Quoting from : Scientific American July 1975 The Manipulation of genes by Stanley Cohen :

Restriction endonucleases (and modification methylases) are widespread in microorganisms; genes for making them were found on viral chromosomes and extrachromosomal plasmid DNA as well as on many bacterial chromosomes.

Why would the genes for making RE be found on viral chromosomes ? Also, could you give some examples where they are found on plasmids ?


3 Answers 3


There are many proposals for the ecological role of restriction-modification (RM) systems, and why they would exist on mobile genetic elements (e.g. plasmids and viruses). In this case, I am specifically talking about the viruses that infect bacteria (aka, bacteriophage).

1) RM systems may have an anti-viral function. Normally we would think of such systems as being part of the host chromosome, but as Alan Boyd suggested, once a temperate phage has integrated itself into the chromosome, its fitness is tied to its host. Therefore, a RM system found in a prophage could prevent infections by additional phage. See here for a discussion of this issue: http://www.ncbi.nlm.nih.gov/pubmed/23471617

2) RM systems may have "addictive" properties by acting as toxin-antitoxin systems. Basically, this means that if the genes for the RM system are lost, the host cell dies. This can provide selection for the maintenance of plasmids and prophage inside of the host cell. See here for a discussion of this issue: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3874152/

3) Finally, viruses need to destroy the host genome -- both to suppress any anti-viral response and to release nutrients for viral replication. While this is typically achieved with nucleases other than restriction endonucleases, there is at least one situation where a RE seems to be involved. This is discussed in the first link above (see the section entitled "Role in Nutrition")


If you have a dsDNA genome and replicate it by rolling circle like most bacteriophages and many plasmids do, an endonuclease is pretty much a necessity.

enter image description here

  • 1
    $\begingroup$ Could you elaborate your answer and include some references ? $\endgroup$
    – biogirl
    Commented May 23, 2014 at 16:01
  • $\begingroup$ this is a nice question and post! the RE nicks the phage genome so that it can be replicated by the polymerase. Without it there would be no copying off of the chromosome. $\endgroup$
    – shigeta
    Commented May 23, 2014 at 17:17
  • 3
    $\begingroup$ But the nicking isn't carried out by a restriction endonuclease but by a specific protein which remains linked to the 5' end of the DNA. If this was accomplished by an R-M system on the phage the site would presumably be modified to protect it from double-stranded cutting. $\endgroup$
    – Alan Boyd
    Commented May 23, 2014 at 18:08
  • $\begingroup$ I generally think of REs as producing double-stranded cuts. If I understand correctly, this answer only applies to those that create single-stranded cuts. Are they also called RE? $\endgroup$
    – adam.r
    Commented May 23, 2014 at 19:31
  • 1
    $\begingroup$ i just wonder which has an older origin- CRISPR-Cas or RE? $\endgroup$
    Commented May 24, 2014 at 6:24

The familiar restriction enzyme EcoRI is plasmid encoded.

Betlach et al. (1976) A restriction endonuclease analysis of the bacterial plasmid controlling the EcoRI restriction and modification of DNA. Fed. Proc. 35:2037 - 43.

For an example of a phage encoded system see:

Dempsey et al. (2005) Sau421, a BcgI-like restriction-modification system encoded by the Staphylococcus aureus quadruple-converting phage Phi42. Microbiology 151: 1301-1311

This second paper suggests an answer to why a phage would have such a system: lysogens of Phi42 are resistant to infection by all 23 members of a standard set of S. aureus phages. Once a phage has lysogenised it is in its interest to prevent lysis of its host.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .