Importance of trmE for Growth of the Psychrophile Pseudomonas syringae at Low Temperatures

Transposon mutagenesis of Pseudomonas syringae Lz4W, a psychrophilicbacterium capable of growing at temperatures between 2 and 30°C,yielded 30 cold-sensitive mutants, and CSM1, one of these cold-sensitivemutants, was characterized. Growth of CSM1 was retarded whenit was cultured at 4°C but not when it was cultured at 22°Cand 28°C compared to the growth of wild-type cells, indicatingthat CSM1 is a cold-sensitive mutant of P. syringae Lz4W. Themutated gene in CSM1 was identified as trmE (coding for tRNAmodification GTPase), and evidence is provided that this geneis induced at low temperatures. Further, the cold-induciblenature of the trmE promoter was demonstrated. In addition, thetranscription start site and the various regulatory elementsof the trmE promoter, such as the –10 region, –35region, UP element, cold box, and DEAD box, were identified,and the importance of these regulatory elements in promoteractivity were confirmed. The importance of trmE in rapid adaptationto growth at low temperatures was further highlighted by plasmid-mediatedcomplementation that alleviated the cold-sensitive phenotypeof CSM1.

INTRODUCTION

Top

INTRODUCTION

Psychrophilic bacteria (57) constitute a sizeable proportionof bacterial diversity because a large proportion of Earth'sbiosphere (75%) is either transiently or permanently cold (temperature,<5°C) (2). Studies have indicated that psychrophilesadapt to low temperatures by being able to sense changes intemperature (41, 48, 59), by modulating membrane fluidity (11-13,28, 29), and because they possess enzymes and genes which areactive at low temperatures (8, 10, 19, 35, 50, 60, 64). In psychrophilicbacteria pnp (encoding polynucleotide phosphorylase) (23), oppA(mediation of the transport of oligonucleotides) (5), and recD(51) have been identified as genes required for low-temperaturegrowth. In contrast, in mesophilic bacteria many genes are inducedfollowing a downshift in temperature; these genes include genesfor fatty acid desaturases and other enzymes (26, 32, 62), coldshock genes (33, 47), and genes involved in replication transcriptionand translation (3, 9, 26, 33, 69). The question is whethersuch genes are induced in psychrophiles, which, unlike mesophiles,are not cold stressed but are cold adapted. The present studyinvestigated the role of trmE in low-temperature growth.

MATERIALS AND METHODS

Top

MATERIALS AND METHODS

Generation of cold-sensitive mutants.
Psychrophilic Pseudomonas syringae Lz4W (referred to as P.syringaebelow) and Escherichia coli strains DH-5 ${alpha}$ and S-17-1 (Table 1)were grown in Antarctic bacterial medium (58) or Luria-Bertanimedium (66). P. syringae was mutagenized with a Tn5 transposon-basedsuicide plasmid vector (pOT182) (35, 42), and cold-sensitivemutants were identified based on their inability to grow ortheir delayed growth on plates incubated at 4°C for 1 week.Growth characteristics were also analyzed using a UV-visiblespectrophotometer (Shimadzu, Kyoto, Japan) (36, 37).

View this table:
[in this window]
[in a new window]

TABLE 1. Bacterial strains and plasmids used in this study

Identification of the disrupted gene.
Southern analysis using genomic DNA of CSM1 (one of the cold-sensitivemutants) and ${alpha}$ -³²P-labeled pOT182 (35, 36) showed that CSM1 producedtwo bands, confirming that a single copy of the transposon wasintegrated (data not shown). The disrupted gene was identifiedby inverse PCR (46) using 35 cycles and transposon-specificoutward primers (primers Ash 1 and Ash 3 for upstream regionsand primers Ash 1 and Ash 2 for downstream regions) (Table 2).PCR-amplified products were purified and sequenced with a ABIPrism 3700 DNA sequencer (Applied Biosystems, California) (12,13, 36, 37) using Ash 1 and Ash 3.

View this table:
[in this window]
[in a new window]

TABLE 2. Primers used in this study^a

Molecular biology techniques.
Genomic DNA of P. syringae was isolated (36), fragmented (35to 40 kb), and cloned in pCC1FOS (Epicentre Technologies, Madison,WI). All other DNA manipulation techniques, including colonyhybridization, were performed as described by Sambrook et al.(54).

