118]), R. prowazekii, C. jejuni, P. multocida, Thermotoga maritima, Deinococcus radiodurans, B. burgdorferi, and Gram-positive bacteria. When bacteria contain putative HME-RND proteins at all, most of these are putative copper/silver exporters of the HME4 group (Tables 1 and 2). Examples are A. aeolicus (two genes, HME3b and HME4), Magnetococcus (HME4), Helicobacter pylori (HME3b and HME4) and
FEMSRE 781 2-6-03
323
E. coli (HME4 = CusA). A second group of bacteria harbored members of the highly unusual HME5 cluster, the cyanobacteria Synechocystis (one HME5 gene), Nostoc (two HME5 genes), Legionella pneumophila, and P. gingivalis (one HME5 gene). The HME5 cluster contains members of the HME-RND protein family by homology. These proteins exhibit the ‘DD’ motif in the TMH IV that is characteristic of HAE proteins, but contain some conserved polar residues in their TMH IV unique to this group (Fig. 2). There is only indirect evidence that these HME5 proteins might transport heavy metals, especially Ni2þ . A putative HME5 protein (sequence slr0794, http://www.kazusa.or.jp/cyano/Synechocystis) is part of an operon designated nrsBACD [172]. Expression of this operon is controlled by nickel via a two-component regulatory system NrsRS ( = RppAB) encoded upstream of nrsBACD on the other DNA strand. The genes in the nrsBACD operon refer to the genomic designations slr0793 (MFP), slr0794 (HME5), slr0795 (hypothetical protein), and slr0796 (putative NreB protein). It has been shown for NreB-like proteins that they are able to exert nickel resistance alone ([173], see below), so the nickel resistance assigned to the nrsBACD may be mediated by nrsD. It is not clear if nrsC is an OMF and if a complete CBA e¥ux complex can be encoded by this operon or parts of it. Therefore, con¢rmation of the HME5 proteins as nickel-exporting RND proteins has to await further experimental evidence. A third group of bacteria that contain putative HMERND proteins may harbor proteins of the HME1 cluster highly related to CzcA. In the analyzed genomes and the current databases, 10 HME1 sequences were found. These were proteins from R. metallidurans (CzcA from di¡erent strains [55,56]), P. aeruginosa (CzrA [101]), Pseudomonas £uorescens (CztA), a fragment from R. metallidurans (gene 848 on contig 456), R. solanacearum (Rsp 0493), Caulobacter (CC2390), two proteins from either pathovar of Xanthomonas campestris and HelA from L. pneumophila. Only these eight bacteria can be expected to exhibit Co2þ / Zn2þ /Cd2þ resistance based on RND-driven transenvelope e¥ux. The Cnr-related HME2 group is much smaller and contains only four members, two of them from R. metallidurans strain CH34 (CnrA and NccA), one from strain 31A (NccA [67]) and one from Caulobacter (CC2724). These three bacteria should exhibit RND-based nickel/cobalt/ cadmium resistance. Group HME4 contains 10 sequences of putative copper/ silver transport systems. These sequences came from E. coli (CusA [90,94,97]), Salmonella enterica pv. typhimurium (SilA [145]), two sequences from R. metallidurans and one sequence from R. solanacearum, Pseudomonas syringae, H. pylori, L. pneumophila, A. aeolicus and Magnetococcus sp. CusA has been introduced above and by Grass and Rensing [88], SilA by Silver [92]. The H. pylori copper export system has also been characterized [174]. Because
Cyaan Magenta Geel Zwart
324
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
of the similarity of the other eight sequences to CusA and SilA, these bacteria can be expected to detoxify copper and/or silver in a way that has been shown for CusA and SilA. The remaining HME3 cluster contains all sequences not closely associated with HME1, HME2 or HME4, but with a unique DSS motif in TMH IV (Fig. 2). There are two subgroups in this cluster (Table 2). Subgroup HME3a exhibits the DFG consensus motif and the proteins in this subgroup may transport divalent heavy metal cations. The ¢ve members of HME3a come from R. solanacearum, Xanthomonas axonopolis pv. citri, and three sequences from R. metallidurans strain CH34. The other subgroup, HME3b, shows mostly the AIG motif characteristic of transporters of monovalent cations. The ¢ve proteins of this cluster come from R. solanacearum, H. pylori, A. aeolicus and again three sequences from R. metallidurans. The sequence from A. aeolicus is unusual since it contains a GIG motif instead of AIG, which gives this putative protein (aq_469) an ancient touch. The relationship of RND proteins leads to a scenario of how heavy metal-transporting HME proteins may have evolved: the earliest branch of HME proteins are the HME5 proteins that still contain the DD motif of RND proteins, and which are much more diverse and widely distributed than HME proteins. Have HME proteins originated from HAE proteins ? The other four groups, HME1 to HME4, are currently found only in proteobacteria and A. aeolicus. This could indicate that the divalent heavy metal transport systems originated from an ancient monovalent transport system such as aq_469 from A. aeolicus during evolution of the proteobacteria.
3. The second layer of metal resistance: cation di¡usion facilitators (CDF family) serve as secondary cation ¢lters in bacteria CDF proteins form a protein family of metal transporters (TC 2.A.4.1.1^2 [46]) occurring in all three domains of life [175]. Primary substrates of CDF proteins are Zn2þ , but also Co2þ , Ni2þ , Cd2þ , and Fe2þ . CDF-mediated transport is driven by a concentration gradient, a chemiosmotic gradient, v8, vpH or a potassium gradient [176^ 178] (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished). First, the physiological function, biochemistry and substrate speci¢cities of the CDF protein family will be analyzed, before the distribution of CDF proteins in sequenced bacterial genomes will be compared to that of R. metallidurans strain CH34. 3.1. Physiological function and substrate speci¢city of CDF proteins All CDF proteins described so far in bacteria are involved in resistance to Zn2þ and other heavy metal cati-
FEMSRE 781 2-6-03
ons. The archetype of the family, CzcD from R. metallidurans strain CH34, was ¢rst described as a regulator of expression of the CzcCBA high resistance system [60], but CzcD is also able to mediate a small degree of Zn2þ /Co2þ / Cd2þ resistance in the absence of the high resistance CzcCBA system [179]. The presence of CzcD decreases the cytoplasmic concentration of these metals [179] which serve as inducers for czc expression (C. GroMe, A. Anton, T. Ho¡mann, S. Franke, G. Schleuder and D.H. Nies, unpublished). Heterologously expressed in E. coli, CzcD was driven by the proton motive force and e¥uxes Zn2þ and Cd2þ , but not Ni2þ and Co2þ (. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished). Cross-linking experiments indicated that CzcD is probably a dimer in vivo. Although CDF-encoding genes are found on the chromosomes of many sequenced organisms, the function of the proteins has only been characterized in a few bacteria. In Bacillus subtilis, the czcD gene is in an operon together with the gene for a TrkA dehydrogenase [180] and mutant studies show some involvement in cobalt resistance [177]. CzcD from B. subtilis was driven by the proton motive force as well as by a potassium gradient, and the exchange seemed to be electroneutral. Substrates for this CzcD protein were Zn2þ , Co2þ and Cd2þ [176]. Two groups cloned from Staphylococcus aureus a small determinant containing a CDF-encoding gene and a regulator. The CDF protein was named ZntA (now RzcB [181]) or CzrB [182]. Both proteins are nearly the same size (325 or 326 aa) and di¡er only in nine positions; they are essentially identical. CzcD from S. aureus is involved in zinc and cobalt resistance. A CDF protein from Thermus thermophilus was shown to mediate zinc and cadmium, but not cobalt resistance [183]. Recently, one of the two CDF proteins from E. coli was characterized, ZitB (the ybgR gene product), which is also a zinc resistance system [184]. This protein diminishes the cellular accumulation of Zn2þ and is driven by a potassium gradient in addition to the proton motive force [185]. E. coli contains another putative CDF-encoding gene, the enigmatic yiiP. Expression of this protein is induced by zinc, deletion of the gene does not lead to an increase in zinc resistance, but overexpression has no e¡ect [184]. The function of this protein is therefore unknown. The bacterial CDF proteins mentioned so far are mainly Zn2þ /Co2þ /Cd2þ -exporting cytoplasmic membrane proteins. Magnetotactic bacteria are able to sense the magnetic ¢eld of the Earth with the help of speci¢c intracellular organelles, the magnetosomes, which contain the mineral magnetite (Fe3 O4 ) as ¢eld-sensing material. MamB was described as a CDF protein located in the magnetosome membrane [186]. Its function is probably to transport iron into this organelle. Magnetosome-like magnetite was discussed as a marker for remnants of ancient life forms from the planet Mars [187,188]. That would increase the distribution of CDF proteins to anoth-
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
er planet! However, a magnetic ¢eld is the prerequisite for the occurrence of magnetotactic life forms. Since the magnetic ¢eld of Mars was only 1/30 as strong as the ¢eld of our Earth when the putative Martian fossils were liberated by a giant impact [189] and this is rarely su⁄cient to drive any magnetotactic behavior, we probably have to wait for better evidence for life forms and CDF proteins from other planets. Usually, bacteria contain one or a few CDF-encoding genes. In eukaryotes, the number of CDF-encoding genes increases in parallel with evolution from yeasts to animals and plants. Saccharomyces cerevisiae contains ¢ve CDFencoding genes [190]. ZRC1 was initially described in the same year as CzcD and this protein mediates resistance to zinc and cadmium [191]. Another yeast CDF, COT1, mediates cobalt resistance [192], but ZRC1 and COT1 can complement each other [193]. Both can even substitute for CzcD in its transport and regulatory function in R. metallidurans [179]. ZRC1 and COT1 are located in the vacuolar membrane and transport the cations Zn2þ , Cd2þ and Co2þ into this compartment for detoxi¢cation and possible future re-use [194^196]. This explains why both proteins are also needed to survive zinc starvation [194,197,198]. Moreover, since the cytoplasm is not detoxi¢ed in ZRC1 knock-out strains, the cell reacts with an increased glutathione level [199]. Similar to the function of CzcD from R. metallidurans, the postulated regulatory function of ZRC1 is therefore an indirect consequence of its transport function. A third S. cerevisiae protein, MSC2 (YDR205W), is also involved in zinc transport [200]. The respective deletion strain, however, accumulated excess zinc in the nucleus-enriched membrane fractions. Therefore, a function of MSC2 in zinc homeostasis of the nucleus was suggested [200]. The other two CDF proteins from S. cerevisiae are not involved in Zn2þ /Cd2þ /Co2þ transport, but these two proteins, MFT1 and MFT2, are probably Fe2þ transport systems located in the mitochondrial membrane [201]. That adds to the evidence that CDF proteins may transport Fe2þ . The ZRC1-related CDF protein Zhf from the ¢ssion yeast Schizosaccharomyces pombe is located in the endoplasmic reticulum and the nuclear envelope like MSC2 [202]. Deletion of zhf leads to increased zinc and cobalt sensitivity, but to increased cadmium and nickel resistance. The cytoplasmic content of zinc, cobalt and cadmium was diminished and detoxi¢cation of cadmium by phytochelatin synthesis was not increased [202]. Thus, this protein may be responsible for supply of the endoplasmic reticulum and the nucleus with zinc and/or cobalt. Involuntary transport of cadmium and nickel into the same compartment by Zhf could mediate increased toxicity of these cations. The currently known CDF proteins in higher animals are all zinc transport systems. ZnT1 is located in the cytoplasmic membrane and seems to be a zinc e¥ux system
FEMSRE 781 2-6-03
325
[203]. In contrast, ZnT2 transports Zn2þ into vesicles and is therefore comparable in function to ZRC1 and COT1 [204]. ZnT3 also transports Zn2þ into vesicles, but exclusively into synaptic vesicles [205]. The protein is most abundant in the zinc-enriched mossy ¢bers that project from the dentate granule cells to hilar and CA3 pyramidal neurons in the brain. Thus, ZnT3 seems to be responsible for the ability of these neurons to release zinc upon excitation [206]. Age-dependent low expression of ZnT3 leads to de¢ciencies of Zn2þ in synaptic vesicles of the mossy ¢ber pathway and induces glutamatergic excitotoxicity in the hippocampal neurons and the deterioration of learning and memory [207]. The fourth known CDF protein from higher animals, ZnT4, is similar to ZnT3 with respect to zinc transport and abundant expression in the brain, but this protein is also expressed in the mammary epithelia and responsible for excretion of zinc into mother’s milk [208]. The CDF family in mammals kept growing with ZnT5 that transports zinc into secretory granules in pancreatic L cells [209] and also plays an important role in maturation of osteoblasts and in maintenance of the cells involved in the cardiac conduction system [210]. ZnT5 proteins from mouse and man are twice the size of other CDF proteins. Since dimerization has been indicated for CzcD (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) and ZnT4 [211], gene duplication followed by gene fusion could have been responsible for the occurrence of ZnT5 and other large CDF proteins. Another large CDF protein from mammals was named ZTL1 (ZnT-like protein). It is located in the enterocyte apical membrane, e.g. in kidney [212]. Comparison of the 765-aa ZnT5 protein with the 523-aa ZTL1 proteins displays a 495-aa middle part that di¡ers in only two amino acid residues (99.6% identity). This middle part starts at M172 of ZnT5 that equals M1 of ZTL1, and extends to K666 of ZnT5. The 100-aa ZnT5 and 28-aa ZTL1 carboxy-termini are completely di¡erent. The genomic sequence encoding the ZTL1 protein consists of 11 exons separated by large introns with the ¢rst ones encoding possible alternative splicing [212]. The ZnT5 protein has a predicted size of 84 kDa, but its size in vivo was determined by Western blot analysis to be 55 kDa [209], which is similar to the predicted size of ZTL1 of 57.3 kDa. Thus, the original gene may encode at least two proteins by splicing of the RNA and post-translational processing of a pre-protein, and the two proteins may di¡er in function and localization. Mammals contain at least two more CDF proteins, ZnT6 and ZnT7. The protein ZnT7 is currently only putative (AY094606) awaiting publication of more data. ZnT6 may transport cytoplasmic zinc into the Golgi apparatus and its expression is regulated by zinc in kidney cells [213]. All these results suggest that intracellular zinc homeostasis in mammals is mediated by a variety of ZnTlike proteins that di¡er in intracellular localization and
Cyaan Magenta Geel Zwart
326
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
organ-speci¢c expression. These proteins transport zinc from the cytoplasm into another compartment, to the outside (ZnT1, ZnT4) like bacterial transport systems, or into vesicles (ZnT2, ZnT3, ZnT5) like ZRC1 and COT1, or into the Golgi apparatus (ZnT6) like Zhf. However, experiments with oocytes expressing ZTL1 indicate that CDF proteins may also be involved in uptake into the cell [212]. The ¢rst CDF protein from plants was ZAT [214]. Arabidopsis thaliana plants overexpressing ZAT (AtCDF1) exhibited enhanced Zn2þ resistance and strongly increased Zn2þ content in the roots under high Zn2þ exposure. Antisense mRNA-producing plants were viable, with a wildtype level of Zn2þ resistance and content [214]. The ZAT protein was puri¢ed and functioned as a zinc uptake system in several microbial model systems [178]. ZAT did not transport other heavy metal cations with the exception of a slow transport of cadmium. A. thaliana contains at least seven more CDF proteins [215]. As in mammals, a variety of CDF proteins may also be responsible for zinc homeostasis in plants. A ZAT-like gene was also cloned from the heavy metal hyperaccumulator plant Thlaspi caerulescens [216]. From the related plant T. goesingense, a CDF-encoding gene with interesting properties was cloned. Transcription of the MTP1 gene from this plant leads to a spliced and an unspliced mRNA. Both mRNAs encode proteins which di¡er within a histidine-rich putative metal binding domain and in their substrate speci¢city [217]. The protein translated from the unspliced mRNA confers high level tolerance to Cd2þ , Co2þ and Zn2þ , whereas the product of the spliced mRNA confers nickel resistance [217]. This is reminiscent of the ZnT5/ZTL1 situation in animals and of nickel as a possible substrate of Zhf. 3.2. Biochemistry of CDF proteins With transport of Ni2þ by the T. goesingense protein, the substrate spectrum of CDF proteins includes Fe2þ , Co2þ , Ni2þ , Zn2þ and Cd2þ . The ionic radii of these cations are 76 pm, 74 pm, 72 pm, 74 pm and 97 pm, respectively [21]. Of the other divalent heavy metal cations of biological importance, Mn2þ (80 pm) may also fall in the size range of possible CDF substrates. In contrast, Mg2þ (65 pm) and Cu2þ (69 pm) may be too small and Ca2þ (99 pm) and Hg2þ (110 pm) too big to serve as CDF substrates in general. Additionally, a⁄nity for a substrate binding site should also di¡er between substrate cations. The cations of the ¢rst transition period usually have a higher a⁄nity for free electron pairs of oxygen and nitrogen atoms, whereas the cations of the second and third transition periods prefer the ‘soft’ electrons of sulfur atoms. The majority of the CDF substrates are divalent metal cations of the ¢rst transition period with sizes of 74 Z 2 pm diameter with Cd2þ being a piggy-back rider due to its chemical similarity to Zn2þ .
