- Email: [email protected]
Functional conservation between the argininosuccinate lyase of the archaeon Methanococcus maripaludis and the corresponding bacterial and eukaryal genes
FEMS Microbiology Letters 173 (1999) 231^238
Functional conservation between the argininosuccinate lyase of the archaeon Methanococcus maripaludis and the corresponding bacterial and eukaryal genes R. Cohen-Kupiec a , M. Kupiec b; *, K. Sandbeck a , J.A. Leigh b
a
a Department of Microbiology, University of Washington, Seattle, WA 98195-7242, USA Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Ramat-Aviv 69978, Israel
Received 31 December 1998; received in revised form 31 January 1999; accepted 31 January 1999
Abstract The argH gene encoding argininosuccinate lyase (ASL) of Methanococcus maripaludis was cloned on a 4.7-kb HindIII genomic fragment. The gene is preceded by a short open reading frame (ORF149), which encodes a polypeptide with an unknown function. The two genes are co-transcribed. The ASL of M. maripaludis shares a high amino acid identity with ASLs from both bacterial and eukaryal origins and was able to complement both an argH Escherichia coli mutant and an arg4 yeast mutant, showing its extraordinary evolutionary conservation. Attempts to create an argH auxotroph of M. maripaludis by disrupting the genomic allele were unsuccessful: although a knockout allele of argH was integrated into the M. maripaludis chromosome by homologous recombination, the intact copy was not excluded, suggesting that the argH gene is essential. z 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Argininosuccinate lyase ; Trans-kingdom complementation ; Archaeon; Methanococcus maripaludis
1. Introduction Archaea share distant phylogenetic relatedness to the two other domains, Bacteria and Eukarya. Archaeal gene structure resembles that of Bacteria [1], whereas promoter and RNA polymerase structures resemble those of eukaryotes [2]. The promoters of most archaeal genes isolated so far contain TATAlike binding sequences. Archaea appear to have only one RNA polymerase and not three separate en-
* Corresponding author. Tel.: +972 (3) 640-9031; Fax: +972 (3) 640-9407; E-mail: [email protected]
zymes as in Eukarya. However, the archaeal RNA polymerase is composed of multiple subunits, some of them homologous to their eukaryal counterparts [3]. Methanococcus maripaludis is one of a few archaeal species for which e¡ective genetic methods are emerging. This mesophilic methanogen inhabits marine environments, is an obligate anaerobe and is capable of diazotrophic growth [4]. We have reported in the past on the development of genetic and molecular tools to study gene regulation in M. maripaludis [4,5]. To date, only one selectable marker is currently used in methanogens, the puromycin resistance-conferring gene. This fact puts serious limi-
0378-1097 / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 9 9 ) 0 0 0 8 1 - 6
FEMSLE 8680 17-3-99
232
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
tations to the number and complexity of manipulations that can be carried out in further developing tools to study molecular fundamentals of Archaea. Thus, ¢nding new selectable markers is necessary. In an e¡ort to create auxotrophic mutants of M. maripaludis, we have isolated the argH gene, encoding the enzyme argininosuccinate lyase (ASL: EC 4.3.2.1). ASL catalyzes the reversible breakdown of argininosuccinic acid into arginine and fumarate, the last step in the biosynthesis of arginine in microorganisms and one of the last in the urea cycle, the major pathway for the detoxi¢cation of ammonia in Eukarya. The disruption of the gene was attempted, and functional complementation in bacteria and yeast was investigated.
2. Materials and methods 2.1. Strains and media Escherichia coli DH5K [6] was used for the propagation of plasmids. The clones were maintained by growth in LB broth with ampicillin (100 Wg/ml). The E. coli argH mutant W3678 was kindly supplied by B.J. Bachmann (Yale University, New Haven, CT, USA). For complementation studies, minimal medium (M9) was used with or without 2 mM (¢nal concentration) of L-arginine. M. maripaludis strains were grown under strict anaerobic conditions in McC medium [5]. Puromycin resistant M. maripaludis transformants were maintained in McC with 2.5 Wg/ml puromycin and 2 mM L-arginine. The Saccharomyces cerevisiae strain SY59: MATa ura3 leu2 trp1 his3 arg4 vade1 (M. Kupiec laboratory stock) was used. Yeast cells were grown at 30³C in YPD (1% yeast extract, 2% Bacto peptone, 2% glucose) or SD (0.67% yeast nitrogen base, 2% glucose and the appropriate nutrients added) media. Bacto Agar (1.8%) was added for solid media. SD3Leu and SD3Arg designate dropout media with all the amino acids added, except for leucine and arginine, respectively. 2.2. Transformation and DNA preparation Preparation and transformation of E. coli competent cells were carried out as described [6]. M. maripaludis transformation was carried out with 5^10 Wg
DNA as described elsewhere [5]. Transformants were plated on McC agar plates with 2 mM arginine and 2.5 Wg/ml puromycin. Yeast cells were transformed by standard procedures [7]. Plasmid pMK407, which contains a 3.0-kb HindIII fragment that carries the ARG4 gene of S. cerevisiae, was used as a positive control. 2.3. Plasmid construction A 4.7-kb HindIII fragment carrying the M. maripaludis argH gene was cloned into pGEM7 in which the EcoRI site was ¢lled in, to generate pRC8. The argH fragment was partially digested with EcoRI and ¢lled in. The only EcoRI site left was then used to clone the puromycin (pur) resistance gene £anked by the promoter and terminator sequences of the mcr gene from Methanococcus voltae [8]. The resulting plasmid, pRC110, was digested partially with EcoRI and ¢lled in, so that only the EcoRI site downstream to the pur fragment was left. In this site the lacZYA operon of E. coli (a SalI-BamHI fragment from plasmid pSK202 [9]), fused to the mcr promoter, was cloned, to generate plasmid pRCZ113 (see Fig. 3A). pRCA8 is the yeast-bacterial shuttle vector YEplac181 carrying the PstI-HindIII M. maripaludis argH fragment. 2.4. Other molecular techniques Southern analysis was performed on a nylon-based membrane (Zeta-probe, Bio-Rad) according to the protocol supplied by the manufacturer. RNA was extracted from M. maripaludis using the guanidinethiocyanate method, followed by phenol-chloroform extractions and ethanol precipitation. The RNA was then digested with RNase-free DNase (Promega), phenol-chloroform extracted and ethanol precipitated. For RT-PCR, 2 Wg RNA were combined with a reaction mix of the Titan One Tube RTPCR System (Boehringer, Germany) and cDNA synthesis was carried out for 45 min at 42 ³C, followed by 30 PCR cycles: 94 ³C, 1 min; 42 ³C, 1 min; 68 ³C, 2 min plus 5 s/cycle. For negative controls, 2 Wg RNA were combined with the PCR mix (1Ureaction bu¡er, 2.5 mM MgCl, 0.2 mM dNTP mix, forward and reverse primers (200 ng each) and two units of Taq polymerase (Promega). The control tubes were
FEMSLE 8680 17-3-99
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
233
Fig. 1. Alignment of ASLs from M. maripaludis, M. jannaschii, E. coli, Haemophilus in£uenza, S. cerevisiae, Candida albicans, Chlamidomonas reinhardtii and Homo sapiens. Only the regions that contain the ASL active site are shown. The full sequence can be obtained from GenBank, at accession number Y16584. The highly conserved residues across the ASL superfamily [12] are marked with asterisks.