For reverse transcription PCR RNA was isolated from P. syringae(1, 13, 36), and first-strand cDNA was synthesized by RT. ThecDNA was then amplified using Ash 71 and Ash 72 for the β-galactosidasegene and Ash 5 and Ash 44 for trmE for 21 cycles (Table 2).

Primer extension analysis and cloning of trmE promoter.
Transcription start site mapping of trmE was carried out byperforming primer extension analysis (37) using [ ${gamma}$ -³²P]ATP-end-labeledPextR2 (Table 2) and total RNA of P. syringae grown at 4°Cand 22°C. The putative trmE promoter of P. syringae wasamplified by PCR using Ash57 and Ash59 (Table 2) and then clonedin pKZ27 with the β-galactosidase gene as the reportergene (Promega Corporation, Madison, WI). The resulting pKZ27::promoterconstruct was then electroporated into P syringae (37).

Deletions of a specific promoter element were obtained by performingoverlap extension PCR (65) (see Fig. 4A) using primers listedin Table 2. PCR products were cloned in pKZ27, and deletionswere confirmed by DNA sequencing. The promoter constructs wereelectroporated into P. syringae, and transformants positivefor β-galactosidase activity (43) were selected.

View larger version (40K):
[in this window]
[in a new window]

FIG. 4. (A) Schematic diagrams of trmE promoter fragment deletion constructs of P. syringae. The deleted regions are indicated by dark gray boxes. The intact trmE promoter is shown on line a, whereas lines b to j show the regions that were deleted, as follows: line b, –10 region (deletion from position –5 to position –10); line c, –35 region (deletion from position –33 to position –38); line d, UP element region (deletion from position –41 to position –54); line e, cold box region (deletion from position 184 to position 194); line f, DEAD box (deletion from position 208 to position 217); line g, cold box and DEAD box regions (deletion from position 184 to position 194 and from position 208 to position 217); line h, conserved region (deletion from position 246 to position 252); line i, long 5'-UTR (deletion from position 9 to position 322); and line j, the entire promoter. (B and C) trmE promoter activities in P. syringae harboring the promoter deletions indicated above, assayed using cells grown at 22°C (B) and at 4°C (C), based on the activity of the β-galactosidase gene, the reporter gene.

Complementation of CSM1 with trmE.
The open reading frame (ORF) of trmE, along with its promoter,was amplified from P. syringae using primers Ash 57 and Ash47 and cloned in pGL10 (3). The resulting pGL10::promoter-trmEconstruct was mobilized into CSM1 (52), and the complementedstrain was checked for recovery of the phenotype.

RESULTS AND DISCUSSION

Top

RESULTS AND DISCUSSION

Cold-sensitive mutants of P. syringae.
A total of 3,500 mutants of P. syringae were generated usinga Tn5 transposon-based suicide plasmid vector (pOT182) (47).Thirty of these mutants were cold sensitive and exhibited delayedgrowth at 4°C but not at 22°C or 28°C (Fig. 1).Growth analysis of CSM1, one of the cold-sensitive mutants,indicated that its growth is not altered at 22°C and 28°Ccompared to the growth of the P. syringae wild-type strain (Fig.1). But at 4°C, CSM1 exhibited a prolonged 60-h lag period,compared to the 20-h lag period for the wild type. Thus, CSM1,like the pnp mutants of the psychrophile Yersinia enterocolitica(24), is a "psychrotrophy-defective" mutant since it exhibitsdelayed growth at 4°C but not at 22°C and 28°C comparedto the growth of the parental strain.

View larger version (17K):
[in this window]
[in a new window]

FIG. 1. Growth of the psychrophile P. syringae ( ${blacksquare}$ ), P. syringae harboring pGL10 ( ${triangleup}$ ), a cold-sensitive mutant of P. syringae (CSM1) ( ${blacktriangleup}$ ), complemented P. syringae cold-sensitive mutant CSM1 harboring the pGL10::trmE promoter plus trmE ( ${square}$ ), and P. syringae cold-sensitive mutant CSM1 harboring only pGL10 ( ${blacklozenge}$ ) cultured at 4°C (A), 22°C (B), and 28°C (C). OD_600nm, optical density at 600 nm.

