- Email: [email protected]
Green fluorescent protein – a bright idea for the study of bacterial protein localization
FEMS Microbiology Letters 204 (2001) 9^18
www.fems-microbiology.org
MiniReview
Green £uorescent protein ^ a bright idea for the study of bacterial protein localization Gregory J. Phillips * Department of Microbiology, 207 Science I Building, Iowa State University, Ames, IA 50011, USA Received 20 February 2001; received in revised form 26 July 2001 ; accepted 26 July 2001 First published online 6 September 2001
Abstract Use of the green fluorescent protein (GFP) of Aequorea victoria as a reporter for protein and DNA localization has provided sensitive, new approaches for studying the organization of the bacterial cell, leading to new insights into diverse cellular processes. GFP has many characteristics that make it useful for localization studies in bacteria, primarily its ability to fluoresce when fused to target polypeptides without the addition of exogenously added substrates. As an alternative to immunofluorescence microscopy, the expression of gfp gene fusions has been used to probe the function of cellular components fundamental for DNA replication, translation, protein export, and signal transduction, that heretofore have been difficult to study in living cells. Moreover, protein and DNA localization can now be monitored in real time, revealing that several proteins important for cell division, development and sporulation are dynamically localized throughout the cell cycle. The use of additional GFP variants that permit the labeling of multiple components within the same cell, and the use of GFP for genetic screens, should continue to make this a valuable tool for addressing complex questions about the bacterial cell. ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Green £uorescent protein; Protein localization; Gene fusion; Hybrid protein ; Fluorescence microscopy
1. Introduction The use of lacZ gene fusions, encoding hybrid proteins composed of a target polypeptide fused to Escherichia coli L-galactosidase, has revolutionized our ability to study bacterial gene expression and protein export [1]. A similar impact is being made by application of green £uorescent protein (GFP) from Aequorea victoria to study macromolecular localization in living cells, including bacteria [2]. Many important questions about the subcellular organization of bacterial cells have been di¤cult to address due, in part, to their small size. GFP has, however, provided new experimental approaches to investigate the localization of both protein and DNA molecules in prokaryotes. The traditional approach to study macromolecular localization in bacteria includes techniques of immunogold labeling and immuno£uorescence microscopy (IFM). Although these methods have been e¡ectively used to vis-
* Tel. : +1 (515) 294-1525; Fax: +1 (515) 294-6019. E-mail address : [email protected] (G.J. Phillips).
ualize the targeting of speci¢c cellular components, their use can be limited by the requirement for antibodies directed against the target protein. Antibodies against proteins found in low abundance can be di¤cult to obtain, and the level of detection of this class of protein is relatively insensitive. They also require cell ¢xation that precludes investigation of living cells. Moreover, these techniques can be technically demanding and costly. Gene fusions using gfp, on the other hand, can be constructed with relative ease (as discussed further below), and their use, in general, heightens the sensitivity of molecular detection. Since the GFP chromophore is formed by autocatalytic cyclization of three amino acids [3], and £uorescence is triggered by excitation with speci¢c wavelength light, neither invasive sample preparation nor substrate addition is required, making it possible to study protein localization in living cells. In addition, since GFP is relatively small (238 amino acids) and can remain functional even when fused to a target protein, it allows protein localization to be monitored in real time. As a result, studies using GFP have bolstered our understanding of a variety of cellular processes such as cell division, chromosome replication and partitioning,
0378-1097 / 01 / $20.00 ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 0 1 ) 0 0 3 5 8 - 5
FEMSLE 10108 17-10-01
10
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
sporulation, development and signal transduction, and have changed our view of the organization of the bacterial cell in general. The bacterial cell, once considered by analogy to be a `bag of enzymes', is now known to accommodate the segregation and sequestration of a variety of molecules into speci¢c subcellular regions. Surprisingly, many of the cellular components are dynamic in nature, assembling and disassembling into higher order structures, and migrating in a directed manner between distinct locations in the bacterial cell, revealing more similarities between prokaryotes and eukaryotes that had previously been imagined [4^6]. 2. Creation and use of GFP fusions Early studies in which wild-type GFP was expressed in E. coli revealed a number of problems that hindered its use as a reporter for protein localization. For example, formation of the GFP chromophore is slow, requiring up to 2 h post-synthesis to form [7]. Wild-type GFP also tends to form inactive inclusion bodies, especially at elevated temperatures. In addition, the £uorescent intensity of GFP is relatively low, and the protein has a major absorbance peak at 395 nm, a value that requires specialized ¢lter sets for optimal excitation. Fortunately, a number of variants of GFP have been isolated that overcome many of these limitations [7^11]. Cormack et al. [10] isolated a particularly valuable set of GFP variants by screening for mutants that showed increased brightness upon excitation at 488 nm. In addition to improved solubility in E. coli, an amino acid change within the chromophore resulted in a `red-shift' in excitation from 395 to 488 nm [11,12]. These variants can be detected with good sensitivity by £uorescein isothiocyanate ¢lter sets commonly found on £uorescence microscopes. Derivatives of these mutants are used almost exclusively in protein localization studies. A number of vectors have been constructed to facilitate creation of gene fusions with di¡erent gfp variants. Many of these vectors have either been published [8,13,14], or are commercially available (e.g. Clontech, Palo Alto, CA, USA). A general strategy for constructing GFP fusions is to use PCR, or other cloning strategies [15] to engineer restriction enzyme sites into both gfp and the target gene to create in-frame fusions on recombinant plasmids. The gfp reporter may be fused to either the 5P or 3P end of the target gene, to generate either amino- or carboxy-terminal fusions, respectively. This choice is often made by predicting which region of the protein will most likely tolerate the addition of extra amino acids and still remain functional. Also, when fusing GFP to membrane proteins, GFP must be joined to a domain that is exposed to the cytoplasmic side of the inner membrane. Where necessary, linker sequences may also be added between the target protein and GFP to improve £uorescence output [16].
GFP is not without its limitations, however. For example, formation of the chromophore requires molecular oxygen, and all forms of GFP display reduced £uorescence at elevated growth temperatures [7]. Consequently, GFP cannot be used under all growth conditions under which microorganisms normally grow. Certain types of proteins also are not suitable for detection by GFP. For example, because of the relatively slow formation of the chromophore, even in the GFP variants, proteins with short halflives may not be detectable. Also, proteins must still be present in su¤cient levels since detection of only a few molecules of GFP is technically di¤cult, with results easily misinterpreted. As with any fusion protein, caution must also be used when interpreting the results of protein localization studies since the addition of the GFP moiety can alter the function, or location of the target polypeptide. A genetic test for altered function is to determine if expression of a GFP fusion can complement a knockout mutation in the gene encoding the target protein. Fortunately, there are several examples, as discussed below, of proteins that remain fully functional, as determined by complementation studies, when fused to GFP. There are also instances where GFP does disrupt normal protein function without interfering with its localization. It is di¤cult to assess from the literature, however, the number of instances where GFP fusion proteins are not active, or have altered localization properties, and each case must be assessed empirically. Foci of GFP £uorescence must also be carefully scrutinized to ensure that they are the result of proteins targeted to a speci¢c subcellular location, and not merely caused by protein aggregation. Also, £uorescent intensity may be arti¢cially skewed by elevated concentrations of GFP within small volumes of subcellular compartments [17]. Another complication that can arise is that synthesis of elevated levels of hybrid proteins can yield incorrect or ambiguous interpretations about protein localization. For example, overexpression of cell division proteins can themselves disrupt normal cell division. In addition, highlevel protein expression can result in protein aggregation or mislocalization. One solution has been provided by Boyd et al. [18] who developed a convenient method to introduce plasmid-borne gfp constructs to the E. coli chromosome, thus allowing expression of the fusions at more physiological relevant levels. Inconsistencies in results between laboratories using GFP fusions have also been reported (compare [19,20] and [21,22]). It is likely that di¡erences in bacterial strains, growth conditions, growth rates, or fusion constructs may explain these discrepancies. In these cases the use of alternative techniques, such as IFM, may be helpful to test conclusions reached by the use of GFP. These limitations notwithstanding, GFP has been successfully used to study protein and DNA localization in a variety of di¡erent microbial systems. A summary of these applications is shown in Table 1 with many of the speci¢c
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
11
Table 1 GFP fusions used to characterize bacterial cellular processes Cellular process
Target protein and function (source)
Reference
DivIVA ^ cell division protein (B. subtilis) FtsA ^ cell division protein (E. coli) FtsH ^ cell division protein (B. subtilis) FtsI ^ cell division protein (E. coli) FtsK ^ cell division protein (E. coli) FtsL ^ cell division protein (E. coli) FtsL ^ cell division protein (B. subtilis) FtsQ ^ cell division protein (E. coli) FtsZ ^ cell division protein (E. coli) MinC ^ cell division protein (E. coli) MinD ^ cell division protein (E. coli) MinD ^ cell division protein (B. subtilis) MinE ^ cell division protein (E. coli) ZipA ^ cell division protein (E. coli) B. subtilis E. coli ParC (topoisomerase IV) (B. subtilis) SopB ^ F factor partitioning (E. coli) DNA polymerase III subunits, PolC, d, NP (B. subtilis) Hbsu ^ histone-like protein (B. subtilis) HU ^ DNA binding protein SeqA ^ replication initiation (E. coli) SMC ^ nucleoid-associated protein (B. subtilis) Replication terminator protein (RTP) (B. subtilis) F, P1 (E. coli) Multicopy plasmids (E. coli) R1 (E. coli) Sec-dependent export (E. coli) TAT-dependent export (E. coli) CckA ^ kinase (C. crescentus) Che proteins (E. coli) DivJ and PleC (C. crescentus) DivK (C. crescentus) PilS ^ signal transduction (Pseudomonas aeruginosa) CotE ^ spore coat (B. subtilis) DacF ^ prespore-speci¢c protein (B. subtilis) SigE ^ sigma factor (B. subtilis) Soj ^ chromosome segregation and transcription (B. subtilis) SpoOJ ^ chromosome segregation (B. subtilis) SpoIIE ^ phosphatase (Bacillus) SpoIIGA ^ protease (B. subtilis) SpoIVA ^ spore development (B. subtilis) SpoIVFB ^ proteolytic activation (B. subtilis) Ribosomal protein S2 (B. subtilis) RNA polymerase subunit LP (B. subtilis)
[76] [33] [77] [35] [37] [36] [78] [38] [33,34] [44] [43] [79] [45^47] [40] [53] [55] [80] [81] [21,22] [82] [83] [19,20] [84] [71] [53] [58] [85] [68] [70,86] [65] [64] [66] [67] [87] [26] [25] [88] [61,62] [59,60] [24,28] [30] [25,27] [89] [51] [51]
Cell division and septation
Chromosome partitioning Chromosome and plasmid partitioning proteins DNA replication and structure
Plasmid partitioning
Protein export Signal transduction
Sporulation
Transcription and translation
GFP fusions discussed in further detail in the following sections. Further discussion of experimental methods for using GFP in macromolecule localization studies in bacteria has been previously published [23]. 3. Sporulation Among the ¢rst applications of GFP in bacteria was to characterize sporulation in Bacillus subtilis [24^26]. Sporulation in Bacillus involves formation of a polar septum that partitions the sporangium (the developing cell) into
the forespore and the mother cell. Assembly of these distinct subcellular compartments requires both di¡erential gene expression and protein localization for which the use of GFP is well suited. For example, GFP fusions between CotA, a protein involved in formation of the spore coat, and DacF, a protein know to be localized to the prespore compartment, were targeted as faithfully as the wild-type proteins [25,26]. SpoIVA is a protein important for development of the mature spore and whose localization was studied using time-lapse photomicroscopy, in combination with deconvolution microscopy. This approach provided a spatio/
FEMSLE 10108 17-10-01
12
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
temporal perspective of SpoIVA localization during sporulation and revealed that SpoIVA localized in a shell-like structure at the mother cell membrane that surrounds the forespore [27]. Localization was found to begin at the time of polar division and continued until completion of the forespore engulfment by the mother cell membrane. SpoIVA was also shown to be present in the mature spore. These studies provided an explanation for the role of SpoIVA in recruitment of coat proteins to the surface of the developing spore [27]. The sporulation-speci¢c transcription factors cE and cF are responsible for induction of gene expression in the mother cell and forespore, respectively. Although both of these sigma factors are synthesized in the sporangium, their activation in their respective cellular compartments requires proper localization of additional components of the sporulation process. One of these components includes SpoIIE, a membrane-bound phosphatase that functions in activation of cF within the forespore. SpoIIE^GFP fusions were used to develop models to explain how cF is selectively activated in the developing spore [17,24,28,29]. Time-lapse observation of these fusions con¢rmed that SpoIIIE was localized near both cell poles and is further subject to additional stages of regulation to ensure that the phosphatase activity is triggered only in the appropriate compartment. Continued use of GFP fusions, in combination with genetic and biochemical techniques, should further help to understand how SpoIIIE activity is sequestered in the developing spore. cE is activated in the mother cell by proteolytic cleavage of an amino-terminal extension that tethers the transcription factor to the cytoplasmic membrane, and eventually to the polar septum. Activation of cE is performed by a speci¢c membrane-bound protease, SpoIIGA. Consistent with its role in cE activation, SpoIIGA^GFP is localized to the asymmetric septum separating the mother cell from the forespore [30]. It has further been reported that fusions between sigE, encoding cE , and gfp direct synthesis of a product that is sequestered to the mother cell side of the asymmetric septum [31], thus providing an explanation of how induction of gene expression can occur speci¢cally in the mother cell. Additional insights into sporulation provided by GFP are discussed further below in Section 6, and are summarized in Table 1. 4. Cell division Perhaps some of the most surprising new insights into cellular processes, especially with respect to proteins that are dynamically localized throughout the cell cycle, have resulted from the ability to directly visualize components of the cell division machinery. Among the ¢rst applications of GFP to study cell division was its application to characterize localization of FtsZ, a highly conserved component of the cell division machinery. Previous studies
using IFM had shown that FtsZ forms a ring at the mid-cell division site [32]. FtsZ^GFP fusions, likewise, formed ring-like structures at the site of septal ring formation in living cells [33]. Moreover, subsequent studies were used to monitor the rapid assembly and disassembly of FtsZ polymers in individual, growing cells throughout the complete cell cycle [34]. In E. coli, at least nine di¡erent proteins are known to be important for construction of the cell division septum. Although the function of most of these proteins in cell wall synthesis remains unclear, the use of GFP fusions has provided means to determine their order of assembly, and to identify features of the proteins important for proper targeting. Ma et al. [33], for example, reported that FtsA colocalized with FtsZ to the site of cell division at mid-cell. Expression of deletion derivatives of an ftsA^ gfp gene fusion further revealed structural features of FtsA important for its proper localization. GFP, along with IFM, has contributed to determining the order of assembly of the cell division apparatus by expressing chimeric proteins in di¡erent mutant backgrounds, or in the presence of cell division inhibitors. For instance, proper targeting of fusions between GFP and the cytoplasmic domains of FtsI [35], FtsL [36], FtsK [37], FtsQ [38], and ZipA [39] all required prior localization of FtsZ, consistent with this protein's involvement early in the cell division process. In most of these cases the GFP fusions were fully functional in haploid, further supporting the experimental conclusions. The result that ZipA can assemble in the absence of the other cell division proteins, but does require FtsZ [39], indicated that ZipA is an early recruit to the cell division site, but apparently does not function as the initiator of FtsZ ring formation as had been previously proposed [40]. The signal to determine the potential site of cell division remains unknown. FtsL and FtsI are considered to be late additions to the septal ring, since both required that the remaining cell division components are expressed for their proper localization. Fig. 1 shows an example of how GFP fusions can be used to investigate cell division in E. coli. Although disruption of the ftsE gene results in elongated cell morphology, the role of this gene in proper cell division is not known [41]. To better characterize this mutant, localization of ZipA^GFP was monitored in an ftsE mutant. As shown, ZipA is localized to discrete foci in both wild-type and mutant cells, indicating that the block in cell division in the mutant does not occur at an early stage in the cell division process. The use of additional GFP fusions will be helpful to identify the defect in cell division imposed by the ftsE mutation. A similar approach has been taken to identify the block in cell division imposed by mutants that lacked phosphatidylethanolamine [42]. Characterizing E. coli strains expressing GFP fusions to the MinC, D, and E proteins led to the striking discovery that these cell division components are dynamically local-
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
Fig. 1. Localization of ZipA^GFP in an E. coli ftsE mutant. A: GFP £uorescence of E. coli MG1655 expressing a ZipA^GFP hybrid protein. B: An isogenic strain carrying the ftsE: :kan allele [41]. Some elongation of the control cells is observed as a result of increased expression of ZipA^GFP. The images were taken with a Nikon Eclipse inverted microscope ¢tted with a 100U Plan Apo 1.4 NA objective coupled to a Prairie Technologies Scanning confocal microscope. Laser excitation was at 488 nm wavelength and the GFP emission was collected with a photomultiplier in the confocal.
ized within the bacterial cell. MinCDE functions collectively to ensure that the septal ring is formed exclusively at the mid-cell position. MinCD speci¢cally blocks formation of the septal ring while MinE provides topological speci¢city to the system by preventing the MinCD inhibitor complex from blocking septation at mid-cell. Rather than performing their roles statically, however, MinD^ GFP and MinC^GFP oscillate rapidly between poles with a periodicity of only tens of seconds [43,44]. MinE was initially thought to form a stationary ring structure near mid-cell [45], but recent analysis has shown that the MinE ring migrates from this position toward one of the poles [46,47]. More speci¢cally, MinE migrates toward the pole where MinCD is residing and as it does so triggers the rapid relocation of MinCD to the opposite pole. Although the exact signi¢cance of the surprising behavior of these proteins for cell division remains unclear, its discovery resulted from the ability to monitor the motion of GFP-tagged proteins in real time. The use of multiple GFP variants to label distinctively the di¡erent Min proteins should provide a better understanding of their function in the cell division process, as well as reveal details about their interaction in vivo. 5. Replication, transcription and translation Due to the lack of adequate experimental tools, a fundamental question that has remained unanswered for decades is whether the replication machinery moves along a
13
stationary chromosome (perhaps the most common way of visualizing the process used in textbooks), or does the replication complex remain at a ¢xed position with the chromosomal DNA threaded through the polymerase `factory' style [48] ? Lemon and Grossman [22] gained a unique perspective of DNA synthesis by the visualizing GFP-tagged replication proteins in living cells. A gene fusion between polC, encoding a subunit of DNA polymerase, and gfp was introduced to the B. subtilis chromosome. The PolC^GFP protein functioned normally in DNA replication even when the hybrid represented the sole source of PolC in the cell. Remarkably, discrete £uorescent foci were visualized near the mid-point of exponentially growing cells. More £uorescent foci were observed as the cellular growth rate increased, consistent with the requirement for multiple replication forks in rapidly growing cells. Similar results were also seen when two additional DNA polymerase subunits were fused to GFP. From these observations, the authors concluded that the DNA polymerase complex of B. subtilis is found at ¢xed positions near mid-cell rather than being distributed throughout the nucleoid. Although other studies did not yield identical results [21], additional studies by Lemon et al. [49] incorporating a reversible block to DNA replication further supported the stationary polymerase complex model. Upon release of the replication block, regions of the newly replicated chromosomes, also visualized by GFP, were oriented away from the centrally located replication complex and toward opposite poles of the cell. Also, in a synchronous population of cells, the chromosomal region located between the origin and terminus migrated toward the replication machinery prior to duplication, again consistent with a model of DNA synthesis where the chromosomal DNA is the dynamic component of the process. These ¢ndings are signi¢cant in that they facilitate development of speci¢c theoretical models of replication and chromosome segregation that can be subject to further testing [50]. Using GFP to monitor localization of the transcriptional and translational machinery has provided new insights into these fundamental cellular processes as well. Inspection of RNA polymerase^GFP fusions expressed in B. subtilis revealed that transcription occurred primarily within the nucleoid, rather than on the periphery [51]. At high growth rates, RNA polymerase became further concentrated at distinct foci, likely representing areas of rRNA synthesis and reminiscent of nucleoli in eukaryotic systems. By contrast, translation, as monitored by ribosomal protein^GFP fusions, was found localized to subcellular regions distinct from transcription, including the cell poles and future division sites [51]. These observations suggest that the bacterial cell is organized in a manner more like that of eukaryotic cells than previously thought. Further work will be required to determine the biological signi¢cance of the spatial separation between transcription and translation in bacteria.
FEMSLE 10108 17-10-01
14
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
6. Chromosome and plasmid segregation The segregation of chromosomes and plasmids is a highly accurate process in bacteria, ensuring that sister molecules are accurately distributed to newly divided cells. Understanding how this process unfolds in living cells has been di¤cult to determine because of the small size of the bacterial cell, and the dynamic nature of the chromosome during replication and cell division. Two strategies to investigate chromosome behavior in bacteria include the use of GFP-fused partitioning proteins that bind speci¢cally to the origin of replication (ori), and by using a GFP^LacI reporter/repressor hybrid protein. This latter protein retains the ability to bind to lacO operator sites that are introduced in tandem to speci¢c locations on the bacterial chromosome, or on plasmids [52]. It has been used to visualize directly DNA molecules as they undergo segregation, revealing that chromosomal partitioning remains a highly ordered process throughout the cell cycle [53^56]. Fluorescent foci representing lacO sequences introduced near the ori of both B. subtilis [54] and E. coli [53] revealed that the ori is preferentially associated at or near the cell poles early in the cell cycle, while a region near the chromosome terminus remained near mid-cell [53]. Observing the kinetics of chromosome motion through a complete cell cycle revealed that, following duplication, one sister ori stayed near the cell pole while the other migrated toward the site of formation of the new pole in the daughter cell [53,57]. In contrast, the low copy number E. coli plasmids F and P1 were found at mid-cell in newly divided cells and then abruptly migrated to the one quarter and three quarter cell positions following replication, suggesting di¡erent mechanisms for partitioning of the chromosome and plasmids [53]. Models of segregation of high copy number plasmids suggest that they are free to di¡use throughout the cytoplasm. However, recent observations by Pogliano et al. indicated that GFP-tagged high copy number plasmids also appeared as distinct foci in the cells at mid-cell or near one of the quarter cell positions, rather than being distributed throughout the cytoplasm [58]. Moreover, foci migrated away from each other during cell growth, reminiscent of the behavior of lower copy number plasmids with active partitioning systems. The correlation between the location and dynamics of high copy plasmids with that observed for the replication machinery is striking, suggesting plasmids may associate with replication proteins to partition themselves e¤ciently. These results have prompted reevaluation of models of passive segregation of high copy number plasmids. In B. subtilis, sporulation is intimately linked with chromosome segregation to ensure that the replicated chromosomes are appropriately localized to opposite cell poles prior to asymmetric cell division. A system that coordinates chromosome segregation and expression of sporulation-speci¢c genes utilizes SpoOJ and Soj, a pair of pro-
teins with homologies to plasmid partitioning proteins. SpoOJ^GFP foci were localized to speci¢c sites near the ori and did so in both vegetatively growing cells as well as cells undergoing asymmetric cell division during sporulation [56,59,60]. The localization and movement of SpoOJ mimic that of the ori [55] and were mobilized to opposite poles during the cell cycle [56]. SpoOJ has been proposed to compose a mitotic-like apparatus for proper orientation and segregation of the Bacillus chromosome [60]. The role of Soj has also been investigated by determining the cellular location of GFP^Soj hybrid proteins. It was found that the protein localized near the cell poles, as well as being distributed freely throughout the cytoplasm of vegetative cells [61,62]. However, the hybrid protein, in a SpoOJ-dependent manner, was found to oscillate between the poles in a manner reminiscent of MinCD, described above. Soj is apparently a dual functioning protein, required both for aggregation of SpoOJ^chromosome complexes, as well as serving as a repressor of sporulation genes. Although it remains unclear how the oscillatory behavior of Soj contributes to its function in sporulation, it is yet another example of how events that comprise a complex cellular process can be coordinated by the dynamic motion of speci¢c proteins. The SeqA protein of E. coli binds to hemi-methylated DNA at GATC adenine methylation sites of newly replicated DNA to sequester the ori from continually initiating new rounds of replication that are out of synchrony with the cell cycle. Similar to that observed with DNA polymerase, SeqA foci were localized near the mid-point of growing cells and subsequently migrated to ¢xed positions one quarter of the distance to the cell poles [20]. Subsequent observations suggested that the location of SeqA coincided with that of the replication forks as predicted by the factory model of DNA replication [19]. Although additional experiments will be required to reconcile di¡erences in the dynamic localization of SeqA in independent studies [19,20,63], the results are consistent with the function of membrane-bound SeqA in tethering the replication forks to the cell center to ensure accurate segregation of the nucleoid by directing newly replicated chromosomal DNA away from the replication machinery [19]. 7. Signal transduction Bacteria respond rapidly to changes in environmental conditions both by immediately changing patterns of gene expression, as well as their direction of motility. Many bacteria have the ability to migrate either toward an attractant, or away from a toxic repellent by continually monitoring their surrounding environment and respond accordingly. A number of membrane-bound proteins function as receptors for speci¢c attractants and repellent molecules. These receptors, in association with additional cellular components, initiate a phospho-relay
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
15
system that results in altered cell motility. To better understand the function of this response system in E. coli, GFP fusions were constructed to determine the localization of components of the chemotaxis machinery. Sourjik and Berg [64] used a version of GFP emitting yellow light (YFP) to tag three chemotaxis proteins, CheA, CheY and CheZ, as well as the motor component FliM. The chemotaxis proteins were localized predominantly to the cell poles in association with the chemotaxis receptor proteins, consistent with previous IMF studies. By expressing the Che^YFP fusions in di¡erent mutant backgrounds, the YFP £uorescence was shown to be dependent upon other components of the chemotaxis complex, including the chemoreceptors. Di¡erential protein localization is also known to be important for controlling prokaryotic development in Caulobacter crescentus. CtrA is a master regulatory protein that controls a number of events in the cell cycle of C. crescentus, and whose activity is controlled by a membranebound histidine kinase, CckA. The spatial distribution of CckA^GFP fusions was found to alter in di¡erent cell cycle stages, becoming localized to a cell pole in early predivisional cells, and then dispersing upon cell division [65]. CckA apparently regulates progression through the developmental cycle of Caulobacter by di¡erential localization of histidine kinase activity. Additional two-component regulatory proteins that undergo di¡erential localization include the histidine kinases DivJ and PleC [66], and the response regulatory DivK [67]. All of these proteins show polar localization at speci¢c stages in the cell cycle, with DivJ and PleC being responsible for phosphorylation of DivK. Importantly, these results indicate that spatial control of the two-component regulatory proteins is superimposed on temporal regulation as a means to coordinate key events in the Caulobacter cell cycle.
