Skip to main content
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 1998 Nov 10;95(23):13923–13928. doi: 10.1073/pnas.95.23.13923

A family of phase-variable restriction enzymes with differing specificities generated by high-frequency gene rearrangements

Kevin Dybvig †,‡,*, Ramakrishnan Sitaraman , C Todd French
PMCID: PMC24968  PMID: 9811902

Abstract

The hsd genes of Mycoplasma pulmonis encode restriction and modification enzymes exhibiting a high degree of sequence similarity to the type I enzymes of enteric bacteria. The S subunits of type I systems dictate the DNA sequence specificity of the holoenzyme and are required for both the restriction and the modification reactions. The M. pulmonis chromosome has two hsd loci, both of which contain two hsdS genes each and are complex, site-specific DNA inversion systems. Embedded within the coding region of each hsdS gene are a minimum of three sites at which DNA inversions occur to generate extensive amino acid sequence variations in the predicted S subunits. We show that the polymorphic hsdS genes produced by gene rearrangement encode a family of functional S subunits with differing DNA sequence specificities. In addition to creating polymorphisms in hsdS sequences, DNA inversions regulate the phase-variable production of restriction activity because the other genes required for restriction activity (hsdR and hsdM) are expressed only from loci that are oriented appropriately in the chromosome relative to the hsd promoter. These data cast doubt on the prevailing paradigms that restriction systems are either selfish or function to confer protection from invasion by foreign DNA.

Keywords: phase variation/mycoplasmas/DNA inversion/antigenic variation/mycoplasma virus


Bacterial restriction and modification (R-M) systems function as a defense mechanism conferring protection from phage infection and other types of DNA invasion. Incoming DNA lacking the appropriate base modifications is cleaved by the restriction endonuclease while host DNA is protected by the presence of sequence-specific base modifications introduced by the activity of the DNA methyltransferase (MTase). These systems usually confer incomplete protection from phage infection because some phage DNA molecules are modified by the MTase before the endonucleolytic cleavage can occur, giving rise to progeny phage containing DNA modifications identical to that of the host and, therefore, resistant to restriction activity.

The type I restriction enzymes are considerably more complex than the more prevalent type II enzymes. Type II restriction activity and MTase activity are performed by two distinct enzymes encoded by gene pairs. The endonuclease reaction is sequence-specific and occurs at or very near to the site modified by the MTase. In the type I systems, the nuclease and MTase activities are performed by the same holoenzyme, which consists of three types of subunits. The S subunit dictates the sequence specificity of the MTase activity. Both the S and M subunits are necessary and sufficient for MTase activity, which occurs at specific adenine residues. All three types of subunits (S, R, and M) are required for the nuclease reaction, which occurs at essentially random sites up to 7 kilobases from the recognition sequence (1). Translocation of the holoenzyme from the recognition site to the cleavage site occurs through the ATP-dependent helicase activity of the R subunit.

It has been proposed that some restriction systems are selfish because the daughter cells that have lost the R-M genes may be killed if the MTase activity is lost more rapidly than the nuclease activity (2, 3). Therefore, it is possible that some restriction systems persist in bacteria because of selfish behavior and are not maintained for cellular defense. Data supporting the selfish behavior hypothesis are based on type II systems. In contrast, a recent study concluded that type I systems may not be selfish and that cellular defense is a possible function (4).

Mycoplasma pulmonis possesses the most complex hsd genes described thus far (5, 6). The M. pulmonis chromosome has two hsd loci, each encoding S, M, and R subunits highly homologous to type I restriction systems. Both loci are site-specific DNA inversion systems containing two hsdS genes flanking hsdR and hsdM. DNA inversions occur within the coding regions of the hsdS genes and result in extensive genetic polymorphisms. We show here that DNA inversions regulate the phase-variable production of restriction activity and that polymorphic hsdS genes encode S subunits of differing specificity. The phase-variable nature of the system strongly argues against the selfish behavior hypothesis and raises the issue of how cells with unmodified genomic DNA survive when restriction activity is induced. We conclude that cultures of M. pulmonis contain a significant subpopulation of cells in which the hsdR and hsdM genes are not transcribed. Such cells lack detectable restriction activity and are highly susceptible to phage infection. Therefore, the hsd loci in M. pulmonis are ineffective as barriers against invasion by foreign DNA and may serve alternative functions.