FEMSRE 781 2-6-03
Most CDF proteins have a size of 300^400 aa. CzcD from R. metallidurans exhibits six transmembrane spans [179] and this can also be assumed as the rule for most other CDF proteins. CDF proteins contain many histidine residues that are located at the amino-terminus, the carboxy-terminus, or between transmembrane helices IV and V, or at all three locations. Histidine residues of CzcD bound two to three Zn2þ per CzcD polypeptide (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) and were also essential for transport of zinc by ZitB from E. coli [185]. Although the histidine-rich carboxy-termini of CzcD and ZitB were not essential for transport (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) [185], they could have a function as regulators of transport activity of these proteins. The CzcDs from B. subtilis and R. metallidurans and ZitB from E. coli are e¥ux pumps driven by the proton motive force or a potassium gradient (A. Anton, G. Grass, C. Rensing and D.H. Nies, unpublished) [176,177,180,185] and function of ZRC1 from S. cerevisiae also required a proton gradient [196]. In eukaryotic cells, however, not all membranes harbor a proton, sodium or potassium gradient. Therefore, some CDF proteins may function as simple di¡usion facilitators in these organisms. Cation transport by the ZAT protein from A. thaliana was not stimulated in proteoliposomes by a proton gradient. Moreover, ZAT functioned as an uptake system in several microbial cells [178]. Therefore, cation transport by CDF proteins could be either an antiport reaction against a chemiosmotic gradient or a facilitated di¡usion driven by the concentration gradient of the respective cation. 3.3. Distribution of CDF proteins in the living world In a similar manner to the RND proteins, the distribution of CDF proteins in sequenced bacterial genomes was analyzed (Table 1). The proteins in these genomes and in current databases were compared in multiple alignments, and used to group and subgroup these proteins (Table 3). Three groups of CDF proteins could be established. Group 1 contained two subgroups, 1a and 1b. Cluster 1b is composed of the large proteins ZnT5, ZTL1 and MSC2, together with three other sequences. This assignment may be an artefact of their size and the multiple cluster algorithm used. Cluster 1a, on the other hand, consists of six sequences of putative proteins from proteobacteria signi¢cantly related to each other. This might indicate a common mechanism of function. The remaining CDF proteins fall into two groups, each containing several subgroups. Group 2 contains most of the known zinc-transporting CDF proteins in bacteria (CzcDs, ZitB), eukaryotes and archaea (Table 3). Cluster 2d occurs exclusively in eukaryotes and also contains the nickel-transporting plant protein that is highly related to ZAT from A. thaliana [215]. Group 2a, which branches o¡ ¢rst, is found in Crenarchaeota (Table 3). This might in-
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
dicate that the mainly zinc-transporting CDF proteins may have originated very early in evolution. Group 3 is assembled around YiiP from E. coli. No data are available for proteins in this group with the exception of MamB, the probable Fe2þ -transporting CDF protein from magnetosome membranes of magnetotactic bacteria [186]. In aerobic environments, iron occurs mainly as the insoluble Fe3þ , however, in anaerobic or micro-aerobic environments, a considerable concentration of Fe2þ may be present. Fe2þ can also be a substrate of CorA-like magnesium uptake systems [198]. Similar to Zn2þ , a second-step ¢ltering of Fe2þ against Mg2þ could be required, which yields a speculative initial idea for the function of the group 3 CDF proteins. When the distribution of CDF proteins is analyzed (Table 1), most prokaryotes contain two putative CDF proteins, one from group 2 and one from group 3. Of the 64 genomes analyzed, 24 contained two putative CDF proteins, 15 one, 16 none, seven three and only two contained four. Prokaryotes with two putative CDF proteins always harbored one member of group 3 plus a putative CDF from group 2 (75%), group 1 (12.5%), or a second protein from group 3 (also 12.5%). Thus, the general out¢t of a bacterium is one group 3 and one group 2 CDF protein. The prokaryotes with three or four CDF proteins always contained a group 3 protein, exclusively or in addition to a group 2 or a group 1 protein. This could indicate that the gene for a group 3 CDF protein is less likely to have been sacri¢ced during evolution than that of a group 2 zinctransporting CDF protein. If group 3 proteins are indeed
327
iron transporters, this may mirror the higher biological importance of iron in comparison to zinc.
4. The third layer of heavy metal resistance : P-type ATPases are the basic defense against heavy metal cations P-type ATPases constitute a superfamily of transport proteins that are driven by ATP hydrolysis [218]. Members of this family occur in all three kingdoms of life. Substrates are inorganic cations such as Hþ , Naþ , Kþ , Mg2þ , Ca2þ , Cuþ , Agþ , Zn2þ and Cd2þ [144]. An individual P-type ATPase can either import its substrate from the outside or from the periplasm to the cytoplasm (importing P-type ATPase) or export it from the cytoplasm to the outside/periplasm (exporting P-type ATPase). With respect to heavy metal homeostasis, P-type ATPases may be important for two reasons. Firstly, import systems for macroelements such as Mg2þ may also import heavy metals [219]. Secondly, exporting P-type ATPases may detoxify heavy metal cations by e¥ux. Members of the huge P-type ATPase superfamily that transport heavy metal cations carry a conserved proline residue, preceded and/or followed by a cysteine residue. This motif is essential for the function of at least one member of this family [220]. Work of the last few years on these ‘CPx-type’ or soft metal-transporting ATPases has been reviewed before [89] and will be partially covered in other reviews in this issue. A survey of heavy metal
Table 3 Groups of CDF proteins Name CDF1 1a 1b CDF2 2a 2b 2c 2d 2e 2f 2g 2h CDF3 3a 3b 3c 3d 3e 3f
Occurrence
Predicted substratea
Examplesb
Proteobacteria (K, L, Q) Bacteria, yeasts, mammals
Zn2þ ? Zn2þ
PA1297 MSC2, ZnT5, ZTL1, ZnT7
Crenarchaeota Euryarchaeota, £exible bacteria, Gram-positive bacteria Bacteria and thermophilic Euryarchaeota Multicellular eukaryotes Proteo- and cyanobacteria Gram-positive bacteria Bacteria Bacteria
Zn2þ ???? Zn2þ ??? Zn2þ ?? Zn2þ , (Ni2þ ) Zn2þ ? Zn2þ ? Zn2þ Zn2þ , Co2þ , Cd2þ
PAE2715 Spr1672 PH0896 ZAT, ZnT2 and 3 All7610 SA0155 ZitB, CzcD_Bacsu CzcD_Ralme
Bacteria Proteobacteria (K, N) Bacteria Gram-positive bacteria (low G+C) Thermotoga and Gram-positive bacteria (low G+C) Prokaryotes
unknown unknown unknown unknown unknown Fe2þ ?
Aq_2073 YiiP DR1236 EF0859 YeaB, YbdO MamB
a
The number of question marks after the predicted substrates indicates the probability of this prediction ; the more question marks, the lower the probability for the putative substrate. b Representative sequences are from P. aeruginosa (PA1297), S. cerevisiae (MSC2), Pyrobaculum aerophilum (PAE2715), Streptococcus pneumoniae (Spr1672), Pyrococcus horikoshii (PH0896), A. thaliana (ZAT), mammals (ZnT2, ZnT3, ZnT5, ZnT7, ZTL1), Nostoc (All7610), S. aureus (SA0155), E. coli (ZitB, YiiP), B. subtilis (CzcD_Bacsu, YeaB, YbdO), R. metallidurans (CzcD_Ralme), A. aeolicus (Aq_2073), D. radiodurans (DR1236), Enterococcus faecalis (EF0859), Magnetococcus gryphiswaldensis (MamB).
FEMSRE 781 2-6-03
Cyaan Magenta Geel Zwart
328
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
cation-transporting CPx-type ATPases in this section will therefore concentrate on data that are important in understanding the interconnections between CPx-type ATPases and members of the other two metal e¥ux protein families, CDF and RND. 4.1. Families of CPx-type ATPases and their substrates One family of CPx-type ATPases is involved in export of Cuþ (but not Cu2þ [220]) and Agþ (Cu-CPx-type ATPases [88,89]. Members of this group were characterized in a variety of organisms, in the archaeon Archaeoglobus fulgidus [221], the cyanobacterium Synechocystis [222], the Gram-positive bacteria B. subtilis [223], Enterococcus faecium [224], Enterococcus hirae [225^227], Streptococcus mutans [228], Lactobacillus sakei [229], the proteobacteria Rhizobium leguminosarum, Sinorhizobium meliloti [230], P. putida [231] and E. coli [87]. Cu-CPx-type ATPases also occur in lower eukaryotes as Dictyostelium [232], Cryptosporidium parvum [233] and Candida albicans [234]. Seven of the 45 genes from A. thaliana that may encode Ptype ATPases are putative CPx-type ATPases [235]. One of these proteins may transport copper and is required for ethylene signaling in the plant [236]. In mammals including human, miss-expression or mutant proteins of CuCPx-type ATPases were assigned to the copper homeostasis defects Menkes’ and Wilson’s diseases [237,238]. This demonstrates a broad distribution of Cu-CPx-type ATPases in the living world. A second family transports Zn2þ , Cd2þ and Pb2þ (ZnCPx-type ATPases), again, mainly from the cytoplasm to the outside or to the periplasm. The very ¢rst member of the CPx-type ATPases described was CadA, a cadmium resistance-mediating protein encoded on a plasmid from S. aureus [13,239]. Examples from other bacteria such as B. subtilis [240], Listeria monocytogenes [241], P. putida [242], S. maltophilia [243], E. coli [244^248], and the yeast S. cerevisiae [249] followed, but the distribution of ZnCPx-ATPases seems to be not as broad as the distribution of the Cu-CPx-type ATPases. The substrate speci¢city of CPx-type ATPases is Cuþ / Agþ or Zn2þ /Cd2þ /Pb2þ . A CPx-type ATPase from the cyanobacterium Oscillatoria brevis uses heavy metal cations from either group as substrates [250]. Overexpression of CadA from H. pylori has some e¡ect on cobalt resistance, so Co2þ may also be a substrate of Zn-CPx-type ATPases [251], as has been suggested before for a P-type ATPase from Synechocystis [252]. There is, however, no clear evidence for the transport of Ni2þ by a CPx-type ATPase. A publication claiming the cloning of a P-type ATPase [253] involved in nickel resistance deals with a possible nickel uptake protein from a completely di¡erent protein family (NiCoT) not known to be involved in metal resistance, but in metal uptake [254^256]. The direction of transport of CPx-type ATPases in bacteria is mostly export to the outside or the periplasm, but
FEMSRE 781 2-6-03
the other direction of transport also seems to occur. The ¢rst examples of Cu-CPx-ATPases came from the Grampositive bacterium E. hirae. Two Cu-CPx-type ATPases were found, CopA involved in copper uptake and CopB in copper resistance by e¥ux [257^261]. These are reviewed in [283]. Physiological evidence indicates that two CPx-type ATPases from the cyanobacterium Synechocystis, CtaA and PacS, may be involved in copper uptake [284]. They facilitate the synthesis of the copper-dependent electron transfer factor plastocyanin [222], although disruption of pacS decreased copper tolerance. Disruption of ctaA diminished the amount of cell-bound copper [222], clearly indicating a function of CtaA in copper uptake. There is also evidence from B. subtilis that the ZnCPx-type ATPase ZosA is a zinc uptake system [262]. It seems to be easy for evolution to change an e¥ux system into an uptake system when this is required by the speci¢c environment the cell struggles to survive in. 4.2. Structure and function The membrane topology of CPx-type ATPases was analyzed exemplarily for the ¢rst described member, CadA from S. aureus [263]. The protein contains eight transmembrane segments with the conserved, eponymous CPC motif located in the sixth of these segments. The ATP binding domain is situated in a large cytoplasmic domain that follows this transmembrane domain. The amino-terminal 109 aa carry a CXXC motif that is also present in most members of the CPx-ATPases [263]. Although heavy metal cations are able to bind to the amino-terminal domain [264], this domain is not essential for transport [248] and a regulatory function has been discussed [241]. When the HME-RND proteins were discussed above, the problems in detoxifying thiol-bound heavy metal cations were mentioned. The complex binding constant for Cuþ ^glutathionate complexes is 1039 [75]. The energy to remove the Cuþ cation from this complex is vGPo = 3RT ln1039 = 3226 kJ mol31 at 30‡C. Thus, to detoxify heavy metal cations bound to thiols, paired cysteine residues in the substrate binding or substrate funneling site are required and additionally a lot of energy. CPx-type ATPases exhibit both features and it has already been demonstrated that the substrates in vivo are likely metal^thiolate complexes rather than the ‘free’ metals [247]. 4.3. Distribution of CPx-type ATPases in prokaryotes If CPx-type ATPases clean out the thiol-bound heavy metal pool, they should occur even more frequently in the living world than CDF or RND proteins. In a similar manner to the approach described above for the other two protein families, the distribution of putative CPxtype ATPase in prokaryotic genomes was analyzed. A number of 16 groups of Cu-CPx-type ATPases and 12 groups of Zn-CPx-type ATPases were found by multiple
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
alignments (data not shown). Members of the other families of P-type ATPases such as MgtA-like magnesium transporters and KdpG-like potassium transporters were only considered as examples. Generally, the KdpG-like proteins were more related to the CPx-type ATPases than the MgtA-like proteins (data not shown). Starting from the 64 prokaryotes analyzed, only eight species did not contain a gene for a putative P-type ATPase at all (Table 1). Five species contained at least one P-type ATPase, but none of the CPx-type. Most (29 species) had CPx-type ATPases of both families, the Cu-CPxtype and the Zn-CPx-type ATPases. If prokaryotes harbored only a member of one of these two families, this was mostly a member of the Cu-CPx-type ATPases (18 species) and in just four cases only a member of the Zn-CPx-type ATPases. Thus, nearly half of the species analyzed contained members of both CPx families, which indicates the importance of the resistance mediated by these proteins. As with the RND proteins, export of copper/silver was more important than e¥ux of zinc/cadmium/lead, mirroring the di¡erent toxicities of these cations. R. metallidurans contains genes or candidate genes for two Cu-CPx-type ATPases, three Zn-CPx-like ATPases and ¢ve other P-type ATPases giving a total number of 10 (Table 1). One of the Zn-CPx-type ATPases, PbrA, is part of a plasmid-mediated lead resistance determinant in R. metallidurans [265], the other two, ZntA and CadA, are involved in resistance to Zn2þ and Cd2þ [68]. This complement of CPx-type ATPases is not unusually high for a bacterium; cyanobacteria, mycobacteria or Rhizobium species exhibited similar numbers (Table 1).
5. Other heavy metal export systems: CHR protein family, NreB, CnrT Before the data about RND, CDF, and CPx proteins are assembled into a picture that describes heavy metal detoxi¢cation in prokaryotic cells, three small groups of transporters will be introduced. All three are linked to resistance determinants that may allow a bacterium to survive in the speci¢c nickel-, cobalt- and chromate-rich serpentine environment [266]. ChrA-like proteins of the CHR protein family [267] detoxify the oxyanion chromate, while NreB and CnrT are nickel e¥ux systems.
329
mode of their action [267]. CHR proteins are probably chromate e¥ux pumps [268,269] driven by the chemiosmotic gradient. Since the export of an anion is in the direction of the electric ¢eld of the proton motive force, v8 alone is su⁄cient to drive chromate e¥ux. The CHR proteins of 64 prokaryotic genomes were compared by multiple alignments and grouped into ¢ve categories that may mirror di¡erences in the physiological or biochemical modes of action. Only a quarter of the bacterial genomes contained at least one putative CHR member (Table 1). Two CHR-encoding genes were found in R. solanacearum and Bacillus halodurans, four in R. metallidurans. Again, R. metallidurans contained the highest number of members of a protein family involved in heavy metal export. Of the 21 putative CHR proteins analyzed, involvement in chromate e¥ux has only been shown for two of the four proteins from R. metallidurans [267,270^273], the ChrA protein from P. aeruginosa [1,268,269,274], and Synechococcus, a cyanobacterium roughly related to Synechocystis [275]. 5.2. NreB-like proteins In three bacterial species (Synechocystis sp., Nostoc sp., S. meliloti), the presence of a CHR protein was accompanied by the presence of a NreB-like protein in the genome (Table 1), in Magnetococcus it appeared without a CHR protein. The NreB protein has been characterized in R. metallidurans strain 31A [173]. This bacterium contains two distinct nickel resistance determinants on its native megaplasmids pTOM8 and pTOM9: ncc [67], which was mentioned above, and nreB [173]. Expression of the nreB gene was speci¢cally induced by nickel and conferred nickel resistance in both strain 31A and E. coli. Cells expressing nreB showed decreased accumulation of Ni2þ , suggesting that NreB mediates nickel e¥ux. The NreB-like proteins have a size of about 400 aa and contain probably 12 transmembrane spans. One putative protein from the enterobacterium Hafnia alvei is much smaller and may be the result of a sequencing error. The other six NreB-like proteins found in current databases all exhibit a characteristic, histidine-rich carboxy-terminus that resembles a natural His tag. 5.3. CnrT-like proteins
5.1. CHR proteins detoxify chromate The CHR family contains proteins that are bound to the cytoplasmic membrane by 10 transmembrane spans [267]. The protein probably originated by gene duplication followed by gene fusion. Several bacteria contain operons that comprise two successive genes encoding the aminoand the carboxy-terminal parts, respectively, of a fulllength CHR protein. This led to the hypothesis that both halves of the full-length proteins may di¡er in the
FEMSRE 781 2-6-03
Only three proteins in the analyzed genomes and the current databases were similar to CnrT from R. metallidurans. These were putative proteins from Yersinia pestis, R. solanacearum and M. loti (Table 1). The cnrT gene is located directly downstream of the cnr cobalt^nickel resistance determinant in plasmid pMOL28 of R. metallidurans and encodes a small degree of nickel resistance by itself (G. Grass, S. Ku«hnemund, B. Fricke, K. Sutter, G. KrauM and D.H. Nies, unpublished) [65]. CnrT-like proteins ex-
Cyaan Magenta Geel Zwart
330
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
hibit 12 or 14 transmembrane spans and, like NreB proteins, a histidine-rich carboxy-terminus. CnrT- and NreB-like proteins do not exhibit any potential ATP binding sites and are thus probably driven by the proton motive force, like CDF proteins. Moreover, all three groups of proteins contain cytoplasmic loops of the polypeptide chain rich in putative metal binding regions and export their substrates across the cytoplasmic membrane. Therefore, CnrT- and NreB-like proteins may act as secondary ¢lter proteins for Ni2þ similar to CDF proteins that may act as secondary ¢lter proteins for Zn2þ , Co2þ and Cd2þ .
6. Heavy metal resistance in prokaryotes revisited Which modes of heavy metal e¥ux do most bacteria have? Export of zinc and cadmium is accomplished at least by CDF2, Zn-CPx and HME1-RND proteins. Of the 64 prokaryotic genomes analyzed, there are approximately equal numbers of species harboring no system (17 species), only one CDF2 protein (13 species), only one ZnCPx (15 species), or a CDF2 plus a Zn-CPx protein (13 species). Only a minority of prokaryotes analyzed have all three proteins (three species), an ATPase plus a HME1RND protein (two species), a CDF2 plus a HME1-RND protein (one species) or exclusively an RND protein (no species). Thus, the presence of a HME1-RND protein is the exception and marks high level resistance to zinc, cobalt and cadmium, while possession of a CDF2 and/or a Zn-CPx-type ATPase is the general case and therefore part of the cellular homeostasis system. Moreover, since no CBA transport system occurs without the presence of a helper protein that imports the cations into the periplasm, this shows that periplasmic access to the CBA pump may also be important for resistance to this group of heavy metal cations. Bacteria relying exclusively on a CDF2 protein for their zinc homeostasis are more adapted to an anaerobic or micro-aerobic environment than to higher oxygen concentrations. The need to remove metals from the thiol pool by a CPx-type ATPase may be smaller when the additional oxidation stress by molecular oxygen is absent. As far as we know today, CPx-type ATPases or CDF proteins detoxify nickel only in rare cases. RND proteins known to detoxify nickel and cobalt (HME2 group) or candidates for this process (HME5 group) are rare and mark species as adapted to nickel-rich environments. Moreover, NreB- and CnrT-like nickel transport systems occur in only a few species. Nickel is used seldom as a cofactor [18] and if needed, is imported into the cell by chemiosmotically driven slow and speci¢c uptake systems of the NiCoT family [46,254^256,276^278] or ABC transport systems [279]. The concentration of nickel in seawater and its toxicity to E. coli is comparable to that of zinc, but this comparison has to be viewed in the light of two
FEMSRE 781 2-6-03
known zinc-exporting systems in E. coli (Table 1) and no known nickel e¥ux system. The toxicity of nickel seems to be a smaller problem than the toxicity of zinc or, especially, copper. Copper is mainly detoxi¢ed by a Cu-CPx-type ATPase. Of the analyzed 64 genomes, 41 contained the gene for a putative Cu-CPx-ATPase, 17 no such protein, and six an additional HME4-RND protein. The presence of such a RND protein may mark a demand for a supplementary export of the periplasmic copper, maybe under anaerobic conditions [93]. The presence of an RND-driven CBA pump is again an exception, indicating increased heavy metal resistance. However, the highest degree of copper resistance is mediated by plasmid-bound copper resistance determinants [9,79^81,83,84] that are mentioned in the article by Grass and Rensing [88]. Seventeen species did not contain any known zinc export system and the same number no known copper e¥ux system (Table 1). The two groups overlap, but not completely ; nine species contained neither zinc- nor copperexporting systems. Four of them, the pathogenic spirochete B. burgdorferi, two species of Rickettsia and the methanogenic archaeon M. jannaschii, contained at least a CDF3 and/or a CHR protein indicating some skills in export of heavy metals: the unknown CDF3 substrate and/or chromate. The remaining ¢ve species without known competence to detoxify heavy metals are the cellwall-free pathogenic Mycoplasma and Ureaplasma species and the aphid symbiont Buchnera. In their pathogenic or symbiotic way of life, these bacteria are probably not confronted with heavy metal stress, rendering metal homeostasis genes super£uous.
7. Concluding remarks: heavy metal resistance is the product of the interaction of many systems We may now be able to answer the question we started with: what makes a heavy metal resistant bacterium like R. metallidurans heavy metal resistant? Firstly, there is a multitude of resistance systems. When R. metallidurans is compared to R. solanacearum, an increase in the numbers of the CPx-type, CDF and RND proteins is observed (Table 1). These additional genes are the product either of horizontal gene transfer or of a duplication of genes already present in the common Ralstonia ancestor. If the duplicated genes in R. metallidurans were closely related to a respective gene in R. solanacearum, this would indicate a possible gene duplication that had happened during the evolution of the genus R. metallidurans. If this is not the case and if the genes are additionally harbored by a plasmid, this would point towards horizontal gene transfer. The number of CPx-type ATPases increases from R. solanacearum to R. metallidurans from one to three Zn-CPx proteins, while the number of Cu-CPx-ATPases is constant at two. The four Zn-CPx proteins from both
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
Ralstonia species belong to the same group of proteins, Zn6, which indicates an origin by gene multiplication. CHR proteins duplicated from two to four. R. metallidurans contains one CDF protein more than R. solanacearum. Both species harbor a CDF1 and a CDF3 protein, the additional protein in R. metallidurans is the CDF2 protein CzcD that is encoded by plasmid pMOL30. The number of HME-RND proteins increased from four in R. solanacearum to an outstanding number of 12 in R. metallidurans. When the various clusters of HME proteins are compared, the number of HME1 and HME4 proteins is duplicated, the number of two HME3 proteins in R. solanacearum is even tripled to six proteins in R. metallidurans. In all cases, predicted proteins from R. metallidurans and R. solanacearum group closely with each other indicating an origin from a common ancestral gene. The remaining two HME2 proteins, CnrA and NccA, which occur only in R. metallidurans and not in R. solanacearum, are both encoded on plasmids that may have been acquired by horizontal gene transfer. Secondly, there is di¡erentiation of the function of metal resistance systems after the duplication event has occurred. The three Zn-CPx-type ATPases of R. metallidurans became lead-, cadmium- and zinc-speci¢c e¥ux pumps. Thirdly, there is combination of the specialized proteins with others to form resistance systems with increased e⁄ciency. Of the two determinants that encode a HME2 pump, ncc on plasmid pMOL30 of strain CH34 became inactivated while cnr combined with the cnrT gene to form an e⁄cient nickel/cobalt pump. The HME3 proteins that may detoxify copper and silver combined with other copper resistance systems. The additional CDF protein in R. metallidurans, CzcD, became attached to the Czc system and is involved in regulation of expression and provides additionally a minor degree of metal resistance. Other genes also became part of the Czc system that is intriguing and unique in its complexity. The three steps, (1) gene multiplication by duplication and horizontal transfer, (2) di¡erentiation of function and (3) combination of genes into highly e⁄cient operons, created the outstanding heavy metal resistance of R. metallidurans.
Acknowledgements This work was funded by the Deutsche Forschungsgemeinschaft, the state Sachsen-Anhalt and by the Fonds der Chemischen Industrie. I am grateful to Chris Rensing, Gregor Grass and Cornelia GroMe for their comments on this review. The R. metallidurans CH34 web site of DOE/ JGI (http://www.jgi.doe.gov/JGI_microbial/html/ralstonia/ ralston_homepage.html) and the comprehensive microbial resource http://www.tigr.org/tigr-scripts/CMR2/CMR HomePage.spl of The Institute of Genomic Research (TIGR) were used intensively to compile Table 1. I would also like to thank all the colleagues, mentors and students
FEMSRE 781 2-6-03
331
in my lab who in the last 20 years have contributed to the progress of the CH34 story.
