Article

ISSN 0006-2979, Biochemistry (Moscow), 2025, Vol. 90, No. 4, pp. 513-521 © Pleiades Publishing, Ltd., 2025.

Published in Russian in Biokhimiya, 2025, Vol. 90, No. 4, pp. 571-579.

513

Restriction–Modification Systems

Specific toward GGATC, GATGC, and GATGG.

Part 2. Functionality and Structure

Sergey Spirin

1,2,3,a

*, Alexander Grishin

4,5

, Ivan Rusinov

Andrei Alexeevski

1,3

, and Anna Karyagina

1,4,5

Belozersky Institute of Physico-Chemical Biology, Lomonosov Moscow State University,

119234 Moscow, Russia

Higher School of Economics National Research University, 109028 Moscow, Russia

NRC “Kurchatov Institute” - SRISA, 117218 Moscow, Russia

Gamaleya National Research Center for Epidemiology and Microbiology,

Ministry of Healthcare of the Russian Federation, 123098 Moscow, Russia

All-Russia Research Institute of Agricultural Biotechnology, 127550 Moscow, Russia

e-mail: sas@belozersky.msu.ru

Received January 21, 2025

Revised March 23, 2025

Accepted March 25, 2025

Abstract— The structural and functional basics of protein functionality of restriction–modification systems

recognizing GGATC/GATCC, GATGC/GCATC, and GATGG/CCATC sites have been studied using bioinformatics

methods. Such systems include a single restriction endonuclease and either two separate DNA methyltrans-

ferases or a single fusion DNA methyltransferase with two catalytic domains. It is known that some of these

systems methylate both adenines in the recognition sites to 6-methyladenine, but the role of each of the

two DNA methyltransferases remained unknown. In this work, we proved the functionality of most known

systems. Based on the analysis of structures of related DNA methyltransferases, we hypothesized which of

the adenines within the recognition site is modified by each of the DNA methyltransferases and suggested

a possible molecular mechanism of changes in the DNA methyltransferase specificity from GATGG to GATGC

during horizontal transfer of its gene.

DOI: 10.1134/S0006297925600152

Keywords: restriction–modification system, 3D structure, DNA methyltransferase, restriction endonuclease, hor-

izontal gene transfer

* To whom correspondence should be addressed.

INTRODUCTION

Restriction–modification (RM) systems protect

prokaryotic cells from the invasion of foreign (e.g.,

viral) DNA [1]. RM systems have been traditionally

divided into several types [2]. Type  II RM systems en-

code two or more proteins; one is a restriction en-

donuclease (REase), which specifically recognizes and

cleaves a cognate DNA sequence, while the other pro-

tein is a methyltransferase (MTase), which modifies

the host DNA and prevents its hydrolysis by the REase.

Unlike most Type II RM systems, subtype  IIA RM sys-

tems are characterized by non-palindromic asymmet-

ric recognition sequences, which necessitates the pres-

ence of either two different MTases or a single fused

MTase with two catalytic centres to methylate both

DNA strands [3, 4].

In Part 1 of this work [Evolution and Ecolo-

gy, Biochemistry (Moscow), vol.  90, issue  4], we de-

scribed the evolution of subtype IIA RM systems

with the specificity towards GGATC/GATCC, GATGC/

GCATC, or GATGG/CCATC. The REases of all such sys-

tems are homologous to each other, which is also true

for their MTases. These MTases methylate adenines

SPIRIN et al.514

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

to N

-methyladenines (6mA) in both DNA strands of

the recognition sequence. Some of these systems pos-

sess two separate MTases, while others carry a sin-

gle fused MTase with two catalytic MTase domains.