MATERIALS AND METHODS

Mycoplasmas.

M. pulmonis was propagated in mycoplasma medium as described (7). Subclones of M. pulmonis (Table 1) were derived by using filter clone methodology (8). In brief, cell cultures were passed gently through a 0.2-μm filter immediately before assaying for colony-forming units. This technique removes cell aggregates, and the resulting colonies are presumed to be derived from single cells. The resulting colonies (filter clones) were selected randomly for propagation and analysis for R-M activity.

Table 1.

M. pulmonis strains used in this study

Strain Parent strain Restriction group, enzymes Reference
KD735-15 KD735 I, no MpuU enzymes 6
KD735-16K KD735 II, MpuUI and MpuUII 5
KD735-16H KD735 VIII, MpuUI This study
KD117 KD735-16H II, MpuUI and MpuUII This study
KD129 KD735-16H III, MpuUI and MpuUIV This study
KD131 KD735-16H IV, MpuUIII and MpuUII This study
KD136 KD735-16H II, MpuUI and MpuUII This study
KD208 KD735-16H V, MpuUV and MpuUVI This study
KD231 KD735-16H VI, MpuUV and MpuUIV This study
KD255 KD735-16H VIII, MpuUI This study
KD262 KD735-16H VII, MpuUI and MpuUVIII This study
KD297 KD735-16H IV, MpuUIII and MpuUII This study
T3 KD736-16K V, MpuUV and MpuUVI This study
KD117-18 KD117 VIII, MpuUI This study

R-M Assays.

R-M properties of M. pulmonis strains were assessed by determining the plaquing efficiency of mycoplasma virus P1 on lawns of host cells as described (6, 9).

Gene Analysis.

Genomic DNA was isolated as described (5). For Southern blot analysis, HindIII-digested chromosomal DNA fragments were subjected to agarose gel electrophoresis, were transferred to MSI MagnaCharge nylon membrane (Fisher), and were hybridized by using normal (high) stringency conditions (8). The probe was a 362-bp PCR product derived from sequences located within the conserved hsdS domain illustrated in Fig. 1 by black shading and was obtained by PCR amplifying DNA from strain KD735–15 using primers 5′-CAATTCATATGACTTTAATGGTG-3′ and 5′-TGTTGAAGACTTTTTTGCTCAC-3′. The probe was radiolabeled with 32P by the random primer method by using Ready-To-Go DNA Labeling Beads (Pharmacia).

Figure 1.

Figure 1

Organization of the M. pulmonis hsd1 (Upper) and hsd2 (Lower) loci. To better illustrate features within the hsdS genes, the hsdM and hsdR genes were shortened and were not drawn to scale. Homologous regions within hsdS are shown by like shading. The sites labeled α, β, γ, and δ refer to vip DNA inversion sites. The sites labeled hrs1, hrs2, and hrs3 are additional sites for DNA inversion. The single promoter driving hsd transcription is labeled by P. The orientations of primer binding sites for PCR analysis are indicated by arrows.

The hsdS genes were isolated from subclones of KD735–16H by PCR (Fig. 1). Primer o.4 (5′-CATCAAGACTAGTGTTAAATTTTTGTAAC-3′) specifically binds to a 65-bp sequence located in the promoter region of hsd2 but absent altogether from hsd1. Therefore, primers o.4 and o.2 (5′-GTGCTTGAAAAGAAAAAAGAGAAGAGG-3′) were used to amplify specifically the transcribed hsdS gene of hsd2 from subclones having hsd2 oriented as shown in Fig. 1, and primers o.4 and o.3 (5′-GCGAATCAAATCTTTTACCC-3′) were used to amplify the transcribed hsdS gene from subclones having the locus in the inverted orientation. No primers were available for the specific PCR amplification of the expressed hsdS gene from hsd1 because hsd1 and hsd2 have identical promoter regions other than the 65-bp sequence that is unique to hsd2. Therefore, the hsd1-specific primer o.1 (5′-ATTTTTGAACAAATACCAGAGAC-3′) and primer o.3 were used to amplify the silent hsdS gene of hsd1 from subclones having hsd1 oriented as shown in Fig. 1. Primers o.1 and o.2 were used for subclones having hsd1 in the inverted orientation. For some experiments, the hsd2-specific primer o.5 (5′-ATCAAGGAATTTATTCTGTTATATC-3′) was paired with o.3 or o.2, depending on the orientation of the locus, to amplify the silent hsdS gene from hsd2.