The gene that was disrupted in CSM1 following amplificationyielded a 700-bp fragment which, as determined by sequencing,exhibited high levels of similarity (98 to 99%) with trmE (encodingtRNA modification GTPase) of Halomonas cupida, Brevibacteriumlinens, and Bacillus megaterium.

Characteristics of trmE of P. syringae.
Colony hybridization screening of a genomic library of P. syringaeusing the ³²P-labeled 700-bp PCR product of trmE identifiedtwo positive clones (5pCC1FOS and 11pCC1FOS). Sequencing ofthe insert of 5pCC1FOS (35 to 40 kb) using Ash 5, Ash 44, andAsh 59 yielded a 2,371-bp sequence corresponding to the completeORF of trmE (1,377 bp) (accession number AM944531), an upstreamincomplete ORF of the putative inner membrane protein translocasecomponent yidC gene subunit (587 bp), and a downstream incompleteORF of the abc transporter ATP-binding protein gene (407 bp).abc was in the same orientation as trmE, and transcriptionalterminators were not detected downstream of trmE. The transposoninsertion site was identified between bp 829 and 830 in trmE.

trmE codes for a 458-amino-acid protein (accession number CAQ16331).BLAST analysis (www.ncbi.nlm.nih.gov) showed that the aminoacid and nucleotide sequences of trmE of P. syringae were similarto the trmE sequences of various bacterial species, and thehighest level of similarity was 99% similarity with H. cupida(nucleotide sequence accession number AM944535 and amino acidsequence accession number CAQ16333).

CLUSTAL W analysis (www.chembnet.org/software/clustals.html)indicated that in TrmE of P. syringae the GTP binding (G-1,G-3, and G-4) and effector molecule binding (G-2) motifs involvedin GTPase activity are conserved (8). In addition, four otherregions (regions I, II, III, and IV) are conserved (see Fig.S1 in the supplemental material).

Complementation of trmE in CSM1.
The growth of complemented strain CSM1 (harboring pGL10::promoter-trmE)matched the growth of the wild type and the mutant at 22°Cand 28°C. But at 4°C the complemented strain grew fasterthan the mutant, and the phenotype was rescued significantly(Fig. 1). The lag period for the complemented strain was 32h, compared to 60 h for the mutant. Thus, it appears that thecold-sensitive phenotype was due to inactivation of trmE. However,the mutant did not recover totally compared to the wild type(Fig. 1). The slower growth of the complemented mutant was notdue to the vector since analysis of wild-type cells transformedeither with the vector (Fig. 1) or with the vector with theinsert did not show any effect on the growth kinetics (datanot shown). The inability of the complemented mutant to recovertotally may have been due to the polar effect of another genethat is downstream of trmE (22, 53) or due to a gene dosageeffect, or trmE may not have been expressed properly.

TrmE helps modify the uridine (U34) at the wobble position to5-methylaminomethyl-uridine (72). This modification influencesframeshift during the translation process (4, 6), thus causinga transient block in initiation of translation, as a consequenceof which growth at low temperatures is affected, as observedfor E. coli. Further, many genes involved in translation, suchas fus and genes for RNA helicases (63) and ribosomal proteins(14, 26, 33, 69), have been implicated as genes that may berequired for growth of mesophilic bacteria at low temperatures.trmE should be added to this list of genes.

Studies of E. coli have led to identification of a number ofcold shock proteins and cold-inducible genes (20, 24, 25). CspA,the major cold shock protein in E. coli, belongs to a familyof nine homologous proteins (27), but not all nine proteinsare cold inducible (38, 45, 56). cspA has been identified invarious psychrophilic bacteria, and this gene is constitutivelyexpressed at 4°C and 22°C (49). Interestingly, noneof the cold-shock-inducible genes are singularly responsiblefor cold adaptation (70).