We have shown that the fusion and separation of the

two MTases have occurred multiple times during the

evolution of such RM systems. While all the MTase

domains of these systems belong to the same Methyl-

transf12 protein family (according to the Pfam data-

base) [5], they can be classified based on the sequence

similarity into two well defined groups, which we des-

ignated as A and B. In every RM system selected for

this study, the two MTase domains always belonged

to different groups. Evidently, these two different do-

mains are responsible for the modification of different

DNA strands. However, the experimental information

on which MTase methylates which DNA strand is lack-

ing. In this article, which is Part 2 of our work, we

analyzed available biochemical data, amino acid se-

quences, and 3D structures of homologous MTases to

determine specific nucleotides methylated by each of

the MTases, predicted MTase regions responsible for

the recognition of cognate DNA sequences, and pro-

posed a hypothetical mechanism that explains chang-

es in the specificity of one of the MTases during its

evolution, presumably, in the course of its horizontal

transfer from an RM systems with a different speci-

ficity.

MATERIALS AND METHODS

The list of RM systems was extracted from RE-

BASE, v.  303 as of 28.02.2023  [6]. The amino acid se-

quences were aligned using Muscle [7], and the align-

ments were visualized in Jalview [8]. The boundaries

between the N- and C-terminal domains of fused

MTases were determined by comparing their se-

quences with the sequences of single-domain MTases.

The evolutionary domains in the sequences of RM

system proteins and domain families were identi-

fied with HMM profiles from the Pfam database  [5].

Protein phylogenies were inferred using FastME  [9]

with midpoint rooting of the resulting phylogenetic

tree. The trees were visualized with MEGA7  [10] and

iTOL [11]. CD-HIT  [12] was used for clustering proteins

with desired sequence identity levels. The structures

of MTases and MTase-DNA complexes were predicted

with AlphaFold2 using ColabFold [13,  14] and visual-

ized and analyzed in PyMOL [15]. Sequence LOGOs

were generated by WebLogo [16].

Following the terminology introduced in the

Part 1 of this study [Evolution and Ecology, Biochem-

istry (Moscow), vol. 90, issue 4], we will further refer

to the RM systems specific toward GGATC/GATCC as

‘red’, GATGG/CCATC – as ‘green’, and GATGC/GCATC –

as ‘blue’. Hereinafter (except Fig. 4), the recognised

sites for ‘red’, ‘green’, and ‘blue’ systems will be des-

ignated as GGATC, GATGG, and GATGC, respectively,

i.e., we will use the sequence of the chain containing

adenine that, according to our results, is methylated

by a groupA MTase.

RESULTS AND DISCUSSION

In Part 1 of our study [Evolution and Ecology, Bio-

chemistry (Moscow), Issue 4, vol. 90], we determined

that all RM systems specific toward GGATC, GATGC, or

GATGG contained one REase domain of the RE_AlwI

family and two MTase domains of the Methyltransf12

family. In total, REBASE v. 303 contained 493 such

systems with two single-domain MTases and 227 sys-

tems with fused two-domain MTases. Based on their

sequence similarity, all single-domain MTases, as well

as the N- and C-terminal domains of fused two-domain

MTases, were divided into two groups designated  A

and  B. Two MTases of the same system or two do-

mains of a fused MTase always belonged to the dif-

ferent groups.

Functionality of the RM systems. The enzymatic

activity of several REases belonging to the studied RM

systems has been demonstrated experimentally. Thus,

the ‘blue’ REase SfaNI hydrolyzed DNA five nucleo-

tides downstream of the recognition sequence GCATC

in the top strand and nine nucleotides downstream in

the bottom strand, which is conventionally designated

as GCATC (5/9) [17]. According to REBASE, the ‘red’

REase AlwI hydrolyzes GGATC (4/5). The ‘green’ REase

McaCI recognizes the sequence CCATC, although the

exact site of hydrolysis remains unknown.

It was shown that mutations of the amino acid

residues E418, D456, E469, and E482 in the nickase

Nt.BstNBI, which is homologous to the studied REases

(Fig.1), resulted in the loss of its catalytic activity [18].

Amino acid residues corresponding to D456 and E482

of Nt.BstNBI were absolutely conserved in 411 refer-

ence sequences chosen from the clusters of studied

REases with 98% sequence identity; the residues cor-

responding to E418 were conserved in 410 sequences

and replaced by D in only one sequence. Amino acid

residues corresponding to E469 were conserved in

407 sequences, replaced by K in two closely related

REases, are deleted in two other sequences.