of the green £uorescence phenotype in genetic approaches to identify mutants expressing altered levels of GFP [69]. In contrast to the results just described, when GFP was fused to proteins exported via the twin-arginine (TAT) pathway, £uorescence was observed in the periplasmic space [70]. This result is consistent with the prediction that the TAT pathway exports fully folded proteins, and that GFP can function in the periplasmic space if delivered in a properly folded conformation.
8. Protein export
The research in the author's laboratory is supported by the National Institutes of Health. The author thanks Craig Aspinwall, Bob Doyle, and Becky Graham for assistance with £uorescent microscopy, and Joe Lutkenhaus for the ZipA^GFP-expressing plasmid.
In addition to visualizing location of speci¢c classes of proteins within the con¢nes of the bacterial cytoplasm, GFP has also been used to study the export of proteins outside of the cytoplasm. Gene fusion technology has played important roles in dissecting bacterial protein export pathways. Fusions between an exported protein and L-galactosidase, for example, have provided new genetic selections and screens to isolate mutants defective in the export process. Using a similar approach, Feilmeier et al. [68] constructed fusions between GFP and proteins destined for localization to the periplasmic space. In contrast to the studies reported above, GFP was inactive in the periplasmic space due to its inability to fold properly in this subcellular compartment. The authors exploited this phenotype to isolate mutants that blocked export of the fusion [68]. Future work will likely see further exploitation
9. Conclusions Improvements in GFP technology are continually being made that enhance the sensitivity and utility of this reporter system, including new GFP variants that display an expanded spectrum of excitation and emission wavelengths. These variants have successfully been used to study localization of di¡erent cellular components within the same cell [49,71]. In addition, £uorescent proteins from other species have been identi¢ed to further expand the repertoire of reporter systems available for localization studies [72,73]. These variants, coupled with improvements in image analysis [74,75], should continue to enhance the sensitivity of GFP detection, further eliminating artifacts resulting from overexpressed or aggregated proteins. Clearly, the continued use of GFP as a localization reporter, in combination with other proven techniques, such as IFM, cell fractionation, and genetic approaches using other reporter genes, will result in new insights about the architecture and machinery of the bacterial cell, in addition to raising new questions about the function of some of its most fundamental components. Acknowledgements
References [1] Silhavy, T.J. (2000) Gene fusions. J. Bacteriol. 182, 5935^5938. [2] Chal¢e, M., Tu, Y., Euskirchen, G., Ward, W.W. and Prasher, D.C. (1994) Green £uorescent protein as a marker for gene expression. Science 263, 802^805. [3] Cody, C.W., Prasher, D.C., Westler, W.M., Pendergast, F.G. and Ward, W.W. (1993) Chemical structure of the hexapeptide chromophore of the Aequorea green-£uorescent protein. Biochemistry 32, 1212^1218. [4] Shapiro, L. and Losick, R. (2000) Dynamic spatial regulation in the bacterial cell. Cell 100, 89^98. [5] Losick, R. and Shapiro, L. (1999) Changing views on the nature of
FEMSLE 10108 17-10-01
16
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15] [16]
[17]
[18]
[19]
[20]
[21]
[22]
[23] [24]
[25]
[26]
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18 the bacterial cell: from biochemistry to cytology. J. Bacteriol. 181, 4135^4143. Shapiro, L. and Losick, R. (1997) Protein localization and cell fate in bacteria. Science 276, 712^718. Heim, R., Prasher, D.C. and Tsien, R.Y. (1994) Wavelength mutations and posttranslational autoxidation of green £uorescent protein. Proc. Natl. Acad. Sci. USA 91, 12501^12504. Crameri, A., Whitehorn, E.A., Tate, E. and Stemmer, W.P. (1996) Improved green £uorescent protein by molecular evolution using DNA shu¥ing. Nat. Biotechnol. 14, 315^319. Ehrig, T., O'Kane, D.J. and Prendergast, F.G. (1995) Green-£uorescent protein mutants with altered £uorescence excitation spectra. FEBS Lett. 367, 163^166. Cormack, B.P., Valdivia, R.H. and Falkow, S. (1996) FACS-optimized mutants of the green £uorescent protein (GFP). Gene 173, 33^38. Heim, R. and Tsien, R.Y. (1996) Engineering green £uorescent protein for improved brightness, longer wavelengths and £uorescence resonance energy transfer. Curr. Biol. 6, 178^182. Delagrave, S., Hawtin, R.E., Silva, C.M., Yang, M.M. and Youvan, D.C. (1995) Red-shifted excitation mutants of the green £uorescent protein. BioTechnology 13, 151^154. Feucht, A. and Lewis, P.J. (2001) Improved plasmid vectors for the production of multiple £uorescent protein fusions in Bacillus subtilis. Gene 264, 289^297. Miller, W.G. and Lindow, S.E. (1997) An improved GFP cloning cassette designed for prokaryotic transcriptional fusions. Gene 191, 149^153. Phillips, G.J. (1999) Gene cassettes for insertional mutagenesis and alteration of open reading frames. Anal. Biochem. 269, 207^210. Prescott, M., Nowakowski, S., Nagley, P. and Devenish, R.J. (1999) The length of polypeptide linker a¡ects the stability of green £uorescent protein fusion proteins. Anal. Biochem. 273, 305^307. King, N., Dreesen, O., Stragier, P., Pogliano, K. and Losick, R. (1999) Septation, dephosphorylation, and the activation of cF during sporulation in Bacillus subtilis. Genes Dev. 13, 1156^1167. Boyd, D., Weiss, D.S., Chen, J.C. and Beckwith, J. (2000) Towards single-copy gene expression systems making gene cloning physiologically relevant : Lambda InCh, a simple Escherichia coli plasmid^ chromosome shuttle system. J. Bacteriol. 182, 842^847. Brendler, T., Sawitzke, J., Sergueev, K. and Austin, S. (2000) A case for sliding SeqA tracts at anchored replication forks during Escherichia coli chromosome replication and segregation. EMBO J. 19, 6249^6258. Onogi, T., Niki, H., Yamozoe, M. and Hiraga, S. (1999) The assembly and migration of SeqA^GFP fusion in living cells of Escherichia coli. Mol. Microbiol. 31, 1775^1782. Imai, Y., Ogasawara, N., Ishigo-Oka, D., Kadoya, R., Daito, T. and Moriya, S. (2000) Subcellular localization of Dna-initiation proteins of Bacillus subtilis: evidence that chromosome replication begins at either edge of the nucleoids. Mol. Microbiol. 36, 1037^1048. Lemon, K.P. and Grossman, A.D. (1998) Localization of bacterial DNA polymerase : evidence for a factory model of replication. Science 282, 1516^1519. Margolin, W. (2000) Green £uorescent protein as a reporter for macromolecular localization in bacterial cells. Methods 20, 62^72. Arigoni, F., Pogliano, K., Webb, C.D., Stragier, P. and Losick, R. (1995) Localization of protein implicated in establishment of cell type to sites of asymmetric division. Science 270, 637. Lewis, P.J. and Errington, J. (1996) Use of green £uorescent protein for detection of cell-speci¢c gene expression and subcellular protein localization during sporulation in Bacillus subtilis. Microbiology 142, 733^740. Webb, C.D., Decatur, A., Teleman, A. and Losick, R. (1995) Use of green £uorescent protein for visualization of cell-speci¢c gene expres-
[27]
[28]
[29]
[30]
[31]
[32] [33]
[34] [35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
sion and subcellular protein localization during sporulation in Bacillus subtilis. J. Bacteriol. 177, 5906^5911. Price, K.D. and Losick, R. (1999) A four-dimensional view of assembly of a morphogenetic protein during sporulation in Bacillus subtilis. J. Bacteriol. 181, 781^790. Barak, I., Behari, J., Olmedo, G., Guzman, P., Brown, D.P., Castro, E., Walker, D., Westpheling, J. and Youngman, P. (1996) Structure and function of the Bacillus SpoIIE protein and its localization to sites of sporulation septum assembly. Mol. Microbiol. 19, 1047^1060. Wu, L.J., Feucht, A. and Errington, J. (1998) Prespore-speci¢c gene expression in Bacillus subtilis is driven by sequestration of SpoIIE phosphatase to the prespore side of the asymmetric septum. Genes Dev. 12, 1371^1380. Fawcett, P., Melnikov, A. and Youngman, P. (1998) The Bacillus SpoIIGA protein is targeted to sites of spore septum formation in a SpoIIE-independent manner. Mol. Microbiol. 28, 931^943. Ju, J. and Haldenwang, W.G. (1999) The `pro' sequence of the sporulation-speci¢c c transcription factor cE directs it to the mother cell side of the sporulation septum. J. Bacteriol. 181, 6171^6175. Bi, E. and Lutkenhaus, J. (1991) FtsZ ring structure associated with division in Escherichia coli. Nature 354, 161^164. Ma, X., Ehrdardt, D.W. and Margolin, W. (1996) Colocalization of cell division proteins FtsZ and FtsA to cytoskeletal structures in living Escherichia coli cells by using green £uorescent protein. Proc. Natl. Acad. Sci. USA 93, 12998^13003. Sun, Q. and Margolin, W. (1998) FtsZ dynamics during the division cycle of live Escherichia coli cells. J. Bacteriol. 180, 2050^2056. Weiss, D.S., Chen, J.C., Ghigo, J.M., Boyd, D. and Beckwith, J. (1999) Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z ring, FtsA, FtsQ, and FtsL. J. Bacteriol. 181, 508^520. Ghigo, J.M., Weiss, D.S., Chen, J.C., Yarrow, J.C. and Beckwith, J. (1999) Localization of FtsL to the Escherichia coli septal ring. Mol. Microbiol. 31, 725^737. Yu, X.C., Tran, A.H., Sun, Q. and Margolin, W. (1998) Localization of cell division protein FtsK to the Escherichia coli septum and identi¢cation of a potential N-terminal targeting domain. J. Bacteriol. 180, 1296^1304. Chen, J.C., Weiss, D.S., Ghigo, J.M. and Beckwith, J. (1999) Septal localization of FtsQ, an essential cell division protein in Escherichia coli. J. Bacteriol. 181, 521^530. Hale, C.A. and de Boer, P.A. (1999) Recruitment of ZipA to the septal ring of Escherichia coli is dependent on FtsZ and independent of FtsA. J. Bacteriol. 181, 167^176. Hale, C.A. and de Boer, P.A. (1997) Direct binding of FtsZ to ZipA, an essential component of the septal ring structure that mediates cell division in E. coli. Cell 88, 175^185. de Leeuw, E., Graham, B., Phillips, G.J., ten Hagen-Jongman, C.M., Oudega, B. and Luirink, J. (1999) Molecular characterization of Escherichia coli FtsE and FtsX. Mol. Microbiol. 31, 983^993. Mileykovskaya, E., Sun, Q., Margolin, W. and Dowhan, W. (1998) Localization and function of early cell division proteins in ¢lamentous Escherichia coli cells lacking phosphatidylethanolamine. J. Bacteriol. 180, 4252^4257. Raskin, D.M. and de Boer, P.A. (1999) Rapid pole-to-pole oscillation of a protein required for directing division to the middle of Escherichia coli. Proc. Natl. Acad. Sci. USA 96, 4971^4976. Hu, Z. and Lutkenhaus, J. (2000) Topological regulation of cell division in Escherichia coli involves rapid pole to pole oscillation of the division inhibitor MinC under the control of MinD and MinE. Mol. Microbiol. 34, 82^90. Raskin, D.M. and de Boer, P.A. (1997) The MinE ring: an FtsZindependent cell structure required for selection of the correct division site in E. coli. Cell 91, 685^694. Fu, X., Shih, Y.-L., Zhang, Y. and Roth¢eld, L.I. (2001) The MinE