References [1] Ohtake, H., Cervantes, C. and Silver, S. (1987) Decreased chromate uptake in Pseudomonas £uorescens carrying a chromate resistance plasmid. J. Bacteriol. 169, 3853^3856. [2] Silver, S. and Phung, L.T. (1996) Bacterial heavy metal resistance: New surprises. Annu. Rev. Microbiol. 50, 753^789. [3] Broer, S., Ji, G., Broer, A. and Silver, S. (1993) Arsenic e¥ux governed by the arsenic resistance determinant of Staphylococcus aureus plasmid pI258. J. Bacteriol. 175, 3480^3485. [4] Brown, N.L., Misra, T.K., Winnie, J.N., Schmidt, A., Sei¡, M. and Silver, S. (1986) The nucleotide sequence of the mercuric resistance operons of plasmid R100 and transposon Tn501: further evidence for mer genes which enhance the activity of the mercuric ion detoxi¢cation system. Mol. Gen. Genet. 202, 143^151. [5] Endo, G. and Silver, S. (1995) CadC, the transcriptional regulatory protein of the cadmium resistance system of Staphylococcus aureus plasmid pI258. J. Bacteriol. 177, 4437^4441. [6] Silver, S., Budd, K., Leahy, K.M., Shaw, W.V., Hammond, D., Novick, R.P., Willsky, G.R., Malamy, M.H. and Rosenberg, H. (1981) Inducible plasmid-determined resistance to arsenate, arsenite, and antimony(III) in Escherichia coli and Staphylococcus aureus. J. Bacteriol. 146, 983^996. [7] Silver, S., Misra, T.K. and Laddaga, R.A. (1989) DNA sequence analysis of bacterial toxic heavy metal resistances. Biol. Trace Elem. Res. 21, 145^163. [8] Silver, S. (1992) Plasmid-determined metal resistance mechanisms : range and overview. Plasmid 27, 1^3. [9] Silver, S., Lee, B.T.O., Brown, N.L. and Cooksey, D.A. (1993) Bacterial plasmid resistances to copper, cadmium, and zinc. In: Chemistry of Copper and Zinc Triads (Welch, A.J. and Chapman, S.K., Eds.), pp. 38^53. The Royal Society of Chemistry, London. [10] Silver, S. and Ji, G. (1994) Newer systems for bacterial resistances to toxic heavy metals. Environ. Health Perspect. 102, 107^113. [11] Silver, S. and Walderhaug, M. (1995) Bacterial plasmid-mediated resistances to mercury, cadmium and copper. In: Toxicology of Metals. Biochemical Aspects (Goyer, R.A. and Cherian, M.G., Eds.), pp. 435^458. Springer Verlag, Berlin. [12] Summers, A.O. and Silver, S. (1972) Mercury resistance in a plasmidbearing strain of Escherichia coli. J. Bacteriol. 112, 1228^1236. [13] Nucifora, G., Chu, L., Misra, T.K. and Silver, S. (1989) Cadmium resistance from Staphylococcus aureus plasmid pI258 cadA gene results from a cadmium-e¥ux ATPase. Proc. Natl. Acad. Sci. USA 86, 3544^3548. [14] Nies, D.H. (2000) Heavy metal resistant bacteria as extremophiles: molecular physiology and biotechnological use of Ralstonia spec. CH34. Extremophiles 4, 77^82. [15] Goris, J., De Vos, P., Coenye, T., Hoste, B., Janssens, D., Brim, H., Diels, L., Mergeay, M., Kersters, K. and Vandamme, P. (2001) Classi¢cation of metal-resistant bacteria from industrial biotopes as Ralstonia campinensis sp. nov., Ralstonia metallidurans sp. nov. and Ralstonia basilensis Steinle et al., 1998 emend. Int. J. Syst. Evol. Microbiol. 51, 1773^1782. [16] Diels, L. and Mergeay, M. (1990) DNA probe-mediated detection of resistant bacteria from soils highy polluted by heavy metals. Appl. Environ. Microbiol. 56, 1485^1491. [17] Mergeay, M., Monchy, S., Vallaeys, T., Auquier, V., Bentomane, A., Bertin, P., Taghavi, S., Dunn, J., van der Lelie, D. and Wattiez, R. (2003) Ralstonia metallidurans, a bacterium speci¢cally adapted to toxic metals: towards a catalogue of metal-responsive genes. FEMS Microbiol. Rev. 27, 385^410.
Cyaan Magenta Geel Zwart
332
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[18] Nies, D.H. (1999) Microbial heavy metal resistance. Appl. Microbiol. Biotechnol. 51, 730^750. [19] Mergeay, M., Nies, D., Schlegel, H.G., Gerits, J., Charles, P. and van Gijsegem, F. (1985) Alcaligenes eutrophus CH34 is a facultative chemolithotroph with plasmid-bound resistance to heavy metals. J. Bacteriol. 162, 328^334. [20] Mukhopadhyay, R., Rosen, B.P., Pung, L.T. and Silver, S. (2002) Microbial arsenic: from geocycles to genes and enzymes. FEMS Microbiol. Rev. 26, 311^325. [21] Weast, R.C. (1984) CRC Handbook of Chemistry and Physics, 64th edn. CRC Press, Boca Raton, FL. [22] Saier Jr., M.H., Tam, R., Reizer, A. and Reizer, J. (1994) Two novel families of bacterial membrane proteins concerned with nodulation, cell division and transport. Mol. Microbiol. 11, 841^847. [23] Tseng, T.-T., Gratwick, K.S., Kollman, J., Park, D., Nies, D.H., Go¡eau, A. and Saier, M.H.J. (1999) The RND superfamily : an ancient, ubiquitous and diverse family that includes human disease and development proteins. J. Mol. Microbiol. Biotechnol. 1, 107^ 125. [24] Brown, D.E., Thrall, M.A., Walkley, S.U., Wurzelmann, S., Wenger, D.A., Allison, R.W. and Just, C.A. (1996) Metabolic abnormalities in feline Niemann-Pick type C heterozygotes. J. Inherit. Metab. Dis. 19, 319^330. [25] Campo, J.V., Stowe, R., Slomka, G., Byler, D. and Gracious, B. (1998) Psychosis as a presentation of physical disease in adolescence: a case of Niemann-Pick disease, type C. Dev. Med. Child. Neurol. 40, 126^129. [26] Carstea, E.D., Morris, J.A., Coleman, K.G., Loftus, S.K., Zhang, D., Cummings, C., Gu, J., Rosenfeld, M.A., Pavan, W.J., Krizman, D.B., Nagle, J., Polymeropoulos, M.H., Sturley, S.L., Ioannou, Y.A., Higgins, M.E., Comly, M., Cooney, A., Brown, A., Kaneski, C.R., Blanchette Mackie, E.J., Dwyer, N.K., Neufeld, E.B., Chang, T.Y., Liscum, L. and Tagle, D.A. (1997) Niemann-Pick C1 disease gene: homology to mediators of cholesterol homeostasis. Science 277, 228^231. [27] Erickson, R.P., Aviles, R.A., Zhang, J., Kozloski, M.A., Garver, W.S. and Heidenreich, R.A. (1997) High-resolution mapping of the spm (Niemann-Pick type C) locus on mouse chromosome 18. Mamm. Genome 8, 355^356. [28] Garver, W.S., Erickson, R.P., Wilson, J.M., Colton, T.L., Hossain, G.S., Kozloski, M.A. and Heidenreich, R.A. (1997) Altered expression of caveolin-1 and increased cholesterol in detergent insoluble membrane fractions from liver in mice with Niemann-Pick disease type C. Biochim. Biophys. Acta 1361, 272^280. [29] Garver, W.S., Hsu, S.C., Erickson, R.P., Greer, W.L., Byers, D.M. and Heidenreich, R.A. (1997) Increased expression of caveolin-1 in heterozygous Niemann-Pick type II human ¢broblasts. Biochem. Biophys. Res. Commun. 236, 189^193. [30] Gillan, T.L., Byers, D.M., Riddell, D.C., Neumann, P.E. and Greer, W.L. (1997) Limiting the Niemann-Pick type C critical region to a 1-cM interval. Clin. Invest. Med. 20, 339^343. [31] Goodrum, J.F. and Pentchev, P.G. (1997) Cholesterol reutilization during myelination of regenerating PNS axons is impaired in Niemann-Pick disease type C mice. J. Neurosci. Res. 49, 389^392. [32] Greer, W.L., Riddell, D.C., Byers, D.M., Welch, J.P., Girouard, G.S., Sparrow, S.M., Gillan, T.L. and Neumann, P.E. (1997) Linkage of Niemann-Pick disease type D to the same region of human chromosome 18 as Niemann-Pick disease type C. Am. J. Hum. Genet. 61, 139^142. [33] Greer, W.L., Riddell, D.C., Gillan, T.L., Girouard, G.S., Sparrow, S.M., Byers, D.M., Dobson, M.J. and Neumann, P.E. (1998) The Nova Scotia (type D) form of Niemann-Pick disease is caused by a G3097CT transversion in NPC1. Am. J. Hum. Genet. 63, 52^54. [34] Gu, J.Z., Carstea, E.D., Cummings, C., Morris, J.A., Loftus, S.K., Zhang, D., Coleman, K.G., Cooney, A.M., Comly, M.E., Fandino, L., Ro¡, C., Tagle, D.A., Pavan, W.J., Pentchev, P.G. and Rosenfeld, M.A. (1997) Substantial narrowing of the Niemann-Pick C
FEMSRE 781 2-6-03
[35]
[36] [37] [38]
[39] [40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
candidate interval by yeast arti¢cial chromosome complementation. Proc. Natl. Acad. Sci. USA 94, 7378^7383. Kovesi, T.A., Lee, J., Shuckett, B., Clarke, J.T., Callahon, J.W. and Phillips, M.J. (1996) Pulmonary in¢ltration in Niemann-Pick disease type C. J. Inherit. Metab. Dis. 19, 792^793. Lange, Y. and Steck, T.L. (1998) Four cholesterol-sensing proteins. Curr. Opin. Struct. Biol. 8, 435^439. Liscum, L. and Klansek, J.J. (1998) Niemann-Pick disease type C. Curr. Opin. Lipidol. 9, 131^135. Osborne, T.F. and Rosenfeld, J.M. (1998) Related membrane domains in proteins of sterol sensing and cell signaling provide a glimpse of treasures still buried within the dynamic realm of intracellular metabolic regulation. Curr. Opin. Lipidol. 9, 137^140. Schi¡mann, R. (1996) Niemann-Pick disease type C. From bench to bedside. J. Am. Med. assoc. 276, 561^564. Schedin, S., Sindelar, P.J., Pentchev, P., Brunk, U. and Dallner, G. (1997) Peroxisomal impairment in Niemann-Pick type C disease. J. Biol. Chem. 272, 6245^6251. Suresh, S., Yan, Z., Patel, R.C., Patel, Y.C. and Patel, S.C. (1998) Cellular cholesterol storage in the Niemann-Pick disease type C mouse is associated with increased expression and defective processing of apolipoprotein D. J. Neurochem. 70, 242^251. Schofer, O., Mischo, B., Puschel, W., Harzer, K. and Vanier, M.T. (1998) Early-lethal pulmonary form of Niemann-Pick type C disease belonging to a second, rare genetic complementation group. Eur. J. Pediatr. 157, 45^49. Sequeira, J.S., Vellodi, A., Vanier, M.T. and Clayton, P.T. (1998) Niemann-Pick disease type C and defective peroxisomal beta-oxidation of branched-chain substrates. J. Inherit. Metab. Dis. 21, 149^ 154. Steinberg, S.J., Mondal, D. and Fensom, A.H. (1996) Co-cultivation of Niemann-Pick disease type C ¢broblasts belonging to complementation groups alpha and beta stimulates LDL-derived cholesterol esteri¢cation. J. Inherit. Metab. Dis. 19, 769^774. Lange, Y., Ye, J. and Steck, T.L. (1998) Circulation of cholesterol between lysosomes and the plasma membrane. J. Biol. Chem. 273, 18915^18922. Saier, M.H.J. (2000) A functional-phylogenetic classi¢cation system for transmembrane solute transporters. Microbiol. Mol. Biol. Rev. 64, 354^411. Johnson, J.M. and Church, G.M. (1999) Alignment and structure prediction of divergent protein families: periplasmic and outer membrane proteins of bacterial e¥ux pumps. J. Mol. Biol. 287, 695^715. Andersen, C., Hughes, C. and Koronakis, V. (2001) Protein export and drug e¥ux through bacterial channel-tunnels. Curr. Opin. Cell Biol. 13, 412^416. Paulsen, I.T., Park, J.H., Choi, P.S. and Saier, M.H.J. (1997) A family of Gram-negative bacterial outer membrane factors that function in the export of proteins, carbohydrates, drugs and heavy metals from Gram-negative bacteria. FEMS Microbiol. Lett. 156, 1^8. Zgurskaya, H.I. and Nikaido, H. (2000) Multidrug resistance mechanisms: drug e¥ux across two membranes. Mol. Microbiol. 37, 219^ 225. Zgurskaya, H.I. and Nikaido, H. (1999) Bypassing the periplasm : Reconstitution of the AcrAB multidrug e¥ux pump of Escherichia coli. Proc. Natl. Acad. Sci. USA 96, 7190^7195. Zgurskaya, H.I. and Nikaido, H. (1999) AcrA is a highly asymmetric protein capable of spanning the periplasm. J. Mol. Biol. 285, 409^ 420. Zgurskaya, H.I. and Nikaido, H. (2000) Cross-linked complex between oligomeric periplasmic lipoprotein AcrA and the inner-membrane-associated multidrug e¥ux pump AcrB from Escherichia coli. J. Bacteriol. 182, 4264^4267. Nies, D., Mergeay, M., Friedrich, B. and Schlegel, H.G. (1987) Cloning of plasmid genes encoding resistance to cadmium, zinc, and cobalt in Alcaligenes eutrophus CH34. J. Bacteriol. 169, 4865^4868. Nies, D.H., Nies, A., Chu, L. and Silver, S. (1989) Expression and
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71] [72]
[73] [74]
[75]