Based on the X-ray analysis of the structure

of Nt.BspD6I nickase, which is 100% identical to

Nt.BstNBI, it was proposed that H489 is another res-

idue essential for the nickase enzymatic activity  [19].

The corresponding amino acid residue in the sequenc-

es of studied REases were strictly conserved. Altogeth-

er, these data indicate that the majority of REases

studied in this work are functional.

RM SYSTEMS: FUNCTIONALITY AND STRUCTURE 515

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

Fig. 1. Fragment of the multiple sequence alignment of the studied REases and Nt.BstNBI nickase. Red asterisks indicate

amino acid residues E418, D456, E469, and E482 of nickase Nt.BstNBI that were experimentally confirmed as important for

its catalytic activity; yellow asterisk indicates H489 presumably participating in the catalysis.

The MTases studied in our work were 6mA

MTases that belong to the α group according to the

classification based on the mutual position of the

conserved motifs within their sequences [20]. The

absolute majority of the studied MTases possessed

both the S-adenosylmethionine-binding motif F-x-G-x-

[G/A] and the catalytic motif D-[P/T]-P-Y, which indi-

cates the functionality of these enzymes. The ability

to methylate DNA has been experimentally demon-

strated by PacBio for almost a hundred of these

MTases.

Hypothetical mechanism of DNA recognition by

MTases and prediction of the methylation sites. The

3D structures are available for three MTases from the

MethyltrasfD12 family: M.EcoT4Dam (PDB IDs:1YFJ,

1YFL, 1YF3, 1Q0S, 1Q0T) [21], M1.DpnII (2DPM) [22],

and M.EcoKDam (2G1P, 2ORE, 4GOL, 4GOM, 4GON,

4GOO, 4GBE, and 4RTJ-4RTS) [23]. All three MTases

are homologous to group B MTases from the studied

RM systems (see Fig. S1 in the Online Resource 1).

M.EcoT4Dam [21] contained several groups of con-

served amino acid residues that formed three clus-

ters on the protein surface (see Fig. 2b in the paper

[21]). The first group included residues located near

the catalytic site (green asterisks in Fig.S1 in the On-

line Resource1), the second and the third groups con-

tained conserved residues from the target-recognition

domain (TRD; yellow asterisk in Fig. S1 in the Online

Resource 1) and conserved β-hairpin (red asterisks

in Fig. S1 in the Online Resource 1), respectively.

Fig. 2. Phylogenetic tree of 12 MTases specific toward GGATC (red), GATGC (green), and GATGG (blue), MTases M.EcoKDam,

M.EcoT4Dam, and M1.DpnII with known 3D structures, and well-studied MTases M.EcoRV, M.FokI, and M1.Bst19I. The phy-

logeny for the fused two-domain MTases was inferred for the N- and C-terminal domains separately. Numbers indicate the

bootstrap support values; branches with lower than 25% support were removed (collapsed).

SPIRIN et al.516

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

Fig. 3. Gene organization and target sequences of representative MTases recognizing GGATG, GATGG, GATGC, and GGATC

sites. Based on the information from REBASE website, red and yellow flags indicate positions of conserved motifs F-x-G-x-G

and D-P-P-Y, respectively; blue asterisks indicate non-perfect matches between the canonical motifs and the actual sequenc-

es (F-x-G-x-A and D-T-P-Y, respectively). Two possible methylation variants are shown for GGATC, with pink background

indicating a more probable variant.

