Behavior of restriction-modification systems as selfish mobile elements and their impact on genome evolution

Restriction–modification (RM) systems are composed of genes that encode a restriction enzyme and a modification methylase. RM systems sometimes behave as discrete units of life, like viruses and transposons. RM complexes attack invading DNA that has not been properly modified and thus may serve as a tool of defense for bacterial cells. However, any threat to their maintenance, such as a challenge by a competing genetic element (an incompatible plasmid or an allelic homologous stretch of DNA, for example) can lead to cell death through restriction breakage in the genome. This post-segregational or post-disturbance cell killing may provide the RM complexes (and any DNA linked with them) with a competitive advantage. There is evidence that they have undergone extensive horizontal transfer between genomes, as inferred from their sequence homology, codon usage bias and GC content difference. They are often linked with mobile genetic elements such as plasmids, viruses, transposons and integrons. The comparison of closely related bacterial genomes also suggests that, at times, RM genes themselves behave as mobile elements and cause genome rearrangements. Indeed some bacterial genomes that survived post-disturbance attack by an RM gene complex in the laboratory have experienced genome rearrangements. The avoidance of some restriction sites by bacterial genomes may result from selection by past restriction attacks. Both bacteriophages and bacteria also appear to use homologous recombination to cope with the selfish behavior of RM systems. RM systems compete with each other in several ways. One is competition for recognition sequences in post-segregational killing. Another is super-infection exclusion, that is, the killing of the cell carrying an RM system when it is infected with another RM system of the same regulatory specificity but of a different sequence specificity. The capacity of RM systems to act as selfish, mobile genetic elements may underlie the structure and function of RM enzymes.     

Cellular responses to postsegregational killing by restriction-modification genes

Plasmids that carry one of several type II restriction modification gene complexes are known to show increased stability. The underlying mechanism was proposed to be the lethal attack by restriction enzyme at chromosomal recognition sites in cells that had lost the restriction modification gene complex. In order to examine bacterial responses to this postsegregational cell killing, we analyzed the cellular processes following loss of the EcoRI restriction modification gene complex carried by a temperature-sensitive plasmid in an Escherichia coli strain that is wild type with respect to DNA repair. A shift to the nonpermissive temperature blocked plasmid replication, reduced the increase in viable cell counts and resulted in loss of cell viability. Many cells formed long filaments, some of which were multinucleated and others anucleated. In a mutant defective in RecBCD exonuclease/recombinase, these cell death symptoms were more severe and cleaved chromosomes accumulated. Growth inhibition was also more severe in recA, ruvAB, ruvC, recG, and recN mutants. The cells induced the SOS response in a RecBC-dependent manner. These observations strongly suggest that bacterial cells die as a result of chromosome cleavage after loss of a restriction modification gene complex and that the bacterial RecBCD/RecA machinery helps the cells to survive, at least to some extent, by repairing the cleaved chromosomes. These and previous results have led us to hypothesize that the RecBCD/Chi/RecA system serves to destroy restricted "nonself" DNA and repair restricted "self" DNA.     

The plasmid carrying the temperature-sensitive mutation in the DNa-methylase gene of the PStI system: effect on host cells at nonpermissive temperature]

Temperature-sensitive (ts) derivatives of plasmid pRMP1, the derivative of PBR322 containing restriction and modification (RM) genes of the PstI system, were obtained using hydroxylamine mutagenesis. One of the isolated plasmids responsible for the inhibition of Escherichia coli cell growth at 42 degrees C, pRMPts, was analyzed in this work. Cells of Rec+ strains carrying this plasmid were unable to divide at 42 degrees C and formed long non-septated filaments that died upon prolonged cultivation. Cells of the RecA- strains carrying pRMPts did not form filaments at 42 degrees C and rapidly disappeared. On agar media with or without ampicillin, Rec+ and RecA- strains with this plasmid formed colonies of temperature-resistant (tr) derivatives with frequencies ranging from 1.5 x 10(-4) to 4 x 10(-6) in independent clones. The structure of plasmids from cells of tr-derivatives of Rec+ and RecA- strains carrying plasmid pRMPts was analyzed by the set of restriction enzymes. Reversions to the temperature-resistant phenotype were shown to result from the following events: (1) the insertional inactivation of the PstI restriction enzyme gene in pRMPts (the insertion of the IS1 element); (2) deletions in plasmid DNA fragments that partially or completely cover the restriction enzyme gene; (3) point mutations; and (4) others. The effect of the chromosomal sulA mutation on the maintenance of the ts-plasmid in bacterial cells was studied at 42 degrees C. High efficiency loss of the plasmid was detected in pRMPts-carrying Rec+ cells with the sulA::Tn5 mutation grown in liquid and solid nutrient media at this temperature. Under similar conditions, plasmid loss was not detected in SulA+ cells. On the basis of the data obtained, it is concluded that the ts-mutation is located in the DNA-methylase gene of plasmid pRMPts. Mutant DNA methylase was unable to methylate all sites in the chromosomal DNA at 42 degrees C. Some of the unmethylated sites can be digested with the PstI enzyme, which leads to the induction of SOS response in Rec+ cells or to total mortality in cells with the recA phenotype.     