nucleotide sequence of a plasmid-determined divalent cation e¥ux system from Alcaligenes eutrophus. Proc. Natl. Acad. Sci. USA 86, 7351^7355. Kunito, T., Kusano, T., Oyaizu, H., Senoo, K., Kanazawa, S. and Matsumoto, S. (1996) Cloning and sequence analysis of czc genes in Alcaligenes sp. strain CT14. Biosci. Biotechnol. Biochem. 60, 699^ 704. van der Lelie, D., Schwuchow, T., Schwidetzky, U., Wuertz, S., Baeyens, W., Mergeay, M. and Nies, D.H. (1997) Two component regulatory system involved in transcriptional control of heavy metal homoeostasis in Alcaligenes eutrophus. Mol. Microbiol. 23, 493^ 503. GroMe, C., Grass, G., Anton, A., Franke, S., Navarrete Santos, A., Lawley, B., Brown, N.L. and Nies, D.H. (1999) Transcriptional organization of the czc heavy metal homoeostasis determinant from Alcaligenes eutrophus. J. Bacteriol. 181, 2385^2393. Nies, D.H. and Brown, N.L. (1998) Two-component systems in the regulation of heavy metal resistance. In: Metal Ions in Gene Regulation (Silver, S. and Walden, W., Eds.), pp. 77^103. Chapman and Hall, London. Nies, D.H. (1992) CzcR and CzcD, gene products a¡ecting regulation of resistance to cobalt, zinc and cadmium (czc system) in Alcaligenes eutrophus. J. Bacteriol. 174, 8102^8110. Nies, D.H. and Silver, S. (1989) Plasmid-determined inducible e¥ux is responsible for resistance to cadmium, zinc, and cobalt in Alcaligenes eutrophus. J. Bacteriol. 171, 896^900. Nies, D.H. (1995) The cobalt, zinc, and cadmium e¥ux system CzcABC from Alcaligenes eutrophus functions as a cation-protonantiporter in Escherichia coli. J. Bacteriol. 177, 2707^2712. Sensfuss, C. and Schlegel, H.G. (1988) Plasmid pMOL28-encoded resistance to nickel is due to speci¢c e¥ux. FEMS Microbiol. Lett. 55, 295^298. Liesegang, H., Lemke, K., Siddiqui, R.A. and Schlegel, H.-G. (1993) Characterization of the inducible nickel and cobalt resistance determinant cnr from pMOL28 of Alcaligenes eutrophus CH34. J. Bacteriol. 175, 767^778. Grass, G., GroMe, C. and Nies, D.H. (2000) Regulation of the cnr cobalt/nickel resistance determinant from Ralstonia sp. CH34. J. Bacteriol. 182, 1390^1398. Tibazarwa, C., Wuertz, S., Mergeay, M., Wyns, L. and van der Lelie, D. (2000) Regulation of the cnr cobalt and nickel resistance determinant of Ralstonia eutropha (Alcaligenes eutrophus) CH34. J. Bacteriol. 182, 1399^1409. Schmidt, T. and Schlegel, H.G. (1994) Combined nickel-cobalt-cadmium resistance encoded by the ncc locus of Alcaligenes xylosoxidans 31A. J. Bacteriol. 176, 7045^7054. Legatzki, A., Franke, S., Lucke, S., Ho¡mann, T., Anton, A., Neumann, D. and Nies, D.H. (2003) First step towards a quantitative model describing Czc-mediated heavy metal resistance in Ralstonia metallidurans. Biodegradation (in press). Goldberg, M., Pribyl, T., Juhnke, S. and Nies, D.H. (1999) Energetics and topology of CzcA, a cation/proton antiporter of the RND protein family. J. Biol. Chem. 274, 26065^26070. Outten, C.E. and O’Halloran, T.V. (2001) Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 292, 2488^2492. Fahey, R.C., Wc, B., Adams, W.B. and Worsham, M.B. (1978) Occurrence of glutathione in bacteria. J. Bacteriol. 133, 1126^1129. McLaggan, D., Logan, T.M., Lynn, D.G. and W, E. (1990) Involvement of gamma-glutamyl peptides in osmoadaptation of Escherichia coli. J. Bacteriol. 172, 3631^3636. Fahey, R.C. (2001) Novel thiols of prokaryotes. Annu. Rev. Microbiol. 55, 333^356. Oram, P.D., Fang, X., Fernando, Q., Letkeman, P. and Letkeman, D. (1996) The formation constants of mercury(II)-glutathione complexes. Chem. Res. Toxicol. 9, 709^712. Osterberg, R., Ligaarden, R. and Persson, D. (1979) Copper(I) com-
FEMSRE 781 2-6-03
[76]
[77]
[78]
[79] [80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89] [90]
[91]
[92]
[93]
[94]
333
plexes of penicillamine and glutathione. J. Inorg. Biochem. 10, 341^ 355. Rae, T.D., Schmidt, P.J., Pufahl, R.A., Culotta, V.C. and O’ Halloran, T.V. (1999) Undetectable intracellular free copper: The requirement of a copper chaperone for superoxide dismutase. Science 284, 805^808. Dominey, L.A. and Kustin, K. (1983) Kinetics and mechanism of Zn(II) complexation with reduced glutathione. J. Inorg. Biochem. 18, 153^160. Brown, N.L., Camakaris, J., Lee, B.T., Williams, T., Morby, A.P., Parkhill, J. and Rouch, D.A. (1991) Bacterial resistances to mercury and copper. J. Cell. Biochem. 46, 106^114. Brown, N.L., Rouch, D.A. and Lee, B.T.O. (1992) Copper resistance determinants in bacteria. Plasmid 27, 41^51. Brown, N.L., Lee, B.T.O. and Silver, S. (1993) Bacterial transport of and resistance to copper. In: Metal Ions in Biological Systems, Vol. 30 (Sigel, H. and Sigel, A., Eds.), pp. 405^430. Marcel Dekker, New York. Brown, N.L., Barrett, S.R., Camakaris, J., Lee, B.T. and Rouch, D.A. (1995) Molecular genetics and transport analysis of the copper-resistance determinant (pco) from Escherichia coli plasmid pRJ1004. Mol. Microbiol. 17, 1153^1166. Bryson, J.W., O’Halloran, T.V., Rouch, D.A., Brown, N.L., Camakaris, J. and Lee, B.T.O. (1993) Chemical and genetic studies of copper resistance in E. coli. In: Bioinorganic Chemistry of Copper (Karlin, K.D. and Tylekar, Z., Eds.), pp. 101^109. Chapman and Hall, New York. Lee, S.M., Grass, G., Rensing, C., Barrett, S.R., Yates, C.J., Stoyanov, J.V. and Brown, N.L. (2002) The Pco proteins are involved in periplasmic copper handling in Escherichia coli. Biochem. Biophys. Res. Commun. 295, 616^620. Lee, B.T.O., Brown, N.L., Rogers, S., Bergemann, A., Camakaris, J. and Rouch, D.A. (1990) Bacterial response to copper in the environment: Escherichia coli as a model system. In: Metal Speciation in the Environment (Gucer, S. and Broeckaert, J.A.C., Eds.). Springer-Verlag, New York. Rouch, D.A. and Brown, N.L. (1997) Copper-inducible transcriptional regulation at two promoters in the Escherichia coli copper resistance determinant pco. Microbiology 141, 1191^1202. Williams, J.R., Morgan, A., Rouch, D.A., Brown, N.L. and Lee, B.T.O. (1993) Copper resistant enteric bacteria from United Kingdom and Australian piggeries. Appl. Environ. Microbiol. 59, 2531^ 2537. Rensing, C., Fan, B., Sharma, R., Mitra, B. and Rosen, B.P. (2000) CopA: An Escherichia coli Cu(I)-translocating P-type ATPase. Proc. Natl. Acad. Sci. USA 97, 652^656. Rensing, C. and Grass, G. (2003) Escherichia coli mechanisms of copper homeostasis in a changing environment. FEMS Microbiol. Rev. 27, 197^213. Rensing, C., Ghosh, M. and Rosen, B.P. (1999) Families of softmetal-ion-transporting ATPases. J. Bacteriol. 181, 5891^5897. Franke, S., Grass, G. and Nies, D.H. (2001) The product of the ybdE gene of the Escherichia coli chromosome is involved in detoxi¢cation of silver ions. Microbiology 147, 965^972. Munson, G.P., Lam, D.L., Outten, F.W. and O’Halloran, T.V. (2000) Identi¢cation of a copper-responsive two-component system on the chromosome of Escherichia coli K-12. J. Bacteriol. 182, 5864^ 5871. Silver, S. (2003) Bacterial silver resistance: Molecular biology and uses and misuses of silver compounds. FEMS Microbiol. Rev. 27, 341^354. Outten, F.W., Hu¡man, D.L., Hale, J.A. and O’Halloran, T.V. (2001) The independent cue and cus systems confer copper tolerance during aerobic and anaerobic growth in Escherichia coli. J. Biol. Chem. 276, 30670^30677. Grass, G. and Rensing, C. (2001) Genes involved in copper homeostasis in Escherichia coli. J. Bacteriol. 183, 2145^2147.
Cyaan Magenta Geel Zwart
334
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[95] Grass, G. and Rensing, C. (2001) CueO is a multi-copper oxidase that confers copper tolerance in Escherichia coli. Biochem. Biophys. Res. Commun. 286, 902^908. [96] Hu¡man, D.L., Huyett, J., Outten, F.W., Doan, P.E., Finney, L.A., Ho¡man, B.M. and O’Halloran, T.V. (2002) Spectroscopy of Cu(II)-PcoC and the multicopper oxidase function of PcoA, two essential components of Escherichia coli pco copper resistance operon. Biochemistry 41, 10046^10055. [97] Franke, S., Grass, G., Rensing, C. and Nies, D.H. (2003) Molecular analysis of the copper-transporting CusCFBA e¥ux system from Escherichia coli. J. Bacteriol. (in press). [98] Nikaido, H. (1998) Multiple antibiotic resistance and e¥ux. Curr. Opin. Microbiol. 1, 516^523. [99] Kohler, T., Pechere, J.C. and Plesiat, P. (1999) Bacterial antibiotic e¥ux systems of medical importance. Cell. Mol. Life Sci. 56, 771^ 778. [100] Ziha Zari¢, I., Llanes, C., Kohler, T., Pechere, J.C. and Plesiat, P. (1999) In vivo emergence of multidrug-resistant mutants of Pseudomonas aeruginosa overexpressing the active e¥ux system MexAMexB-OprM. Antimicrob. Agents Chemother. 43, 287^291. [101] Hassan, M.E.T., van der Lelie, D., Springael, D., Ro«mling, U., Ahmed, N. and Mergeay, M. (1999) Identi¢cation of a gene cluster, czr, involved in cadmium and zinc resistance in Pseudomonas aeruginosa. Gene 238, 417^425. [102] Srikumar, R., Kon, T., Gotoh, N. and Poole, K. (1998) Expression of Pseudomonas aeruginosa multidrug e¥ux pumps MexA-MexBOprM and MexC-MexD-OprJ in a multidrug-sensitive Escherichia coli strain. Antimicrob. Agents. Chemother. 42, 65^71. [103] Li, X.Z. and Poole, K. (1999) Organic solvent-tolerant mutants of Pseudomonas aeruginosa display multiple antibiotic resistance. Can. J. Microbiol. 45, 18^22. [104] Li, X.Z., Zhang, L. and Poole, K. (1998) Role of the multidrug e¥ux systems of Pseudomonas aeruginosa in organic solvent tolerance. J. Bacteriol. 180, 2987^2991. [105] Gotoh, N., Tsujimoto, H., Poole, K., Yamagishi, J.I. and Nishino, T. (1995) The outer membrane protein OprM of Pseudomonas aeruginosa is encoded by oprK of the mexA-mexB-oprK multidrug resistance operon. Antimicrob. Agents. Chemother. 39, 2567^ 2569. [106] Maseda, H., Yoneyama, H. and Nakae, T. (2000) Assignment of the substrate-selective subunits of the MexEF-OprN multidrug e¥ux pump of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 44, 658^664. [107] Chuanchuen, R., Narasaki, C.T. and Schweizer, H.P. (2002) The MexJK e¥ux pump of Pseudomonas aeruginosa requires OprM for antibiotic e¥ux but not for e¥ux of triclosan. J. Bacteriol. 184, 5036^5044. [108] Aendekerk, S., Ghysels, B., Cornelis, C. and Baysse, C. (2002) Characterization of a new e¥ux pump, MexGHI-OpmD, from Pseudomonas aeruginosa that confers resistance to vanadium. Microbiology 148, 2371^2381. [109] Morita, Y., Komori, Y., Mima, T., Kuroda, T., Mizushima, T. and Tsuchiya, T. (2001) Construction of a series of mutants lacking all of the four major mex operons for multidrug e¥ux pumps or possessing each one of the operons from Pseudomonas aeruginosa PAO1: MexCD-OprJ is an inducible pump. FEMS Microbiol. Lett. 202, 139^143. [110] Nishino, K. and Yamaguchi, A. (2001) Analysis of a complete library of putative drug transporter genes in Escherichia coli. J. Bacteriol. 183, 5803^5812. [111] Sulavik, M.C., Houseweart, C., Cramer, C., Jiwani, N., Murgolo, N., Greene, J., DiDomenico, B., Shaw, K.J., Miller, G.H., Hare, R. and Shimer, G. (2001) Antibiotic susceptibility pro¢les of Escherichia coli strains lacking multidrug e¥ux pump genes. Antimicrob. Agents Chemother. 45, 1126^1136. [112] White, D.G., Goldman, J.D., Demple, B. and Levy, S.B. (1997) Role of the acrAB locus inorganic solvent tolerance mediated by
FEMSRE 781 2-6-03
[113] [114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
[126]
[127]
[128]
[129]
[130]
[131]
expression of marA, soxS, or robA in Escherichia coli. J. Bacteriol. 1997, 6122^6126. Thanassi, D.G., Cheng, L.W. and Nikaido, H. (1997) Active e¥ux of bile salts by Escherichia coli. J. Bacteriol. 179, 2512^2518. Aono, R. (1998) Improvement of organic solvent tolerance level of Escherichia coli by overexpression of stress-responsive genes. Extremophiles 2, 239^248. Aono, R., Tsukagoshi, N. and Yamamoto, M. (1998) Involvement of outer membrane protein TolC, a possible member of the mar-sox regulon, in maintenance and improvement of organic solvent tolerance of Escherichia coli K-12. J. Bacteriol. 180, 938^944. Rand, J.D., Danby, S.G., Greenway, D.L.A. and England, R.R. (2002) Increased expression of the multidrug e¥ux genes acrAB occurs during slow growth of Escherichia coli. FEMS Microbiol. Lett. 207, 91^95. Rosenberg, E.Y., Ma, D. and Nikaido, H. (2000) AcrD of Escherichia coli is an aminoglycoside e¥ux pump. J. Bacteriol. 182, 1754^ 1756. Sa¤nchez, L., Pan, W., Vin‹as, M. and Nikaido, H. (1998) The acrAB homolog of Haemophilus in£uenzae codes for a functional multidrug e¥ux pump. J. Bacteriol. 179, 6855^6857. Hagman, K.E., Lucas, C.E., Balthazar, J.T., Snyder, L., Nilles, M., Judd, R.C. and Shafer, W.M. (1997) The MtrD protein of Neisseria gonorrhoeae is a member of the resistance/nodulation/division protein family constituting part of an e¥ux system. Microbiology 7, 2117^2125. Veal, W.L., Nicholas, R.A. and Shafer, W.M. (2002) Overexpression of the MtrC-MtrD-MtrE e¥ux pump due to an mtrR mutation is required for chromosomally mediated penicillin resistance in Neisseria gonorrhoeae. J. Bacteriol. 184, 5619^5624. Ramos, J.L., Duque, E., Godoy, P. and Segura, A. (1998) E¥ux pumps involved in toluene tolerance in Pseudomonas putida DOTT1E. J. Bacteriol. 180, 3323^3329. Mosqueda, G. and Ramos, J.L. (2000) A set of genes encoding a second toluene e¥ux system in Pseudomonas. J. Bacteriol. 182, 937^ 943. Lin, J., Michel, L.O. and Zhang, Q.J. (2002) CmeABC functions as a multidrug e¥ux system in Campylobacter jejuni. Antimicrob. Agents Chemother. 46, 2124^2131. Pumbwe, L. and Piddock, L.J.V. (2002) Identi¢cation and molecular characterisation of CmeB, a Campylobacter jejuni multidrug e¥ux pump. FEMS Microbiol. Lett. 206, 185^189. Magnet, S., Courvalin, P. and Lambert, T. (2001) Resistance-nodulation-cell division-type e¥ux pump involved in aminoglycoside resistance in Acinetobacter baumannii strain BM4454. Antimicrob. Agents Chemother. 45, 3375^3380. Alonso, A. and Martinez, J.L. (2000) Cloning and characterization of SmeDEF, a novel multidrug e¥ux pump from Stenotrophomonas maltophilia. Antimicrob. Agents Chemother. 44, 3079^3086. Palumbo, J.D., Kado, C.I. and Phillips, D.A. (1998) An iso£avonoid-inducible e¥ux pump in Agrobacterium tumefaciens is involved in competitive colonization of roots. J. Bacteriol. 180, 3107^ 3113. Peng, W.T. and Nester, E.W. (2001) Characterization of a putative RND-type e¥ux system in Agrobacterium tumefaciens. Gene 270, 245^252. Braibant, M., Guilloteau, L. and Zygmunt, M.S. (2002) Functional characterization of Brucella melitensis NorMI, an e¥ux pump belonging to the multidrug and toxic compound extrusion family. Antimicrob. Agents Chemother. 46, 3050^3053. Ikeda, T. and Yoshimura, F. (2002) A resistance-nodulation-cell division family xenobiotic e¥ux pump in an obligate anaerobe, Porphyromonas gingivalis. Antimicrob. Agents Chemother. 46, 3257^ 3260. Kumar, A. and Worobec, E.A. (2002) Fluoroquinolone resistance of Serratia marcescens : involvement of a proton gradient-dependent e¥ux pump. J. Antimicrob. Chemother. 50, 593^596.