Both DNA strands in the recognition sequences

of the studied RM systems contained one adenine

residue each. Several 3D structures of MethyltrasfD12

family MTases, namely M.EcoT4Dam (PDB IDs: 1YFJ,

1YFL, 1YF3, and 1Q0T), and M.EcoKDam (PDBID 2G1P,

and 4RTJ-4RTS) were obtained in complex with DNA

[21, 23]. Since these MTases were homologous to the

group B MTases studied in our work (Fig. 2 and S1

in the Online Resource 1), it was possible to predict

which adenine is methylated by which MTase in the

RM systems recognizing GATGG (‘green’) and GATGC

(‘blue’) sequences. Specifically, multiple sequence

alignment of group  B MTases with M.EcoKDam and

M.EcoT4Dam identified a conserved arginine residue

(R116 in M.EcoT4Dam and R124 in M.EcoKDam; the

first red asterisk in Fig.S1 in the Online Resource1).

In M.EcoKDam, this residue is responsible for the

recognition of the guanine residue in the DNA strand

complementary to the strand containing methylated

adenine of the GATC recognition sequence [23]. Pre-

sumably, this means that group  B MTases methylate

adenine in the ATC subsequence. This hypothesis was

supported by the experimental data on the nucleo-

tides methylated by the ‘green’ MTases M1.Hpy300VI

(group  A) and M2.Hpy300VI (group B) [24], M.FokI [25]

specific towards GGATG (the N- and C-terminal do-

mains of this enzyme are homologous to group A and

group B MTases, respectively), and MTase M1.Bst19I

which is similar to group B MTases [26]. Although

M1.Bst19I recognizes GATGC site, its sequence differs

significantly from other MTases with the same speci-

ficity. M1.Bst19I did not cluster together with the rest

of ‘blue’ MTases in the phylogenetic tree (Fig. 2). The

RM system containing M1.Bst19I was not included in

the list of RM systems studied in this work, because

the amino acid sequence of the corresponding REase

Bst19I was not available.

Therefore, we predict that group B MTases be-

longing to the systems with the GATGG and GATGC

recognition sites (‘green’ and ‘blue’, respectively)

methylate adenines that are complementary to the

third (T) nucleotide. Group A MTases likely methylate

the second (A) nucleotide in these sequences. Howev-

er, no such prediction can be made for the systems

recognizing GGATC (red), since in this case, both DNA

strands contain the ATC subsequence (Fig. 3).

To determine the most probable methylation pat-

tern for the MTases with GGATC recognition sequence,

we analyzed the 3D structures of the MTase complex-

es with DNA (Fig.  4). The models of all proteins, in-

cluding M.EcoKDam and M.EcoT4Dam, were gener-

ated with ColabFold, since the crystal structures for

RM SYSTEMS: FUNCTIONALITY AND STRUCTURE 517

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

Fig.  4. 3D models of group B MTases and their homologs in complex with DNA. Cyan, MTase catalytic domain; green, TRD;

red, loop potentially involved in the recognition of the nucleotide pairs shown to the right in the figure; magenta, region

presumably involved in the recognition of nucleotides shown to the left in the figure. In the DNA molecule, pale pink indi-

cates four nucleotides that correspond to the GATC recognition sequence of M.EcoKDam; light lilac and pink correspond to

flanking nucleotide pairs. Nucleotides designated with letters and 5′ and 3′ denotations are given for the strands containing

adenine residues methylated by MTases B

M.EcoKDam and M.EcoT4Dam lacked coordinates for

several loops. For M.EcoKDam and M.EcoT4Dam, the

regions present in the crystal structures did not differ

significantly from the obtained models (RMSD values

for comparison of M.EcoKDam model with its X-ray

structure (2G1P) and of M.EcoT4DAM model with its

structure (1YFL) were 0.225 and 0.633  Å, respective-

ly). The positions of DNA and S-adenosylhomocysteine

were modelled from 2G1P. MTase models were aligned

with the 2G1P structure over the catalytic domain (co-

loured cyan in Fig.4) to unify their spatial orientation.

We hypothesized that the conserved arginine res-

idue (R124 in M.EcoKDam) of group  B MTases always

interacts with guanine in the DNA strand complemen-

tary to the strand containing the methylated adenine

in the ATC motif. If this adenine is a part of the GGATC

sequence, then group B MTase (e.g., N-terminal domain

of M.AlwI) would have amino acid residues specifi-

cally recognizing the two 5′ nucleotides; on the other

hand, if this adenine is a part of the complementary

GATCC sequence, then the group B MTase would con-

tain residues recognizing both 5′ and 3′ nucleotides.

As shown in Fig.  4, the structure of the N-terminal

domain of M.AlwI possesses a loop located close to the

nucleotides adjacent to the methylated adenine from

the 3′ direction (the loop is shown in red in Fig.  4

and underlined with red in the sequence alignment

in Fig.S1 in the Online Resource1). At the same time,

SPIRIN et al.518

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

Fig.  5. LOGO diagrams of the sequence alignments of

group B MTases loops (shown in red in Fig. 4). a)  Dia-

gram constructed for the loops of GGATC-specific MTases,

homologous to residues 199-218 of M.AlwI N-terminal

domain (VPISEYSDFKRYTKEQFYLE). b) Diagram con-

structed for the loops of GGATG-specific MTases, homol-

ogous to residues 552-571 of M.FokI C-terminal domain

(LITTGSYNDGNRGFKDWNRL). The recognition sequences of

the respective MTases are shown under each LOGO diagram;

the nucleotide pair presumably recognized by the protein

loop is highlighted in bold; G residue indicated by the arrow

likely binds to the conserved arginine.

Fig.  6. LOGO diagrams of the sequence alignments of

group  B MTase loops that are shown in magenta in Fig.  4.

a) Diagram constructed for loops of GATGC-specific MTases

(‘blue’), homologous to residues 490-509 of M.SfaNI C-termi-

nal domain (LSNSKMYGYNYYKTSSAKGL). b and c)Diagrams

constructed for the 23-residue (b) and 24-residue (c) loops

of GATGG-specific MTases (‘green’), homologous to residues

111-133 of M2.McaCI (LSCSYLSITVPDELKKKYVKTYY). The

recognition sequences of the respective MTases are shown

under each LOGO diagram; the nucleotide pair presumably

recognized by the protein loop is highlighted in bold.

M.AlwI lacks protein segments coloured in magenta

in Fig.4; these segments are present in the structures

of MTases (‘green’ M2.McaCI and C-terminal domain

of ‘blue’ M.SfaNI) whose recognition sequences extend

toward the 5′ end relative to the methylated adenine

(underlined with magenta in the sequence alignment

in Fig.S1 in the Online Resource1). Therefore, it seems

more plausible that the N-terminal domain of M.AlwI

methylates adenine in GATCC and not in GGATC.

Group B MTases that recognize GGATC (‘red’) or

GGATG (M.FokI-like) contain the conserved motif Y-x-

D-x-x-R (Fig. 5) potentially involved in the decoding

of guanine marked with an arrow in Fig.5 that could

possibly interact with the conserved arginine residue

of this motif.

Group B MTases recognizing GATGC and GATGG

(‘blue’ and ‘green’, respectively) contain the L-S-x-[S/T]

motif (Fig.  6) at the beginning of the loop (coloured

in magenta in Fig.  4). The rest of the loop sequences

differ between the ‘blue’ and ‘green’ group B MTases,

probably because they recognise GC nucleotide pairs

oriented differently with respect to the methylated

adenines. These GC pairs are probably recognized

by some residues from the G-Y-x–x-Y-x-x-x-S motif in

‘blue’ MTases (Fig.  6a) and K-x-x-x-x-K-T-Y-[F/Y] motif

in ‘green’ MTases (Fig.6, b and c). ‘Green’ MTases con-

tain two variants of the loop (23 and 24 amino acid

residues in length). The sequences of these variants

are similar at the edges but differ in the central part.

Changes in the MTase specificity after hori-

zontal gene transfer. In several RM systems, includ-

ing Cup11541IV, Hfe11613I, and Hmu12714II whose

specificity towards GATGC was confirmed by PacBio,

group  A MTases was more similar to the MTases

recognizing GATGG (‘green’) than to the majority of

MTases with the GATGC specificity (‘blue’). Group B

MTases and REases from these systems, however, did

not demonstrate such anomalous clustering (note

the positions of Hmu12714II and Hfe11614I proteins

on the trees in Fig.  7, a-c). Presumably, an ancestral

groupA MTase has changed its specificity from GATGG

to GATGC upon the horizontal transfer of its gene.

To suggest the molecular mechanisms responsible for

this change, we compared the sequences of group  A

MTases recognizing GATGG and GATGC (Fig.  7). The

specificity of these MTases was confirmed by PacBio.

All ‘blue’ group A MTases, including those more simi-

lar to the ‘green’ group A enzymes in majority of their

RM SYSTEMS: FUNCTIONALITY AND STRUCTURE 519

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

Fig.  7. Phylogenetic trees for the representative group A MTases(a), group B MTases(b), and REases(c) recognizing GATGC

(‘blue’) or GATGG (‘green’), and the 3D model of group A MTase M1.Hmu12714II(d). The GC pair corresponding to the 3′-end

cytosine in GATGC is shown in red. S-adenosylhomocysteine and side chains of amino acid residues from the N-x-R-S-N

motif are shown as ball-and-stick models; the catalytic domain and the TRD indicated with cyan and green, respectively.

amino acid sequences, contained in the C-terminal

region the conserved motif N-x-R-S-N, which ‘green’

MTases lacked. As can be seen from the model of

M1.Hmu12714II complex with DNA (Fig.  7d), this motif

is positioned in the DNA major groove, with arginine

side chain located close to the guanine complementary

to the 3′-end C in GATGC. It can be speculated that the

arginine residue of this motif is responsible for the

enzyme specificity.

The N-x-R-S-N motif could have emerged inde-

pendently as a result of convergent evolution. More

likely, however, that this motif has been obtained by

an MTase with the GATGG specificity through recom-

bination with an MTase recognizing GATGC.

CONCLUSION

In this work, we analyzed the sequences and 3D

structures of proteins from the RM systems recogniz-

ing GGATC, GATGC, and GATGG sequences, as well as

several related proteins, to establish the functionality

of the majority of these systems and to propose which

adenine base is methylated by each of the two MTases

in a RM system and how the MTase recognition se-

quence can change from GATGG to GATGC upon the

horizontal transfer of an MTase gene.

Abbreviations. MTase, DNA methyltransferase;

REase, restriction endonuclease; RM system, restric-

tion–modification system; TRD, target recognition do-

main.

Supplementary information. The online version

contains supplementary material available at https://

doi.org/10.1134/S0006297925600152.

Contributions. S.S. and A.K. developed the con-

cept and supervised the study; S.S., A.G., I.R., and A.K.

curated the data, developed the software, and ana-

lyzed the data; S.S., A.G., and A.K. wrote the manu-

script.

Funding. This work was supported by the Rus-

sian Foundation for Basic Research (project no.21-14-

00135).

Ethics approval and consent to participate.

This work does not contain any studies involving hu-

man or animal subjects.

Conflict of interest. The authors of this work de-

clare that they have no conflicts of interest.

SPIRIN et al.520

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

REFERENCES

1. Williams, R.J. (2003) Restriction endonucleases: clas-

sification, properties, and applications, Mol. Biotech-

nol., 23, 225-244, https://doi.org/10.1385/mb:23:3:225.

2. Roberts, R. J. (2003) A nomenclature for restriction

enzymes, DNA methyltransferases, homing endonu-

cleases and their genes, Nucleic Acids Res., 31, 1805-

1812, https://doi.org/10.1093/nar/gkg274.

3. Madhusoodanan, U.K., and Rao, D.N. (2010) Diversity

of DNA methyltransferases that recognize asymmetric

target sequences, Crit. Rev. Biochem. Mol. Biol., 45,

125-145, https://doi.org/10.3109/10409231003628007.

4. Vasu,K., and Nagaraja,V. (2013) Diverse functions of

restriction-modification systems in addition to cel-

lular defense, Microbiol. Mol. Biol. Rev., 77, 53-72,

https://doi.org/10.1128/mmbr.00044-12.