Stability of EcoRI restriction-modification enzymes in vivo differentiates the EcoRI restriction-modification system from other postsegregational cell killing systems

Certain type II restriction modification gene systems can kill host cells when these gene systems are eliminated from the host cells. Such ability to cause postsegregational killing of host cells is the feature of bacterial addiction modules, each of which consists of toxin and antitoxin genes. With these addiction modules, the differential stability of toxin and antitoxin molecules in cells plays an essential role in the execution of postsegregational killing. We here examined in vivo stability of the EcoRI restriction enzyme (toxin) and modification enzyme (antitoxin), the gene system of which has previously been shown to cause postsegregational host killing in Escherichia coli. Using two different methods, namely, quantitative Western blot analysis and pulse-chase immunoprecipitation analysis, we demonstrated that both the EcoRI restriction enzyme and modification enzyme are as stable as bulk cellular proteins and that there is no marked difference in their stability. The numbers of EcoRI restriction and modification enzyme molecules present in a host cell during the steady-state growth were estimated. We monitored changes in cellular levels of the EcoRI restriction and modification enzymes during the postsegregational killing. Results from these analyses together suggest that the EcoRI gene system does not rely on differential stability between the toxin and the antitoxin molecules for execution of postsegregational cell killing. Our results provide insights into the mechanism of postsegregational killing by restriction-modification systems, which seems to be distinct from mechanisms of postsegregational killing by other bacterial addiction modules.     

A functional bacteria-derived restriction modification system in the mitochondrion of a heterotrophic protist

The overarching trend in mitochondrial genome evolution is functional streamlining coupled with gene loss. Therefore, gene acquisition by mitochondria is considered to be exceedingly rare. Selfish elements in the form of self-splicing introns occur in many organellar genomes, but the wider diversity of selfish elements, and how they persist in the DNA of organelles, has not been explored. In the mitochondrial genome of a marine heterotrophic katablepharid protist, we identify a functional type II restriction modification (RM) system originating from a horizontal gene transfer (HGT) event involving bacteria related to flavobacteria. This RM system consists of an HpaII-like endonuclease and a cognate cytosine methyltransferase (CM). We demonstrate that these proteins are functional by heterologous expression in both bacterial and eukaryotic cells. These results suggest that a mitochondrion-encoded RM system can function as a toxin–antitoxin selfish element, and that such elements could be co-opted by eukaryotic genomes to drive biased organellar inheritance.

This study reveals that a functional type II restriction modification system of flavobacterial ancestry has been horizontally transferred into the mitochondrion of a marine protist and is capable of encoding potent function, perhaps allowing it to play a role in inter-organellar warfare or protection against further integration of foreign DNA.     

Maintenance forced by a restriction-modification system can be modulated by a region in its modification enzyme not essential for methyltransferase activity

