Genes for restriction-modification enzymes are often clustered together in a region called an "immigration control region". The immigration island from E. coli K-12 is about 14 Kbp. The six restriction-modification genes in this region are organized as shown below:
The percentages shown in the figure above represent the %GC of the DNA in that region of the immigration control region. In contrast, the E. coli chromosome is about 50% GC overall. The differences in %GC between the genes in the immigration control region relative to the chromosome suggests that these genes were inherited via horizontal transfer -- that is, these genes probably originally came from other organisms. Furthermore, because the %GC varies so much between the different groups of genes, it seems likely that the immigration control region is a mosaic of multiple gene transfer events from organisms with different %GC contents. Large blocks of genes that seem to differ from the host chromosome are often called "islands", so sometimes the immigration control region is called an "immigration island". In E. coli and related bacteria the immigration island is located near 99 min on the chromosome.
The immigration control region varies widely between different strains of E. coli, Salmonella, and related bacteria. Different restriction phenotypes can be due to different alleles of the genes for one restriction system or the substitution of genes for a different restriction system. Some strains lack the entire immigration control region. For example, E. coli C strains are missing the entire 14 Kb immigration control region and thus are restriction-minus.
The genetic variation in the immigration control region can arise in several ways. The HsdS protein from Type I restriction systems can interact with the HsdM and HsdR proteins from other Type I restriction systems that are in the same gene family. For example, the restriction specificity of E. coli B can be changed by substituting the hsdS gene from E. coli B with the hsdS gene from E. coli K-12.
In addition, the genes for the HsdS protein of Type I restriction enzymes in the same gene family share sufficient DNA sequence homology that they can recombine. Recombination can yield new HsdS proteins that recognize a different restriction site than either of the two parents. This can occur in several ways, but one example is shown in the figure below. Sty SPI and Sty LtIII are naturally occuring restriction systems, that can recombine to produce new restriction systems with different site specificity.
The regions labeled 2 and 3 contain the protein domains that recognize the restriction site -- domain 2 interacts with the bases shown to the left of the (N6) sequence and domain 3 interacts with the bases shown to the right of (N6). The DNA sequence of the region labeled 2 is sufficiently conserved that the genes can recombine in this region, thereby switching the two specificity domains.
In E. coli K-12 the immigration control region encodes three restriction systems.
The hsdR gene product is the restriction subunit that cleaves the DNA if this site is unmethylated. The hsdM gene product is a methyltransferase that transfers a methyl group from S-adenosylmethionine (SAM) to the DNA at the indicated A residues.
Note that the EcoK system only recognized unmethylated DNA while the other two restriction systems only recognize modified DNA. Thus, these three systems provide a way to attack any foreign DNA that is either modified or unmodified in a way that would identify it as foreign.
In contrast to these restriction systems which are located on the bacterial chromosome, most Type II and Type III restriction systems are located on plasmids or phages.
For a comprehensive list of restriction enzymes with a brief description of their properties and relevant references, check out the NEB restriction enzyme resource.
lease send comments, suggestions, or questions to firstname.lastname@example.org
Last modified July 15, 2002