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
[47]
[48] [49] [50]
[51]
[52]
[53]
[54]
[55]
[56]
[57] [58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
ring required for proper placement of the division site is a mobile structure that changes its cellular location during Escherichia coli division cycle. Proc. Natl. Acad. Sci. USA 98, 980^985. Hale, C.A., Meinhardt and de Boer, P.A. (2001) Dynamic localization cycle of the cell division regulator MinE in Escherichia coli. EMBO J. 20, 1563^1572. Dingman, C.W. (1974) Bidirectional chromosome replication : some topological considerations. J. Theor. Biol. 42, 187^195. Lemon, K.P. and Grossman, A.D. (2000) Movement of replicating DNA through a stationary replisome. Mol. Cell 6, 1321^1330. Sawitzke, J. and Austin, S. (2001) An analysis of the factory model for chromosome replication and segregation in bacteria. Mol. Microbiol. 40, 786^794. Lewis, P.J., Thaker, S.D. and Errington, J. (2000) Compartmentalization of transcription and translation in Bacillus subtilis. EMBO J. 19, 710^718. Straight, A.F., Belmont, A.S., Robinett, C.C. and Murray, A.W. (1996) GFP tagging of budding yeast chromosomes reveals that protein^protein interactions can mediate sister chromatid cohesion. Curr. Biol. 6, 1599^1608. Gordon, G.S., Sitnikov, D., Webb, C.D., Teleman, A., Straight, A., Losick, R., Murray, A.W. and Wright, A. (1997) Chromosome and low copy plasmid segregation in E. coli : visual evidence for distinct mechanisms. Cell 90, 1113^1121. Webb, C.D., Teleman, A., Gordon, S., Straight, A., Belmont, A., Lin, D.C., Grossman, A.D., Wright, A. and Losick, R. (1997) Bipolar localization of the replication origin regions of chromosomes in vegetative and sporulating cells of B. subtilis. Cell 88, 667^674. Webb, C.D., Graumann, P.L., Kahana, J.A., Teleman, A.A., Silver, P.A. and Losick, R. (1998) Use of time-lapse microscopy to visualize rapid movement of the replication origin region of the chromosome during the cell cycle in Bacillus subtilis. Mol. Microbiol. 28, 883^ 892. Teleman, A.A., Graumann, P.L., Lin, D.C.H., Grossman, A.D. and Losick, R. (1998) Chromosome arrangement within a bacterium. Curr. Biol. 8, 1102^1109. Gordon, G.S. and Wright, A. (2000) DNA segregation in bacteria. Annu. Rev. Microbiol. 54, 681^708. Pogliano, J., Ho, T.Q., Zhong, Z. and Helinski, D.R. (2001) Multicopy plasmids are clustered and localized in Escherichia coli. Proc. Natl. Acad. Sci. USA 98, 4486^4491. Glaser, P., Sharpe, M.E., Raether, B., Perego, M., Ohlsen, K. and Errington, J. (1997) Dynamic, mitotic-like behavior of a bacterial protein required for accurate chromosome partitioning. Genes Dev. 11, 1160^1168. Lin, D.C.-H., Levine, P.A. and Grossman, A.D. (1997) Bipolar localization of a chromosome partition protein in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 94, 4721^4726. Marston, A.L. and Errington, J. (1999) Dynamic movement of the ParA-like Soj protein of B. subtilis and its dual role in nucleoid organization and developmental regulation. Mol. Cell 4, 673^682. Quisel, J.D., Lin, D.C. and Grossman, A.D. (1999) Control of development by altered localization of a transcription factor in B. subtilis. Mol. Cell 4, 665^672. Hiraga, S., Ichinose, C., Niki, H. and Yajazoe, M. (1998) Cell cycledependent duplication and bi-directional migration of SeqA-associated DNA^protein complexes in Escherichia coli. Mol. Cell 1, 381^ 387. Sourjik, V. and Berg, H.C. (2000) Localization of components of the chemotaxis machinery of Escherichia coli using £uorescent protein fusions. Mol. Microbiol. 37, 740^751. Jacobs, C., Bomian, I.J., Maddock, J.R. and Shapiro, L. (1999) Cell cycle-dependent polar localization of an essential bacterial histidine kinase that controls DNA replication and cell division. Cell 97, 111^ 120. Wheeler, R.T. and Shapiro, L. (1999) Di¡erential localization of two
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
17
histidine kinases controlling bacterial cell di¡erentiation. Mol. Cell 4, 683^694. Jacobs, C., Hung, D. and Shapiro, L. (2001) Dynamic localization of a cytoplasmic signal transduction response regulator controls morphogenesis during the Caulobacter cell cycle. Proc. Natl. Acad. Sci. USA 98, 4095^4100. Feilmeier, B.J., Iseminger, G., Schroeder, D., Webber, H. and Phillips, G.J. (2000) Green £uorescent protein functions as a reporter for protein localization in Escherichia coli. J. Bacteriol. 182, 4068^ 4076. Valdivia, R.H. and Falkow, S. (1996) Bacterial genetics by £ow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by di¡erential £uorescence induction. Mol. Microbiol. 22, 367^378. Santini, C.L., Bernadac, A., Zhang, M., Chanal, A., Ize, B., Blanco, C. and Wu, L.F. (2000) Translocation of jelly¢sh green £uorescent protein via the TAT system of Escherichia coli and change of its periplasmic localization in response to osmotic up-shock. J. Biol. Chem. 276, 8159^8164. Lemon, K.P., Kurtser, I. and Grossman, A.D. (2001) E¡ects of replication termination mutants on chromosome partitioning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 98, 212^217. Baird, G.S., Zacharias, D.A. and Tsien, R.Y. (2000) Biochemistry, mutagenesis, and oligomerization of DsRed, a red £uorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11984^11989. Gross, L.A., Baird, G.S., Ho¡man, R.C., Baldridge, K.K. and Tsien, R.Y. (2000) The structure of the chromophore within DsRed, a red £uorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11990^11995. Dittrich, P., Malvezzi-Campeggi, F., Jahnz, M. and Schwille, P. (2001) Accessing molecular dynamics in cells by £uorescence correlation spectroscopy. Biol. Chem. 382, 491^494. Brock, R., Vamosi, G., Vereb, G. and Jovin, T.M. (1999) Rapid characterization of green £uorescent protein fusion proteins on the molecular and cellular level by £uorescence correlation microscopy. Proc. Natl. Acad. Sci. USA 96, 10123^10128. Edwards, D.H., Thomaides, H.B. and Errington, J. (2000) Promiscuous targeting of Bacillus subtilis cell division protein DivIVA to division sites in Escherichia coli and ¢ssion yeast. EMBO J. 19, 2719^ 2727. Wehrl, W., Niederweis, M. and Schumann, W. (2000) The FtsH protein accumulates at the septum of Bacillus subtilis during cell division and sporulation. J. Bacteriol. 182, 3870^3873. Sievers, J. and Errington, J. (2000) The Bacillus subtilis cell division protein FtsL localizes to sites of septation and interacts with DivIC. Mol. Microbiol. 36, 846^855. Marston, A.L., Thomaides, H.B., Edwards, D.H., Sharpe, M.E. and Errington, J. (1998) Polar localization of the MinD protein of Bacillus subtilis and its role in selection of the mid-cell division site. Genes Dev. 12, 3419^3430. Huang, W.M., Libbey, J.L., van der Hoeven, P. and Yu, S.X. (1998) Bipolar localization of Bacillus subtilis topoisomerase IV, an enzyme required for chromosome segregation. Proc. Natl. Acad. Sci. USA 95, 4652^4657. Kim, S.K. and Wang, J.C. (1998) Localization of F plasmid SopB protein to positions near the poles of Escherichia coli cells. Proc. Natl. Acad. Sci. USA 95, 1523^1527. Kohler, P. and Marahiel, M.A. (1997) Association of the histone-like protein HBsu with the nucleoid of Bacillus subtilis. J. Bacteriol. 179, 2060^2064. Wery, M., Woldringh, C.L. and Rouviere-Yaniv, J. (2001) HU^GFP and DAPI co-localize on the Escherichia coli nucleoid. Biochimie 83, 193^200. Britton, R.A., Lin, D.C. and Grossman, A.D. (1998) Characterization of a prokaryotic SMC protein involved in chromosome partitioning. Genes Dev. 12, 1254^1259.
FEMSLE 10108 17-10-01
18
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
[85] Jensen, R.B. and Gerdes, K. (1999) Mechanism of DNA segregation in prokaryotes: ParM partitioning protein of plasmid R1 co-localizes with its replicon during the cell cycle. EMBO J. 18, 4076^4084. [86] Thomas, J.D., Daniel, R.A., Errington, J. and Robinson, C. (2001) Export of active green £uorescent protein to the periplasm by the twin-arginine translocase (Tat) pathway in Escherichia coli. Mol. Microbiol. 39, 47^53. [87] Boyd, J.M. (2000) Localization of the histidine kinase PilA to the
poles of Pseudomonas aeruginosa and identi¢cation of a localization domain. Mol. Microbiol. 36, 153^162. [88] Ju, J., Tingqiu, L. and Haldenwang, W.G. (1998) Forespore expression and processing of the SigE transcription factor in wild-type and mutant Bacillus subtilis. J. Bacteriol. 180, 1681^1693. [89] Resnekov, O., Alper, S. and Losick, R. (1996) Subcellular localization of proteins governing the proteolytic activation of a developmental transcription factor in Bacillus subtilis. Genes Cells 1, 529^542.