PCR cycling conditions were as described (5), but the number of cycles was reduced to 25. PCR products were gel-purified and cloned by using the pGEM-T Easy vector (Promega). DNA sequencing was performed by automated fluorescent dye terminator methods at the Sequencing Core Facility, University of Alabama at Birmingham. Primers were obtained from the Oligonucleotide Synthesis Core Facility at the University of Alabama at Birmingham.

RESULTS

R-M Properties of M. pulmonis Strains.

Strain KD735–16 has been maintained in the laboratory for several years and has been shown both to restrict and modify mycoplasma virus P1 (6). The restriction enzyme produced in KD735–16 has been designated in the REBASE database as MpuUI (10). Southern hybridization analysis using probes specific for the hsdS genes demonstrated that KD735–16 is considerably heterogeneous, possessing subpopulations in which the hsd loci have undergone gene rearrangements (Fig. 2). Because stocks of KD735–16 that had been maintained in the laboratory by separate individuals may contain different cell subpopulations, the stocks were assigned new strain designations (e.g., KD735–16H and KD735–16K). Experiments were undertaken to determine whether some subpopulations possessed R-M enzymes with specificities different from MpuUI. To this end, 147 filter clones of KD735–16H were isolated.

Figure 2.

Figure 2

Southern blot analysis of HindIII-digested M. pulmonis chromosomal DNA hybridized with the hsdS-specific probe described in Materials and Methods. Strain abbreviations: 15, KD735–15; 16, KD735–16H; 17, KD117; 129, KD129; etc.

R-M properties were examined by assaying plaque-forming units of mycoplasma virus P1. KD735–15 is a strain of M. pulmonis that lacks detectable R-M activity and has hsd1 inverted in the chromosome relative to KD735–16 (6). Stocks of P1 virus prepared by infection of KD735–15 (designated P1•15) are susceptible to restriction by the MpuUI enzyme when assayed on lawns of KD735–16H. Of the 147 subclones of KD735–16H that were analyzed, 17 (12%) failed to restrict P1•15. The phenotype and the hsd1 orientation of these 17 subclones were identical to KD735–15. Because hsd1 inversions leading to the KD735–15 phenotype have been described, the current study focused on the 130 subclones of KD735–16H that retained R-M activity. Of these 130 subclones, 96 (65%) did not restrict the P1 virus that had been propagated on KD735–16H (P1•16). One of these subclones (KD255) was chosen for further study and was found to be indistinguishable (in terms of R-M properties and configurations of the hsd loci) from the majority cell population in KD735–16H. Southern hybridization analysis confirmed that KD255 is more homogeneous than KD735–16H (Fig. 2).

The subclones of KD735–16H that restricted both P1•15 and P1•16 were studied in detail because they likely possessed R-M activity with specificity different from MpuUI. Stocks of P1 virus propagated on each of these subclones were prepared and used to assess R-M properties as summarized in Table 2. Each subclone of KD735–16H was assigned to one of eight distinct restriction groups. Group I consisted of the 17 subclones that failed to restrict P1•15 and had restriction properties identical to KD735–15. Group II consisted of 26 subclones (20%) typified by KD117, having R-M properties similar to a stock of KD735–16 that had been maintained in the laboratory for several years and had been assigned the strain designation KD735–16K. Groups III through VII consisted of subclones that possess previously unobserved R-M activities. Members of each of these groups restricted stocks of P1 virus propagated on any of the other groups. Group VIII consisted of the 96 subclones typified by KD255 that restricted P1•15 but not P1•16. Group VIII subclones were unusual in that they restricted virus propagated on some subclones (groups I, IV, V, and VI) but not others. Southern hybridization analysis indicated that most restriction groups had a distinct hybridization profile, indicating that differences in R-M activity resulted from gene rearrangements (Fig. 2).