added to the PCR cycler for the PCR part only. Forward primers (sense) were (numbers are as in the submitted GenBank sequence, accession number Y16584): rckI, ACAATTCAAACCTTGG (810^ 825), rckII, GTGCATCATGACTAC (14^28), 2.5F, TTGACGACATTCACATGG (327^344); reverse
primers (antisense) were: 2.5R (is complementary to 2.5F), CCATGTGAATGTCGTCAA (327^344) and 2.5SP6, CATTGCACCGCATCCTAA (671^688). DNA sequencing was performed with nested primers, using an automatic system (Perkin-Elmer DNA sequencer).
FEMSLE 8680 17-3-99
234
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
3. Results 3.1. Cloning and sequencing of the M. maripaludis argH gene A 4.7-kb HindIII genomic DNA fragment from M. maripaludis [10] was sequenced and found to contain two open reading frames (ORFs): a large one, 482 aa long, preceded by a short ORF, 149 aa long. The 482-aa ORF showed high identity to the ASL encoding gene of Methanococcus jannaschii (75% aa sequence identity), E. coli (47.6%) and other prokaryotes (Fig. 1). The M. maripaludis ASL homologue showed high identity also to argininosuccinate lyases from Eukarya, such as rat (46.8% aa sequence identity), the yeast Saccharomyces cerevisiae (47%), human (45%) and from the green alga Chlamydomonas reinhardtii (46%). The 149-aa ORF (designated ORF149) showed low ( 6 10%) homology to other prokaryotic proteins with no identi¢ed function. Among these were a 144aa hypothetical protein from the archaeon Pyrococcus horikoshii, a 137-aa putative protein from Aquifex aeolicus, and ORF YneT from Bacillus subtilis. ORF149 and the argH gene are separated by 41 nucleotides. The sequences GGTGG and GGTGA are complementary to the 16S rRNA 3P sequences of methanogens. These ribosome binding sequences were located four and ¢ve nucleotides upstream from the ¢rst ATGs of ORF149 and argH, respectively, indicating that their translation occurs separately. A sequence reminiscent of a consensus promoter sequence of methanogens (AAANNTTTATATA) could be detected 74 nucleotides upstream to the ¢rst ATG of ORF149. 3.2. argH constitutes an operon with ORF149 In order to determine the size of the argH transcript and the site of transcription initiation, we carried northern and primer extension analyses. The argH mRNA, however, could not be detected by either of those methods. Therefore, we performed an RT-PCR analysis which enabled us to amplify the argH transcript and check whether ORF149 and the argH gene constitute an operon. For that purpose, ¢ve primers were used. Three primers, rckI, rckII and 2.5F, corresponded to the sense
Fig. 2. RT-PCR ampli¢cation results : with primers 2.5SP6+rckII (lane 1), 2.5R+rckII (lane2), 2.5SP6+2.5F (lane 3), 2.5R+rckI (lane 4), PCR reaction (no RT step) with RNA and primers 2.5R+rckI (lane 5), 2 Wg RNA that were used in the RT-PCR and PCR reactions (lane 6), molecular standards DNA marker (lane 7). The samples were run in a 1.2% agarose gel. Numbers on the right represent the size (in kb) of the DNA marker.
strand of the putative operon and two primers, 2.5R and 2.5SP6, to its antisense strand. The last three primers are within the argH gene sequences. If the two ORFs were transcribed separately, we would observe an RT-PCR product only with primer pairs internal to a gene (2.5SP6 and 2.5F, Fig. 2, lane 3). However, all the reactions yielded an ampli¢ed product (Fig. 2). Thus, ampli¢cation with primers rckII and 2.5SP6 yielded a 675-bp product (Fig. 2, lane 1), and ampli¢cation with primers rckII or rckI with 2.5R (Fig. 2, lanes 2 and 4, respectively) yielded 316-bp and 723-bp products. Because both rckI and rckII primers are located in ORF149, the RT-PCR results indicate that argH is co-transcribed as an operon with ORF149. 3.3. Disruption of the M. maripaludis argH Despite the fact that M. maripaludis is one of the few archaea for which molecular tools have been developed in recent years, to the best of our knowledge, no auxotrophic mutants have been established to date in this species. We sought to disrupt the argH locus in order to create an auxotrophic mutant of M. maripaludis that will depend on exogenous arginine for growth. Such a mutant could enable us to use the argH gene as a selectable marker in M. maripaludis.