FEMSLE 10108 17-10-01
www.fems-microbiology.org
MiniReview
Green £uorescent protein ^ a bright idea for the study of bacterial protein localization Gregory J. Phillips * Department of Microbiology, 207 Science I Building, Iowa State University, Ames, IA 50011, USA Received 20 February 2001; received in revised form 26 July 2001 ; accepted 26 July 2001 First published online 6 September 2001
Abstract Use of the green fluorescent protein (GFP) of Aequorea victoria as a reporter for protein and DNA localization has provided sensitive, new approaches for studying the organization of the bacterial cell, leading to new insights into diverse cellular processes. GFP has many characteristics that make it useful for localization studies in bacteria, primarily its ability to fluoresce when fused to target polypeptides without the addition of exogenously added substrates. As an alternative to immunofluorescence microscopy, the expression of gfp gene fusions has been used to probe the function of cellular components fundamental for DNA replication, translation, protein export, and signal transduction, that heretofore have been difficult to study in living cells. Moreover, protein and DNA localization can now be monitored in real time, revealing that several proteins important for cell division, development and sporulation are dynamically localized throughout the cell cycle. The use of additional GFP variants that permit the labeling of multiple components within the same cell, and the use of GFP for genetic screens, should continue to make this a valuable tool for addressing complex questions about the bacterial cell. ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Green £uorescent protein; Protein localization; Gene fusion; Hybrid protein ; Fluorescence microscopy
1. Introduction The use of lacZ gene fusions, encoding hybrid proteins composed of a target polypeptide fused to Escherichia coli L-galactosidase, has revolutionized our ability to study bacterial gene expression and protein export [1]. A similar impact is being made by application of green £uorescent protein (GFP) from Aequorea victoria to study macromolecular localization in living cells, including bacteria [2]. Many important questions about the subcellular organization of bacterial cells have been di¤cult to address due, in part, to their small size. GFP has, however, provided new experimental approaches to investigate the localization of both protein and DNA molecules in prokaryotes. The traditional approach to study macromolecular localization in bacteria includes techniques of immunogold labeling and immuno£uorescence microscopy (IFM). Although these methods have been e¡ectively used to vis-
* Tel. : +1 (515) 294-1525; Fax: +1 (515) 294-6019. E-mail address : [email protected] (G.J. Phillips).
ualize the targeting of speci¢c cellular components, their use can be limited by the requirement for antibodies directed against the target protein. Antibodies against proteins found in low abundance can be di¤cult to obtain, and the level of detection of this class of protein is relatively insensitive. They also require cell ¢xation that precludes investigation of living cells. Moreover, these techniques can be technically demanding and costly. Gene fusions using gfp, on the other hand, can be constructed with relative ease (as discussed further below), and their use, in general, heightens the sensitivity of molecular detection. Since the GFP chromophore is formed by autocatalytic cyclization of three amino acids [3], and £uorescence is triggered by excitation with speci¢c wavelength light, neither invasive sample preparation nor substrate addition is required, making it possible to study protein localization in living cells. In addition, since GFP is relatively small (238 amino acids) and can remain functional even when fused to a target protein, it allows protein localization to be monitored in real time. As a result, studies using GFP have bolstered our understanding of a variety of cellular processes such as cell division, chromosome replication and partitioning,
0378-1097 / 01 / $20.00 ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 0 1 ) 0 0 3 5 8 - 5
FEMSLE 10108 17-10-01
10
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
sporulation, development and signal transduction, and have changed our view of the organization of the bacterial cell in general. The bacterial cell, once considered by analogy to be a `bag of enzymes', is now known to accommodate the segregation and sequestration of a variety of molecules into speci¢c subcellular regions. Surprisingly, many of the cellular components are dynamic in nature, assembling and disassembling into higher order structures, and migrating in a directed manner between distinct locations in the bacterial cell, revealing more similarities between prokaryotes and eukaryotes that had previously been imagined [4^6]. 2. Creation and use of GFP fusions Early studies in which wild-type GFP was expressed in E. coli revealed a number of problems that hindered its use as a reporter for protein localization. For example, formation of the GFP chromophore is slow, requiring up to 2 h post-synthesis to form [7]. Wild-type GFP also tends to form inactive inclusion bodies, especially at elevated temperatures. In addition, the £uorescent intensity of GFP is relatively low, and the protein has a major absorbance peak at 395 nm, a value that requires specialized ¢lter sets for optimal excitation. Fortunately, a number of variants of GFP have been isolated that overcome many of these limitations [7^11]. Cormack et al. [10] isolated a particularly valuable set of GFP variants by screening for mutants that showed increased brightness upon excitation at 488 nm. In addition to improved solubility in E. coli, an amino acid change within the chromophore resulted in a `red-shift' in excitation from 395 to 488 nm [11,12]. These variants can be detected with good sensitivity by £uorescein isothiocyanate ¢lter sets commonly found on £uorescence microscopes. Derivatives of these mutants are used almost exclusively in protein localization studies. A number of vectors have been constructed to facilitate creation of gene fusions with di¡erent gfp variants. Many of these vectors have either been published [8,13,14], or are commercially available (e.g. Clontech, Palo Alto, CA, USA). A general strategy for constructing GFP fusions is to use PCR, or other cloning strategies [15] to engineer restriction enzyme sites into both gfp and the target gene to create in-frame fusions on recombinant plasmids. The gfp reporter may be fused to either the 5P or 3P end of the target gene, to generate either amino- or carboxy-terminal fusions, respectively. This choice is often made by predicting which region of the protein will most likely tolerate the addition of extra amino acids and still remain functional. Also, when fusing GFP to membrane proteins, GFP must be joined to a domain that is exposed to the cytoplasmic side of the inner membrane. Where necessary, linker sequences may also be added between the target protein and GFP to improve £uorescence output [16].
GFP is not without its limitations, however. For example, formation of the chromophore requires molecular oxygen, and all forms of GFP display reduced £uorescence at elevated growth temperatures [7]. Consequently, GFP cannot be used under all growth conditions under which microorganisms normally grow. Certain types of proteins also are not suitable for detection by GFP. For example, because of the relatively slow formation of the chromophore, even in the GFP variants, proteins with short halflives may not be detectable. Also, proteins must still be present in su¤cient levels since detection of only a few molecules of GFP is technically di¤cult, with results easily misinterpreted. As with any fusion protein, caution must also be used when interpreting the results of protein localization studies since the addition of the GFP moiety can alter the function, or location of the target polypeptide. A genetic test for altered function is to determine if expression of a GFP fusion can complement a knockout mutation in the gene encoding the target protein. Fortunately, there are several examples, as discussed below, of proteins that remain fully functional, as determined by complementation studies, when fused to GFP. There are also instances where GFP does disrupt normal protein function without interfering with its localization. It is di¤cult to assess from the literature, however, the number of instances where GFP fusion proteins are not active, or have altered localization properties, and each case must be assessed empirically. Foci of GFP £uorescence must also be carefully scrutinized to ensure that they are the result of proteins targeted to a speci¢c subcellular location, and not merely caused by protein aggregation. Also, £uorescent intensity may be arti¢cially skewed by elevated concentrations of GFP within small volumes of subcellular compartments [17]. Another complication that can arise is that synthesis of elevated levels of hybrid proteins can yield incorrect or ambiguous interpretations about protein localization. For example, overexpression of cell division proteins can themselves disrupt normal cell division. In addition, highlevel protein expression can result in protein aggregation or mislocalization. One solution has been provided by Boyd et al. [18] who developed a convenient method to introduce plasmid-borne gfp constructs to the E. coli chromosome, thus allowing expression of the fusions at more physiological relevant levels. Inconsistencies in results between laboratories using GFP fusions have also been reported (compare [19,20] and [21,22]). It is likely that di¡erences in bacterial strains, growth conditions, growth rates, or fusion constructs may explain these discrepancies. In these cases the use of alternative techniques, such as IFM, may be helpful to test conclusions reached by the use of GFP. These limitations notwithstanding, GFP has been successfully used to study protein and DNA localization in a variety of di¡erent microbial systems. A summary of these applications is shown in Table 1 with many of the speci¢c
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
11
Table 1 GFP fusions used to characterize bacterial cellular processes Cellular process
Target protein and function (source)
Reference
DivIVA ^ cell division protein (B. subtilis) FtsA ^ cell division protein (E. coli) FtsH ^ cell division protein (B. subtilis) FtsI ^ cell division protein (E. coli) FtsK ^ cell division protein (E. coli) FtsL ^ cell division protein (E. coli) FtsL ^ cell division protein (B. subtilis) FtsQ ^ cell division protein (E. coli) FtsZ ^ cell division protein (E. coli) MinC ^ cell division protein (E. coli) MinD ^ cell division protein (E. coli) MinD ^ cell division protein (B. subtilis) MinE ^ cell division protein (E. coli) ZipA ^ cell division protein (E. coli) B. subtilis E. coli ParC (topoisomerase IV) (B. subtilis) SopB ^ F factor partitioning (E. coli) DNA polymerase III subunits, PolC, d, NP (B. subtilis) Hbsu ^ histone-like protein (B. subtilis) HU ^ DNA binding protein SeqA ^ replication initiation (E. coli) SMC ^ nucleoid-associated protein (B. subtilis) Replication terminator protein (RTP) (B. subtilis) F, P1 (E. coli) Multicopy plasmids (E. coli) R1 (E. coli) Sec-dependent export (E. coli) TAT-dependent export (E. coli) CckA ^ kinase (C. crescentus) Che proteins (E. coli) DivJ and PleC (C. crescentus) DivK (C. crescentus) PilS ^ signal transduction (Pseudomonas aeruginosa) CotE ^ spore coat (B. subtilis) DacF ^ prespore-speci¢c protein (B. subtilis) SigE ^ sigma factor (B. subtilis) Soj ^ chromosome segregation and transcription (B. subtilis) SpoOJ ^ chromosome segregation (B. subtilis) SpoIIE ^ phosphatase (Bacillus) SpoIIGA ^ protease (B. subtilis) SpoIVA ^ spore development (B. subtilis) SpoIVFB ^ proteolytic activation (B. subtilis) Ribosomal protein S2 (B. subtilis) RNA polymerase subunit LP (B. subtilis)
[76] [33] [77] [35] [37] [36] [78] [38] [33,34] [44] [43] [79] [45^47] [40] [53] [55] [80] [81] [21,22] [82] [83] [19,20] [84] [71] [53] [58] [85] [68] [70,86] [65] [64] [66] [67] [87] [26] [25] [88] [61,62] [59,60] [24,28] [30] [25,27] [89] [51] [51]
Cell division and septation
Chromosome partitioning Chromosome and plasmid partitioning proteins DNA replication and structure
Plasmid partitioning
Protein export Signal transduction
Sporulation
Transcription and translation
GFP fusions discussed in further detail in the following sections. Further discussion of experimental methods for using GFP in macromolecule localization studies in bacteria has been previously published [23]. 3. Sporulation Among the ¢rst applications of GFP in bacteria was to characterize sporulation in Bacillus subtilis [24^26]. Sporulation in Bacillus involves formation of a polar septum that partitions the sporangium (the developing cell) into
the forespore and the mother cell. Assembly of these distinct subcellular compartments requires both di¡erential gene expression and protein localization for which the use of GFP is well suited. For example, GFP fusions between CotA, a protein involved in formation of the spore coat, and DacF, a protein know to be localized to the prespore compartment, were targeted as faithfully as the wild-type proteins [25,26]. SpoIVA is a protein important for development of the mature spore and whose localization was studied using time-lapse photomicroscopy, in combination with deconvolution microscopy. This approach provided a spatio/
FEMSLE 10108 17-10-01
12
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
temporal perspective of SpoIVA localization during sporulation and revealed that SpoIVA localized in a shell-like structure at the mother cell membrane that surrounds the forespore [27]. Localization was found to begin at the time of polar division and continued until completion of the forespore engulfment by the mother cell membrane. SpoIVA was also shown to be present in the mature spore. These studies provided an explanation for the role of SpoIVA in recruitment of coat proteins to the surface of the developing spore [27]. The sporulation-speci¢c transcription factors cE and cF are responsible for induction of gene expression in the mother cell and forespore, respectively. Although both of these sigma factors are synthesized in the sporangium, their activation in their respective cellular compartments requires proper localization of additional components of the sporulation process. One of these components includes SpoIIE, a membrane-bound phosphatase that functions in activation of cF within the forespore. SpoIIE^GFP fusions were used to develop models to explain how cF is selectively activated in the developing spore [17,24,28,29]. Time-lapse observation of these fusions con¢rmed that SpoIIIE was localized near both cell poles and is further subject to additional stages of regulation to ensure that the phosphatase activity is triggered only in the appropriate compartment. Continued use of GFP fusions, in combination with genetic and biochemical techniques, should further help to understand how SpoIIIE activity is sequestered in the developing spore. cE is activated in the mother cell by proteolytic cleavage of an amino-terminal extension that tethers the transcription factor to the cytoplasmic membrane, and eventually to the polar septum. Activation of cE is performed by a speci¢c membrane-bound protease, SpoIIGA. Consistent with its role in cE activation, SpoIIGA^GFP is localized to the asymmetric septum separating the mother cell from the forespore [30]. It has further been reported that fusions between sigE, encoding cE , and gfp direct synthesis of a product that is sequestered to the mother cell side of the asymmetric septum [31], thus providing an explanation of how induction of gene expression can occur speci¢cally in the mother cell. Additional insights into sporulation provided by GFP are discussed further below in Section 6, and are summarized in Table 1. 4. Cell division Perhaps some of the most surprising new insights into cellular processes, especially with respect to proteins that are dynamically localized throughout the cell cycle, have resulted from the ability to directly visualize components of the cell division machinery. Among the ¢rst applications of GFP to study cell division was its application to characterize localization of FtsZ, a highly conserved component of the cell division machinery. Previous studies
