The N⁵-Glutamine S-Adenosyl-l-Methionine-Dependent Methyltransferase PrmC/HemK in Chlamydia trachomatis Methylates Class 1 Release Factors

Abstract

The gene prmC, encoding the putative S-adenosyl-l-methionine (AdoMet)-dependent methyltransferase (MTase) of release factors (RFs) of the obligate intracellular pathogen Chlamydia trachomatis, was functionally analyzed. Chlamydial PrmC expression suppresses the growth defect of a prmC knockout strain of Escherichia coli K-12, suggesting an interaction of chlamydial PrmC with E. coli RFs in vivo. In vivo methylation assays carried out with recombinant PrmC and RFs of chlamydial origin demonstrated that PrmC methylates RFs within the tryptic fragment containing the universally conserved sequence motif Gly-Gly-Gln. This is consistent with the enzymatic properties of PrmC of E. coli origin. We conclude that C. trachomatis PrmC functions as an N⁵-glutamine AdoMet-dependent MTase, involved in methylation of RFs.

S-Adenosyl-l-methionine (AdoMet)-dependent methyltransferases (MTases) are a large family of ubiquitous enzymes that transfer methyl groups from AdoMet to nitrogen, carbon, or oxygen compounds of a wide variety of substrates, such as nucleic acids, proteins, and many small molecules (24). The addition of methyl groups to DNA and posttranslational methylation of proteins as mediated by MTases provides a mechanism by which epigenetic information is imparted to these molecules. Such information can alter the targeting and timing of gene expression and activity of certain enzymes (6).

Methylation of DNA in bacteria serves to distinguish between self and nonself DNA, directs postreplicative mismatch repair, and controls the DNA replication and cell cycle. In addition, N⁶-adenine methylation of DNA plays an important role in the regulation of gene expression of virulence factors of a growing list of bacterial pathogens. For example, alterations in the levels of the N⁶-adenine methylator Dam attenuate the virulence of Salmonella spp., Yersinia pseudotuberculosis, and Vibrio cholerae (14). A hallmark of N⁶-adenine-specific and N⁴-cytosine DNA MTases is the common signature (N/D)PPY (4, 15).

Posttranslational methylation of proteins has been implicated in RNA transport and processing (arginine modification), chromatin regulation (lysine modification), and in chemotaxis of bacteria (glutamate modification) (3, 11, 13). It has recently been demonstrated that the methylase HemK of Escherichia coli, initially misidentified as a component of the heme biosynthetic pathway (17), modifies the class 1 release factors, RF1 and RF2, by N⁵-methylation of the glutamine residue of the universally conserved sequence motif Gly-Gly-Gln (GGQ) (5, 16). RF1 and RF2 recognize stop codons in mRNA, leading to the release of the completed polypeptide chain (8, 9). RF1 recognizes UAA and UAG stop codons, and RF2 recognizes UAA and UGA (26). HemK-mediated methylation of class 1 RFs has an important stimulatory effect on the release activity in vitro of RF2 of E. coli K-12 but not on RF1 (2). Inactivation of hemK in E. coli K-12 leads to increased stop codon read-through, an induction of the oxidative stress response, and severely retarded growth, indicating the importance of the posttranslational methylation event. Overexpression of HemK is toxic to cells (5, 16). The discovery that HemK actually methylates RFs was unexpected because HemK contains the hallmark sequence signature thought to be specific to N⁶-adenine DNA MTases, and no other members of this family have been previously reported to methylate proteins. On the basis of the recently elucidated enzyme activity, it has been suggested that the gene name hemK be replaced by prmC (protein release factor methylation) (5). In this work we will refer to the gene as prmC.

Chlamydia trachomatis is an obligate intracellular bacterial pathogen responsible for blinding trachoma and sexually transmitted disease (23). The role of methylation in the regulation of gene expression and protein activity in C. trachomatis is poorly understood. Genomes of Chlamydia spp. apparently lack the homologues of genes encoding well-characterized DNA MTases, as well as the metK gene, which encodes the enzyme AdoMet synthetase (7, 20, 21, 28). This raised the question of whether DNA and/or protein methylation events in Chlamydia do occur. To assess this, the sequenced chlamydial genomes were searched for possible prmC homologues. On the basis of homology with the E. coli prmC gene, CT024 was identified as a putative prmC gene. The PrmC function of CT024 was established by complementation of an E. coli prmC knockout mutant and by the ability of the protein to methylate RFs.

MATERIALS AND METHODS

Bacterial strains and plasmids.

E. coli K-12 strain SC5 [Δ(lac-pro. argE ara gyrA rpoB thi prfA-His₆-Tc ΔhemK) (5) was used for complementation assays and overexpression of PrmC of E. coli and chlamydial origin and overexpression of RF1 and RF2 from C. trachomatis. E. coli K-12 strain TOP10F′ (Invitrogen) was used for cloning purposes and propagation of plasmids. C. trachomatis bv. Lymphogranuloma venereum L2/434/Bu was originally obtained from P. B. Wyrick (East Tennessee State University). Plasmid pQE-80L (QIAGEN) was used for overexpression of chlamydial PrmC, RF1, and RF2. Plasmids pLV1 and pLV-hemK have been described previously (5). Plasmid pWSK129 (31) was used as a low-copy-number expression vector. Plasmid PCR2.1 (Invitrogen) was used for cloning and sequencing of PCR products.

Media and growth conditions.

E. coli strains were grown in Luria broth (LB) medium or on LB plates at the temperatures indicated, supplemented with the appropriate antibiotics at the following concentrations: tetracycline, 12.5 μg/ml; kanamycin, 50 μg/ml; and ampicillin, 100 μg/ml. Expression of recombinant proteins was induced by the addition of isopropyl-beta-d-thiogalactopyranoside (IPTG) at a final concentration of 1 mM. Growth comparisons were made by measuring the optical density at 600 nm (OD₆₀₀) of liquid cultures at intervals. C. trachomatis L2/434/Bu was propagated in HeLa 229 cells (ATCC CCL2.1) essentially as described previously (19).

Recombinant DNA techniques.

DNA, purified from C. trachomatis as described previously (19), was used as a template for PCR. Chlamydial prmC was amplified with primers 5′-CCCCGGATCCATGAAGAAACTGCTTAG-3′ and 5′-CCCCGAGCTCTCATCTCTCGGAAAAGCC-3′. Chlamydial prfA, encoding RF1, was amplified with primers 5′-CCCCGGTACCATGGAAATAAAAGTTTTAGAGTGTT-3′ and 5′-CCCCGTCGACCTAAGCAGTTTCTTCAT-3′. Chlamydial prfB, encoding RF2, was amplified with primers 5′-CCCCGGATCCATGCATGAGAATTTTGACAA-3′ and 5′-CCCCCTGCAGTCATGTAATTTCTCCATAAT-3′.

For sequencing, PCR products were directly cloned into PCR2.1 (Invitrogen). For other purposes, PCR products were digested with restriction enzymes (BamHI and SacI for prmC; KpnI and SalI for prfA; BamHI and PstI for prfB) (Roche), gel purified, and ligated in-frame into expression vectors. This way, recombinant proteins of chlamydial PrmC, RF1, and RF2, with a His₆ tag fused to the N terminus, were created using pQE-80L (QIAGEN), and the clones were designated pCtPrmC, pCtRF1, and pCtRF2, respectively. The DNA fragment encoding His₆-tagged chlamydial RFs was obtained by digestion of pCtRF1 and pCtRF2 with XhoI and PstI and subcloned into pWSK129 predigested with the same enzymes, thereby creating pWSK129-CtRF1 and pWSK129-CtRF2. All recombinant techniques and plasmid isolations were carried out essentially as described previously (22). DNA sequencing was performed by using the Thermo Sequenase II dye terminator sequencing premix kit (Amersham) according to instructions by the manufacturer. Sequences were obtained with an ABI370 automated sequencer (Perkin-Elmer).

Purification of His₆-tagged RF1 and RF2 of C. trachomatis.

For in vivo methylation experiments, recombinant His₆-tagged RF1 and RF2 were purified by nickel affinity chromatography, using nickel-nitrilotriacetic acid (Ni-NTA) columns (QIAGEN).

SC5 transformed with pWSK129-CtRF1 or pWSK129-CtRF2 and cotransformed with pCtPrmC or pQE-80L was grown in 0.5 or 4 liters of LB, respectively, supplemented with tetracycline, kanamycin, and ampicillin. Cultures were grown to an OD₆₀₀ of 0.5, and expression of recombinant protein was induced by addition of IPTG at a final concentration of 1 mM. Two hours after induction, cells were harvested by centrifugation at 10,000 × g and resuspended in 1 ml (RF1) or 5 ml (RF2) of resuspension buffer (30 mM Tris-HCl [pH 8.0], 1 mM dithiothreitol [DTT], 0.5 mM phenylmethylsulfonyl fluoride, 5 mM imidazole, containing 5 μg of DNase and 1 mg of lyzozyme) and incubated on ice for 30 min. Then, cells were lysed by sonification, cell debris was removed by centrifugation at 13,000 × g for 30 min, and the supernatant was loaded on to Ni-NTA columns. Recombinant proteins were eluted by using an imidazole gradient from 5 to 300 mM in 30 mM Tris-HCl (pH 8), 1 mM DTT, and 0.5 mM phenylmethylsulfonyl fluoride.

Mass spectrometric analysis.

Methylation of the tryptic fragment containing the GGQ domain of RFs of C. trachomatis was determined by mass spectrometry. His₆-tagged chlamydial RFs, methylated in vivo and Ni-NTA purified, were further purified by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), staining with Coomassie and excising the bands corresponding to chlamydial RF1 and RF2. For matrix-assisted laser desorption ionization (MALDI) analysis, protein-containing gel slices were treated essentially as described previously (27). Briefly, gel pieces were equilibrated for 15 min at 56°C in 50 mM NH₄HCO₃ and 1% (wt/vol) DTT, followed by a 30-min incubation, replacing DTT by 2.5% (wt/vol) iodoacetamide at room temperature. Twice, the pieces were washed with 100 mM NH₄HCO₃ followed by 100% acetonitrile. After removing all liquid, gel pieces were dried in a vacuum centrifuge and swollen on ice in digestion buffer containing 50 mM NH₄HCO₃ and 12.5 ng of trypsin (sequencing grade; Roche Diagnostics, Indianapolis, Ind.)/μl. After 30 min, the supernatant was removed and replaced with buffer without trypsin, hydrating the pieces during overnight incubation at 37°C. Overlay was collected, and peptides were extracted with 10 mM NH₄HCO₃, followed by 100% acetonitrile at room temperature. The pooled extracts were dried in a vacuum centrifuge. Peptides were resuspended in 5 μl of 60% acetonitrile-5% (vol/vol) formic acid. All peptide solutions were mixed 1:1 with a solution containing 52 mM α-cyano-4-hydroxycinnamic acid (Sigma-Aldrich) in 50% (vol/vol) ethanol-48% acetonitrile-2% (vol/vol) trifluoroacetic acid and 1 mM ammonium acetate. Prior to dissolving, the α-cyano-4-hydroxycinnamic acid was washed briefly with acetone. The mixture was spotted on a M@LDI target plate (Micromass, Wythenshawe, United Kingdom) and dried at room temperature. Reflectron MALDI-time of flight (MALDI-TOF) mass spectrometry spectra were collected on a Micromass M@LDI R instrument. The resulting spectra were used to search the following databases by means of the MassLynx ProteinProbe software package (Micromass): the Non-Redundant Protein Database (National Center for Biotechnology Information) and the International Protein Index (European Bioinformatics Institute, Cambridge, United Kingdom) and compared to the predicted mass profiles of the recombinant RF proteins.

Nucleotide sequence accession numbers.

The DNA sequences of prmC, prfA, and prfB and of C. trachomatis strain L2/434/Bu have been submitted to the DDBJ/EMBL/GenBank databases under accession numbers AY600244 (prmC), AY600245 (prfA), and AY600246 (prfB).

RESULTS

Identification of the putative chlamydial prmC homologue.

Analysis of the genome sequence of C. trachomatis serovar D strain D/UW-3/Cx revealed one open reading frame, CT024, encoding a putative protein containing the hallmark sequence of N⁶-adenine-specific DNA MTases (28). CT024 was originally designated hemK/yfcB, encoding an N⁶-adenine-specific DNA methylase, and was apparently thought to be functionally involved in some step of heme biosynthesis. CT024 of C. trachomatis serovar D is constitutively expressed during the intracellular developmental cycle of the bacterium (1, 18). The initiation codon of the gene overlaps the termination codon of prfA, the gene encoding RF1. This tandem, overlapping organization of prfA and putative prmC in one operon is conserved in all chlamydial species, as well as in E. coli and many other bacterial species for which the complete genome sequence is available (www.tigr.org). CT024 of C. trachomatis serovar D (290 amino acids) shows a significant homology (31% identity in a 269-amino-acid overlap) with PrmC of E. coli; hence, we refer to this gene as prmC. The amino acid sequence identity of CT024 with putative PrmC sequences of other chlamydial species available in public databases ranges from 80 to 57% identity, and sequencing of the corresponding gene of the C. trachomatis serovar L2 strain (L2/434/Bu) revealed only one amino acid substitution compared to the sequence of serovar D (data not shown).

C. trachomatis PrmC complements prmC knockout in E. coli.

The potential PrmC function of the CT024 product was studied by testing the ability of the chlamydial protein to complement the growth deficiency of an E. coli K-12 prmC knockout strain (SC5). In SC5, the part of prmC encoding the AdoMet binding motif is deleted, and the remaining part of the gene is truncated due to the insertion of a cassette encoding tetracycline resistance (5).