Expression of trmE and mapping of the transcription start site.
In vivo expression of trmE in P. syringae that was shifted from22°C to 4°C increased transiently more than twofoldup to 30 min after the shift (Fig. 2A to C), and the maximumincrease, about 2.5-fold, was observed at 20 min after the shift.This temporal increase was consistently observed in three independentexperiments using RNA isolated specifically in each experiment.In contrast, cells which were grown only at 22°C (control)or cells which were shifted from 22°C to 28°C did notshow any change in the transcript levels for up to 60 min afterthe shift. These results imply that trmE expression is underthe control of a cold-inducible promoter. Expression of pnpand oppA in the psychrophiles Y. enterocolitica and Listeriamonocytogenes, respectively, was also reported to increase incells growing at 5°C compared to cells growing at 30°C(5, 23).

View larger version (57K):
[in this window]
[in a new window]

FIG. 2. RT-PCR analysis of trmE expression in P. syringae. RT-PCR was carried out using total RNA isolated from 3 ml of P. syringae grown to an optical density at 600 nm of $~$ 1.1 at 22°C and shifted to 4°C, 22°C, and 28°C for 0, 10, 20, 30, 45, and 60 min. Lane 1 contained the culture grown at 22°C (before the shift); lanes 2 to 6 contained the culture grown at 22°C and shifted to 4°C for 10, 20, 30, 45, and 60 min, respectively; lanes 7 to 11 contained the culture grown at 22°C and shifted to 22°C for 10, 20, 30, 45, and 60 min, respectively; and lanes 12 to 16 contained the culture grown at 22°C and shifted to 28°C for 10, 20, 30, 45, and 60 min, respectively. Lane M contained markers. (B) Amount of total RNA used in each of the RT-PCRs. (C) Ratios of the trmE transcript to total RNA (as determined by densitometry) for lanes 1 to 16 in panels A and B. The experiment was done three times, and the data are the results of a representative experiment.

Primer extension analysis revealed a primer extension productof an expected size (100 bp) in P. syringae grown either at4°C or at 22°C (Fig. 3A). The transcription start sitewas identified as a C (Fig. 3B).

View larger version (48K):
[in this window]
[in a new window]

FIG. 3. (A) Primer extension analysis of the trmE transcript of P. syringae cells grown to mid-log phase (optical density at 600 nm, $~$ 1.1) at 4°C and 22°C. The primer extension reactions were carried out using 5'-end-labeled primer PextR2 and reverse transcriptase. Primer extension products R4 and R22 are from cultures grown at 4°C and 22°C, respectively, and represent the transcription start site (TSS) (arrow). (B) Nucleotide sequence of the trmE promoter of P. syringae and its regulatory elements. The transcription start site at position +1 (C), –10 region (TGGATT), –35 region (TGAAAT), UP element (TACTTCTGGAAAGT), cold box (TGAACAACTGC), DEAD box (AACAGTGGTA), conserved region (CAAAAA), Shine-Dalgarno sequence (SDS) (GAGG), and translation start site (ATG) are highlighted. The arrows indicate the direction of the primers, the transcription start site, and the translation start site.

Activity and expression of the trmE promoter.
Primers Ash 57 and Ash 59 amplified a 655-bp product, whichcorresponded to 591 bp upstream and 64 bp downstream of thetranslation start site (ATG). Between the transcription andtranslation start sites, a long 345-bp 5' untranslated region(5'-UTR) was identified (Fig. 3B), which is one of the longest5'-UTRs in cold-inducible genes (Table 3). Further, the –10region, the –35 region, the UP element, the cold box,the DEAD box, and a conserved region (CAAAAAA) were identified(Fig. 3B; see Fig. S2 in the supplemental material). These promoterelements are highly conserved in the trmE promoters of otherbacteria (see Fig. S2 in the supplemental material). When theputative promoter (655-bp fragment) upstream of the trmE genewas ligated upstream of the β-galactosidase gene and clonedin P. syringae, it exhibited β-galactosidase activity (Fig.4A to C). But when the promoter elements were deleted (exceptfor deletion of the –35 region, which resulted in 50 and70% reductions in activity at 22 and 4°C, and deletion ofCAAAAA, which had no effect), a drastic reduction in β-galactosidaseactivity was observed irrespective of whether the transformantswere cultured at 4°C or 22°C (Fig. 4A to C). The CAAAAAAsequence was reported previously for the hut (30) and cti (37)promoters of P. syringae (see Fig. S2 in the supplemental material),but it does not appear to influence the promoter activity oftrmE. The present study demonstrates that the 5'-UTR, –10region, –35 region, UP element, cold box, and DEAD box(Fig. 3B) are essential for the activity of the trmE promoter,as they are for other genes (18, 31, 39, 44, 47, 66, 67) (Fig.4A to C). Further, the trmE promoter elements of P. syringaehad sequences identical to those in B. linens and Marinomonasprimoryensis, which are psychrophiles (see Fig. S2 in the supplementalmaterial).

View this table:
[in this window]
[in a new window]

TABLE 3. Length of 5'-UTRs in some cold-inducible genes of bacteria

The sequences of the –10 and –35 regions were similarto the sequences of the –10 (TATA/GAT/C) and –35(TA/GGA/GTA/T) regions of 11 promoters of the psychrophile Pseudoalteromonashaloplanktis but differed from the consensus sequences of theE. coli –10 and –35 regions (TATAAT and TTGACA,respectively) based on 300 promoters (40). Whether the observedsimilarity is a reflection of the cold-inducible nature of thepromoter is difficult to predict. The UP element (17) in P.syringae is an AT-rich region (64.3%) and shows limited similaritywith E. coli (21) and P. haloplanktis sequences (15) (see Fig.S2 in the supplemental material). The activity of the trmE promoterwas also analyzed by monitoring the levels of β-galactosidase(reporter) transcripts in P. syringae harboring pKZ27::promoter.The β-galactosidase gene transcript level was unalteredin control cultures, which were cultured continuously at 22°C,and in cultures shifted from 22°C to 28°C (Fig. 5A to C).But when cultures were shifted from 22°C to 4°C, a transientincrease in the β-galactosidase gene transcript level wasobserved for up to 30 min (Fig. 5A to C), confirming that thetrmE promoter is under transcriptional regulation at low temperatures.

View larger version (64K):
[in this window]
[in a new window]

FIG. 5. RT-PCR analysis of the expression of the trmE promoter based on expression of the β-galactosidase gene, the reporter gene in P. syringae (harboring pKZ27::promoter). (A) Lane 1, β-galactosidase gene transcript level in a culture grown at 22°C (before a shift); lanes 2 to 6, transcript levels in cultures grown at 22°C and shifted to 4°C for 10, 20, 30, 45, and 60 min, respectively; lanes 7 to 11, transcript levels in cultures grown at 22°C and shifted to 22°C for 10, 20, 30, 45, and 60 min; lanes 12 to 16, cultures grown at 22°C and shifted to 28°C for 10, 20, 30, 45, and 60 min, respectively; lane M, markers. (B) Amount of total RNA used in each of the RT-PCRs in panel A. (C) Ratio of the β-galactosidase gene transcript to the total RNA (as determined by densitometry) for lanes 1 to 16 in panels A and B. The experiment was done three times, and the data are the results of a representative experiment.

This temperature-dependent regulation is consistent with theconclusion that trmE plays a significant role in the adaptationto cold temperatures. Whereas previous studies have shown thatTrmE modifies tRNA and has GTPase activity, this report is thefirst report to identify a low-temperature function (7, 55,71). When TrmE activity is inhibited in E. coli, the loss ofU34 modification causes a transient block in the initiationof translation (16). Additional studies are needed to determinethe exact relationship between protein translation and the TrmE-mediatedmechanism of cold resistance. This report is the first reportdemonstrating the importance of trmE for growth at low temperatures.

ACKNOWLEDGMENTS

A.K.S. thanks the Council of Scientific and Industrial Research(Government of India, New Delhi, India) for a Junior and SeniorResearch Fellowship. S.S. thanks CSIR for funding under theNetwork Project "Exploitation of India's rich microbial diversity."

FOOTNOTES

* Corresponding author. Mailing address: Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad 500 007, India. Phone: 91-40-27192504. Fax: 91-40-27160591. E-mail: shivas{at}ccmb.res.in

Published ahead of print on 8 May 2009.

Supplemental material for this article may be found at http://aem.asm.org/.

REFERENCES

Top