using IFM had shown that FtsZ forms a ring at the mid-cell division site [32]. FtsZ^GFP fusions, likewise, formed ring-like structures at the site of septal ring formation in living cells [33]. Moreover, subsequent studies were used to monitor the rapid assembly and disassembly of FtsZ polymers in individual, growing cells throughout the complete cell cycle [34]. In E. coli, at least nine di¡erent proteins are known to be important for construction of the cell division septum. Although the function of most of these proteins in cell wall synthesis remains unclear, the use of GFP fusions has provided means to determine their order of assembly, and to identify features of the proteins important for proper targeting. Ma et al. [33], for example, reported that FtsA colocalized with FtsZ to the site of cell division at mid-cell. Expression of deletion derivatives of an ftsA^ gfp gene fusion further revealed structural features of FtsA important for its proper localization. GFP, along with IFM, has contributed to determining the order of assembly of the cell division apparatus by expressing chimeric proteins in di¡erent mutant backgrounds, or in the presence of cell division inhibitors. For instance, proper targeting of fusions between GFP and the cytoplasmic domains of FtsI [35], FtsL [36], FtsK [37], FtsQ [38], and ZipA [39] all required prior localization of FtsZ, consistent with this protein's involvement early in the cell division process. In most of these cases the GFP fusions were fully functional in haploid, further supporting the experimental conclusions. The result that ZipA can assemble in the absence of the other cell division proteins, but does require FtsZ [39], indicated that ZipA is an early recruit to the cell division site, but apparently does not function as the initiator of FtsZ ring formation as had been previously proposed [40]. The signal to determine the potential site of cell division remains unknown. FtsL and FtsI are considered to be late additions to the septal ring, since both required that the remaining cell division components are expressed for their proper localization. Fig. 1 shows an example of how GFP fusions can be used to investigate cell division in E. coli. Although disruption of the ftsE gene results in elongated cell morphology, the role of this gene in proper cell division is not known [41]. To better characterize this mutant, localization of ZipA^GFP was monitored in an ftsE mutant. As shown, ZipA is localized to discrete foci in both wild-type and mutant cells, indicating that the block in cell division in the mutant does not occur at an early stage in the cell division process. The use of additional GFP fusions will be helpful to identify the defect in cell division imposed by the ftsE mutation. A similar approach has been taken to identify the block in cell division imposed by mutants that lacked phosphatidylethanolamine [42]. Characterizing E. coli strains expressing GFP fusions to the MinC, D, and E proteins led to the striking discovery that these cell division components are dynamically local-
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
Fig. 1. Localization of ZipA^GFP in an E. coli ftsE mutant. A: GFP £uorescence of E. coli MG1655 expressing a ZipA^GFP hybrid protein. B: An isogenic strain carrying the ftsE: :kan allele [41]. Some elongation of the control cells is observed as a result of increased expression of ZipA^GFP. The images were taken with a Nikon Eclipse inverted microscope ¢tted with a 100U Plan Apo 1.4 NA objective coupled to a Prairie Technologies Scanning confocal microscope. Laser excitation was at 488 nm wavelength and the GFP emission was collected with a photomultiplier in the confocal.
ized within the bacterial cell. MinCDE functions collectively to ensure that the septal ring is formed exclusively at the mid-cell position. MinCD speci¢cally blocks formation of the septal ring while MinE provides topological speci¢city to the system by preventing the MinCD inhibitor complex from blocking septation at mid-cell. Rather than performing their roles statically, however, MinD^ GFP and MinC^GFP oscillate rapidly between poles with a periodicity of only tens of seconds [43,44]. MinE was initially thought to form a stationary ring structure near mid-cell [45], but recent analysis has shown that the MinE ring migrates from this position toward one of the poles [46,47]. More speci¢cally, MinE migrates toward the pole where MinCD is residing and as it does so triggers the rapid relocation of MinCD to the opposite pole. Although the exact signi¢cance of the surprising behavior of these proteins for cell division remains unclear, its discovery resulted from the ability to monitor the motion of GFP-tagged proteins in real time. The use of multiple GFP variants to label distinctively the di¡erent Min proteins should provide a better understanding of their function in the cell division process, as well as reveal details about their interaction in vivo. 5. Replication, transcription and translation Due to the lack of adequate experimental tools, a fundamental question that has remained unanswered for decades is whether the replication machinery moves along a
13
stationary chromosome (perhaps the most common way of visualizing the process used in textbooks), or does the replication complex remain at a ¢xed position with the chromosomal DNA threaded through the polymerase `factory' style [48] ? Lemon and Grossman [22] gained a unique perspective of DNA synthesis by the visualizing GFP-tagged replication proteins in living cells. A gene fusion between polC, encoding a subunit of DNA polymerase, and gfp was introduced to the B. subtilis chromosome. The PolC^GFP protein functioned normally in DNA replication even when the hybrid represented the sole source of PolC in the cell. Remarkably, discrete £uorescent foci were visualized near the mid-point of exponentially growing cells. More £uorescent foci were observed as the cellular growth rate increased, consistent with the requirement for multiple replication forks in rapidly growing cells. Similar results were also seen when two additional DNA polymerase subunits were fused to GFP. From these observations, the authors concluded that the DNA polymerase complex of B. subtilis is found at ¢xed positions near mid-cell rather than being distributed throughout the nucleoid. Although other studies did not yield identical results [21], additional studies by Lemon et al. [49] incorporating a reversible block to DNA replication further supported the stationary polymerase complex model. Upon release of the replication block, regions of the newly replicated chromosomes, also visualized by GFP, were oriented away from the centrally located replication complex and toward opposite poles of the cell. Also, in a synchronous population of cells, the chromosomal region located between the origin and terminus migrated toward the replication machinery prior to duplication, again consistent with a model of DNA synthesis where the chromosomal DNA is the dynamic component of the process. These ¢ndings are signi¢cant in that they facilitate development of speci¢c theoretical models of replication and chromosome segregation that can be subject to further testing [50]. Using GFP to monitor localization of the transcriptional and translational machinery has provided new insights into these fundamental cellular processes as well. Inspection of RNA polymerase^GFP fusions expressed in B. subtilis revealed that transcription occurred primarily within the nucleoid, rather than on the periphery [51]. At high growth rates, RNA polymerase became further concentrated at distinct foci, likely representing areas of rRNA synthesis and reminiscent of nucleoli in eukaryotic systems. By contrast, translation, as monitored by ribosomal protein^GFP fusions, was found localized to subcellular regions distinct from transcription, including the cell poles and future division sites [51]. These observations suggest that the bacterial cell is organized in a manner more like that of eukaryotic cells than previously thought. Further work will be required to determine the biological signi¢cance of the spatial separation between transcription and translation in bacteria.
FEMSLE 10108 17-10-01
14
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
6. Chromosome and plasmid segregation The segregation of chromosomes and plasmids is a highly accurate process in bacteria, ensuring that sister molecules are accurately distributed to newly divided cells. Understanding how this process unfolds in living cells has been di¤cult to determine because of the small size of the bacterial cell, and the dynamic nature of the chromosome during replication and cell division. Two strategies to investigate chromosome behavior in bacteria include the use of GFP-fused partitioning proteins that bind speci¢cally to the origin of replication (ori), and by using a GFP^LacI reporter/repressor hybrid protein. This latter protein retains the ability to bind to lacO operator sites that are introduced in tandem to speci¢c locations on the bacterial chromosome, or on plasmids [52]. It has been used to visualize directly DNA molecules as they undergo segregation, revealing that chromosomal partitioning remains a highly ordered process throughout the cell cycle [53^56]. Fluorescent foci representing lacO sequences introduced near the ori of both B. subtilis [54] and E. coli [53] revealed that the ori is preferentially associated at or near the cell poles early in the cell cycle, while a region near the chromosome terminus remained near mid-cell [53]. Observing the kinetics of chromosome motion through a complete cell cycle revealed that, following duplication, one sister ori stayed near the cell pole while the other migrated toward the site of formation of the new pole in the daughter cell [53,57]. In contrast, the low copy number E. coli plasmids F and P1 were found at mid-cell in newly divided cells and then abruptly migrated to the one quarter and three quarter cell positions following replication, suggesting di¡erent mechanisms for partitioning of the chromosome and plasmids [53]. Models of segregation of high copy number plasmids suggest that they are free to di¡use throughout the cytoplasm. However, recent observations by Pogliano et al. indicated that GFP-tagged high copy number plasmids also appeared as distinct foci in the cells at mid-cell or near one of the quarter cell positions, rather than being distributed throughout the cytoplasm [58]. Moreover, foci migrated away from each other during cell growth, reminiscent of the behavior of lower copy number plasmids with active partitioning systems. The correlation between the location and dynamics of high copy plasmids with that observed for the replication machinery is striking, suggesting plasmids may associate with replication proteins to partition themselves e¤ciently. These results have prompted reevaluation of models of passive segregation of high copy number plasmids. In B. subtilis, sporulation is intimately linked with chromosome segregation to ensure that the replicated chromosomes are appropriately localized to opposite cell poles prior to asymmetric cell division. A system that coordinates chromosome segregation and expression of sporulation-speci¢c genes utilizes SpoOJ and Soj, a pair of pro-
teins with homologies to plasmid partitioning proteins. SpoOJ^GFP foci were localized to speci¢c sites near the ori and did so in both vegetatively growing cells as well as cells undergoing asymmetric cell division during sporulation [56,59,60]. The localization and movement of SpoOJ mimic that of the ori [55] and were mobilized to opposite poles during the cell cycle [56]. SpoOJ has been proposed to compose a mitotic-like apparatus for proper orientation and segregation of the Bacillus chromosome [60]. The role of Soj has also been investigated by determining the cellular location of GFP^Soj hybrid proteins. It was found that the protein localized near the cell poles, as well as being distributed freely throughout the cytoplasm of vegetative cells [61,62]. However, the hybrid protein, in a SpoOJ-dependent manner, was found to oscillate between the poles in a manner reminiscent of MinCD, described above. Soj is apparently a dual functioning protein, required both for aggregation of SpoOJ^chromosome complexes, as well as serving as a repressor of sporulation genes. Although it remains unclear how the oscillatory behavior of Soj contributes to its function in sporulation, it is yet another example of how events that comprise a complex cellular process can be coordinated by the dynamic motion of speci¢c proteins. The SeqA protein of E. coli binds to hemi-methylated DNA at GATC adenine methylation sites of newly replicated DNA to sequester the ori from continually initiating new rounds of replication that are out of synchrony with the cell cycle. Similar to that observed with DNA polymerase, SeqA foci were localized near the mid-point of growing cells and subsequently migrated to ¢xed positions one quarter of the distance to the cell poles [20]. Subsequent observations suggested that the location of SeqA coincided with that of the replication forks as predicted by the factory model of DNA replication [19]. Although additional experiments will be required to reconcile di¡erences in the dynamic localization of SeqA in independent studies [19,20,63], the results are consistent with the function of membrane-bound SeqA in tethering the replication forks to the cell center to ensure accurate segregation of the nucleoid by directing newly replicated chromosomal DNA away from the replication machinery [19]. 7. Signal transduction Bacteria respond rapidly to changes in environmental conditions both by immediately changing patterns of gene expression, as well as their direction of motility. Many bacteria have the ability to migrate either toward an attractant, or away from a toxic repellent by continually monitoring their surrounding environment and respond accordingly. A number of membrane-bound proteins function as receptors for speci¢c attractants and repellent molecules. These receptors, in association with additional cellular components, initiate a phospho-relay
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
15
system that results in altered cell motility. To better understand the function of this response system in E. coli, GFP fusions were constructed to determine the localization of components of the chemotaxis machinery. Sourjik and Berg [64] used a version of GFP emitting yellow light (YFP) to tag three chemotaxis proteins, CheA, CheY and CheZ, as well as the motor component FliM. The chemotaxis proteins were localized predominantly to the cell poles in association with the chemotaxis receptor proteins, consistent with previous IMF studies. By expressing the Che^YFP fusions in di¡erent mutant backgrounds, the YFP £uorescence was shown to be dependent upon other components of the chemotaxis complex, including the chemoreceptors. Di¡erential protein localization is also known to be important for controlling prokaryotic development in Caulobacter crescentus. CtrA is a master regulatory protein that controls a number of events in the cell cycle of C. crescentus, and whose activity is controlled by a membranebound histidine kinase, CckA. The spatial distribution of CckA^GFP fusions was found to alter in di¡erent cell cycle stages, becoming localized to a cell pole in early predivisional cells, and then dispersing upon cell division [65]. CckA apparently regulates progression through the developmental cycle of Caulobacter by di¡erential localization of histidine kinase activity. Additional two-component regulatory proteins that undergo di¡erential localization include the histidine kinases DivJ and PleC [66], and the response regulatory DivK [67]. All of these proteins show polar localization at speci¢c stages in the cell cycle, with DivJ and PleC being responsible for phosphorylation of DivK. Importantly, these results indicate that spatial control of the two-component regulatory proteins is superimposed on temporal regulation as a means to coordinate key events in the Caulobacter cell cycle.