Table 2.

Summary of restriction and modification properties of M. pulmonis subclones

Infecting virus Stock Relative plaque-forming units on M. pulmonis strain*
Group I KD735-15 Group II KD117 KD136 Group III KD129 Group IV KD131 KD297 Group V KD208 T3 Group VI KD231 Group VII KD262 Group VIII KD255 KD117-18
 P1•15 1 2  ×  10−5 1  ×  10−4 3  ×  10−3 1  ×  10−3 2  ×  10−3 1  ×  10−4 2  ×  10−3
 P1•117/P1•136 1 1 4  ×  10−3 6  ×  10−4 7  ×  10−3 1  ×  10−3 3  ×  10−4 1
 P1•129 1 2  ×  10−3 1 8  ×  10−4 2  ×  10−2 2  ×  10−1 6  ×  10−3 1
 P1•131/P1•297 1 <10−4 <10−4 1 9  ×  10−3 2  ×  10−3 <10−5 1  ×  10−4
 P1•208/P1•T3 1 2  ×  10−5 8  ×  10−6 1  ×  10−3 1 8  ×  10−4 2  ×  10−5 2  ×  10−3
 P1•231 1 1  ×  10−6 3  ×  10−3 2  ×  10−3 3  ×  10−2 1 2  ×  10−5 2  ×  10−3
 P1•262 1 2  ×  10−3 2  ×  10−4 3  ×  10−3 1  ×  10−2 5  ×  10−3 1 1
 P1•255 1 4  ×  10−2 5  ×  10−3 3  ×  10−3 1  ×  10−3 5  ×  10−3 5  ×  10−3 1
*

Similar results were obtained in four independent experiments. 

Stocks of P1 virus are abbreviated as follows: P1•15, P1 virus grown on strain KD735-15; P1•117, P1 virus grown on strain KD117; etc. 

Sequence Analysis of hsdS Genes.

The hsd gene rearrangements identified in previous studies were site-specific DNA inversions that occurred at sites designated vip. Each hsd locus has four vip sites (two per hsdS gene) referred to as vipα, vipβ, vipδ, and vipγ (Fig. 1). Previously, four types of DNA inversions were identified, resulting from recombination between vipα and vipγ (α-γ inversion), vipα and vipδ (α-δ inversion), vipβ and vipγ (β-γ inversion), and vipβ and vipδ (β-δ inversion) (5). For representative members of each restriction group, the nucleotide sequence of one of the hsdS genes from each hsd locus was determined and used to deduce the particular type of DNA inversion that had occurred, if any, along with the sequence of the complementary hsdS gene. For each hsd locus, one hsdS gene is expressed (Fig. 1, left end) and the other is transcriptionally silent (Fig. 1, right end) (5). The expressed gene from hsd2 and the silent gene from hsd1 were PCR amplified as described in Materials and Methods. The nucleotide sequences of the PCR products derived from these two genes permitted the identification of the type of DNA inversion (α-γ, α-δ, β-γ, or β-δ) that had occurred at each locus. By using this strategy, the particular DNA inversions giving rise to subclones within restriction groups II, IV, V, VII, and VIII were identified as summarized in Fig. 3. These subclones had all undergone predictable DNA inversions at vip sites.

Figure 3.

Figure 3

Schematic diagrams of the hsdS genes associated with each restriction group. Shaded segments refer to regions of homology as shown in Fig. 1. Recombination sites (vip and hrs) and promoter (P) sites are as in Fig. 1.

Identification of New Site-Specific Gene Rearrangements in hsdS.

The PCR products derived from the hsd2 locus of group III (KD129) and group VI (KD208) subclones had unexpected sequences requiring further study. The hsd2 locus from these subclones was oriented such that the transcribed hsdS gene was PCR amplified by using primers o.4 and o.3. To document the configuration of the hsd2 locus in these subclones, the silent hsdS gene was also PCR amplified by using the hsd2-specific primer o.5 paired with primer o.2 (Fig. 1). Sequence comparison of the silent and expressed hsdS genes of KD129 and KD208 indicated that the hsd2 locus of both subclones had undergone the same gene rearrangement, which apparently consisted of a DNA inversion at a site other than vip. The inversion occurred between sites designated in Fig. 3 as hrs1 (hsd recombination site) and hrs2. There are three hrs sites per hsd locus, and these sites had been recognized as 20-bp direct repeats present at each end of a 99-bp region that is present only in the genes labeled in Fig. 1 as hsdS1A and hsdS2B (5, 6). Although the nucleotide sequences of the hrs sites appear distinct from the sequences of vip, the vipγ and hrs3 sites are adjacent in the chromosome with some overlap, as are vipα and hrs1 (Fig. 1). An alignment of the hsd2 vip and hrs sites is shown in Fig. 4.

Figure 4.

Figure 4

Alignment of the vip and hrs DNA inversion sites of the hsd2 locus. Sequences are from strains KD735–15 or KD129 with the abbreviated strain designation in parenthesis. Boxes are drawn around the 12-nt vip sequences and the 20-nt hrs sequences.

MpuU Enzymes and Nomenclature.

Nearly all of the hsdS genes expressed by subclones within each restriction group encode functional S subunits. The S subunits encoded by hsd1 are identified with the odd-numbered restriction enzymes MpuUI, MpuUIII, etc., and the subunits encoded by hsd2 are identified with the even-numbered enzymes MpuUII, MpuUIV, etc. (Fig. 3). The predicted HsdS proteins of group II subclones (e.g., KD117) are the S subunits of MpuUI and MpuUII. Group III subclones produce the enzymes MpuUI and MpuUIV. Groups II and III both produce MpuUI. Yet, each of these groups restricts P1 virus propagated on the other group. The different restriction properties exhibited by groups II and III are attributed to MpuUII and MpuUIV, indicating that both of these enzymes are functional but with DNA recognition sequence specificities distinct from one another and from MpuUI. Group IV subclones produce MpuUIII and MpuUII. Because group IV and group II subclones both produce MpuUII, the different restriction properties exhibited by these groups indicate that MpuUI and MpuUIII are functionally distinct from one another and from MpuUII. A comparison of groups V and VI, both of which produce MpuUV, indicates that MpuUVI is functional and distinct from MpuUV and MpuUIV. A comparison of groups III and VI indicates that MpuUV is functional and distinct from MpuUI and MpuUIV. A comparison of group VII to groups II and III indicates that MpuUVIII is functional and distinct from MpuUI, MpuUII, and MpuUIV. This latter comparison is of interest because the sequences of MpuUIV and MpuUVIII are identical but for the presence of the 33-aa region in MpuUIV that is encoded by the 99-nt region bounded by hrs2 and hrs3. From the available data, it is not possible to conclude that all MpuU enzymes are functionally distinct. It is possible that MpuUIII and MpuUV are functionally equivalent, and MpuUVI may be equivalent to MpuUII. The S subunits of these enzyme pairs differ only at their carboxyl-terminal regions with undetermined functional consequence.

Lack of R-M Activity in Group I Subclones.

Group I subclones are devoid of detectable R-M activity but should produce the S subunits of MpuUV and MpuUII [compare the expressed hsdS genes illustrated in Fig. 1 (left end of the hsd loci) with the MpuUV and MpuUII S subunits shown in Fig. 3]. Because MpuUV and MpuUII are functional, as discussed above, one might predict that KD735–15 should exhibit R-M activity. However, only one promoter (see Fig. 1) has been identified in each hsd locus (5). The orientations of hsd1 and hsd2 in the KD735–15 chromosome are such that hsdR and hsdM would not be transcribed from this promoter. Therefore, it is plausible that KD735–15 lacks R-M activity because R and M subunits are not produced.