Several type II restriction-modification gene complexes can force their maintenance on their host bacteria by killing cells that have lost them in a process called postsegregational killing or genetic addiction. It is likely to proceed by dilution of the modification enzyme molecule during rounds of cell division following the gene loss, which exposes unmethylated recognition sites on the newly replicated chromosomes to lethal attack by the remaining restriction enzyme molecules. This process is in apparent contrast to the process of the classical types of postsegregational killing systems, in which built-in metabolic instability of the antitoxin allows release of the toxin for lethal action after the gene loss. In the present study, we characterize a mutant form of the EcoRII gene complex that shows stronger capacity in such maintenance. This phenotype is conferred by an L80P amino acid substitution (T239C nucleotide substitution) mutation in the modification enzyme. This mutant enzyme showed decreased DNA methyltransferase activity at a higher temperature in vivo and in vitro than the nonmutated enzyme, although a deletion mutant lacking the N-terminal 83 amino acids did not lose activity at either of the temperatures tested. Under a condition of inhibited protein synthesis, the activity of the L80P mutant was completely lost at a high temperature. In parallel, the L80P mutant protein disappeared more rapidly than the wild-type protein. These results demonstrate that the capability of a restriction-modification system in forcing maintenance on its host can be modulated by a region of its antitoxin, the modification enzyme, as in the classical postsegregational killing systems.     

Methyl-cytosine specific restriction in Escherichia coli K-12

Experiments on transformation of Escherichia coli K-12 cells by plasmids carrying RM systems with different recognition sites containing 5-methylcytosine have shown that the gene mcrB determines the function of restriction. The data obtained made it possible to believe that E. coli possesses no restriction system recognizing specifically cytosine methylated in position 4.     

Evolution of sequence recognition by restriction-modification enzymes: selective pressure for specificity decrease

Several type II restriction-modification (RM) gene complexes kill host bacterial cells that have lost them, through attack on the chromosomal recognition sites of these cells. Two RM gene complexes recognizing the same sequence cannot simultaneously enjoy such stabilization through postsegregational host killing, because one will defend chromosomal sites from attack by the other. In the present work, we analyzed intrahost competition between two RM gene complexes when the recognition sequence of one was included in that of the other. When the EcoRII gene complex, recognizing 5'-CCWGG (W = A, T), is lost from the host, the SsoII gene complex, which recognizes 5'-CCNGG (N = A, T, G, C), will prevent host death by protecting CCWGG sites on the chromosome. However, when the SsoII (CCNGG) gene complex is lost, the EcoRII (CCWGG) gene complex will be unable to prevent host death through attack by SsoII on 5'-CCSGG (S = C, G) sites. These predictions were verified in our experiments, in which we analyzed plasmid maintenance, cell growth, cell shape, and chromosomal DNA. Our results demonstrate the presence of selective pressure for decrease in the specificity of recognition sequence of RM systems in the absence of invading DNA.     

DNA restriction dependent on two recognition sites: activities of the SfiI restriction-modification system in Escherichia coli

In contrast to many type II restriction enzymes, dimeric proteins that cleave DNA at individual recognition sites 4-6 bp long, the SfiI endonuclease is a tetrameric protein that binds to two copies of an elongated sequence before cutting the DNA at both sites. The mode of action of the SfiI endonuclease thus seems more appropriate for DNA rearrangements than for restriction. To elucidate its biological function, strains of Escherichia coli expressing the SfiI restriction-modification system were transformed with plasmids carrying SfiI sites. The SfiI system often failed to restrict the survival of a plasmid with one SfiI site, but plasmids with two or more sites were restricted efficiently. Plasmids containing methylated SfI sites were not restricted. No rearrangements of the plasmids carrying SfiI sites were detected among the transformants. Hence, provided the target DNA contains at least two recognition sites, SfiI displays all of the hallmarks of a restriction-modification system as opposed to a recombination system in E. coli cells. The properties of the system in vivo match those of the enzyme in vitro. For both restriction in vivo and DNA cleavage in vitro, SfiI operates best with two recognition sites on the same DNA.     

Multiplication of a restriction-modification gene complex

Previous works have suggested that some gene complexes encoding a restriction (R) enzyme and a cognate modification (M) enzyme may behave as selfish mobile genetic elements. RM gene complexes, which destroy 'non-self' elements marked by the absence of proper methylation, are often associated with mobile genetic elements and are involved in various genome rearrangements. Here, we found amplification of a restriction-modification gene complex. BamHI gene complex inserted into the Bacillus chromosome showed resistance to replacement by a homologous stretch of DNA. Some cells became transformed with the donor without losing BamHI. In most of these transformants, multiple copies of BamHI and the donor allele were arranged as tandem repeats. When a clone carrying one copy of each allele was propagated, extensive amplification of BamHI and the donor unit was observed in a manner dependent on restriction enzyme gene. This suggests that restriction cutting of the genome participates in the amplification. Visualization by fluorescent in situ hybridization revealed that the amplification occurred in single cells in a burst-like fashion that is reminiscent of induction of provirus replication. The multiplication ability in a bacterium with natural capacity for DNA release, uptake and transformation will be discussed in relation to spreading of RM gene -complexes.     