FEMSLE 8680 17-3-99
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
235
Fig. 3. A: Schematic presentations of plasmids pRC8 and pRCZ113. The 4.7-kb HindIII fragment that served as a probe for the Southern analysis shown in C and the size of the internal EcoRI fragment that hybridized to the M. maripaludis DNA digested with this enzyme are indicated. Also indicated are the approximate locations of ORF149 (small arrow) and argH in this fragment (large arrow) and the size of the disrupted argH locus in pRCZ113; H, HindIII; R, EcoRI; P, PstI; XR, a ¢lled in EcoRI site; B, BamHI. B : Schematic representation of the two possibilities for integration. 1: a single event that integrates one disrupted copy and leaves an intact one; 2: a replacement of the intact copy by a disrupted one. C: Southern analysis of wild-type M. maripaludis DNA, digested with the indicated restriction enzymes and DNA from transformants 21 and 28 digested with HindIII, probed with the 4.7-kb HindIII fragment shown in A. On the right, size standards in kb.
Plasmid pRC8, carrying the 4.7-kb HindIII argH fragment in a pGEM vector was disrupted by a lacpur cassette, yielding plasmid pRCZ113 (Fig. 3A). The lac-pur cassette consists of the lac operon of E. coli and a puromycin resistance gene, both transcriptionally fused to the mrc gene promoter of M. voltae. Plasmid pRCZ113 could integrate in the M. maripaludis genome either via a single crossing-over event between the genomic argH locus and one of the homologous £anking sequences in the plasmid (Fig. 3B1), or replace the wild-type argH locus in the genome with the disrupted one carried on the plasmid,
through a double crossing-over or a gene conversion event (Fig. 3B2). M. maripaludis cells were transformed with pRCZ113 and selected for resistance to puromycin. The transformants were supplemented with L-arginine. Twenty-eight out of 30 colonies analyzed showed a single crossing-over integration of the plasmid (Fig. 3C, transformant 21), such that a disrupted and a wild-type copy were present in tandem in the chromosome, separated by pGEM vector sequence. The two remaining colonies contained a fully disrupted copy of the argH locus, together with an in-
FEMSLE 8680 17-3-99
236
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
Fig. 4. Complementation of SY59, an arg4 S. cerevisiae strain, with pRCA8 (M. maripaludis argH) (A), YEplac181 (vector) (B) and pMK407 (yeast ARG4 gene) (C) on SD plus arginine (SD3Leu) and SD minus arginine (SD3Arg) plates.
tact one. The absence of the 3-kb pGEM vector DNA from the genomic DNA of these transformants suggested the existence of two copies of the M. maripaludis chromosome within a single cell (Fig. 3C, transformant 28), one with the disrupted argH locus, and another with an intact copy. Southern analysis of EcoRI- and BamHI-digested genomic DNAs from the M. maripaludis transformants con-
¢rmed the co-existence of a disrupted and an intact copy of the M. maripaludis chromosome within the same cells. These results suggested that the argH gene can not be disrupted in M. maripaludis, even when L-arginine is provided. Cultivation of the transformants from both clones at either higher puromycin concentration or without selective pressure did not result in the loss of the wild-type copy of the
Fig. 5. Complementation of an argH E. coli mutant with pGEM7 (vector) (A), pRC8 (pGEM+M. maripaludis ArgH) (B), pRCA8 (YEplac181+M. maripaludis argH) (C); in D, the mutant strain was streaked. Strains were streaked on M9 plates (MIN) and with the addition of arginine (MIN+Arg).
FEMSLE 8680 17-3-99
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
argH gene (data not shown). We conclude that the argH gene is essential for M. maripaludis growth and can not be mutated. 3.4. The M. maripaludis argH complements E. coli and yeast mutants Because the ASL of M. maripaludis showed high homology both to bacterial and to eukaryotic enzymes we wanted to check whether the archaeal argH could be expressed in bacteria and in a simple eukaryotic organism and complement the corresponding mutations. We used a S. cerevisiae strain defective in the ARG4 gene, which encodes argininosuccinate lyase. A PstI-HindIII fragment carrying the M. maripaludis argH locus was isolated from pRC8 (Fig. 3A) and cloned in YEplac181, a plasmid that can be propagated in E. coli and in S. cerevisiae. Competent yeast cells were transformed with the resulting plasmid, pRCA8. The transformants were grown on plates that contained either SD or SD3arginine media. While all transformants grew on the SD plates, only transformants that contained either plasmid pRCA8 or the control plasmid pMK407 (carrying S. cerevisiae wild-type ARG4 gene) grew on the SD3arginine plates (Fig. 4A). The argH of M. maripaludis was also tested for its ability to complement an E. coli argH mutant. An E. coli strain containing a Tn5 insertion at the argH locus could not grow on plates with minimal medium without Larginine. Only transformants with plasmid pRC8 or pRCA8 could grow on these plates (Fig. 5). These results demonstrate that the argH gene of the archaeon M. maripaludis could be transcribed, translated correctly and complement the corresponding mutations in both the bacterium E. coli and the eukaryote S. cerevisiae.
4. Discussion The gene encoding the homologue of ASL in M. maripaludis was sequenced and found to be preceded by a short ORF. The two genes are co-transcribed, but translated separately, as indicated by the presence of ribosome binding sites upstream to their coding sequences. This gene organization does not occur in M. jannaschii, a related methanogen, in which the
237
argH gene is preceded by an NADH dehydrogenase [11], and no sequences homologous to ORF149 were found in this organism. ORF149 does not contain a transmembrane domain, eliminating the possibility that it is active in the transport of arginine or other substrates of the arginine biosynthesis pathway; nor does it contain any other known motifs. In the microorganisms that contained a homologous protein, no physical relation to ASL exists. The ASL of M. maripaludis shares high amino acid identity with ASLs from prokaryotic and eukaryotic origins, suggesting that the gene in all three living domains evolved from a common ancestor. The fact that the M. maripaludis ASL was able to complement both E. coli and S. cerevisiae implies that the gene was correctly translated, and the protein was folded and assembled to form a functional tetramer [12]. It is possible that the ORF149 and the argH genes were co-transcribed in E. coli from a spurious prokaryotic promoter present in the clone. We note that a sequence resembling the 310 box of bacterial promoters is located downstream to the archaeal putative promoter. In yeast, translation is initiated at the ¢rst AUG of the mRNA [13] and does not depend on a ribosome binding sequence. If the two ORFs were co-transcribed in yeast, argH translation would not be correct. Therefore, we assume that transcription occurred from a eukaryotic promoter-like sequence, upstream to the argH gene. Attempts to disrupt the argH locus were not successful, even though arginine was added, suggesting that M. maripaludis does not transport arginine and that the gene is essential. The glnA gene of M. maripaludis also could not be disrupted, even when glutamine was added to the growth medium [14]. M. maripaludis can utilize alanine ([15]; our unpublished results), this fact excludes the possibility that M. maripaludis cannot import amino acids in general. A possible explanation for the indispensability of these genes is the role of their products in keeping the cellular ammonia and glutamate content balanced. ASL participates in the urea cycle, an important cycle for the detoxi¢cation of ammonia and the maintenance of a balanced nitrogen concentration. The urea cycle is generally regarded as a function characteristic of multicellular organisms. However, a complete urea cycle has been described in several prokaryotes such as Agrobacterium sp. [16], Bacillus
FEMSLE 8680 17-3-99
238
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
sp. [17], cyanobacteria [18], E. coli [19], and Streptomyces sp. [20]. All these organisms possess arginase activity, which catabolyzes arginine into urea and ornithine, urea is then hydrolyzed to ammonia and CO2 . It is not known yet if M. maripaludis contains arginase or urease homologues. As an autotroph capable of the energy-consuming nitrogen ¢xation process, M. maripaludis may assimilate glutamate and thus use the arginine biosynthetic pathway indirectly for the maintenance of a balanced nitrogen content in the cells.