Limited R-M Activity of Group VIII Subclones.

Group VIII subclones (Table 2) are unusual in that they restrict viruses propagated on groups I, IV, V, and VI but not on groups II, III, and VII. The hsdS nucleotide sequences of group VIII and group II subclones were identical, and both groups are predicted by sequence analysis to produce MpuUI and MpuUII. The subclones that modify P1 virus, rendering it resistant to restriction by group VIII subclones, all produce MpuUI whereas subclones lacking MpuUI produce P1 virus that is restricted by group VIII. This implies that group VIII subclones produce functional MpuUI. Also, P1 virus propagated on subclones producing MpuUII but not MpuUI (group IV) are restricted by group VIII subclones, further indicating that group VIII subclones produce functional MpuUI. P1 virus propagated on KD255 (P1•255) is not modified as predicted from sequence analysis because P1•255 was restricted by group II subclones. Because group VIII subclones evidently produce MpuUI, we conclude that group VIII subclones fail to produce MpuUII even though the hsdS sequence data predict otherwise.

Analysis of Additional Subclones Derived from Groups II and VIII.

KD735–16H was not the only stock of M. pulmonis possessing subpopulations with differing R-M properties. For example, strain T3 is a group V subclone isolated from KD735–16K (group II). Strain KD117–18 (group VIII) was 1 of 20 subclones of KD117 (group II) that were analyzed. Therefore, cultures of group II subclones probably contain a significant group VIII subpopulation, implying that group VIII subclones do not arise by rare mutational events.

DISCUSSION

Phase-variable R-M enzymes probably are not limited to M. pulmonis and may be widespread. Helicobacter pylori, Neisseria gonorrhoeae, and Haemophilus influenzae possess genes that are predicted to encode phase-variable R-M enzymes (11, 12). In these Gram-negative bacterial systems, phase variation would result from frameshift errors occurring during DNA replication by slipped strand mispairing within a reiterated domain. The genes encoding the phase-variable R-M enzymes of M. pulmonis and other bacteria are clearly not selfish.

What protects the chromosome from endonucleolytic attack when restriction activity is induced? The transfer of hsd genes from one strain of Escherichia coli to another is possible because R-M activity is regulated such that the chromosome of recipient cells is modified before restriction activity is produced (13). The hsdR gene of E. coli is transcribed from a separate promoter than are hsdM and hsdS. In contrast, hsdR precedes hsdM in the hsd loci of M. pulmonis, suggesting that MTase activity is not induced before restriction activity. For bacterial species possessing natural transformation systems, the induction of phase-variable restriction activity may be part of an autolysis process that releases DNA into the environment for uptake by other cells. This is plausible but not generally applicable to mycoplasmas that lack natural transformation (14). It has been noted that KD735–15 (group I, lacking R-M activity) is more stable with respect to hsd inversions than is KD735–16 (6). DNA from KD735–16 is significantly more heterogeneous than DNA from KD735–15 (Fig. 2), supporting this observation. This is in spite of KD735–16 having been propagated less frequently in our laboratory than has KD735–15. Perhaps the apparent stability of KD735–15 results from the frequent death of progeny that have undergone induction of restriction activity via hsd inversion. Such a self-destructive system would not be maintained in a cell population unless phase-variable R-M activity had an essential function.

R-M activity in M. pulmonis is complex and may be modulated by other factors in addition to hsd inversions. As described above, MpuUII is produced in group II but not in group VIII subclones. The nucleotide sequences of the coding regions and promoter regions of the hsdS genes were identical for groups II and VIII, and the reason for the failure of group VIII to produce MpuUII is undetermined. A 65-bp sequence is present in the 5′ untranslated region of the hsdS mRNA from hsd2 but is absent in hsd1 (5). Therefore, hsd2 gene expression may be regulated differently from hsd1, and an undetermined factor may repress MpuUII production in group VIII subclones.