Sequence of the 68,869 bp IncP-1alpha plasmid pTB11 from a waste-water treatment plant reveals a highly conserved backbone, a Tn402-like integron and other transposable elements

To analyse the significance of conjugative broad-host-range IncP-1alpha plasmids for the spread of antibiotic resistance determinants in waste-water treatment plants we isolated and characterised five different IncP-1alpha plasmids from bacteria of activated sludge and the final effluents of a municipal waste-water treatment plant. These plasmids mediate resistance to ampicillin, cefaclor, cefuroxime, gentamicin, kanamycin, spectinomycin, streptomycin, tetracycline, tobramycin, and trimethoprim. The complete 68,869 bp DNA-sequence of the IncP-1alpha plasmid pTB11 was determined. The pTB11 backbone modules for replication (Rep), mating pair formation (Trb), multimer resolution (Mrs), post-segregational killing (Psk), conjugative DNA-transfer (Tra), plasmid control (Ctl), and stable maintenance and inheritance (KilA, KilE, and KilC) are highly conserved as compared to the 'Birmingham' IncP-1alpha plasmids. In contrast to the 'Birmingham' plasmids pTB11 carries an insert of a Tn402-derivative integrating a class 1 integron in the intergenic region between the multimer resolution operon parCBA and the post-segregational killing operon parDE. The integron comprises the resistance gene cassettes oxa2 (beta-lactamase), aacA4 (aminoglycoside-6'N-acetyltransferase), and aadA1 (aminoglycoside-3'-adenylyltransferase) and a complete tniABQR transposition module. Integron-specific sequences were also identified on other IncP-1alpha plasmids analysed in this work. In contrast to the 'Birmingham' plasmids the pTB11 tetracycline resistance module carries a pecM- and a pncA-like gene downstream of the tetracycline resistance gene tetA and contains an insertion of the new insertion sequence element ISTB11. The transposable elements IS21 and Tn1 which disrupted, respectively, orf7 and klcB on the 'Birmingham' plasmids are not present on pTB11. Identification of IncP-1alpha plasmids in bacteria of the waste-water treatment plant's final effluents indicates that bacteria carrying these kind of plasmids are released into the environment.     

The coevolution of toxin and antitoxin genes drives the dynamics of bacterial addiction complexes and intragenomic conflict

Bacterial genomes commonly contain 'addiction' gene complexes that code for both a toxin and a corresponding antitoxin. As long as both genes are expressed, cells carrying the complex can remain healthy. However, loss of the complex (including segregational loss in daughter cells) can entail death of the cell. We develop a theoretical model to explore a number of evolutionary puzzles posed by toxin-antitoxin (TA) population biology. We first extend earlier results demonstrating that TA complexes can spread on plasmids, as an adaptation to plasmid competition in spatially structured environments, and highlight the role of kin selection. We then considered the emergence of TA complexes on plasmids from previously unlinked toxin and antitoxin genes. We find that one of these traits must offer at least initially a direct advantage in some but not all environments encountered by the evolving plasmid population. Finally, our study predicts non-transitive 'rock-paper-scissors' dynamics to be a feature of intragenomic conflict mediated by TA complexes. Intragenomic conflict could be sufficient to select deleterious genes on chromosomes and helps to explain the previously perplexing observation that many TA genes are found on bacterial chromosomes.     