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 [132] Nikaido, H., Basina, M., Nguyen, V. and Rosenberg, E.Y. (1998) Multidrug e¥ux pump AcrAB of Salmonella typhimurium excretes only those L-lactam antibiotics containing lipophilic side chains. J. Bacteriol. 180, 4682^4692. [133] Gotoh, N., Tsujimoto, H., Tsuda, M., Okamoto, K., Nomura, A., Wada, T., Nakahashi, M. and Nishino, T. (1998) Characterization of the MexC-MexD-OprJ multidrug e¥ux system in delta mexAmexB-oprM mutants of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 42, 1938^1943. [134] Lomovskaya, O., Lee, A., Hoshino, K., Ishida, H., Mistry, A., Warren, M.S., Boyer, E., Chamberland, S. and Lee, V.J. (1999) Use of a genetic approach to evaluate the consequences of inhibition of e¥ux pumps in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 43, 1340^1346. [135] Li, X.Z., Zhang, L., Srikumar, R. and Poole, K. (1998) beta-lactamase inhibitors are substrates for the multidrug e¥ux pumps of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 42, 399^ 403. [136] Oethinger, M., Kern, W.V., Jellen-Ritter, A.S., McMurry, L.M. and Levy, S.B. (2000) Ine¡ectiveness of topoisomerase mutations in mediating clinically signi¢cant £uoroquinolone resistance in Escherichia coli in the absence of the AcrAB e¥ux pump. Antimicrob. Agents Chemother. 44, 10^13. [137] Potrykus, J., Baranska, S. and Wegrzyn, G. (2002) Inactivation of the acrA gene is partially responsible for chloramphenicol sensitivity of Escherichia coli CM2555 strain expressing the chloramphenicol acetyltransferase gene. Microb. Drug Resist.-Mech. Epidemiol. Dis. 8, 179^185. [138] Ramos, J.L., Duque, E., Gallegos, M.T., Godoy, P., Ramos-Gonzalez, M.I., Rojas, A., Teran, W. and Segura, A. (2002) Mechanisms of solvent tolerance in gram-negative bacteria. Annu. Rev. Microbiol. 56, 743^768. [139] Kawamura-Sato, K., Shibayama, K., Horii, T., Iimuma, Y., Arakawa, Y. and Ohta, M. (1999) Role of multiple e¥ux pumps in Escherichia coli in indole expulsion. FEMS Microbiol. Lett. 179, 345^352. [140] Evans, K. and Poole, K. (1999) The MexA-MexB-OprM multidrug e¥ux system of Pseudomonas aeruginosa is growth-phase regulated. FEMS Microbiol. Lett. 173, 35^39. [141] Pearson, J.P., Van Delden, C. and Iglewski, B.H. (1999) Active e¥ux and di¡usion are involved in transport of Pseudomonas aeruginosa cell-to-cell signals. J. Bacteriol. 181, 1203^1210. [142] Guan, L., Ehrmann, M., Yoneyama, H. and Nakae, T. (1999) Membrane topology of the xenobiotic-exporting subunit, MexB, of the MexA, B-OprM extrusion pump in Pseudomonas aeruginosa. J. Biol. Chem. 274, 10517^10522. [143] Gotoh, N., Kusumi, T., Tsujimoto, H., Wada, T. and Nishino, T. (1999) Topological analysis of an RND family transporter, MexD of Pseudomonas aeruginosa. FEBS Lett. 458, 32^36. [144] Saier, M.H.J. (1994) Computer-aided analyses of transport protein sequences: gleaning evidence concerning function, structure, biogenesis, and evolution. Microbiol. Rev. 58, 71^93. [145] Gupta, A., Matsui, K., Lo, J.F. and Silver, S. (1999) Molecular basis for resistance to silver in Salmonella. Nat. Med. 5, 183^188. [146] Aires, J.R., Pechere, J.C., Van Delden, C. and Kohler, T. (2002) Amino acid residues essential for function of the MexF e¥ux pump protein of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 46, 2169^2173. [147] Kaback, H.R. (1988) Site-directed mutagenesis and ion-gradient driven active transport: on the path of the proton. Annu. Rev. Physiol. 50, 243^256. [148] Kaback, H.R., Voss, J. and Wu, J. (1997) Helix packing in polytopic membrane proteins: the lactose permease of Escherichia coli. Curr. Opin. Struct. Biol. 7, 537^542. [149] Varela, M.F. and Wilson, T.H. (1996) Molecular biology of the lactose carrier of Escherichia coli. Biochim. Biophys. Acta 276, 21^34.
FEMSRE 781 2-6-03
335
[150] Richter, H.-T., Brown, L.S., Needleman, R. and Lanyi, J.K. (1996) A linkage of the pKa’s of asp-85 and glu-204 forms part of the reprotonation switch of bacteriorhodopsin. Biochemistry 35, 4054^ 4062. [151] Murakami, S., Nakashima, R., Yamashita, R. and Yamaguchi, A. (2002) Crystal structure of bacterial multidrug e¥ux transporter AcrB. Nature 419, 587^593. [152] Pos, K.M. and Diederichs, K. (2002) Puri¢cation, crystallization and preliminary di¡raction studies of AcrB, the inner-membrane multi-drug e¥ux protein. Acta Crystallogr. D 58, 1865^1867. [153] Pribyl, T. (2001) Topologie des CzcCBA-E¥ux-Komplexes aus Ralstonia metallidurans CH34. Thesis, Martin-Luther-Universita«t, Halle-Wittenberg. [154] Lacroix, F.J.C., Cloeckaert, A., Grepinet, O., Pinault, C., Popo¡, M.Y. and Waxin, H. (1996) Salmonella typhimurium AcrB-like gene ^ Identi¢cation and role in resistance to biliary salts and detergents and in murine infection. FEMS Microbiol. Lett. 135, 161^167. [155] Ma, D., Cook, D.N., Alberti, M., Pon, N.G., Kikaido, H. and Hearst, J.E. (1993) Molecular cloning and characterization of acrA and acrE genes of Escherichia coli. J. Bacteriol. 175, 6299^ 6313. [156] Srikumar, R., Li, X.Z. and Poole, K. (1997) Inner membrane e¥ux components are responsible for beta-lactam speci¢city of multidrug e¥ux pumps in Pseudomonas aeruginosa. J. Bacteriol. 179, 7875^ 7881. [157] Yoneyama, H., Ocaktan, A., Gotoh, N., Nishino, T. and Nakae, T. (1998) Subunit swapping in the Mex-extrusion pumps in Pseudomonas aeruginosa. Biochem. Biophys. Res. Commun. 244, 898^902. [158] Murata, T., Kuwagaki, M., Shin, T., Gotoh, N. and Nishino, T. (2002) The substrate speci¢city of tripartite e¥ux systems of Pseudomonas aeruginosa is determined by the RND component. Biochem. Biophys. Res. Commun. 299, 247^251. [159] Cavallo, J.D., Plesiat, P., Couetdic, G. and Leblanc, F. (1997) J. Antimicrob. Chemother. 50, 1039^1043. [160] Koronakis, V., Shar¡, A., Koronakis, E., Luisi, B. and Hughes, C. (2000) Crystal structure of the bacterial membrane protein TolC central to multidrug e¥ux and protein export. Nature 405, 914^ 919. [161] Wong, K.K.Y. and Hancock, R.E.W. (2000) Insertion mutagenesis and membrane topology model of the Pseudomonas aeruginosa outer membrane protein OprM. J. Bacteriol. 182, 2402^2410. [162] Andersen, C., Koronakis, E., Hughes, C. and Koronakis, V. (2002) An aspartate ring at the TolC tunnel entrance determines ion selectivity and presents a target for blocking by large cations. Mol. Microbiol. 44, 1131^1139. [163] Yoshihara, E., Maseda, H. and Saito, K. (2002) The outer membrane component of the multidrug e¥ux pump from Pseudomonas aeruginosa may be a gated channel. Eur. J. Biochem. 269, 4738^ 4745. [164] Kawabe, T., Fujihira, E. and Yamaguchi, A. (2000) Molecular construction of a multidrug exporter system, AcrAB: Molecular interaction between AcrA and AcrB, and cleavage of the N-terminal signal sequence of AcrA. J. Biochem. 128, 195^200. [165] Avila-Sakar, A.J., Misaghi, S., Wilson-Kubalek, E.M., Downing, K.H., Zgurskaya, H., Nikaido, H. and Nogales, E. (2001) Lipidlayer crystallization and preliminary three-dimensional structural analysis of AcrA, the periplasmic component of a bacterial multidrug e¥ux pump. J. Struct. Biol. 136, 81^88. [166] Elkins, C.A. and Nikaido, H. (2002) Substrate speci¢city of the RND-type multidrug e¥ux pumps AcrB and AcrD of Escherichia coli is determined predominately by two large periplasmic loops. J. Bacteriol. 184, 6490^6498. [167] Tikhonova, E.B., Wang, Q.J. and Zgurskaya, H.I. (2002) Chimeric analysis of the multicomponent multidrug e¥ux transporters from gram-negative bacteria. J. Bacteriol. 184, 6499^6507. [168] Mao, W.M., Warren, M.S., Black, D.S., Satou, T., Murata, T., Nishino, T., Gotoh, N. and Lomovskaya, O. (2002) On the mech-
Cyaan Magenta Geel Zwart
336
[169]
[170]
[171]
[172]
[173]
[174]
[175]
[176]
[177]
[178]
[179]
[180]
[181]
[182]
[183]
[184]
[185]
[186]
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 anism of substrate speci¢city by resistance nodulation division (RND)-type multidrug resistance pumps: the large periplasmic loops of MexD from Pseudomonas aeruginosa are involved in substrate recognition. Mol. Microbiol. 46, 889^901. Maseda, H., Kitao, M., Eda, S., Yoshihara, E. and Nakae, T. (2002) A novel assembly process of the multicomponent xenobiotic e¥ux pump in Pseudomonas aeruginosa. Mol. Microbiol. 46, 677^686. Peterson, J.D., Umayam, L.A., Dickinson, T., Hickey, E.K. and White, O. (2001) The comprehensive microbial resource. Nucleic Acids Res. 29, 123^125. Rensing, C., Pribyl, T. and Nies, D.H. (1997) New functions for the three subunits of the CzcCBA cation-proton-antiporter. J. Bacteriol. 179, 6871^6879. Lopez-Maury, L., Garcia-Dominguez, M., Florencio, F.J. and Reyes, J.C. (2002) A two-component signal transduction system involved in nickel sensing in the cyanobacterium Synechocystis sp. PCC 6803. Mol. Microbiol. 43, 247^256. Grass, G., Fan, B., Rosen, B.P., Lemke, K., Schlegel, H.G. and Rensing, C. (2001) NreB from Achromobacter xylosoxidans 31A is a nickel-induced transporter conferring nickel resistance. J. Bacteriol. 183, 2803^2807. Waidner, B., Melchers, K., Ivanov, I., Loferer, H., Bensch, K.W., Kist, M. and Bereswill, S. (2002) Identi¢cation by RNA pro¢ling and mutational analysis of the novel copper resistance determinants CrdA (HP1326), CrdB (HP1327), and CzcB (HP1328) in Helicobacter pylori. J. Bacteriol. 184, 6700^6708. Paulsen, I.T. and Saier Jr., M.H. (1997) A novel family of ubiquitous heavy metal ion transport proteins. J. Membr. Biol. 156, 99^ 103. Gu¡anti, A.A., Wei, Y., Rood, S.V. and Krulwich, T.A. (2002) An antiport mechanism for a member of the cation di¡usion facilitator family: divalent cations e¥ux in exchange for Kþ and Hþ . Mol. Microbiol. 45, 145^153. Wang, W., Gu¡anti, A.A., Wei, Y., Ito, M. and Krulwich, T.A. (2000) Two types of Bacillus subtilis tetA(L) deletion strains reveal the physiological importance of TetA(L) in Kþ acquisition as well as in Naþ , alkali, and tetracycline resistance. J. Bacteriol. 182, 2088^ 2095. Bloss, T., Clemens, S. and Nies, D.H. (2002) Characterisation of the ZAT1p zinc transporter from Arabidopsis thaliana in microbial model organisms and reconstituted proteoliposomes. Planta 214, 783^791. Anton, A., GroMe, C., ReiMman, J., Pribyl, T. and Nies, D.H. (1999) CzcD is a heavy metal ion transporter involved in regulation of heavy metal resistance in Ralstonia sp. strain CH34. J. Bacteriol. 181, 6876^6881. Sturr, M.G., Ablooglu, A.J. and Krulwich, T.A. (1997) A Bacillus subtilis locus encoding several gene products a¡ecting transport of cation. Gene 188, 91^94. Xiong, A.M. and Jayaswal, R.K. (1998) Molecular characterization of a chromosomal determinant conferring resistance to zinc and cobalt ions in Staphylococcus aureus. J. Bacteriol. 180, 4024^4029. Kuroda, M., Hayashi, H. and Ohta, T. (1999) Chromosome-determined zinc-responsible operon czr in Staphylococcus aureus strain 912. Microbiol. Immunol. 43, 115^125. Spada, S., Pembroke, J.T. and Wall, J.G. (2002) Isolation of a novel Thermus thermophilus metal e¥ux protein that improves Escherichia coli growth under stress conditions. Extremophiles 6, 301^308. Grass, G., Fan, B., Rosen, B.P., Franke, S., Nies, D.H. and Rensing, C. (2001) ZitB (YbgR), a member of the cation di¡usion facilitator family, is an additional zinc transporter in Escherichia coli. J. Bacteriol. 183, 4664^4667. Lee, S.M., Grass, G., Haney, C.J., Fan, B., Rosen, B.P., Anton, A., Nies, D.H. and Rensing, C. (2002) Functional analysis of the Escherichia coli zinc transporter ZitB. FEMS Microbiol. Lett. 215, 273^278. Gru«nberg, K., Wawer, C., Tebo, B.M. and Schu«ler, D. (2001) A
FEMSRE 781 2-6-03
[187]
[188]
[189]
[190]
[191]
[192]
[193]
[194]
[195]
[196]
[197]
[198]