[9]
[10]
[11]
Acknowledgments [12]
We thank Peter Kessler for his help and J. Bachmann for the strains. This work was supported by Grant 96-35305-3891 from the US Department of Agriculture.
[13]
[14]
References [1] Woese, C.R., Kandler, O. and Wheelis, M.L. (1990) Towards a natural system of organisms : proposal for the domains archaea, bacteria and eucarya. Proc. Natl. Acad. Sci. USA 87, 4576^4579. [2] Baumann, P., Qureshi, S.A. and Jackson, S.P. (1995) Transcription: new insights from studies on archaea. Trends Genet. 11, 279^283. [3] Reeve, J.N. (1992). Molecular biology of methanogens. Annu. Rev. Microbiol. 46, 165^191. [4] Blank, C.E., Kessler, P.S. and Leigh, J.A. (1995) Genetics in methanogens: transposon insertion mutagenesis of a Methanococcus maripaludis nifH gene. J. Bacteriol. 177, 5773^5777. [5] Cohen-Kupiec, R., Blank, C.E. and Leigh J.A. (1997) Transcriptional regulation in archaea : in vivo demonstration of a repressor binding site in a methanogen. Proc. Natl. Acad. Sci. USA 94, 1316^1320. [6] Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557^580. [7] Schiestl, R.H. and Gietz, R.D. (1989) High e¤ciency transformation of intact yeast cells by single stranded nucleic acid as carrier. Curr. Genet. 16, 339-346. [8] Gernhardt, P., Possot, O., Foglino, M., Sibold, L. and Klein,
[15]
[16]
[17]
[18]
[19]
[20]
A. (1990) Construction of an integration vector for use in the archaebacterium Methanococcus voltae and expression of a eubacterial resistance gene. Mol. Gen. Genet. 221, 273^279. Tomich, C.S., Kaytes, P.S., Olsen, M.K. and Patel, H. (1988) Use of lacZ expression to monitor transcription. Plasmid 20, 167^170. Sandbeck, K.A. and Leigh J.A. (1991) Recovery of an integration shuttle vector from tandem repeats in Methanococcus maripaludis. Appl. Environ. Microbiol. 57, 2762^2763. Bult, C.J., White, O., Olsen, G.J., Zhou, L., Fleischmann, R.D., Sutton, G.G., Blake, J.A., FitzGerald, L.M., Clayton, R.A., Gocayne, J.D., Kerlavage, A.R., Dougherty, B.A., Tomb, J.F., Adams, M.D., Reich, C.I., Overbeek, R., Kirkness, E.F., Weinstock, K.G., Merrick, J.M., Glodek, A., Scott, J.L., Geoghagen, N.S.M. and Venter, J.C. (1996) Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii. Science 273, 1058^1073. Turner, M.A., Simpson, A., Mcinnes, R.R. and Howell, P.L. (1997) Human argininosuccinate lyase : a structural basis for intragenic complementation. Proc. Natl. Acad. Sci. USA 94, 9063^9068. Baim, S.B. and Sherman, F. (1988) mRNA structures in£uencing translation in the yeast Saccharomyces cerevisiae. Mol. Cell. Biol. 8, 1591^1601. Cohen-Kupiec, R., Marx, C.J. and Leigh, J.A. (1999) Function and regulation of glnA transcription in the methanogenic archaeon Methanococcus maripaludis. J. Bacteriol. 181, 256^ 261. Whitman, W.B. (1989) Order II. Methanococcales Balch and Wolfe 1981, 216VP . In Bergey's Manual of Systematic Bacteriology (Staley, J.T., Bryant, M.P., Pfenning, N. and Holt, J.G., Eds.), Vol. 3, pp. 2185^2190. Williams and Wilkins, Baltimore, MD. Dessaux, Y., Petit, A., Tempe, J., Demarez, M., Legrain, C. and Wiame, J.M. (1986) Arginine catabolism in Agrobacterium strains : role of the Ti plasmid. J. Bacteriol. 166, 44^50. Patchett, M.L., Daniel, R.M. and Morgan, H.W. (1991) Characterization of arginase from the extreme thermophile Bacillus caldovelox. Biochim. Biophys. Acta 1077, 291^298. Gupta, M. and Carr, N.G. (1981) Enzymology of arginine metabolism in heterocyst forming cyanobacteria. FEMS Microbiol. Lett. 12, 179^181. Sans, N., Schroder, G. and Schroder, J. (1987) The Noc region of Ti plasmid C58 codes for arginase and ornithine cyclodeaminase. Eur. J. Biochem. 167, 81^87. Bascaran, V., Hardisson, C. and Brana, A.F. (1989) Isolation and characterization of nitrogen-deregulated mutants of Streptomyces clavuligerus. J. Gen. Microbiol. 135, 2475^2482.