Comparison of ccd of F, parDE of RP4, and parD of R1 using a novel conditional replication control system of plasmid R1

A number of plasmid-encoded gene systems are thought to stabilize plasmids by killing plasmid-free cells (also termed post-segregational killing or plasmid addiction). Here we analyse the mechanisms of plasmid stabilization by ccd of F, parDE of RP4 and parD of R1, and compare them to hok/sok of R1. To induce synchronous plasmid loss we constructed a novel plasmid replication-arrest system, which possesses the advantage that plasmid replication can be completely arrested by the addition of IPTG, a non-metabolizable inducer. Using isogenic plasmid constructions we have found, for the first time, consistent correlation between the effect on steady-state loss rates and the effect on cell proliferation in the plasmid replication-arrest assay for all three systems. The parDE system had the most pronounced effect both on plasmid stabilization and on plasmid retention after replication arrest. In contrast, ccd and parD both exhibited weaker effects than anticipated from previously published results. Thus, our results indicate that the function and efficiencies of some of the systems should be reconsidered. Our results are consistent with the previously postulated hypothesis that ccd and parDE act by killing plasmid-free segregants, whereas parD seems to act by inhibiting cell division of plasmid-free segregants.     

New restriction enzymes discovered from Escherichia coli clinical strains using a plasmid transformation method

The presence of restriction enzymes in bacterial cells has been predicted by either classical phage restriction-modification (R-M) tests, direct in vitro enzyme assays or more recently from bacterial genome sequence analysis. We have applied phage R-M test principles to the transformation of plasmid DNA and established a plasmid R-M test. To validate this test, six plasmids that contain BamHI fragments of phage lambda DNA were constructed and transformed into Escherichia coli strains containing known R-M systems including: type I (EcoBI, EcoAI, Eco124I), type II (HindIII) and type III (EcoP1I). Plasmid DNA with a single recognition site showed a reduction of relative efficiency of transformation (EOT = 10(-1)-10(-2)). When multiple recognition sites were present, greater reductions in EOT values were observed. Once established in the cell, the plasmids were subjected to modification (EOT = 1.0). We applied this test to screen E.coli clinical strains and detected the presence of restriction enzymes in 93% (14/15) of cells. Using additional subclones and the computer program, RM Search, we identified four new restriction enzymes, Eco377I, Eco585I, Eco646I and Eco777I, along with their recognition sequences, GGA(8N)ATGC, GCC(6N)TGCG, CCA(7N)CTTC, and GGA(6N)TATC, respectively. Eco1158I, an isoschizomer of EcoBI, was also found in this study.     

Occurrence of small Hsd plasmids in Salmonella typhi, Shigella boydii, and Escherichia coli.

The natural occurrence of small Hsd (host specificity for DNA) plasmids was demonstrated in restriction endonuclease-producing strains of Salmonella typhi, Shigella boydii, and Escherichia coli. The five Hsd plasmids isolated were between 5.0 and 12.2 kilobases long. The copy number of all the Hsd plasmids was high (more than 10 copies per cell). Introduction of these small plasmids into E. coli strain 0 drastically lowered the efficiency of plating of the lambda.0 phages (the efficiency of plating was less than 5 X 10(-5) PFU-1). High restriction endonuclease activities were detected in the Hsd plasmid-positive strains because of the elevated copy numbers of the hsdR+ gene. The advantages of using E. coli strains containing the small Hsd plasmids for purification of type II restriction endonucleases are discussed.     

Effect of Bacillus subtilis BsuM restriction-modification on plasmid transfer by polyethylene glycol-induced protoplast fusion

Polyethylene glycol (PEG)-induced cell fusion is a promising method to transfer larger DNA from one cell to another than conventional genetic DNA transfer systems. The laboratory strain Bacillus subtilis 168 contains a restriction (R) and modification (M) system, BsuM, which recognizes the sequence 5'-CTCGAG-3'. To study whether the BsuM system affects DNA transfer by the PEG-induced cell fusion between R(+)M(+) and R(-)M(-) strains, we examined transfer of plasmids pHV33 and pLS32neo carrying no and eight BsuM sites, respectively. It was shown that although the transfer of pLS32neo but not pHV33 from the R(-)M(-) to R(+)M(+) cells was severely restricted, significant levels of transfer of both plasmids from the R(+)M(+) to R(-)M(-) cells were observed. The latter result shows that the chromosomal DNA in the R(-)M(-) cell used as the recipient partially survived restriction from the donor R(+)M(+) cell, indicating that the BsuM R(-)M(-) strain is useful as a host for accepting DNA from cells carrying a restriction system(s). Two such examples were manifested for plasmid transfer from Bacillus circulans and Bacillus stearothermophilus strains to a BsuM-deficient mutant, B. subtilis RM125.