5. Mistry, J., Chuguransky, S., Williams, L., Qureshi, M.,

Salazar, G. A., Sonnhammer, E.L.L., Tosatto, S. C. E.,

Paladin, L., Raj, S., Richardson, L. J., Finn, R. D., and

Bateman, A. (2020) Pfam: the protein families da-

tabase in 2021, Nucleic Acids Res., 49, D412-D419,

https://doi.org/10.1093/nar/gkaa913.

6. Roberts, R. J., Vincze, T., Posfai, J., and Macelis, D.

(2014) REBASE – a database for DNA restriction and

modification: enzymes, genes and genomes, Nucle-

ic Acids Res., 43, D298-D299, https://doi.org/10.1093/

nar/gku1046.

7. Edgar, R. C. (2004) MUSCLE: multiple sequence

alignment with high accuracy and high throughput,

Nucleic Acids Res., 32, 1792-1797, https://doi.org/

10.1093/nar/gkh340.

8. Waterhouse, A. M., Procter, J. B., Martin, D. M. A.,

Clamp,M., and Barton, G.J. (2009) Jalview Version2–

a multiple sequence alignment editor and analysis

workbench, Bioinformatics, 25, 1189-1191, https://

doi.org/10.1093/bioinformatics/btp033.

9. Lefort,V., Desper,R., and Gascuel,O. (2015) FastME2.0:

A comprehensive, accurate, and fast distance-based

phylogeny inference program, Mol. Biol. Evol., 32,

2798-2800, https://doi.org/10.1093/molbev/msv150.

10. Kumar,S., Stecher,G., and Tamura,K. (2016) MEGA7:

Molecular Evolutionary Genetics Analysis version7.0

for bigger datasets, Mol. Biol. Evol., 33, 1870-1874,

https://doi.org/10.1093/molbev/msw054.

11. Letunic, I., and Bork, P. (2021) Interactive Tree Of

Life (iTOL) v5: an online tool for phylogenetic tree

display and annotation, Nucleic Acids Res., 49,

W293-W296, https://doi.org/10.1093/nar/gkab301.

12. Li, W., and Godzik, A. (2006) Cd-hit: a fast program

for clustering and comparing large sets of protein or

nucleotide sequences, Bioinformatics, 22, 1658-1659,

https://doi.org/10.1093/bioinformatics/btl158.

13. Mirdita, M., Schütze, K., Moriwaki, Y., Heo, L.,

Ovchinnikov, S., and Steinegger, M. (2022) Colab-

Fold: making protein folding accessible to all, Nat.

Methods, 19, 679-682, https://doi.org/10.1038/s41592-

022-01488-1.

14. Jumper,J., Evans,R., Pritzel,A., Green,T., Figurnov,M.,

Ronneberger, O., Tunyasuvunakool, K., Bates, R.,

Žídek,A., Potapenko,A., Bridgland,A., Meyer,C., Kohl,

S. A. A., Ballard, A. J., Cowie, A., Romera- Paredes,B.,

Nikolov, S., Jain, R., Adler, J., Back, T., Petersen, S.,

Reiman, D., Clancy, E., Zielinski, M., Steinegger, M.,

Pacholska, M., Berghammer, T., Bodenstein, S.,

Silver, D., Vinyals, O., Senior, A. W., Kavukcuoglu, K.,

Kohli, P., and Hassabis, D. (2021) Highly accurate

protein structure prediction with AlphaFold, Nature,

596, 583-589, https://doi.org/10.1038/s41586-021-

03819-2.

15. DeLano, W.L. (2002) Pymol: An open-source molecu-

lar graphics tool, CCP4 Newsl. Protein Crystallogr, 40,

82-92.

16. Crooks, G.E., Hon,G., Chandonia, J.M., and Brenner,

S. E. (2004) WebLogo: A sequence logo generator,

Genome Res., 14, 1188-1190, https://doi.org/10.1101/

gr.849004.

17. Gingeras, T.R., MIlazzo, J.P., and Roberts, R.J. (1978) A

computer assisted method for the determination of re-

striction enzyme recognition sites, Nucleic Acids Res.,

5, 4105-4127, https://doi.org/10.1093/nar/5.11.4105.

18. Higgins, L. S., Besnier, C., and Kong, H. (2001) The

nicking endonuclease N.BstNBI is closely related to

type IIS restriction endonucleases MlyI and PleI,

Nucleic Acids Res., 29, 2492-2501, https://doi.org/

10.1093/nar/29.12.2492.

19. Kachalova, G. S., Rogulin, E. A., Yunusova, A. K.,

Artyukh, R. I., Perevyazova, T. A., Matvienko, N. I.,

Zheleznaya, L. A., and Bartunik, H. D. (2008) Struc-

tural analysis of the heterodimeric type IIS restric-

tion endonuclease R.BspD6I acting as a complex

between a monomeric site-specific nickase and a

catalytic subunit, J. Mol. Biol., 384, 489-502, https://

doi.org/10.1016/j.jmb.2008.09.033.

20. Malone, T., Blumenthal, R. M., and Cheng, X. (1995)

Structure-guided analysis reveals nine sequence mo-

tifs conserved among DNA amino-methyltransferases,

and suggests a catalytic mechanism for these enzymes,

J. Mol. Biol., 253, 618-632, https://doi.org/10.1006/

jmbi.1995.0577.

21. Yang,Z., Horton, J.R., Zhou,L., Zhang, X.J., Dong,A.,

Zhang,X., Schlagman, S.L., Kossykh,V., Hattman, S.,

and Cheng, X. (2003) Structure of the bacteriophage

T4 DNA adenine methyltransferase, Nat. Struct. Biol.,

10, 849-855, https://doi.org/10.1038/nsb973.

22. Horton, J. R., Liebert,K., Hattman,S., Jeltsch, A., and

Cheng, X. (2005) Transition from nonspecific to spe-

cific DNA interactions along the substrate-recognition

pathway of dam methyltransferase, Cell, 121, 349-361,

https://doi.org/10.1016/j.cell.2005.02.021.

23. Horton, J. R., Liebert, K., Bekes, M., Jeltsch, A., and

Cheng, X. (2006) Structure and substrate recognition

RM SYSTEMS: FUNCTIONALITY AND STRUCTURE 521

BIOCHEMISTRY (Moscow) Vol. 90 No. 4 2025

of the Escherichia coli DNA adenine methyltransfer-

ase, J.Mol. Biol., 358, 559-570, https://doi.org/10.1016/j.

jmb.2006.02.028.

24. Nell,S., Estibariz,I., Krebes,J., Bunk,B., Graham, D.Y.,

Overmann, J., Song, Y., Spröer, C., Yang, I., Wex, T.,

Korlach,J., Malfertheiner,P., and Suerbaum,S. (2018)

Genome and methylome variation in Helicobacter

pylori with a cag pathogenicity island during early

stages of human infection, Gastroenterology, 154,

612-623, https://doi.org/10.1053/j.gastro.2017.10.014.

25. Friedrich, T., Fatemi, M., Gowhar, H., Leismann, O.,

and Jeltsch, A. (2000) Specificity of DNA binding

and methylation by the M.FokI DNA methyltransfer-

ase, Biochim. Biophys. Acta, 1480, 145-159, https://

doi.org/10.1016/s0167-4838(00)00065-0.

26. Tomilova, J.E., Kuznetsov, V.V., Abdurashitov, M. A.,

Netesova, N.A., and Degtyarev, S.K. (2010) Recombi-

nant DNA-methyltransferase M1.Bst19I from Bacillus

stearothermophilus 19: purification, properties, and

amino acid sequence analysis, Mol. Biol., 44, 606-615,

https://doi.org/10.1134/S0026893310040163.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in published

maps and institutional affiliations. AI tools may have

been used in the translation or editing of this article.