In the absence of IPTG, SC5 cells transformed with pCtPrmC (pQE-80L containing C. trachomatis prmC) or pLV-hemK (pLV1 containing E. coli hemK/prmC) clearly restored the growth defect of the prmC knockout strain and exhibited roughly the same growth characteristics (Fig. 1). This observation indicates complementation of PrmC in SC5 by pCtPrmC and hence functionality of the chlamydial protein in a heterologous background. Production of recombinant protein by both transformants cultured for 4 h in the absence of IPTG was undetectable in the protein profiles obtained by SDS-PAGE (data not shown), indicating that low levels of either recombinant PrmC were sufficient to achieve complementation.

FIG. 1.

Growth kinetics of the prmC knockout strain SC5 and complementation of the growth defect by recombinant prmC of C. trachomatis. Overnight cultures of SC5 and corresponding transformants were diluted to about an OD₆₀₀ of 0.05 with LB broth containing the appropriate antibiotics to a volume of 50 ml. Growth of the strains at 37°C was followed by measuring the density of the cultures at intervals in the absence (open symbols) and presence (closed symbols) of IPTG. Squares, SC5 transformed with pCtPrmC; circles, SC5 transformed with pLV-hemK; triangles, SC5 transformed with pQE-80L.

However, a striking difference in growth was observed between SC5 transformed with pCtPrmC and SC5 transformed with pLV-hemK in the presence of IPTG. Overexpression of pLV-hemK strongly retarded growth, to the level of the negative control (SC5 cells transformed with empty pQE-80L) (Fig. 1), confirming previous observations (16). In contrast, IPTG-induced overexpression of pCtPrmC stimulated growth, similar to growth in the absence of inducer. SDS-PAGE analysis revealed that in the presence of IPTG after 4 h of growth, the two recombinant PrmC proteins were produced in comparable amounts (data not shown); hence, the absence of toxicity of the chlamydial prmC product is not explained by less-efficient expression. More likely, the growth differences in the presence of IPTG may reflect differences between chlamydial and E. coli PrmC in their interaction with the substrates, the E. coli RFs.

In vivo methylation of chlamydial RFs by PrmC.

Our findings demonstrated that chlamydial PrmC complements the E. coli prmC knockout, suggestive of a direct interaction with RF2Thr246, and thus imply that chlamydial PrmC directly methylates RFs, in analogy to the function of PrmC of E. coli. PrmC of E. coli targets the glutamine residue of the universally conserved motif GGQ of RFs (5, 16). To determine whether PrmC of C. trachomatis also targets the glutamine residue of the GGQ motif, in vivo methylation assays with E. coli strain SC5 were carried out. To this end, N-terminal His₆-tagged recombinant C. trachomatis RF1 or RF2 was produced in SC5 cotransformed with pCtPrmC. After growth in liquid culture to the mid-exponential phase in the presence of IPTG, cells were harvested and chlamydial RFs were purified by affinity chromatography and SDS-PAGE. Chlamydial N-terminal His₆-tagged RF1 and RF2, purified from SC5 cotransformed with the empty vector pQE-80L and grown in the presence of IPTG, were used as negative controls. PrmC-dependent methylation was assessed by mass spectrometric analysis. The peptide mass fingerprint obtained by MALDI-TOF analysis of the tryptic digests identified the predicted release factor in all instances. In purified His₆-tagged recombinant C. trachomatis RF1 or RF2, a tryptic fragment of m/z 1,640.8 (corresponding to the predicted GGQ-containing fragment of both RFs) was present only in the extract obtained from cotransformation with the empty vector pQE-80L (Fig. 2A). In the extract derived from the cotransformation with pCtPrmC, the tryptic fragment of m/z 1,640.8 was absent. However, the tryptic digest of the RF proteins from this extract contained a new fragment of m/z 1,654.8, exactly corresponding to the addition of a methyl group to the fragment of both RF1 and RF2 containing the putative methylation site (Fig. 2B). These results lead to the conclusion that the gene CT024 of C. trachomatis encodes a functional PrmC that methylates chlamydial RFs within the tryptic fragment containing the universally conserved sequence motif GGQ.

FIG. 2.