of the green £uorescence phenotype in genetic approaches to identify mutants expressing altered levels of GFP [69]. In contrast to the results just described, when GFP was fused to proteins exported via the twin-arginine (TAT) pathway, £uorescence was observed in the periplasmic space [70]. This result is consistent with the prediction that the TAT pathway exports fully folded proteins, and that GFP can function in the periplasmic space if delivered in a properly folded conformation.
8. Protein export
The research in the author's laboratory is supported by the National Institutes of Health. The author thanks Craig Aspinwall, Bob Doyle, and Becky Graham for assistance with £uorescent microscopy, and Joe Lutkenhaus for the ZipA^GFP-expressing plasmid.
In addition to visualizing location of speci¢c classes of proteins within the con¢nes of the bacterial cytoplasm, GFP has also been used to study the export of proteins outside of the cytoplasm. Gene fusion technology has played important roles in dissecting bacterial protein export pathways. Fusions between an exported protein and L-galactosidase, for example, have provided new genetic selections and screens to isolate mutants defective in the export process. Using a similar approach, Feilmeier et al. [68] constructed fusions between GFP and proteins destined for localization to the periplasmic space. In contrast to the studies reported above, GFP was inactive in the periplasmic space due to its inability to fold properly in this subcellular compartment. The authors exploited this phenotype to isolate mutants that blocked export of the fusion [68]. Future work will likely see further exploitation
9. Conclusions Improvements in GFP technology are continually being made that enhance the sensitivity and utility of this reporter system, including new GFP variants that display an expanded spectrum of excitation and emission wavelengths. These variants have successfully been used to study localization of di¡erent cellular components within the same cell [49,71]. In addition, £uorescent proteins from other species have been identi¢ed to further expand the repertoire of reporter systems available for localization studies [72,73]. These variants, coupled with improvements in image analysis [74,75], should continue to enhance the sensitivity of GFP detection, further eliminating artifacts resulting from overexpressed or aggregated proteins. Clearly, the continued use of GFP as a localization reporter, in combination with other proven techniques, such as IFM, cell fractionation, and genetic approaches using other reporter genes, will result in new insights about the architecture and machinery of the bacterial cell, in addition to raising new questions about the function of some of its most fundamental components. Acknowledgements
References [1] Silhavy, T.J. (2000) Gene fusions. J. Bacteriol. 182, 5935^5938. [2] Chal¢e, M., Tu, Y., Euskirchen, G., Ward, W.W. and Prasher, D.C. (1994) Green £uorescent protein as a marker for gene expression. Science 263, 802^805. [3] Cody, C.W., Prasher, D.C., Westler, W.M., Pendergast, F.G. and Ward, W.W. (1993) Chemical structure of the hexapeptide chromophore of the Aequorea green-£uorescent protein. Biochemistry 32, 1212^1218. [4] Shapiro, L. and Losick, R. (2000) Dynamic spatial regulation in the bacterial cell. Cell 100, 89^98. [5] Losick, R. and Shapiro, L. (1999) Changing views on the nature of
FEMSLE 10108 17-10-01
16
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15] [16]
[17]
[18]
[19]
[20]
[21]
[22]
[23] [24]
[25]
[26]
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18 the bacterial cell: from biochemistry to cytology. J. Bacteriol. 181, 4135^4143. Shapiro, L. and Losick, R. (1997) Protein localization and cell fate in bacteria. Science 276, 712^718. Heim, R., Prasher, D.C. and Tsien, R.Y. (1994) Wavelength mutations and posttranslational autoxidation of green £uorescent protein. Proc. Natl. Acad. Sci. USA 91, 12501^12504. Crameri, A., Whitehorn, E.A., Tate, E. and Stemmer, W.P. (1996) Improved green £uorescent protein by molecular evolution using DNA shu¥ing. Nat. Biotechnol. 14, 315^319. Ehrig, T., O'Kane, D.J. and Prendergast, F.G. (1995) Green-£uorescent protein mutants with altered £uorescence excitation spectra. FEBS Lett. 367, 163^166. Cormack, B.P., Valdivia, R.H. and Falkow, S. (1996) FACS-optimized mutants of the green £uorescent protein (GFP). Gene 173, 33^38. Heim, R. and Tsien, R.Y. (1996) Engineering green £uorescent protein for improved brightness, longer wavelengths and £uorescence resonance energy transfer. Curr. Biol. 6, 178^182. Delagrave, S., Hawtin, R.E., Silva, C.M., Yang, M.M. and Youvan, D.C. (1995) Red-shifted excitation mutants of the green £uorescent protein. BioTechnology 13, 151^154. Feucht, A. and Lewis, P.J. (2001) Improved plasmid vectors for the production of multiple £uorescent protein fusions in Bacillus subtilis. Gene 264, 289^297. Miller, W.G. and Lindow, S.E. (1997) An improved GFP cloning cassette designed for prokaryotic transcriptional fusions. Gene 191, 149^153. Phillips, G.J. (1999) Gene cassettes for insertional mutagenesis and alteration of open reading frames. Anal. Biochem. 269, 207^210. Prescott, M., Nowakowski, S., Nagley, P. and Devenish, R.J. (1999) The length of polypeptide linker a¡ects the stability of green £uorescent protein fusion proteins. Anal. Biochem. 273, 305^307. King, N., Dreesen, O., Stragier, P., Pogliano, K. and Losick, R. (1999) Septation, dephosphorylation, and the activation of cF during sporulation in Bacillus subtilis. Genes Dev. 13, 1156^1167. Boyd, D., Weiss, D.S., Chen, J.C. and Beckwith, J. (2000) Towards single-copy gene expression systems making gene cloning physiologically relevant : Lambda InCh, a simple Escherichia coli plasmid^ chromosome shuttle system. J. Bacteriol. 182, 842^847. Brendler, T., Sawitzke, J., Sergueev, K. and Austin, S. (2000) A case for sliding SeqA tracts at anchored replication forks during Escherichia coli chromosome replication and segregation. EMBO J. 19, 6249^6258. Onogi, T., Niki, H., Yamozoe, M. and Hiraga, S. (1999) The assembly and migration of SeqA^GFP fusion in living cells of Escherichia coli. Mol. Microbiol. 31, 1775^1782. Imai, Y., Ogasawara, N., Ishigo-Oka, D., Kadoya, R., Daito, T. and Moriya, S. (2000) Subcellular localization of Dna-initiation proteins of Bacillus subtilis: evidence that chromosome replication begins at either edge of the nucleoids. Mol. Microbiol. 36, 1037^1048. Lemon, K.P. and Grossman, A.D. (1998) Localization of bacterial DNA polymerase : evidence for a factory model of replication. Science 282, 1516^1519. Margolin, W. (2000) Green £uorescent protein as a reporter for macromolecular localization in bacterial cells. Methods 20, 62^72. Arigoni, F., Pogliano, K., Webb, C.D., Stragier, P. and Losick, R. (1995) Localization of protein implicated in establishment of cell type to sites of asymmetric division. Science 270, 637. Lewis, P.J. and Errington, J. (1996) Use of green £uorescent protein for detection of cell-speci¢c gene expression and subcellular protein localization during sporulation in Bacillus subtilis. Microbiology 142, 733^740. Webb, C.D., Decatur, A., Teleman, A. and Losick, R. (1995) Use of green £uorescent protein for visualization of cell-speci¢c gene expres-
[27]
[28]
[29]
[30]
[31]
[32] [33]
[34] [35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
sion and subcellular protein localization during sporulation in Bacillus subtilis. J. Bacteriol. 177, 5906^5911. Price, K.D. and Losick, R. (1999) A four-dimensional view of assembly of a morphogenetic protein during sporulation in Bacillus subtilis. J. Bacteriol. 181, 781^790. Barak, I., Behari, J., Olmedo, G., Guzman, P., Brown, D.P., Castro, E., Walker, D., Westpheling, J. and Youngman, P. (1996) Structure and function of the Bacillus SpoIIE protein and its localization to sites of sporulation septum assembly. Mol. Microbiol. 19, 1047^1060. Wu, L.J., Feucht, A. and Errington, J. (1998) Prespore-speci¢c gene expression in Bacillus subtilis is driven by sequestration of SpoIIE phosphatase to the prespore side of the asymmetric septum. Genes Dev. 12, 1371^1380. Fawcett, P., Melnikov, A. and Youngman, P. (1998) The Bacillus SpoIIGA protein is targeted to sites of spore septum formation in a SpoIIE-independent manner. Mol. Microbiol. 28, 931^943. Ju, J. and Haldenwang, W.G. (1999) The `pro' sequence of the sporulation-speci¢c c transcription factor cE directs it to the mother cell side of the sporulation septum. J. Bacteriol. 181, 6171^6175. Bi, E. and Lutkenhaus, J. (1991) FtsZ ring structure associated with division in Escherichia coli. Nature 354, 161^164. Ma, X., Ehrdardt, D.W. and Margolin, W. (1996) Colocalization of cell division proteins FtsZ and FtsA to cytoskeletal structures in living Escherichia coli cells by using green £uorescent protein. Proc. Natl. Acad. Sci. USA 93, 12998^13003. Sun, Q. and Margolin, W. (1998) FtsZ dynamics during the division cycle of live Escherichia coli cells. J. Bacteriol. 180, 2050^2056. Weiss, D.S., Chen, J.C., Ghigo, J.M., Boyd, D. and Beckwith, J. (1999) Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z ring, FtsA, FtsQ, and FtsL. J. Bacteriol. 181, 508^520. Ghigo, J.M., Weiss, D.S., Chen, J.C., Yarrow, J.C. and Beckwith, J. (1999) Localization of FtsL to the Escherichia coli septal ring. Mol. Microbiol. 31, 725^737. Yu, X.C., Tran, A.H., Sun, Q. and Margolin, W. (1998) Localization of cell division protein FtsK to the Escherichia coli septum and identi¢cation of a potential N-terminal targeting domain. J. Bacteriol. 180, 1296^1304. Chen, J.C., Weiss, D.S., Ghigo, J.M. and Beckwith, J. (1999) Septal localization of FtsQ, an essential cell division protein in Escherichia coli. J. Bacteriol. 181, 521^530. Hale, C.A. and de Boer, P.A. (1999) Recruitment of ZipA to the septal ring of Escherichia coli is dependent on FtsZ and independent of FtsA. J. Bacteriol. 181, 167^176. Hale, C.A. and de Boer, P.A. (1997) Direct binding of FtsZ to ZipA, an essential component of the septal ring structure that mediates cell division in E. coli. Cell 88, 175^185. de Leeuw, E., Graham, B., Phillips, G.J., ten Hagen-Jongman, C.M., Oudega, B. and Luirink, J. (1999) Molecular characterization of Escherichia coli FtsE and FtsX. Mol. Microbiol. 31, 983^993. Mileykovskaya, E., Sun, Q., Margolin, W. and Dowhan, W. (1998) Localization and function of early cell division proteins in ¢lamentous Escherichia coli cells lacking phosphatidylethanolamine. J. Bacteriol. 180, 4252^4257. Raskin, D.M. and de Boer, P.A. (1999) Rapid pole-to-pole oscillation of a protein required for directing division to the middle of Escherichia coli. Proc. Natl. Acad. Sci. USA 96, 4971^4976. Hu, Z. and Lutkenhaus, J. (2000) Topological regulation of cell division in Escherichia coli involves rapid pole to pole oscillation of the division inhibitor MinC under the control of MinD and MinE. Mol. Microbiol. 34, 82^90. Raskin, D.M. and de Boer, P.A. (1997) The MinE ring: an FtsZindependent cell structure required for selection of the correct division site in E. coli. Cell 91, 685^694. Fu, X., Shih, Y.-L., Zhang, Y. and Roth¢eld, L.I. (2001) The MinE