FEMSLE 8680 17-3-99
Functional conservation between the argininosuccinate lyase of the archaeon Methanococcus maripaludis and the corresponding bacterial and eukaryal genes R. Cohen-Kupiec a , M. Kupiec b; *, K. Sandbeck a , J.A. Leigh b
a
a Department of Microbiology, University of Washington, Seattle, WA 98195-7242, USA Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Ramat-Aviv 69978, Israel
Received 31 December 1998; received in revised form 31 January 1999; accepted 31 January 1999
Abstract The argH gene encoding argininosuccinate lyase (ASL) of Methanococcus maripaludis was cloned on a 4.7-kb HindIII genomic fragment. The gene is preceded by a short open reading frame (ORF149), which encodes a polypeptide with an unknown function. The two genes are co-transcribed. The ASL of M. maripaludis shares a high amino acid identity with ASLs from both bacterial and eukaryal origins and was able to complement both an argH Escherichia coli mutant and an arg4 yeast mutant, showing its extraordinary evolutionary conservation. Attempts to create an argH auxotroph of M. maripaludis by disrupting the genomic allele were unsuccessful: although a knockout allele of argH was integrated into the M. maripaludis chromosome by homologous recombination, the intact copy was not excluded, suggesting that the argH gene is essential. z 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Argininosuccinate lyase ; Trans-kingdom complementation ; Archaeon; Methanococcus maripaludis
1. Introduction Archaea share distant phylogenetic relatedness to the two other domains, Bacteria and Eukarya. Archaeal gene structure resembles that of Bacteria [1], whereas promoter and RNA polymerase structures resemble those of eukaryotes [2]. The promoters of most archaeal genes isolated so far contain TATAlike binding sequences. Archaea appear to have only one RNA polymerase and not three separate en-
* Corresponding author. Tel.: +972 (3) 640-9031; Fax: +972 (3) 640-9407; E-mail: [email protected]
zymes as in Eukarya. However, the archaeal RNA polymerase is composed of multiple subunits, some of them homologous to their eukaryal counterparts [3]. Methanococcus maripaludis is one of a few archaeal species for which e¡ective genetic methods are emerging. This mesophilic methanogen inhabits marine environments, is an obligate anaerobe and is capable of diazotrophic growth [4]. We have reported in the past on the development of genetic and molecular tools to study gene regulation in M. maripaludis [4,5]. To date, only one selectable marker is currently used in methanogens, the puromycin resistance-conferring gene. This fact puts serious limi-
0378-1097 / 99 / $20.00 ß 1999 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 9 9 ) 0 0 0 8 1 - 6
FEMSLE 8680 17-3-99
232
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
tations to the number and complexity of manipulations that can be carried out in further developing tools to study molecular fundamentals of Archaea. Thus, ¢nding new selectable markers is necessary. In an e¡ort to create auxotrophic mutants of M. maripaludis, we have isolated the argH gene, encoding the enzyme argininosuccinate lyase (ASL: EC 4.3.2.1). ASL catalyzes the reversible breakdown of argininosuccinic acid into arginine and fumarate, the last step in the biosynthesis of arginine in microorganisms and one of the last in the urea cycle, the major pathway for the detoxi¢cation of ammonia in Eukarya. The disruption of the gene was attempted, and functional complementation in bacteria and yeast was investigated.
2. Materials and methods 2.1. Strains and media Escherichia coli DH5K [6] was used for the propagation of plasmids. The clones were maintained by growth in LB broth with ampicillin (100 Wg/ml). The E. coli argH mutant W3678 was kindly supplied by B.J. Bachmann (Yale University, New Haven, CT, USA). For complementation studies, minimal medium (M9) was used with or without 2 mM (¢nal concentration) of L-arginine. M. maripaludis strains were grown under strict anaerobic conditions in McC medium [5]. Puromycin resistant M. maripaludis transformants were maintained in McC with 2.5 Wg/ml puromycin and 2 mM L-arginine. The Saccharomyces cerevisiae strain SY59: MATa ura3 leu2 trp1 his3 arg4 vade1 (M. Kupiec laboratory stock) was used. Yeast cells were grown at 30³C in YPD (1% yeast extract, 2% Bacto peptone, 2% glucose) or SD (0.67% yeast nitrogen base, 2% glucose and the appropriate nutrients added) media. Bacto Agar (1.8%) was added for solid media. SD3Leu and SD3Arg designate dropout media with all the amino acids added, except for leucine and arginine, respectively. 2.2. Transformation and DNA preparation Preparation and transformation of E. coli competent cells were carried out as described [6]. M. maripaludis transformation was carried out with 5^10 Wg
DNA as described elsewhere [5]. Transformants were plated on McC agar plates with 2 mM arginine and 2.5 Wg/ml puromycin. Yeast cells were transformed by standard procedures [7]. Plasmid pMK407, which contains a 3.0-kb HindIII fragment that carries the ARG4 gene of S. cerevisiae, was used as a positive control. 2.3. Plasmid construction A 4.7-kb HindIII fragment carrying the M. maripaludis argH gene was cloned into pGEM7 in which the EcoRI site was ¢lled in, to generate pRC8. The argH fragment was partially digested with EcoRI and ¢lled in. The only EcoRI site left was then used to clone the puromycin (pur) resistance gene £anked by the promoter and terminator sequences of the mcr gene from Methanococcus voltae [8]. The resulting plasmid, pRC110, was digested partially with EcoRI and ¢lled in, so that only the EcoRI site downstream to the pur fragment was left. In this site the lacZYA operon of E. coli (a SalI-BamHI fragment from plasmid pSK202 [9]), fused to the mcr promoter, was cloned, to generate plasmid pRCZ113 (see Fig. 3A). pRCA8 is the yeast-bacterial shuttle vector YEplac181 carrying the PstI-HindIII M. maripaludis argH fragment. 2.4. Other molecular techniques Southern analysis was performed on a nylon-based membrane (Zeta-probe, Bio-Rad) according to the protocol supplied by the manufacturer. RNA was extracted from M. maripaludis using the guanidinethiocyanate method, followed by phenol-chloroform extractions and ethanol precipitation. The RNA was then digested with RNase-free DNase (Promega), phenol-chloroform extracted and ethanol precipitated. For RT-PCR, 2 Wg RNA were combined with a reaction mix of the Titan One Tube RTPCR System (Boehringer, Germany) and cDNA synthesis was carried out for 45 min at 42 ³C, followed by 30 PCR cycles: 94 ³C, 1 min; 42 ³C, 1 min; 68 ³C, 2 min plus 5 s/cycle. For negative controls, 2 Wg RNA were combined with the PCR mix (1Ureaction bu¡er, 2.5 mM MgCl, 0.2 mM dNTP mix, forward and reverse primers (200 ng each) and two units of Taq polymerase (Promega). The control tubes were
FEMSLE 8680 17-3-99
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
233
Fig. 1. Alignment of ASLs from M. maripaludis, M. jannaschii, E. coli, Haemophilus in£uenza, S. cerevisiae, Candida albicans, Chlamidomonas reinhardtii and Homo sapiens. Only the regions that contain the ASL active site are shown. The full sequence can be obtained from GenBank, at accession number Y16584. The highly conserved residues across the ASL superfamily [12] are marked with asterisks.