Methylation of chlamydial RFs by CtPrmC. Relevant areas of MALDI-TOF mass spectra of tryptic digests of N-terminal His₆-tagged RF2 are shown. Since data obtained with RF1 are exactly the same for the tryptic fragment containing the GGQ motif, only data obtained with RF2 are shown. The calculated masses of the nonmethylated and methylated M+H ion are 1,640.79 and 1,654.80, respectively. Chlamydial RF2 was purified from SC5 transformed with pQE-80L (A) or from SC5 transformed with pCtPrmC (B). Note the total disappearance of the nonmethylated ion in panel B and the appearance of a novel fragment of m/z 1,654.8, corresponding to a methylated fragment containing the GGQ motif.

DISCUSSION

In this work we demonstrate that PrmC of C. trachomatis is involved in methylation of class I RFs. After PrmB, the ribosomal protein L3 glutamine MTase, PrmC is the second functionally well-characterized N⁵-glutamine AdoMet-dependent MTase naturewide (5). CtPrmC is the first described chlamydial MTase involved in the basic biological process of translation termination.

The activity of chlamydial PrmC was determined in the genetic background of E. coli. The relatively high number of molecules of C. trachomatis PrmC needed for complementation in SC5 may be indicative of a lower intrinsic activity of the chlamydial enzyme than of E. coli PrmC. On the other hand, a less-optimal interaction of the enzyme with heterologous RFs might also be the cause of less-efficient methyl transfer. Although the amino acid homology between E. coli and chlamydial PrmCs is only moderate (31%), the active center of the enzyme is very well conserved, as are the residues surrounding this center (25). This makes a lower intrinsic activity of chlamydial PrmC less likely. The primary sequence homology of the RFs of both species (53% for RF1 and 45% for RF2) seems relatively high, but it can be imagined that minor structural differences between chlamydial and E. coli RFs in conjunction with a chlamydial enzyme might hamper the proper positioning of the glutamine residue, possibly resulting in less-efficient methylation of RF2Thr246 by chlamydial PrmC in SC5. In addition, since the N-terminal domain of PrmC is suggested to be involved in substrate binding (25), the presence of a His₆ tag to the N terminus of the recombinant protein might also contribute to a lower transferase efficiency of chlamydial PrmC. Therefore, the observed lower efficiency of chlamydial PrmC in methylating E. coli RFs is more likely due to interaction differences between the enzyme and the substrate than to a lower intrinsic activity. Clearly, more kinetic studies, using purified enzymes and substrates, should be carried out to address this issue in more detail. The availability of recombinant chlamydial PrmC and RFs now enables such studies.

PrmC-mediated methylation of RFs of E. coli K-12 is essential for normal growth of the bacteria (5, 16), but the significance of PrmC for other organisms was unknown. Based on the maintained presence of the gene in small genomes of pathogens like the Mycoplasma, Rickettsia, and the Chlamydia and its occurrence among the proposed minimal set of genes required for life, an important role for PrmC in other organisms seemed likely (10). The functionality of chlamydial PrmC now provides additional strong experimental evidence for the physiological importance of the enzyme in organisms other than E. coli K-12. In addition, the results of the present study indicate that AdoMet-dependent methylation events in Chlamydia can take place in vivo. However, the gene metK, encoding AdoMet synthethase, is absent from the genomes of Chlamydia, implying that Chlamydia, like some Rickettsia spp. (29), must have evolved a transport system to import AdoMet from the host cell. Studies are now under way to investigate AdoMet transport in C. trachomatis and to identify the putative chlamydial AdoMet permease. Since AdoMet is an essential nutrient, involved in so many reactions that it is the second most widely used enzyme substrate after ATP, its chlamydial permease might be a novel target for therapeutic intervention.

Data about AdoMet-dependent methylation events in Chlamydia are scarce. Analysis of genomic DNA from C. trachomatis revealed unmethylated Dam sites and low levels of Dcm methylation (30). So far, sequenced genomes of Chlamydia spp. lack dam and dcm, raising doubts about the observed Dcm methylation. To our knowledge, only one other report has described methylation in C. trachomatis. The question was addressed whether, in analogy with eukaryotic systems (12), AdoMet-dependent methylation of chlamydial histone-like proteins by a chlamydial protein containing a SET domain (SET) occurs (32). Only intramolecular methylation of the SET protein was found, while no histone methylation was detected. Thus, the role of methylation in the regulation of gene expression and protein activity of Chlamydia is a completely uninvestigated field. Its exploration might provide novel insights into the mechanisms used by this pathogen to successfully grow and develop inside host cells. Finally, since C. trachomatis is phylogenetically very distantly related to E. coli, our results strongly indicate that N⁵ methylation of glutamine of the universally conserved sequence motif GGQ of RFs, previously described only for E. coli (5, 16), is not restricted to this species but more likely is a universally conserved phenomenon that contributes to the fine tuning of basic biological processes, such as translation termination.