FEMSLE 10108 17-10-01
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
[47]
[48] [49] [50]
[51]
[52]
[53]
[54]
[55]
[56]
[57] [58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
ring required for proper placement of the division site is a mobile structure that changes its cellular location during Escherichia coli division cycle. Proc. Natl. Acad. Sci. USA 98, 980^985. Hale, C.A., Meinhardt and de Boer, P.A. (2001) Dynamic localization cycle of the cell division regulator MinE in Escherichia coli. EMBO J. 20, 1563^1572. Dingman, C.W. (1974) Bidirectional chromosome replication : some topological considerations. J. Theor. Biol. 42, 187^195. Lemon, K.P. and Grossman, A.D. (2000) Movement of replicating DNA through a stationary replisome. Mol. Cell 6, 1321^1330. Sawitzke, J. and Austin, S. (2001) An analysis of the factory model for chromosome replication and segregation in bacteria. Mol. Microbiol. 40, 786^794. Lewis, P.J., Thaker, S.D. and Errington, J. (2000) Compartmentalization of transcription and translation in Bacillus subtilis. EMBO J. 19, 710^718. Straight, A.F., Belmont, A.S., Robinett, C.C. and Murray, A.W. (1996) GFP tagging of budding yeast chromosomes reveals that protein^protein interactions can mediate sister chromatid cohesion. Curr. Biol. 6, 1599^1608. Gordon, G.S., Sitnikov, D., Webb, C.D., Teleman, A., Straight, A., Losick, R., Murray, A.W. and Wright, A. (1997) Chromosome and low copy plasmid segregation in E. coli : visual evidence for distinct mechanisms. Cell 90, 1113^1121. Webb, C.D., Teleman, A., Gordon, S., Straight, A., Belmont, A., Lin, D.C., Grossman, A.D., Wright, A. and Losick, R. (1997) Bipolar localization of the replication origin regions of chromosomes in vegetative and sporulating cells of B. subtilis. Cell 88, 667^674. Webb, C.D., Graumann, P.L., Kahana, J.A., Teleman, A.A., Silver, P.A. and Losick, R. (1998) Use of time-lapse microscopy to visualize rapid movement of the replication origin region of the chromosome during the cell cycle in Bacillus subtilis. Mol. Microbiol. 28, 883^ 892. Teleman, A.A., Graumann, P.L., Lin, D.C.H., Grossman, A.D. and Losick, R. (1998) Chromosome arrangement within a bacterium. Curr. Biol. 8, 1102^1109. Gordon, G.S. and Wright, A. (2000) DNA segregation in bacteria. Annu. Rev. Microbiol. 54, 681^708. Pogliano, J., Ho, T.Q., Zhong, Z. and Helinski, D.R. (2001) Multicopy plasmids are clustered and localized in Escherichia coli. Proc. Natl. Acad. Sci. USA 98, 4486^4491. Glaser, P., Sharpe, M.E., Raether, B., Perego, M., Ohlsen, K. and Errington, J. (1997) Dynamic, mitotic-like behavior of a bacterial protein required for accurate chromosome partitioning. Genes Dev. 11, 1160^1168. Lin, D.C.-H., Levine, P.A. and Grossman, A.D. (1997) Bipolar localization of a chromosome partition protein in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 94, 4721^4726. Marston, A.L. and Errington, J. (1999) Dynamic movement of the ParA-like Soj protein of B. subtilis and its dual role in nucleoid organization and developmental regulation. Mol. Cell 4, 673^682. Quisel, J.D., Lin, D.C. and Grossman, A.D. (1999) Control of development by altered localization of a transcription factor in B. subtilis. Mol. Cell 4, 665^672. Hiraga, S., Ichinose, C., Niki, H. and Yajazoe, M. (1998) Cell cycledependent duplication and bi-directional migration of SeqA-associated DNA^protein complexes in Escherichia coli. Mol. Cell 1, 381^ 387. Sourjik, V. and Berg, H.C. (2000) Localization of components of the chemotaxis machinery of Escherichia coli using £uorescent protein fusions. Mol. Microbiol. 37, 740^751. Jacobs, C., Bomian, I.J., Maddock, J.R. and Shapiro, L. (1999) Cell cycle-dependent polar localization of an essential bacterial histidine kinase that controls DNA replication and cell division. Cell 97, 111^ 120. Wheeler, R.T. and Shapiro, L. (1999) Di¡erential localization of two
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
17
histidine kinases controlling bacterial cell di¡erentiation. Mol. Cell 4, 683^694. Jacobs, C., Hung, D. and Shapiro, L. (2001) Dynamic localization of a cytoplasmic signal transduction response regulator controls morphogenesis during the Caulobacter cell cycle. Proc. Natl. Acad. Sci. USA 98, 4095^4100. Feilmeier, B.J., Iseminger, G., Schroeder, D., Webber, H. and Phillips, G.J. (2000) Green £uorescent protein functions as a reporter for protein localization in Escherichia coli. J. Bacteriol. 182, 4068^ 4076. Valdivia, R.H. and Falkow, S. (1996) Bacterial genetics by £ow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by di¡erential £uorescence induction. Mol. Microbiol. 22, 367^378. Santini, C.L., Bernadac, A., Zhang, M., Chanal, A., Ize, B., Blanco, C. and Wu, L.F. (2000) Translocation of jelly¢sh green £uorescent protein via the TAT system of Escherichia coli and change of its periplasmic localization in response to osmotic up-shock. J. Biol. Chem. 276, 8159^8164. Lemon, K.P., Kurtser, I. and Grossman, A.D. (2001) E¡ects of replication termination mutants on chromosome partitioning in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 98, 212^217. Baird, G.S., Zacharias, D.A. and Tsien, R.Y. (2000) Biochemistry, mutagenesis, and oligomerization of DsRed, a red £uorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11984^11989. Gross, L.A., Baird, G.S., Ho¡man, R.C., Baldridge, K.K. and Tsien, R.Y. (2000) The structure of the chromophore within DsRed, a red £uorescent protein from coral. Proc. Natl. Acad. Sci. USA 97, 11990^11995. Dittrich, P., Malvezzi-Campeggi, F., Jahnz, M. and Schwille, P. (2001) Accessing molecular dynamics in cells by £uorescence correlation spectroscopy. Biol. Chem. 382, 491^494. Brock, R., Vamosi, G., Vereb, G. and Jovin, T.M. (1999) Rapid characterization of green £uorescent protein fusion proteins on the molecular and cellular level by £uorescence correlation microscopy. Proc. Natl. Acad. Sci. USA 96, 10123^10128. Edwards, D.H., Thomaides, H.B. and Errington, J. (2000) Promiscuous targeting of Bacillus subtilis cell division protein DivIVA to division sites in Escherichia coli and ¢ssion yeast. EMBO J. 19, 2719^ 2727. Wehrl, W., Niederweis, M. and Schumann, W. (2000) The FtsH protein accumulates at the septum of Bacillus subtilis during cell division and sporulation. J. Bacteriol. 182, 3870^3873. Sievers, J. and Errington, J. (2000) The Bacillus subtilis cell division protein FtsL localizes to sites of septation and interacts with DivIC. Mol. Microbiol. 36, 846^855. Marston, A.L., Thomaides, H.B., Edwards, D.H., Sharpe, M.E. and Errington, J. (1998) Polar localization of the MinD protein of Bacillus subtilis and its role in selection of the mid-cell division site. Genes Dev. 12, 3419^3430. Huang, W.M., Libbey, J.L., van der Hoeven, P. and Yu, S.X. (1998) Bipolar localization of Bacillus subtilis topoisomerase IV, an enzyme required for chromosome segregation. Proc. Natl. Acad. Sci. USA 95, 4652^4657. Kim, S.K. and Wang, J.C. (1998) Localization of F plasmid SopB protein to positions near the poles of Escherichia coli cells. Proc. Natl. Acad. Sci. USA 95, 1523^1527. Kohler, P. and Marahiel, M.A. (1997) Association of the histone-like protein HBsu with the nucleoid of Bacillus subtilis. J. Bacteriol. 179, 2060^2064. Wery, M., Woldringh, C.L. and Rouviere-Yaniv, J. (2001) HU^GFP and DAPI co-localize on the Escherichia coli nucleoid. Biochimie 83, 193^200. Britton, R.A., Lin, D.C. and Grossman, A.D. (1998) Characterization of a prokaryotic SMC protein involved in chromosome partitioning. Genes Dev. 12, 1254^1259.
FEMSLE 10108 17-10-01
18
G.J. Phillips / FEMS Microbiology Letters 204 (2001) 9^18
[85] Jensen, R.B. and Gerdes, K. (1999) Mechanism of DNA segregation in prokaryotes: ParM partitioning protein of plasmid R1 co-localizes with its replicon during the cell cycle. EMBO J. 18, 4076^4084. [86] Thomas, J.D., Daniel, R.A., Errington, J. and Robinson, C. (2001) Export of active green £uorescent protein to the periplasm by the twin-arginine translocase (Tat) pathway in Escherichia coli. Mol. Microbiol. 39, 47^53. [87] Boyd, J.M. (2000) Localization of the histidine kinase PilA to the
poles of Pseudomonas aeruginosa and identi¢cation of a localization domain. Mol. Microbiol. 36, 153^162. [88] Ju, J., Tingqiu, L. and Haldenwang, W.G. (1998) Forespore expression and processing of the SigE transcription factor in wild-type and mutant Bacillus subtilis. J. Bacteriol. 180, 1681^1693. [89] Resnekov, O., Alper, S. and Losick, R. (1996) Subcellular localization of proteins governing the proteolytic activation of a developmental transcription factor in Bacillus subtilis. Genes Cells 1, 529^542.
FEMSLE 10108 17-10-01