added to the PCR cycler for the PCR part only. Forward primers (sense) were (numbers are as in the submitted GenBank sequence, accession number Y16584): rckI, ACAATTCAAACCTTGG (810^ 825), rckII, GTGCATCATGACTAC (14^28), 2.5F, TTGACGACATTCACATGG (327^344); reverse
primers (antisense) were: 2.5R (is complementary to 2.5F), CCATGTGAATGTCGTCAA (327^344) and 2.5SP6, CATTGCACCGCATCCTAA (671^688). DNA sequencing was performed with nested primers, using an automatic system (Perkin-Elmer DNA sequencer).
FEMSLE 8680 17-3-99
234
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
3. Results 3.1. Cloning and sequencing of the M. maripaludis argH gene A 4.7-kb HindIII genomic DNA fragment from M. maripaludis [10] was sequenced and found to contain two open reading frames (ORFs): a large one, 482 aa long, preceded by a short ORF, 149 aa long. The 482-aa ORF showed high identity to the ASL encoding gene of Methanococcus jannaschii (75% aa sequence identity), E. coli (47.6%) and other prokaryotes (Fig. 1). The M. maripaludis ASL homologue showed high identity also to argininosuccinate lyases from Eukarya, such as rat (46.8% aa sequence identity), the yeast Saccharomyces cerevisiae (47%), human (45%) and from the green alga Chlamydomonas reinhardtii (46%). The 149-aa ORF (designated ORF149) showed low ( 6 10%) homology to other prokaryotic proteins with no identi¢ed function. Among these were a 144aa hypothetical protein from the archaeon Pyrococcus horikoshii, a 137-aa putative protein from Aquifex aeolicus, and ORF YneT from Bacillus subtilis. ORF149 and the argH gene are separated by 41 nucleotides. The sequences GGTGG and GGTGA are complementary to the 16S rRNA 3P sequences of methanogens. These ribosome binding sequences were located four and ¢ve nucleotides upstream from the ¢rst ATGs of ORF149 and argH, respectively, indicating that their translation occurs separately. A sequence reminiscent of a consensus promoter sequence of methanogens (AAANNTTTATATA) could be detected 74 nucleotides upstream to the ¢rst ATG of ORF149. 3.2. argH constitutes an operon with ORF149 In order to determine the size of the argH transcript and the site of transcription initiation, we carried northern and primer extension analyses. The argH mRNA, however, could not be detected by either of those methods. Therefore, we performed an RT-PCR analysis which enabled us to amplify the argH transcript and check whether ORF149 and the argH gene constitute an operon. For that purpose, ¢ve primers were used. Three primers, rckI, rckII and 2.5F, corresponded to the sense
Fig. 2. RT-PCR ampli¢cation results : with primers 2.5SP6+rckII (lane 1), 2.5R+rckII (lane2), 2.5SP6+2.5F (lane 3), 2.5R+rckI (lane 4), PCR reaction (no RT step) with RNA and primers 2.5R+rckI (lane 5), 2 Wg RNA that were used in the RT-PCR and PCR reactions (lane 6), molecular standards DNA marker (lane 7). The samples were run in a 1.2% agarose gel. Numbers on the right represent the size (in kb) of the DNA marker.
strand of the putative operon and two primers, 2.5R and 2.5SP6, to its antisense strand. The last three primers are within the argH gene sequences. If the two ORFs were transcribed separately, we would observe an RT-PCR product only with primer pairs internal to a gene (2.5SP6 and 2.5F, Fig. 2, lane 3). However, all the reactions yielded an ampli¢ed product (Fig. 2). Thus, ampli¢cation with primers rckII and 2.5SP6 yielded a 675-bp product (Fig. 2, lane 1), and ampli¢cation with primers rckII or rckI with 2.5R (Fig. 2, lanes 2 and 4, respectively) yielded 316-bp and 723-bp products. Because both rckI and rckII primers are located in ORF149, the RT-PCR results indicate that argH is co-transcribed as an operon with ORF149. 3.3. Disruption of the M. maripaludis argH Despite the fact that M. maripaludis is one of the few archaea for which molecular tools have been developed in recent years, to the best of our knowledge, no auxotrophic mutants have been established to date in this species. We sought to disrupt the argH locus in order to create an auxotrophic mutant of M. maripaludis that will depend on exogenous arginine for growth. Such a mutant could enable us to use the argH gene as a selectable marker in M. maripaludis.
FEMSLE 8680 17-3-99
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
235
Fig. 3. A: Schematic presentations of plasmids pRC8 and pRCZ113. The 4.7-kb HindIII fragment that served as a probe for the Southern analysis shown in C and the size of the internal EcoRI fragment that hybridized to the M. maripaludis DNA digested with this enzyme are indicated. Also indicated are the approximate locations of ORF149 (small arrow) and argH in this fragment (large arrow) and the size of the disrupted argH locus in pRCZ113; H, HindIII; R, EcoRI; P, PstI; XR, a ¢lled in EcoRI site; B, BamHI. B : Schematic representation of the two possibilities for integration. 1: a single event that integrates one disrupted copy and leaves an intact one; 2: a replacement of the intact copy by a disrupted one. C: Southern analysis of wild-type M. maripaludis DNA, digested with the indicated restriction enzymes and DNA from transformants 21 and 28 digested with HindIII, probed with the 4.7-kb HindIII fragment shown in A. On the right, size standards in kb.
Plasmid pRC8, carrying the 4.7-kb HindIII argH fragment in a pGEM vector was disrupted by a lacpur cassette, yielding plasmid pRCZ113 (Fig. 3A). The lac-pur cassette consists of the lac operon of E. coli and a puromycin resistance gene, both transcriptionally fused to the mrc gene promoter of M. voltae. Plasmid pRCZ113 could integrate in the M. maripaludis genome either via a single crossing-over event between the genomic argH locus and one of the homologous £anking sequences in the plasmid (Fig. 3B1), or replace the wild-type argH locus in the genome with the disrupted one carried on the plasmid,
through a double crossing-over or a gene conversion event (Fig. 3B2). M. maripaludis cells were transformed with pRCZ113 and selected for resistance to puromycin. The transformants were supplemented with L-arginine. Twenty-eight out of 30 colonies analyzed showed a single crossing-over integration of the plasmid (Fig. 3C, transformant 21), such that a disrupted and a wild-type copy were present in tandem in the chromosome, separated by pGEM vector sequence. The two remaining colonies contained a fully disrupted copy of the argH locus, together with an in-
FEMSLE 8680 17-3-99
236
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
Fig. 4. Complementation of SY59, an arg4 S. cerevisiae strain, with pRCA8 (M. maripaludis argH) (A), YEplac181 (vector) (B) and pMK407 (yeast ARG4 gene) (C) on SD plus arginine (SD3Leu) and SD minus arginine (SD3Arg) plates.
tact one. The absence of the 3-kb pGEM vector DNA from the genomic DNA of these transformants suggested the existence of two copies of the M. maripaludis chromosome within a single cell (Fig. 3C, transformant 28), one with the disrupted argH locus, and another with an intact copy. Southern analysis of EcoRI- and BamHI-digested genomic DNAs from the M. maripaludis transformants con-
¢rmed the co-existence of a disrupted and an intact copy of the M. maripaludis chromosome within the same cells. These results suggested that the argH gene can not be disrupted in M. maripaludis, even when L-arginine is provided. Cultivation of the transformants from both clones at either higher puromycin concentration or without selective pressure did not result in the loss of the wild-type copy of the
Fig. 5. Complementation of an argH E. coli mutant with pGEM7 (vector) (A), pRC8 (pGEM+M. maripaludis ArgH) (B), pRCA8 (YEplac181+M. maripaludis argH) (C); in D, the mutant strain was streaked. Strains were streaked on M9 plates (MIN) and with the addition of arginine (MIN+Arg).
FEMSLE 8680 17-3-99
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
argH gene (data not shown). We conclude that the argH gene is essential for M. maripaludis growth and can not be mutated. 3.4. The M. maripaludis argH complements E. coli and yeast mutants Because the ASL of M. maripaludis showed high homology both to bacterial and to eukaryotic enzymes we wanted to check whether the archaeal argH could be expressed in bacteria and in a simple eukaryotic organism and complement the corresponding mutations. We used a S. cerevisiae strain defective in the ARG4 gene, which encodes argininosuccinate lyase. A PstI-HindIII fragment carrying the M. maripaludis argH locus was isolated from pRC8 (Fig. 3A) and cloned in YEplac181, a plasmid that can be propagated in E. coli and in S. cerevisiae. Competent yeast cells were transformed with the resulting plasmid, pRCA8. The transformants were grown on plates that contained either SD or SD3arginine media. While all transformants grew on the SD plates, only transformants that contained either plasmid pRCA8 or the control plasmid pMK407 (carrying S. cerevisiae wild-type ARG4 gene) grew on the SD3arginine plates (Fig. 4A). The argH of M. maripaludis was also tested for its ability to complement an E. coli argH mutant. An E. coli strain containing a Tn5 insertion at the argH locus could not grow on plates with minimal medium without Larginine. Only transformants with plasmid pRC8 or pRCA8 could grow on these plates (Fig. 5). These results demonstrate that the argH gene of the archaeon M. maripaludis could be transcribed, translated correctly and complement the corresponding mutations in both the bacterium E. coli and the eukaryote S. cerevisiae.
4. Discussion The gene encoding the homologue of ASL in M. maripaludis was sequenced and found to be preceded by a short ORF. The two genes are co-transcribed, but translated separately, as indicated by the presence of ribosome binding sites upstream to their coding sequences. This gene organization does not occur in M. jannaschii, a related methanogen, in which the
237
argH gene is preceded by an NADH dehydrogenase [11], and no sequences homologous to ORF149 were found in this organism. ORF149 does not contain a transmembrane domain, eliminating the possibility that it is active in the transport of arginine or other substrates of the arginine biosynthesis pathway; nor does it contain any other known motifs. In the microorganisms that contained a homologous protein, no physical relation to ASL exists. The ASL of M. maripaludis shares high amino acid identity with ASLs from prokaryotic and eukaryotic origins, suggesting that the gene in all three living domains evolved from a common ancestor. The fact that the M. maripaludis ASL was able to complement both E. coli and S. cerevisiae implies that the gene was correctly translated, and the protein was folded and assembled to form a functional tetramer [12]. It is possible that the ORF149 and the argH genes were co-transcribed in E. coli from a spurious prokaryotic promoter present in the clone. We note that a sequence resembling the 310 box of bacterial promoters is located downstream to the archaeal putative promoter. In yeast, translation is initiated at the ¢rst AUG of the mRNA [13] and does not depend on a ribosome binding sequence. If the two ORFs were co-transcribed in yeast, argH translation would not be correct. Therefore, we assume that transcription occurred from a eukaryotic promoter-like sequence, upstream to the argH gene. Attempts to disrupt the argH locus were not successful, even though arginine was added, suggesting that M. maripaludis does not transport arginine and that the gene is essential. The glnA gene of M. maripaludis also could not be disrupted, even when glutamine was added to the growth medium [14]. M. maripaludis can utilize alanine ([15]; our unpublished results), this fact excludes the possibility that M. maripaludis cannot import amino acids in general. A possible explanation for the indispensability of these genes is the role of their products in keeping the cellular ammonia and glutamate content balanced. ASL participates in the urea cycle, an important cycle for the detoxi¢cation of ammonia and the maintenance of a balanced nitrogen concentration. The urea cycle is generally regarded as a function characteristic of multicellular organisms. However, a complete urea cycle has been described in several prokaryotes such as Agrobacterium sp. [16], Bacillus
FEMSLE 8680 17-3-99
238
R. Cohen-Kupiec et al. / FEMS Microbiology Letters 173 (1999) 231^238
sp. [17], cyanobacteria [18], E. coli [19], and Streptomyces sp. [20]. All these organisms possess arginase activity, which catabolyzes arginine into urea and ornithine, urea is then hydrolyzed to ammonia and CO2 . It is not known yet if M. maripaludis contains arginase or urease homologues. As an autotroph capable of the energy-consuming nitrogen ¢xation process, M. maripaludis may assimilate glutamate and thus use the arginine biosynthetic pathway indirectly for the maintenance of a balanced nitrogen content in the cells.
[9]
[10]
[11]
Acknowledgments [12]
We thank Peter Kessler for his help and J. Bachmann for the strains. This work was supported by Grant 96-35305-3891 from the US Department of Agriculture.
[13]
[14]
References [1] Woese, C.R., Kandler, O. and Wheelis, M.L. (1990) Towards a natural system of organisms : proposal for the domains archaea, bacteria and eucarya. Proc. Natl. Acad. Sci. USA 87, 4576^4579. [2] Baumann, P., Qureshi, S.A. and Jackson, S.P. (1995) Transcription: new insights from studies on archaea. Trends Genet. 11, 279^283. [3] Reeve, J.N. (1992). Molecular biology of methanogens. Annu. Rev. Microbiol. 46, 165^191. [4] Blank, C.E., Kessler, P.S. and Leigh, J.A. (1995) Genetics in methanogens: transposon insertion mutagenesis of a Methanococcus maripaludis nifH gene. J. Bacteriol. 177, 5773^5777. [5] Cohen-Kupiec, R., Blank, C.E. and Leigh J.A. (1997) Transcriptional regulation in archaea : in vivo demonstration of a repressor binding site in a methanogen. Proc. Natl. Acad. Sci. USA 94, 1316^1320. [6] Hanahan, D. (1983) Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557^580. [7] Schiestl, R.H. and Gietz, R.D. (1989) High e¤ciency transformation of intact yeast cells by single stranded nucleic acid as carrier. Curr. Genet. 16, 339-346. [8] Gernhardt, P., Possot, O., Foglino, M., Sibold, L. and Klein,
[15]
[16]
[17]
[18]
[19]
[20]
A. (1990) Construction of an integration vector for use in the archaebacterium Methanococcus voltae and expression of a eubacterial resistance gene. Mol. Gen. Genet. 221, 273^279. Tomich, C.S., Kaytes, P.S., Olsen, M.K. and Patel, H. (1988) Use of lacZ expression to monitor transcription. Plasmid 20, 167^170. Sandbeck, K.A. and Leigh J.A. (1991) Recovery of an integration shuttle vector from tandem repeats in Methanococcus maripaludis. Appl. Environ. Microbiol. 57, 2762^2763. Bult, C.J., White, O., Olsen, G.J., Zhou, L., Fleischmann, R.D., Sutton, G.G., Blake, J.A., FitzGerald, L.M., Clayton, R.A., Gocayne, J.D., Kerlavage, A.R., Dougherty, B.A., Tomb, J.F., Adams, M.D., Reich, C.I., Overbeek, R., Kirkness, E.F., Weinstock, K.G., Merrick, J.M., Glodek, A., Scott, J.L., Geoghagen, N.S.M. and Venter, J.C. (1996) Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii. Science 273, 1058^1073. Turner, M.A., Simpson, A., Mcinnes, R.R. and Howell, P.L. (1997) Human argininosuccinate lyase : a structural basis for intragenic complementation. Proc. Natl. Acad. Sci. USA 94, 9063^9068. Baim, S.B. and Sherman, F. (1988) mRNA structures in£uencing translation in the yeast Saccharomyces cerevisiae. Mol. Cell. Biol. 8, 1591^1601. Cohen-Kupiec, R., Marx, C.J. and Leigh, J.A. (1999) Function and regulation of glnA transcription in the methanogenic archaeon Methanococcus maripaludis. J. Bacteriol. 181, 256^ 261. Whitman, W.B. (1989) Order II. Methanococcales Balch and Wolfe 1981, 216VP . In Bergey's Manual of Systematic Bacteriology (Staley, J.T., Bryant, M.P., Pfenning, N. and Holt, J.G., Eds.), Vol. 3, pp. 2185^2190. Williams and Wilkins, Baltimore, MD. Dessaux, Y., Petit, A., Tempe, J., Demarez, M., Legrain, C. and Wiame, J.M. (1986) Arginine catabolism in Agrobacterium strains : role of the Ti plasmid. J. Bacteriol. 166, 44^50. Patchett, M.L., Daniel, R.M. and Morgan, H.W. (1991) Characterization of arginase from the extreme thermophile Bacillus caldovelox. Biochim. Biophys. Acta 1077, 291^298. Gupta, M. and Carr, N.G. (1981) Enzymology of arginine metabolism in heterocyst forming cyanobacteria. FEMS Microbiol. Lett. 12, 179^181. Sans, N., Schroder, G. and Schroder, J. (1987) The Noc region of Ti plasmid C58 codes for arginase and ornithine cyclodeaminase. Eur. J. Biochem. 167, 81^87. Bascaran, V., Hardisson, C. and Brana, A.F. (1989) Isolation and characterization of nitrogen-deregulated mutants of Streptomyces clavuligerus. J. Gen. Microbiol. 135, 2475^2482.
FEMSLE 8680 17-3-99