![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Journal of Bacteriology, March 2003, p. 2022-2025, Vol. 185, No. 6
0021-9193/03/$08.00+0 DOI: 10.1128/JB.185.6.2022-2025.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Genetics,1 Department of Biochemistry, University of Leicester, Leicester LE1 7RH, United Kingdom2
Received 23 September 2002/ Accepted 16 December 2002
 |
ABSTRACT |
|---|
 Top Abstract TextReferences |
|---|
|
TEXT |
|---|
|
TopAbstractTextReferences |
|---|
Type I R-M systems are prevalent and diverse in their specificities, as indicated by the presence of candidate genes in approximately half of the microbial genomes that have been fully sequenced and by the diversity of target sequences recognized by enzymes from closely related bacteria (12, 14). One ecological role of restriction systems is thought to be the destruction of infecting foreign DNA identified by the lack of cognate modification. Some bacteriophages and plasmids carry genes that alleviate the restriction of DNA by type I systems (16). The prototype of the plasmid antirestriction genes is the ardA locus of the IncI1 plasmid ColIbP-9 (8, 17). The ColIb ardA gene product functions in conjugation to alleviate the restriction of the transferring plasmid. Evidence for such a role includes the effect of mutation of ardA and coregulation of its expression with the plasmid transfer system (17); the location of ardA genes in the leading region, defined as the first portion of the plasmid to enter the recipient cell (7); and the transient induction of ardA in the conjugatively infected cell by a process thought to involve transcription of the transferring single strand of DNA (1, 4).
An appealingly simple hypothesis is that the ArdA system transiently blocks type I restriction systems in the conjugatively infected cell while allowing the immigrant plasmid to acquire protective modification as a prelude to its establishment as a symbiotic element. The hypothesis is inconsistent with reports that under conditions causing overproduction of the protein, ColIb ArdA inhibited the in vivo modification activity of EcoKI in addition to restriction (8, 17). Conserved ardA genes have been found in representatives of four other incompatibility groups (B, FV, K, and N) of enterobacterial plasmids (5, 7). We have reinvestigated the effects of ArdA on EcoKI by using the protein encoded by the IncB plasmid R16. ColIb and R16 are closely related plasmids, and the latter specifies an ArdA homologue that is predicted to be 93% identical at the amino acid sequence level with the ColIb equivalent and to have a similarly high net negative charge (7). Here we report that R16 ArdA selectively inhibited the restriction activity of EcoKI, leaving significant levels of methyltransferase activity under conditions in which restriction was almost completely prevented.
Does R16 ArdA inhibit the restriction and modification activities of EcoKI?
The assay was similar to that used elsewhere (8, 17) and involved virulent (vir) and clear-plaque (cI) mutants of phage 
to probe the R-M activities of EcoKI in strains containing or lacking R16 ArdA protein. The cI ral18 strain provided by N. E. Murray was also used; the ral+ locus is an early gene that increases the efficiency of EcoKI as a de novo methyltransferase to promote efficient modification of DNA generated by the replication of infecting DNA molecules escaping restriction (13). ArdA production was determined by a pUC18 recombinant containing R16 ardA+ on a 2.2-kb PstI fragment (pLG2051) (7). The maintenance stability of the recombinant plasmid was maximized by adding methicillin (50 µg ml-1) to ampicillin (50 µg ml-1)-supplemented liquid media. Methicillin does not traverse the outer membrane of Escherichia coli and binds extracellular ß-lactamase in an inactive complex.
E. coli K-12 strains included 5K (rK- mK+), C600 (rK+ mK+) (17), and a homogenic rK- mK+ derivative, NM816, with a point mutation in hsdR, provided by N. E. Murray. Strain NM710 lacks the defective prophage called Rac: this chromosomal locus, present in C600 and some other K-12 strains, contains lambdoid prophage sequences, including an inducible gene (lar) that is an analogue of ral (11).
Figure 1 shows the effect of R16 ArdA specified by an overproducing recombinant on the de novo modification activity of EcoKI, as measured by the capacity of progeny phages from a vir.0 (unmodified) infection of C600(pLG2051) to form plaques on C600 relative to the capacity of the phages to form plaques on 5K. It is noted that DNA contains five EcoKI sites which require methylation in at least one DNA strand to avoid the restriction response. The data (summarized in Table 1) show that at 60 min, 35% of the phages in a burst of 218 carried full EcoKI modification. Assuming that each of the five EcoKI sites is modified equally, this value equates to an 80% efficiency of modification at each site. In the absence of ArdA protein, total plaque-forming ability was reduced and no burst was observed. These results illustrate the effect of ArdA as a restriction alleviation protein and indicate that, while there is significant inhibition of modification, cells containing the overproduced R16 protein retain substantial levels of EcoKI methyltransferase activity. As shown in Table 1, significant levels of de novo modification of infecting vir.0 also occurred in ArdA+ cells when the host strain was rK- mK+ (NM816) due to an hsdR mutation and when the Rac prophage was deleted (NM710). In the experiments with restricting hosts, Ard+ plasmid-containing cells supported almost complete alleviation of the restriction of incoming phages, as inferred from the efficiency-of-plating data (Table 1, column 3).
|
|
.K (modified) infection would retain modification were pLG2051 to cause complete inhibition of EcoKI methyltransferase activity. This fraction would represent those progeny containing one of the parental strands of the infecting modified phage genome. The results show that when the host strain was r- m+ (NM816) and carried the ArdA+ plasmid, approximately half of the progeny phages in the burst possessed EcoKI modification. The larger burst size detected in the plasmid-containing strain than that in NM816 is attributed to experimental variation and is thought to be biologically insignificant. To investigate whether Lar or Ral protein contributes to the retention of modification, tests were performed with a strain lacking the Rac prophage and a phage with a mutation in ral. The data indicate that neither the Rac prophage nor the ral gene of phage is responsible for the retention of EcoKI modification in the presence of R16 ArdA protein, although the effect of ral+ in enhancing modification can be seen. Overall, the results indicate that high in vivo levels of R16 ArdA preferentially inhibit the restriction activity of EcoKI rather than block R-M functions to the same extent.
|
There is compelling evidence that ArdA functions in the conjugatively infected recipient cell to alleviate type I restriction of the immigrant plasmid (20). Our results are inconsistent with the generalization that ArdA proteins inactivate type I enzymes in that the R16 ArdA homologue showed specificity in targeting the restriction function of EcoKI, leaving significant levels of methyltransferase activity under conditions in which restriction was almost completely prevented. It is noted that our test for determining the fraction of modified phages probably underestimates the extent of EcoKI modification since all five target sites in DNA require methylation to produce the modified response.
There is no obvious explanation of the finding that R16 ArdA preferentially targeted the restriction activity of EcoKI but that the ColIb homologue blocked the restriction and modification activities of the enzyme. We have confirmed the latter response (unpublished data). The cloned plasmid sectors in the ArdA+ recombinants are alike in size, and the proteins have very similar compositions. The cloned ColIb ardA is grossly overexpressed relative to the gene in the native plasmid (1), raising the possibility that ArdA has nonphysiological effects on EcoKI modification dependent on the level of overexpression.
It is unknown how ArdA functions as an antirestriction protein. One possibility is that ArdA blocks a step specific for the restriction reaction, such as the DNA translocation or cleavage process. Alternatively, ArdA proteins might target the R2M2S1 form of the type I enzyme but not the M2S1 complex, which is active as a methyltransferase. Such hypotheses can be tested by purifying ArdA and using it in experiments in vitro.
|
ACKNOWLEDGMENTS |
|---|
We thank Noreen Murray for the gift of strains.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractTextReferences |
|---|
. J. Mol. Biol. 190:11-22.[CrossRef][Medline]
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Appl. Environ. Microbiol. | Infect. Immun. | Eukaryot. Cell |
|---|---|---|
| Mol. Cell. Biol. | J. Virol. | Microbiol. Mol. Biol. Rev. |
| ALL ASM JOURNALS |
2 x 109 cells ml-1. Bacteriophages were added at a multiplicity of infection of 0.05. After 10 min of adsorption at room temperature, bacteria were sedimented and resuspended in nutrient broth at 2 x 107 cells ml-1. Samples were plated immediately (t = 0) with indicator bacteria to determine the number of PFU derived from infected cells and unadsorbed phages. At times thereafter, samples were treated with chloroform and plated with 5K indicator cells (
) to measure the total burst size and with C600 cells (•) to determine the number of EcoKI-modified phages. The number of PFU plating on 5K following infection of C600 lacking pLG2051 is also shown (ArdA-; 
). Data are the means of results from two experiments.