[199]
[200]
[201]
[202]
[203]
[204]
large gene cluster encoding several magnetosome proteins is conserved in di¡erent species of magnetotactic bacteria. Appl. Environ. Microbiol. 67, 4573^4582. Buseck, P.R., Dunin-Borkowski, R.E., Devouard, B., Frankel, R.B., McCartney, M.R., Midgley, P.A., Po¤sfai, M. and Weyland, M. (2001) Magnetite morphology and life on Mars. Proc. Natl. Acad. Sci. USA 98, 13490^13495. Thomas-Keprta, K.L., Clemett, S.J., Bazylinski, D.A., Kirschvink, J.L., McKay, D.S., Wentworth, S.J., Vali, H., Gibson Jr., J.E.K. and Romanek, C.S. (2002) Magnetofossils from ancient Mars: a robust biosignature in the Martian meteorite ALH84001. Appl. Environ. Microbiol. 68, 3663^3672. Luhmann, J.G. and Russell, C.T. (1997) Mars: magnetic ¢eld and magnetosphere. In: Encyclopedia of Planetary Sciences (Shirley, J.H. and Fainbridge, R.W., Eds.), pp. 454^456. Chapman and Hall, New York. Paulsen, I.T., Sliwinski, M.K., Nelissen, B., Go¡eau, A. and Saier, M.H. (1998) Uni¢ed inventory of established and putative transporters encoded within the complete genome of Saccharomyces cerevisiae. FEBS Lett. 430, 116^125. Kamizomo, A., Nishizawa, M., Teranishi, A., Murata, K. and Kimura, A. (1989) Identi¢cation of a gene conferring resistance to zinc and cadmium ions in the yeast Saccharomyces cerevisiae. Mol. Gen. Genet. 219, 161^167. Conklin, D.S., McMaster, J.A., Culbertson, M.R. and Kung, C. (1992) COT1, a gene involved in cobalt accumulation in Saccharomyces cerevisiae. Mol. Cell. Biol. 12, 3678^3688. Conklin, D.S., Culbertson, M.R. and Kung, C. (1994) Interactions between gene products involved in divalent cation transport in Sacchromyces cerevisiae. Mol. Gen. Genet. 244, 303^311. MacDiarmid, C.W., Gaither, L.A. and Eide, D. (2000) Zinc transporters that regulate vacuolar zinc storage in Saccharomyces cerevisiae. EMBO J. 19, 2845^2855. Miyabe, S., Izawa, S. and Inoue, Y. (2001) The Zrc1 is involved in zinc transport system between vacuole and cytosol in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 282, 79^83. MacDiarmid, C.W., Milanick, M.A. and Eide, D.J. (2002) Biochemical properties of vacuolar zinc transport systems of Saccharomyces cerevisiae. J. Biol. Chem. 277, 39187^39194. Miyabe, S., Izawa, S. and Inoue, Y. (2000) Expression of ZRC1 coding for suppressor of zinc toxicity is induced by zinc-starvation stress in Zap1-dependent fashion in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 276, 879^884. MacDiarmid, C.W. and Gardner, R.C. (1998) Overexpression of the Saccharomyces cerevisiae magnesium transport system confers resistance to aluminum ion. J. Biol. Chem. 273, 1727^1732. Kobayashi, S., Miyabe, S., Izawa, S., Inoue, Y. and Kimura, A. (1996) Correlation of the Osr/Zrc1 gene product and the intracellular glutathione levels in Saccharomyces cerevisiae. Biotechnol. Appl. Biochem. 23, 3^6. Li, L. and Kaplan, J. (2001) The yeast gene MSC2, a member of the cation di¡usion facilitator family, a¡ects the cellular distribution of zinc. J. Biol. Chem. 276, 5036^5043. Li, L. and Kaplan, J. (1997) Characterization of two homologous yeast genes that encode mitochondrial iron transporters. J. Biol. Chem. 272, 28485^28493. Clemens, S., Bloss, T., Vess, C., Neumann, D., Nies, D.H. and zur Nieden, U. (2002) A transporter in the endoplasmic reticulum/nuclear envelope of Schizosaccharomyces pombe cells di¡erentially a¡ects transition metal tolerance. J. Biol. Chem. 277, 18215^ 18221. Palmiter, R.D. and Findley, S.D. (1995) Cloning and functional characterization of a mammalian zinc transporter that confers resistance to zinc. EMBO J. 14, 639^649. Palmiter, R.D., Cole, T.B. and Findley, S.D. (1996) ZnT-2, a mammalian protein that confers resistance to zinc by facilitating vesicular sequestration. EMBO J. 15, 1784^1791.
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 [205] Palmiter, R.D., Cole, T.B., Quaife, C.J. and Findley, S.D. (1996) ZnT3, a putative transporter of zinc into synaptic vesicles. Proc. Natl. Acad. Sci. USA 93, 14934^14939. [206] Wenzel, H.J., Cole, T.B., Born, D.E., Schwartzkroin, P.A. and Palmiter, R.D. (1997) Ultrastructral localization of zinc transporter-3 (ZnT-3) to synaptic vesicle membranes within mossy ¢ber boutons in the hippocampus of mouse and monkey. Proc. Natl. Acad. Sci. USA 94, 12676^12681. [207] Saito, T., Takahashi, K., Nakagawa, N., Hosokawa, T., Kurasaki, M., Yamanoshita, O., Yamamoto, Y., Sasaki, H., Nagashima, K. and Fujita, H. (2000) De¢ciencies of hippocampal Zn and ZnT3 accelerate brain aging of rat. Biochem. Biophys. Res. Commun. 279, 505^511. [208] Michalczyk, A.A., Allen, J., Blomeley, R.C. and Ackland, M.L. (2002) Constitutive expression of hZnT4 zinc transporter in human breast epithelial cells. Biochem. J. 364, 105^113. [209] Kambe, T., Narita, H., Yamaguchi-Iwai, Y., Hirose, J., Amano, T., Sugiura, N., Sasaki, R., Mori, K., Iwanaga, T. and Nagao, M. (2002) Cloning and characterization of a novel mammalian zinc transporter, zinc transporter 5, abundantly expressed in pancreatic beta cells. J. Biol. Chem. 277, 19049^19055. [210] Inoue, K., Matsuda, K., Itoh, M., Kawaguchi, H., Tomoike, H., Aoyagi, T., Nagai, R., Hori, M., Nakamura, Y. and Tanaka, T. (2002) Osteopenia and male-speci¢c sudden cardiac death in mice lacking a zinc transporter gene, Znt5. Hum. Mol. Genet. 11, 1775^ 1784. [211] Murgia, C., Vespignani, I., Cerase, J., Nobili, F. and Perozzi, G. (1999) Cloning, expression, and vesicular localization of zinc transporter Dri 27/ZnT4 in intestinal tissue and cells. Am. J. Physiol. Gastrointest. Liver Physiol. 277, G1231^G1239. [212] Cragg, R.A., Christie, G.R., Phillips, S.R., Russi, R.M., Kury, S., Mathers, J.C., Taylor, P.M. and Ford, D. (2002) A novel zinc-regulated human zinc transporter, hZTL1, is localized to the enterocyte apical membrane. J. Biol. Chem. 277, 22789^22797. [213] Huang, L.P., Kirschke, C.P. and Gitschier, J. (2002) Functional characterization of a novel mammalian zinc transporter, ZnT6. J. Biol. Chem. 277, 26389^26395. [214] van der Zaal, B.J., Neuteboom, L.W., Pinas, J.E., Chardonnens, A.N., Schat, H., Verkleij, J.A.C. and Hooykaas, P.J.J. (1999) Overexpression of a novel Arabidopsis gene related to putative zinc-transporter genes from animals can lead to enhanced zinc resistance and accumulation. Plant Physiol. 119, 1047^1055. [215] Ma«ser, P., Thomine, S., Schroeder, J.I., Ward, J.M., Hirschi, K., Sze, H., Talke, I.N., Amtmann, A., Maathuis, F.J.M., Sanders, D., Harper, J.F., Tchieu, J., Gribskov, M., Persans, M.W., Salt, D.E., Kim, S.A. and Guerinot, M.L. (2001) Phylogenetic relationships within cation transporter families of Arabidopsis. Plant Physiol. 126, 1646^1667. [216] Assuncao, A.G.L., Martins, P.D., De Folter, S., Vooijs, R., Schat, H. and Aarts, M.G.M. (2001) Elevated expression of metal transporter genes in three accessions of the metal hyperaccumulator Thlaspi caerulescens. Plant Cell Environ. 24, 217^226. [217] Persans, M.W., Nieman, K. and Salt, D.E. (2001) Functional activity and role of cation-e¥ux family members in Ni hyperaccumulation in Thlaspi goesingense. Proc. Natl. Acad. Sci. USA 98, 9995^ 10000. [218] Fagan, M.J. and Saier Jr., M.H. (1994) P-type ATPases of eukaryotes and bacteria : sequence comparisons and construction of phylogenetic trees. J. Mol. Evol. 38, 57^99. [219] Snavely, M.D., Florer, J.B., Miller, C.G. and Maguire, M.E. (1989) Magnesium transport in Salmonella typhimurium: 28 Mg2þ transport by CorA, MgtA, and MgtB systems. J. Bacteriol. 171, 4761^ 4766. [220] Fan, B. and Rosen, B.P. (2002) Biochemical characterization of CopA, the Escherichia coli Cu(I)-translocating P-type ATPase. J. Biol. Chem. 277, 46987^46992. [221] Mandal, A.K., Cheung, W.D. and Arguello, J.M. (2002) Character-
FEMSRE 781 2-6-03
[222]
[223]
[224]
[225]
[226]
[227]
[228]
[229]
[230]
[231]
[232]
[233]
[234]
[235]
[236]
[237]
[238] [239]
337
ization of a thermophilic P-type Agþ /Cuþ -ATPase from the extremophile Archaeoglobus fulgidus. J. Biol. Chem. 277, 7201^7208. Tottey, S., Rich, P.R., Rondet, S.A.M. and Robinson, N.J. (2001) Two Menkes-type ATPases supply copper for photosynthesis in Synechocystis PCC 6803. J. Biol. Chem. 276, 19999^20004. Banci, L., Bertini, L., Cio¢-Ba¡oni, S., D’Onofrio, M., Gonnelli, L., Marhuenda-Egea, F.C. and Ruiz-Duenas, F.J. (2002) Solution structure of the N-terminal domain of a potential copper-translocating P-type ATPase from Bacillus subtilis in the apo and Cu(I) loaded states. J. Mol. Biol. 317, 415^429. Hasman, H. and Aarestrup, F.M. (2002) tcrB, a gene conferring transferable copper resistance in Enterococcus faecium : occurrence, transferability, and linkage to macrolide and glycopeptide resistance. Antimicrob. Agents Chemother. 46, 1410^1416. Bissig, K.D., Wunderli-Ye, H., Duda, P.W. and Solioz, M. (2001) Structure-function analysis of puri¢ed Enterococcus hirae CopB copper ATPase : e¡ect of Menkes/Wilson disease mutation homologues. Biochem. J. 357, 217^223. Bissig, K.D., Voegelin, T.C. and Solioz, M. (2001) Tetrathiomolybdate inhibition of the Enterococcus hirae CopB copper ATPase. FEBS Lett. 507, 367^370. Wunderli-Ye, H. and Solioz, M. (2001) Puri¢cation and functional analysis of the copper ATPase CopA of Enterococcus hirae. Biochem. Biophys. Res. Commun. 280, 713^719. Vats, N. and Lee, S.F. (2001) Characterization of a copper-transport operon, copYAZ, from Streptococcus mutans. Microbiology 147, 653^662. Stentz, R., Loizel, C., Malleret, C. and Zagorec, M. (2000) Development of genetic tools for Lactobacillus sakei: Disruption of the betagalactosidase gene and use of lacZ as a reporter gene to study regulation of the putative copper ATPase, AtkB. Appl. Environ. Microbiol. 66, 4272^4278. Reeve, W.G., Tiwari, R.P., Kale, N.B., Dilworth, M.J. and Glenn, A.R. (2002) ActP controls copper homeostasis in Rhizobium leguminosarum bv. viciae and Sinorhizobium meliloti preventing low pHinduced copper toxicity. Mol. Microbiol. 43, 981^991. Adaikkalam, V. and Swarup, S. (2002) Molecular characterization of an operon, cueAR, encoding a putative P1-type ATPase and a MerR-type regulatory protein involved in copper homeostasis in Pseudomonas putida. Microbiology (UK) 148, 2857^2867. Burlando, B., Evangelisti, V., Dondero, F., Pons, G., Camakaris, J. and Viarengo, A. (2002) Occurrence of Cu-ATPase in Dictyostelium : possible role in resistance to copper. Biochem. Biophys. Res. Commun. 291, 476^483. LaGier, M.J., Zhu, G. and Keithly, J.S. (2001) Characterization of a heavy metal ATPase from the apicomplexan Cryptosporidium parvum. Gene 266, 25^34. Riggle, P.J. and Kumamoto, C.A. (2000) Role of a Candida albicans P1-type ATPase in resistance to copper and silver ion toxicity. J. Bacteriol. 182, 4899^4905. Axelsen, K.B. and Palmgren, M.G. (2001) Inventory of the superfamily of P-type ion pumps in Arabidopsis. Plant Physiol. 126, 696^ 706. Hirayama, T., Kieber, J.J., Hirayama, N., Kogan, M., Guzman, P., Nourizadeh, S., Alonso, J.M., Dailey, W.P., Dancis, A. and Ecker, J.R. (1999) Responsive-to-antagonist1, a Menkes/Wilson disease-related copper transporter, is required for ethylene signaling in Arabidopsis. Cell 97, 383^393. Larin, D., Mekios, C., Das, K., Ross, B., Yang, A.S. and Gilliam, T.C. (1999) Characterization of the interaction between the Wilson and Menkes disease proteins and the cytoplasmic copper chaperone, HAH1p. J. Biol. Chem. 274, 28497^28504. Butler, P., McIntyre, N. and Mistry, P.K. (2001) Molecular diagnosis of Wilson disease. Mol. Genet. Metab. 72, 223^230. Udo, E.E., Jacob, L.E. and Mathew, B. (2000) A cadmium resistance plasmid, pXU5, in Staphylococcus aureus, strain ATCC25923. FEMS Microbiol. Lett. 189, 79^80.
Cyaan Magenta Geel Zwart
338
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339
[240] Solovieva, I.M. and Entian, K.D. (2002) Investigation of the yvgW Bacillus subtilis chromosomal gene involved in Cd2þ ion resistance. FEMS Microbiol. Lett. 208, 105^109. [241] Bal, N., Mintz, E., Guillain, F. and Catty, P. (2001) A possible regulatory role for the metal-binding domain of CadA, the Listeria monocytogenes Cd2þ -ATPase. FEBS Lett. 506, 249^252. [242] Lee, S.W., Glickmann, E. and Cooksey, D.A. (2001) Chromosomal locus for cadmium resistance in Pseudomonas putida consisting of a cadmium-transporting ATPase and a MerR family response regulator. Appl. Environ. Microbiol. 67, 1437^1444. [243] Alonso, A., Sanchez, P. and Martinez, J.L. (2000) Stenotrophomonas maltophilia D457R contains a cluster of genes from gram-positive bacteria involved in antibiotic and heavy metal resistance. Antimicrob. Agents Chemother. 44, 1778^1782. [244] Beard, S.J., Hashim, R., MembrilloHernandez, J., Hughes, M.N. and Poole, R.K. (1997) Zinc(II) tolerance in Escherichia coli K12: evidence that the zntA gene (o732) encodes a cation transport ATPase. Mol. Microbiol. 25, 883^891. [245] Rensing, C., Mitra, B. and Rosen, B.P. (1997) The zntA gene of Escherichia coli encodes a Zn(II)-translocating P-type ATPase. Proc. Natl. Acad. Sci. USA 94, 14326^14331. [246] Rensing, C., Sun, Y., Mitra, B. and Rosen, B.P. (1998) Pb(II)-translocating P-type ATPases. J. Biol. Chem. 273, 32614^32617. [247] Sharma, R., Rensing, C., Rosen, B.P. and Mitra, B. (2000) The ATP hydrolytic activity of puri¢ed ZntA, a Pb(II)/Cd(II)/Zn(II)-translocating ATPase from Escherichia coli. J. Biol. Chem. 275, 3873^3878. [248] Mitra, B. and Sharma, R. (2001) The cysteine-rich amino-terminal domain of ZntA, a Pb(II)/Zn(II)/Cd(II)-translocating ATPase from Escherichia coli, is not essential for its function. Biochemistry 40, 7694^7699. [249] Shiraishi, E., Inouhe, M., Joho, M. and Tohoyama, H. (2000) The cadmium-resistant gene, CAD2, which is a mutated putative coppertransporter gene (PCA1), controls the intracellular cadmium-level in the yeast S. cerevisiae. Curr. Genet. 37, 79^86. [250] Tong, L., Nakashima, S., Shibasaka, M., Katsuhara, M. and Kasamo, K. (2002) A novel histidine-rich CPx-ATPase from the ¢lamentous cyanobacterium Oscillatoria brevis related to multiple-heavymetal cotolerance. J. Bacteriol. 184, 5027^5035. [251] Herrmann, L., Schwan, D., Garner, R., Mobley, H.L.T., Haas, R., Scha«fer, K.P. and Melchers, K. (1999) Helicobacter pylori cadA encodes an essential Cd(II)-Zn(II)-Co(II) resistance factor in£uencing urease activity. Mol. Microbiol. 33, 524^536. [252] Rutherford, J.C., Cavet, J.S. and Robinson, N.J. (1999) Cobalt-dependent transcriptional switching by a dual-e¡ector MerR-like protein regulates a cobalt-exporting variant CPx-type ATPase. J. Biol. Chem. 274, 25827^25832. [253] Amoroso, M.J., Schubert, D., Mitscherlich, P., Schumann, P. and Kothe, E. (2000) Evidence for high a⁄nity nickel transporter genes in heavy metal resistant Streptomyces spec. J. Basic Microbiol. 40, 295^301. [254] Degen, O. and Eitinger, T. (2002) Substrate speci¢city of nickel/ cobalt permeases: Insights from mutants altered in transmembrane domains I and II. J. Bacteriol. 184, 3569^3577. [255] Eitinger, T. and Friedrich, B. (1994) A topological model for the high-a⁄nity nickel transporter of Alcaligenes eutrophus. Mol. Microbiol. 12, 1025^1032. [256] Eitinger, T., Wolfram, L., Degen, O. and Anthon, C. (1997) A Ni2þ binding motif is the basis of high a⁄nity transport of the Alcaligenes eutrophus nickel permease. J. Biol. Chem. 272, 17139^17144. [257] Odermatt, A., Suter, H., Krapf, R. and Solioz, M. (1992) An ATPase operon involved in copper resistance by Enterococcus hirae. Ann. NY Acad. Sci. 671, 484^486. [258] Odermatt, A., Suter, H., Krapf, R. and Solioz, M. (1993) Primary structure of two P-type ATPases involved in copper homeostasis in Enterococcus hirae. J. Biol. Chem. 268, 12775^12779. [259] Odermatt, A., Krapf, R. and Solioz, M. (1994) Induction of the putative copper ATPases, CopA and CopB, of Enterococcus hirae
FEMSRE 781 2-6-03
[260]
[261]
[262]
[263]
[264]
[265]
[266] [267]
[268]
[269]
[270]
[271]
[272]
[273]
[274]
[275]
[276]
[277]
[278]
[279]
by Agþ and Cu2þ , and Agþ extrusion by CopB. Biochem. Biophys. Res. Commun. 202, 44^48. Odermatt, A. and Solioz, M. (1995) Two trans-acting metalloregulatory proteins controlling expression of the copper-ATPases of Enterococcus hirae. J. Biol. Chem. 270, 4349^4354. Lu, Z.H., Dameron, C.T. and Solioz, M. (2003) The Enterococcus hirae paradigm of copper homeostasis: Copper chaperone turnover, interactions, and transactions. Biometals 16, 137^143. Gaballa, A. and Helmann, J.D. (2002) A peroxide-induced zinc uptake system plays an important role in protection against oxidative stress in Bacillus subtilis. Mol. Microbiol. 45, 997^1005. Tsai, K.J., Lin, Y.F., Wong, M.D., Yang, H.H.C., Fu, H.L. and Rosen, B.P. (2002) Membrane topology of the pl258 CadA Cd(II)/ Pb(II)/Zn(II)-translocating P-type ATPase. J. Bioenerg. Biomembr. 34, 147^156. DiDonato, M., Zhang, J.Y., Que, L. and Sarkar, B. (2002) Zinc binding to the NH2 -terminal domain of the Wilson disease copper-transporting ATPase ^ Implications for in vivo metal ion-mediated regulation of ATPase activity. J. Biol. Chem. 277, 13409^ 13414. Borremans, B., Hobman, J.L., Provoost, A., Brown, N.L. and Van der Lelie, D. (2001) Cloning and functional analysis of the pbr lead resistance determinant of Ralstonia metallidurans CH34. J. Bacteriol. 183, 5651^5658. Baker, A.J.M. (1987) Metal tolerance in plants. New Phytol. 106, 93^111. Nies, D.H., Koch, S., Wachi, S., Peitzsch, N. and Saier, M.H.J. (1998) CHR, a novel family of prokaryotic proton motive forcedriven transporters probably containing chromate/sulfate transporters. J. Bacteriol. 180, 5799^5802. Alvarez, A.H., Moreno-Sanchez, R. and Cervantes, C. (1999) Chromate e¥ux by means of the ChrA chromate resistance protein from Pseudomonas aeruginosa. J. Bacteriol. 181, 7398^7400. Pimentel, B.E., Moreno-Sanchez, R. and Cervantes, C. (2002) E¥ux of chromate by Pseudomonas aeruginosa cells expressing the ChrA protein. FEMS Microbiol. Lett. 212, 249^254. Nies, A., Nies, D.H. and Silver, S. (1990) Nucleotide sequence and expression of a plasmid-encoded chromate resistance determinant from Alcaligenes eutrophus. J. Biol. Chem. 265, 5648^5653. Nies, A., Nies, D.H. and Silver, S. (1989) Cloning and expression of plasmid genes encoding resistances to chromate and cobalt in Alcaligenes eutrophus. J. Bacteriol. 171, 5065^5070. Peitzsch, N., Eberz, G. and Nies, D.H. (1998) Alcaligenes eutrophus as a bacterial chromate sensor. Appl. Environ. Microbiol. 64, 453^ 458. Juhnke, S., Peitzsch, N., Hu«bener, N., GroMe, C. and Nies, D.H. (2002) New genes involved in chromate resistance in Ralstonia metallidurans strain CH34. Arch. Microbiol. 179, 15^25. Cervantes, C., Ohtake, H., Chu, L., Misra, T.K. and Silver, S. (1990) Cloning, nucleotide sequence, and expression of the chromate resistance determinant of Pseudomonas aeruginosa plasmid pUM505. J. Bacteriol. 172, 287^291. Nicholson, M.L. and Laudenbach, D.E. (1995) Genes encoded on a cyanobacterial plasmid are transcriptionally regulated by sulfur availability and cysR. J. Bacteriol. 177, 2143^2150. Lenz, O., Schwartz, E., Dernedde, J., Eitinger, T. and Friedrich, B. (1994) The Alcaligenes eutrophus H16 hoxX gene participates in hydrogenase regulation. J. Bacteriol. 176, 4385^4393. Mobley, H.L.T., Garner, R.M. and Bauerfeind, P. (1995) Helicobacter pylori nickel-transport gene NixA ^ Synthesis of catalytically active urease in Escherichia coli independent of growth conditions. Mol. Microbiol. 16, 97^109. Fulkerson, J.F., Garner, R.M. and Mobley, H.L.T. (1998) Conserved residues and motifs in the NixA protein of Helicobacter pylori are critical for the high a⁄nity transport of nickel ions. J. Biol. Chem. 273, 235^241. Depina, K., Navarro, C., Mcwalter, L., Boxer, D.H., Price, N.C.
Cyaan Magenta Geel Zwart
D.H. Nies / FEMS Microbiology Reviews 27 (2003) 313^339 and Kelly, S.M. (1995) Puri¢cation and characterization of the periplasmic nickel-binding protein NikA of Escherichia coli. Eur. J. Biochem. 227, 857^865. [280] Altschul, S.F., Madden, T.L., Scha«¡er, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSIBLAST : a new generation of protein database search programs. Nucleic Acids Res. 25, 3389^3402. [281] Altschul, K., Altschul, S. and Altschul, S.F. (1990) Methods for assessing the statistical signi¢cance of molecular sequence features by using general scoring schemes. Proc. Natl. Acad. Sci. USA 87, 2264^2268.
FEMSRE 781 2-6-03
339
[282] Altschul, K., Altschul, S. and Altschul, S.F. (1993) Applications and statistics for multiple high-scoring segments in molecular sequences. Proc. Natl. Acad. Sci. USA 90, 5873^5877. [283] Solioz, M. and Stoyanov, J.V. (2003) Copper homeostasis in Enterococcus hirae. FEMS Microbiol. Rev. 27, 183^195. [284] Cavet, J.S., Borrelly, G.P.M. and Robinson, N.J. (2003) Zn, Cu and Co in cyanobacteria : selective control of metal availability. FEMS Microbiol. Rev. 27, 165^181.
Cyaan Magenta Geel Zwart