Removal of a frameshift between the hsdM and hsdS genes of the EcoKI Type IA DNA restriction and modification system produces a new type of system and links the different families of Type I systems

Abstract

The EcoKI DNA methyltransferase is a trimeric protein comprised of two modification subunits (M) and one sequence specificity subunit (S). This enzyme forms the core of the EcoKI restriction/modification (RM) enzyme. The 3' end of the gene encoding the M subunit overlaps by 1 nt the start of the gene for the S subunit. Translation from the two different open reading frames is translationally coupled. Mutagenesis to remove the frameshift and fuse the two subunits together produces a functional RM enzyme in vivo with the same properties as the natural EcoKI system. The fusion protein can be purified and forms an active restriction enzyme upon addition of restriction subunits and of additional M subunit. The Type I RM systems are grouped into families, IA to IE, defined by complementation, hybridization and sequence similarity. The fusion protein forms an evolutionary intermediate form lying between the Type IA family of RM enzymes and the Type IB family of RM enzymes which have the frameshift located at a different part of the gene sequence.

Figure 1.

Structures of Type I MTases showing the HsdS subunit with TRDS (yellow) and conserved helical regions (orange), and the two HsdM subunits (one in cyan, the other with the N-terminal domain in green, the MTase catalytic domain in blue and the C-terminal helical tail in grey). (a) The EcoKI MTase structural model (14). The C terminus of an M subunit and the N terminus of the S subunit are indicated by ‘C’ and ‘N’, respectively, and are close together in space. (b) A schematic of the gene layout for typical Type IA/IC RM systems and for Type IB RM systems. The frameshift at the junction between regions coding for HsdM and HsdS subunits is indicated by showing the hsdM gene above the hsdS gene. The colour scheme is as in part (a) (with the additional inserts in the hsdS gene for the IB system coloured grey for reasons discussed later). A small conserved section at the start of hsdS is not shown for clarity. (c) The circular arrangement of HsdS subunit sequences for a Type IA/IC HsdS subunit, as originally proposed by Kneale (23), for a half-S subunit and for the Type IB HsdS subunit.

Figure 2.

Escherichia coli NM1261(DE3) (r⁺m⁺s⁻) transformed with, from left to right, plasmids pBio2, pJFMS and pMSFusion. Phage were then spotted on these strains with λv.k on left and λv.o on right of each panel and dilutions from 10⁻¹ to 10⁻⁸ as indicated by the diagram at the left.

Figure 3.

Phage recovered from E. coli NM679 spotted on E. coli NM1049(DE3) (r⁺m⁺s⁺), top row, or on E. coli NM1261(DE3) (r⁺m⁺s⁻), bottom row). From left to right are cells transformed with pBIO2, pJFMS or pMSFusion. Phage dilutions are 10⁻¹ to 10⁻⁴ as shown.

Figure 4.

Endonuclease assay for DNA cleavage by MSFusion. Final concentration of DNA (unmethylated pBRsk1) was 3 nM. Final concentration of reconstituted nuclease was 60 nM using at least 2 HsdR and 1 HsdM per M₁S₁ protein or MSFusion protein. The enzymes were reconstituted at 1000 nM and then diluted into the solution containing DNA. M = 1 kb ladder; OC = open circle; L = linear; CCC = covalent closed circle.

Figure 5.

Western blot of cell extracts using antibodies specific to HsdM and HsdS to detect the MSFusion protein and the presence of any HsdM produced by proteolysis or stalled translation of the fusion protein. M = molecular weight marker; CE = cell extract of E. coli NM679; un. = uninduced; in. = induced.

Figure 6.

Two possible evolutionary routes to convert between Type IA/IC RM systems and Type IB RM systems via a Type IMS RM system with HsdM subunit fused to HsdS subunit. The colouring scheme for structural domains is the same as in Figure 1.