The heterogeneity of M. pulmonis cultures in regards to R-M activity must result in virus stocks with heterogeneity in base modifications. One result of the heterogeneity is that the size of plaques that are obtained varies with the restriction conditions. Uniformly large plaques are obtained when plaque-forming units are assayed on lawns of group I subclones. In contrast, when plaque-forming units are assayed on restrictive strains, a mixture of large and small plaques is obtained. We interpret small plaques as arising from the initial infection of a cell that was a member of a different restriction group than the cell majority within the lawn. This initially infected cell would release progeny viruses that are restricted by neighboring cells as the plaque develops, limiting its size. The percentage of group I cells on the lawn particularly would influence the ratio of large to small plaques because these cells are infected efficiently but release unmodified progeny viruses that are restricted by cells from all other restriction groups.

What is the function of the phase-variable R-M systems of M. pulmonis? Mycoplasmas are thought not to possess natural transformation and conjugation systems, and R-M systems therefore would not be required as barriers to these types of gene transfer. The hsd loci may be an exotic system for protection against phage infection, but the presence of a significant cell subpopulation lacking R-M systems suggests that the Hsd enzymes of M. pulmonis may be an ineffectual barrier. Infection of the restriction-negative subpopulation would release high levels of progeny phage, numerically overwhelming the restriction enzymes produced in other cells in the population. The potential for the loss of R-M activity on DNA inversion suggests that M. pulmonis may have evolved mechanisms to maintain R-M-negative subpopulations. For example, immediately upstream of the hsdS1A gene illustrated in Fig. 1 is a transcription terminator signaling the end of the adjacent polC gene (6, 15). Were this transcription terminator deleted during the evolution of the hsd loci, transcription could proceed through polC and into hsd1. This would permit transcription of hsdR1 and hsdM1 regardless of the orientation of hsd1 in the chromosome. Thus, all cells in the population would possess R-M activity, and phage protection would be maximal. The polC transcription terminator might have been maintained during evolution because it is sometimes advantageous to have subpopulations lacking R-M activity.

In M. pulmonis, there is an apparent association between phase variation in R-M activity and antigenic variation. The vsa (variable surface antigen) locus is a large, DNA inversion system that encodes a family of highly reiterated, phase-variable surface proteins known as the V-1 antigens (16, 17). Variation in the V-1 proteins has been associated with disease pathogenesis, and V-1 may facilitate the attachment of M. pulmonis to host tissue (18, 19). Most vsa genes are transcriptionally silent and lack a promoter, ribosome binding site, and the first 700 nucleotides of the vsa coding region. Only a single vsa expression site has been identified thus far, and site-specific DNA inversions regulate phase-variable gene expression by recombining silent genes with this expression site. Pedigree analysis of M. pulmonis subclones has indicated a correlation between vsa inversions and hsd inversions (16, 20). Perhaps inversions at these loci are regulated coordinately. Alternatively, the induction of R-M activity associated with hsd inversion may stimulate vsa inversions. Stimulation of vsa inversions could occur as a result of the introduction of double-stranded DNA breaks in the chromosome by the induced restriction activity or, alternatively, the induced modification activity may serve to regulate expression of the gene encoding the site-specific recombinase responsible for catalyzing vsa inversions. Regardless of the explanation for the association between DNA inversions at hsd and vsa, it seems likely that the hsd loci have some role beyond phage restriction, possibly in the pathogenic process.

Acknowledgments

This work was supported by U.S. Public Health Service grant GM1126 from the National Institutes of Health. The University of Alabama at Birmingham Sequencing Core Facility is a service provided by the Center for AIDS Research (National Institutes of Health Grant 5 P30 AI27767). The University of Alabama at Birmingham oligonucleotide synthesis core facility is supported by National Institutes of Health Grant 5 P50 CA13148.

ABBREVIATIONS

R-M

restriction and modification

MTase

methyltransferase

Footnotes

This paper was submitted directly (Track II) to the Proceedings Office.

Data deposition: The sequences reported in this paper have been deposited in the GenBank database (accession nos. AF076984AF076990).

References

  • 1.Bickle T A, Kruger D H. Microbiol Rev. 1993;57:434–450. doi: 10.1128/mr.57.2.434-450.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Naito T, Kusano K, Kobayashi I. Science. 1995;267:897–899. doi: 10.1126/science.7846533. [DOI] [PubMed] [Google Scholar]
  • 3.Nakayama Y, Kobayashi I. Proc Natl Acad Sci USA. 1998;95:6442–6447. doi: 10.1073/pnas.95.11.6442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.O’Neill M, Chen A, Murray N. Proc Natl Acad Sci USA. 1997;94:14596–14601. doi: 10.1073/pnas.94.26.14596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Sitaraman R, Dybvig K. Mol Microbiol. 1997;26:109–120. doi: 10.1046/j.1365-2958.1997.5571938.x. [DOI] [PubMed] [Google Scholar]
  • 6.Dybvig K, Yu H. Mol Microbiol. 1994;12:547–560. doi: 10.1111/j.1365-2958.1994.tb01041.x. [DOI] [PubMed] [Google Scholar]
  • 7.Dybvig K, Cassell G H. Science. 1987;235:1392–1394. doi: 10.1126/science.3029869. [DOI] [PubMed] [Google Scholar]
  • 8.Bhugra B, Dybvig K. Mol Microbiol. 1992;6:1149–1154. doi: 10.1111/j.1365-2958.1992.tb01553.x. [DOI] [PubMed] [Google Scholar]
  • 9.Dybvig K, Liss A, Alderete J, Cole R M, Cassell G H. Isr J Med Sci. 1987;23:418–422. [PubMed] [Google Scholar]
  • 10.Roberts R J, Macelis D. Nucleic Acids Res. 1994;22:3628–3639. doi: 10.1093/nar/22.17.3628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Saunders N J, Peden J F, Hood D W, Moxon E R. Mol Microbiol. 1998;27:1091–1098. doi: 10.1046/j.1365-2958.1998.00768.x. [DOI] [PubMed] [Google Scholar]
  • 12.Hood D W, Deadman M E, Jennings M P, Bisercic M, Fleischmann R D, Venter J C, Moxon E R. Proc Natl Acad Sci USA. 1996;93:11121–11125. doi: 10.1073/pnas.93.20.11121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kulik E M, Bickle T A. J Mol Biol. 1996;264:891–906. doi: 10.1006/jmbi.1996.0685. [DOI] [PubMed] [Google Scholar]
  • 14.Dybvig K, Voelker L L. Annu Rev Microbiol. 1996;50:25–57. doi: 10.1146/annurev.micro.50.1.25. [DOI] [PubMed] [Google Scholar]
  • 15.Barnes M H, Tarantino P M, Jr, Spacclapoli P, Brown N C, Yu H, Dybvig K. Mol Microbiol. 1994;13:843–854. doi: 10.1111/j.1365-2958.1994.tb00476.x. [DOI] [PubMed] [Google Scholar]
  • 16.Bhugra B, Voelker L L, Zou N, Yu H, Dybvig K. Mol Microbiol. 1995;18:703–714. doi: 10.1111/j.1365-2958.1995.mmi_18040703.x. [DOI] [PubMed] [Google Scholar]
  • 17.Simmons W L, Zuhua C, Glass J I, Simecka J W, Cassell G H, Watson H L. Infect Immun. 1996;64:472–479. doi: 10.1128/iai.64.2.472-479.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Talkington D F, Fallon M T, Watson H L, Thorp R K, Cassell G H. Microb Pathog. 1989;7:429–436. doi: 10.1016/0882-4010(89)90023-5. [DOI] [PubMed] [Google Scholar]
  • 19.Watson, H. L., Zheng, X. & Cassell, G. H. (1993) Clin. Infect. Dis. 17, Suppl. 1, S183–S186. [DOI] [PubMed]
  • 20.Dybvig K. Mol Microbiol. 1993;10:465–471. doi: 10.1111/j.1365-2958.1993.tb00919.x. [DOI] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES