Valery A. Chernukhin, Danila A. Gonchar, Murat A. Abdurashitov, Tatyana N. Nayakshina, Vladimir S. Dedkov, Natalya A. Mikhnenkova, Elena N. Lomakovskaya and Sergey Kh. Degtyarev

SibEnzyme Ltd., Novosibirsk, 630117, Russia; E-mail: gonchar@sibenzyme.ru

We have discovered a bacterial strain Lysinibacillus manganicus An22 that produces the new prototype of restriction endonuclease named LmnI. This enzyme recognizes a nonpalindromic DNA sequence 5’-GCTCC-3’/3’-CGAGG-5’.
The LmnI restriction endonuclease preparation with the concentration of 1000 units/ml was isolated using four chromatographic steps. It was shown that new enzyme cuts its recognition sequence forming 3’-protruding ends as indicated by the arrows: 5’-GCTCCN↓-3’/3’-CGAG↑GN-5’, and can be applied to a IIS type of restriction-modification systems.

Abbreviations: bp – base pairs; DTT – Dithiothreitol; ENase – restriction endonuclease; MTase – DNA-methyltransferase; PAAG – polyacrylamide gel; RM system – restriction-modification system; Tris – Tris(hydroxymethyl)aminomethane.

INTRODUCTION

Site-specific restriction endonuclease (ENase) along with appropriate DNA-methyltransferase (MTase) represent bacterial restriction-modification system. Type II restriction endonucleases are most studied and demanded ones in molecular biology researches, they are widely used in long DNA mapping, gene cloning, DNA-diagnostics and many other fields. Type II restriction endonucleases recognize short sequences of double-stranded DNA, 4 to 8-bp, and specifically cleave them inside or at a fixed distance outside their recognition sequences. The latter recognize asymmetric (non-palindromic) sequences and, according to temporary classification [1], belong to the IIS (from “shift”) subtype. Currently known more than 90 such enzymes, recognizing unique asymmetric sites and appear to be prototypes [2].

This study is dedicated to determination of the substrate specificity of LmnI restriction endonuclease, which was isolated from bacterial strain Lysinibacillus manganicus An22 and appears to be new prototype.

MATERIALS AND METHODS

Reagents from following manufacturers were used in this study: “Sigma” (USA), “Fisher” (USA), “Panreac” (Spain) and “Helicon” (Russia). Next carriers were used for chromatographic purification of the enzyme: Phenyl-sepharose (“Sigma”, USA), Hydroxyapatite and Bio-Gel A-0,5m (“BioRad”, USA). The biomass production of bacterial strain Lysinibacillus manganicus was carried out using components of the nutrient medium made by ”Organotechnie” (France).

Enzymes, DNA, deoxynucleoside triphosphates, synthetic oligonucleotides and molecular weight markers of 1 kb, used in this study, were from “SibEnzyme” Ltd (Russia). γ[³²P]-labeled DNA duplex was used for LmnI cleavage position determination. Initial oligodeoxyribonucleotides were labeled using T4 polynucleotide kinase and γ[³²P]-ATP. To separate the products of enzymatic hydrolysis of labeled oligonucleotides, electrophoresis was carried out in 20% polyacrylamide gel containing 7 M urea.

The morphological and physico-biochemical properties of the strain were studied using known techniques [3]. The type of microorganism was determined according to the Bergey’s manual of determinative bacteriology [4] and according to the results of sequencing of the 16S rRNA gene fragment. pBR322 plasmid DNA [5] isolation was performed using the “QIAGEN GmbH” kit (Germany) according to the manufacturer’s protocols.

The sequencing of the PCR fragment of the 16S rRNA gene was performed by the Sanger method on ABI 3130xI Genetic Analyzer automatic sequencer (Applied Biosystems, USA) according to the manufacturer’s protocols. Calculation of theoretical pictures of the hydrolysis of DNA substrates with a known sequence was made using Vector NTI Suite 7 software.

Growing cells of L. manganicus An22 strain.

Cells of the strain Lysinibacillus manganicus An22 were grown in a thermoshaker at 30^оС in 0.5-liter flasks containing 0.3 L of LB nutrient medium (1% tryptone, 0.5% yeast extract, 0.5% NaCl, pH 7.6) with the addition of 0.1% MgSO₄ and 0.001% thiamine, with stirring at 120 rpm. The culture grew within 18-20 hours to an optical density of A₅₅₀=1.9±0.3. The cells were collected by centrifugation at 8000 g for 15 min in J2-21 “Beckman” centrifuge (USA). Biomass was stored at -20°C.

Lmn I activity assay

To test the activity of LmnI restriction endonuclease in cells, 1 ml of the grown culture was transferred into 1.5 ml Eppendorf tubes and centrifuged at 6000 g for 2 min in Eppendorf 5804 microcentrifuge (Germany). The precipitate was suspended in 180 μl of water, and a crude cell lysate was prepared according to the method described previously [6]. Then 5 μl aliquot of crude lysate was added to 20 μl of the reaction mixture containing 1 μg of phage Lambda DNA in SE-buffer “B” (10 mM Tris-HCl (pH 7.6), 10 mM MgCl₂, 1 mM DTT), and the multiple dilutions in 4 and 16 times was performed. The mixture was incubated at 37^оС for 1 hour. The reaction products were applied on a 1% agarose gel and electrophoresis was performed in Tris-acetate buffer (50 mM Tris-acetate (pH 8.0), 20 mM Sodium acetate, 2 mM EDTA) at a voltage of 150 V. After staining with ethidium bromide, the gel was photographed in UV light.

To test the activity of LmnI in the chromatographic profile, 1 µl aliquots of the fractions were added to 20 µl of the reaction mixture, that was incubated for 10 min at 37^оС, and then applied on a 1% agarose gel and electrophoresis was performed in Tris-acetate buffer.

To determine the sequence specificity of LmnI ENase, 2 μl of the enzyme preparation was added to 20 μl of the reaction mixture containing 1 μg of substrate DNA in reaction SE-buffer “B”. The mixture was incubated at 37^оС for 2 hours, after that 10 μl aliquots were applied onto a 1% agarose gel and electrophoresis was carried out in Tris-acetate buffer. The resulting hydrolysis patterns were compared with theoretically calculated ones.

Determination of LmnI DNA cleavage position

As a substrate for determining the position of hydrolysis by the LmnI ENase on the upper and lower strands of DNA, we used an oligonucleotide duplex consisting of mutually complementary deoxyriboligonucleotides Lmn1d and Lmn2r, each 38 nucleotides in length:

Lmn1d: 5’-CCCTTTCCTCTGCCCGCGGAGCTTGATGATGCTTTCCC-3’
Lmn2r: 5’-GGGAAAGCATCATCAAGCTCCGCGGGCAGAGGAAAGGG-3’

This duplex contains the putative LmnI recognition site and the recognition sites of the control ENases: MnlI, SfaNI and Sfr303I. When setting up the reaction with LmnI and control enzymes, the duplex was previously modified by incorporating a radioactive label into the upper or lower chain. The desired chain was labeled from the 5′-end using T4 Polynucleotide Kinase and γ[³²P]-ATP, and then purified from contaminations by gel-filtration on Sephadex G-50, a complementary unlabeled chain was added, the tube with the reaction mixture was heated for 5 minutes at 95^оС, and then cooled on the desktop to room temperature. The reaction mixture with a volume of 20 μl contained SE-buffer “B”, labeled oligonucleotide duplex at a concentration of 65 nM, and 2 μl of restriction endonuclease. The mixture was incubated for 1 hour at a temperature of 37^оС, after which 10 μl aliquots were applied on 20% polyacrylamide gel with 7 M urea and electrophoresis was performed in Tris-borate buffer. Gel radioautography was performed using a Cyclone Storage System instrument (Packard Instrument Co., USA).

Purification of LmnI ENase

All stages of the enzyme purification were carried out at 4^оС using following solutions:

Buffer A – 20 mM Tris-HCl, pH 7.6, 7 mM mercaptoethanol;
Buffer B – 50 mM Tris-HCl, pH 7.6, 7 mM mercaptoethanol;
Buffer K – 10 mM K-phosphate, pH 7.2, 7 mM mercaptoethanol.

Extraction. 20 g of biomass was suspended in 60 ml of buffer A with the addition of 0.2 M NaCl, 1 mM phenylmethylsulfonyl fluoride, 0.1 mg/ml lysozyme, 0.1% Triton X-100. The cells were sonicated on Soniprep 150 disintegrator (“MSE”, England) with an adapter diameter of 2 cm. The treatment was carried out at an amplitude of 20 μm 10 times in 1 min at 1 min intervals, cooling the suspension in an ice bath. The cell debris was removed by centrifugation at 15000 rpm for 30 minutes in JA-20 rotor using J2-21 centrifuge (Beckman, USA).

The total protein was precipitated from the extract by adding ammonium sulfate to 60% and centrifuged at 12000 rpm for 20 minutes. The precipitate was dissolved in 60 ml of Buffer A.

Gel-filtration. The resulting 60 ml fraction was loaded on a Bio-Gel A-0.5m column (“Bio-Rad”, USA, V = 500 ml), equilibrated with Buffer A with addition of 0.4 M NaCl, 0.1% Triton X-100, and washed with one column volume of the same buffer at a rate of 30 ml/h. 50 fractions of 10 ml were collected. According to the result of the LmnI activity assay, fractions 25-40, containing the target activity, were pooled. The resulting fraction was precipitated by adding ammonium sulfate to 60% and centrifuged for 40 min at 12000 rpm. Then the precipitate was dissolved in 25 ml of Buffer K.

Chromatography on Hydroxyapatite. The resulting fraction was dialyzed against 2 liters of Buffer A for 16 hours, applied to a column with 20 ml of Hydroxyapatite, equilibrated with Buffer K, and washed with two column volumes of the same buffer. Enzyme elution was performed with a linear gradient of K-phosphate buffer, pH 7.2, from 0.04 to 0.25 M K-phosphate with a volume of 350 ml. 60 fractions of 5.8 ml were collected. Active fractions 39-59 (0.18 M K-phosphate) were pooled.

Chromatography on Phenyl-sepharose. 1 M Tris-HCl, pH 7.6, to a concentration of 60 mM, glycerol to a concentration of 5%, 4 M NaCl to a concentration of 0.8 M, dry ammonium sulfate to a concentration of 1.7 M were added to the obtained fraction, and it was applied to a column with 4 ml of Phenyl-sepharose, equilibrated with Buffer B, containing 1.7 M ammonium sulfate. The column was washed with 2 column volumes of the same buffer. Enzyme elution was performed with a linear gradient of ammonium sulfate from 1 to 0 M in Buffer B, with a volume of 160 ml. Of the 40 fractions №29-39 (0.12 M ammonium sulfate), containing the target enzyme, were combined.

Rechromatography on Hydroxyapatite. The obtained fraction was dialyzed against 1 liter of Buffer A for 16 hours, applied to a 1.5 ml Hydroxyapatite column equilibrated with buffer K and washed with two column volumes of buffer K. Enzyme elution was performed with a linear gradient of K-phosphate buffer, pH 7.2, from 0.04 to 0.25 M K-phosphate with a volume of 100 ml. Of the 40 fractions №15-18, containing a peak of LmnI activity, were pooled.

Concentrating and storage of the enzyme preparation. The obtained fraction was dialyzed for 20 hours against 300 ml of buffer A containing 0.25 M NaCl and 50% glycerol, and stored at -20^оC. One unit of LmnI activity was defined as the amount of enzyme required to complete digestion of 1 μg of phage Lambda DNA in 50 μl of the reaction mixture for 1 hour at 37^оС.

RESULTS AND DISCUSSION

During the mass screening of bacterial strains isolated from soil and fresh water samples, the soil strain, originally designated as An22, and producing a restriction endonuclease with unknown specificity, was discovered. Samples of soil and fresh water were selected and provided by the leading engineer of the IGM SB RAS Sergey Dementyev. As a result of microbiological studies and sequencing of the 16S rRNA gene fragment, the discovered bacterial strain was identified as Lysinibacillus manganicus An22, and the restriction endonuclease was named LmnI according to the generally accepted nomenclature [7].

As a result of L. manganicus An22 cells growing, 20 g of biomass (4 g per liter) with an enzyme activity of 8-10 units/g was obtained. After a four-stage chromatographic purification, 2.5 ml of LmnI ENase preparation with a concentration of 1000 units/ml was obtained from this amount of biomass.

The optimal conditions for DNA digestion with LmnI were SE-buffer “B” and a temperature of 37^оС (data not shown). The enzyme was inactivated by heating at 65^оС for 20 minutes.

The specificity of LmnI ENase was determined using phage λ, phage T7 DNAs, as well as pBR322 plasmid DNA. Below the patterns of these DNAs hydrolysis with LmnI, obtained experimentally (Fig. 1a), and theoretically calculated patterns of the same DNA cleavage at the site 5’-GCTCC-3’/3’-CGAGG-5’ (Fig. 1b) are presented. It can be seen that the lengths of experimentally obtained DNA fragments coincide with the calculated ones.

Figure 1. Site-specificity determination of LmnI restriction endonuclease on phage λ, phage Т7 DNAs and pBR322 plasmid DNA.

To confirm the recognition sequence for LmnI ENase and to determine its cleavage position in DNA, we used an oligonucleotide duplex consisting of the deoxyribooligonucleotides Lmn1d and Lmn2r. The sequence of nucleotides in the duplex was chosen in such a way that it included the recognition site of the studied enzyme LmnI, as well as the recognition sites of several control restriction endonucleases with a known cleavage positions. Since LmnI recognizes an asymmetric site, based on literature data, it was highly likely that the positions of hydrolysis by the enzyme of the upper and lower strands of DNA would also be asymmetric relative to the center of the recognizable nucleotide sequence.

Figure 2 shows the structure of the Lmn1d/Lmn2r duplex, the recognition sites of LmnI and control enzymes – SfaNI (5’-GCATC(5/9)-3’), Sfr303I (5’-CCGC↓GG-3’) and MnlI (5’-CCTC(7/6)-3’), are marked by the frames. It should be noted that the recognition sites of Sfr303I and LmnI ENases are overlapping by two nucleotide residues (Fig. 2). At the same time, Sfr303I cleaves the upper strand of the duplex from the 5’-terminus before the LmnI site, and the lower strand splits in two nucleotides to the 3’-terminus of the LmnI site. In turn, SfaNI and MnlI recognition sequences are located far from the LmnI site, but their DNA cleavage positions are either within the sequence recognized by LmnI (SfaNI, upper chain), or directly near this sequence (MnlI, upper and lower chains).

Figure 2. The structure of Lmn1d/Lmn2r DNA duplex.

In a series of experiments, the upper or lower chain of the duplex was labeled from the 5’-end as described in “Materials and Methods”. Then, this duplex was digested with LmnI and control enzymes, and after electrophoresis in denaturating PAAG, the bands corresponding to the products of hydrolysis were compared.

Figure 3 shows the radioautograph of PAAG with the result of the Lmn1d*/Lmn2r and Lmn1d/Lmn2r* duplexes digestion with LmnI and the corresponding control ENases after electrophoresis.

Figure 3. Determination of cleavage positions for LmnI with using of γ[³²P]-labeled oligonucleotide duplex.

Lanes: 1 – intact Lmn1d*/Lmn2r duplex; 2 – Lmn1d*/Lmn2r + LmnI; 3 – Lmn1d*/Lmn2r + Sfr303I; 4 – Lmn1d*/Lmn2r + SfaNI; 5 – intact Lmn1d/Lmn2r* duplex; 6 – Lmn1d/Lmn2r* + MnlI; 7 – Lmn1d/Lmn2r* + LmnI; 8 – Lmn1d/Lmn2r* + Sfr303I. Labeled chain is marked by *. LmnI recognition sequence is marked by the frame. Products were separated in 20 % polyacrylamide gel with 7 M urea.

As can be seen from Figure 3, the products of hydrolysis of the Lmn1d*/Lmn2r duplex, containing the labeled upper chain, coincide for the LmnI and SfaNI (lanes 2 and 4, respectively), and the band corresponding to Sfr303I cleavage product is shifted down one position relative to the bands produced by LmnI and SfaNI (lane 3). Therefore, the sequence 5’-GGAGC-3’ is cleaved by LmnI ENase in the region between the first and second guanine residues: 5’-G↓GAGC-3’.

It can also be seen from Figure 3, the products of digestion of the Lmn1d/Lmn2r* duplex containing the labeled lower chain are the same for the MnlI and LmnI (lanes 6 and 7, respectively), and the band corresponding to the cleavage product of Sfr303I is shifted up one position relative to the bands produced by MnlI and LmnI (lane 8). Therefore, LmnI ENase cleaves the 5’-GCTCC-3’ sequence at a distance of one nucleotide from the 3’-terminus: 5’-GCTCCg↓-3’.

Of the foregoing we can conclude that the new enzyme recognizes the non-palindromic DNA sequence 5’-GCTCCN↓-3’/3’-CGAG↑GN-5’ and hydrolyzes it, as shown by arrows to form 3′-protruding “sticky” two-nucleotide ends. Since no other restriction enzymes with the same recognition sequence have been detected so far, LmnI ENase can be considered as a new prototype. Earlier, we assumed that the active center of the enzyme involved in DNA cleavage is more conservative than the recognition sequence, which allows us to pick out entire restriction enzyme groups that have a common DNA cleavage mechanism [8]. In [8], we proposed a scheme for the arrangement of recognition sequences and DNA cleavage sites for a group of 5 enzymes that form 3′-protruding two-nucleotide ends. In the database in 2019 [2], there are already 7 enzymes (including LmnI) that have an asymmetric recognition site and cleave DNA with the formation of 3’-protruding two-nucleotide ends. In addition to LmnI, these are the enzymes BsmI (5’-GAATGC(1/-1)-3’), BsrI (5’-ACTGG(1/-1)-3’), BsrDI (5’-GCAATG(2/0)-3’), BtsI (5’-GCAGTG(2/0)-3’), BstF5I (5’-GGATG(2/0)-3’) and BtsMutI (5’-CAGTG(2/0)-3’). The layout of the recognition sites and DNA cleavage positions for these seven enzymes are presented in Table 1.

Table 1

The table above shows that although the recognition sequences for these enzymes are very different, inside them a common TS/SA pair is found (where S is G or C) located at the same distance from the cleavage position for all the above-mentioned ENases. It can be assumed that this similarity of recognition sites of these seven enzymes is associated with their general mechanism of DNA substrate cleavage. It is worth noting that all the above homologues belong to the IIS subtype of restriction-modification systems, as well as LmnI RM system that we found.

Thus, as a result of a targeted search, we discovered a new prototype, the restriction endonuclease LmnI from the bacterial strain Lysinibacillus manganicus An22, which recognizes the DNA sequence 5’-GCTCC(1/-1)-3’. The enzyme found belongs to a IIS subtype of restriction-modification systems and can be used for molecular biological and genetic engineering work.

REFERENCES

1. Roberts R.J., Belfort M., Bestor T. et al. A nomenclature for restriction enzymes, DNA-methyltransferases, homing endonucleases and their genes. // Nucleic Acids Res. 2003. V. 31. P. 1805-1812.
2. Roberts R.J. The Restriction Enzyme Database. 2019. http://rebase.neb.com/rebase/link_proto
3. Guide to practical classes in microbiology. Ed. N.S. Yegorova. Moscow State University, Moscow, 1995. 224 p.
4. Bergey’s Manual of Determinative Bacteriology. Ninth edition. V.1. Ed. Holt J.G., Krieg N.R., Sneath P.H.A., Staley J.T., Williams S.T. Mir, Moscow. 1997. 432 p.
5. Bolivar,F., Rodriguez,R.L., Greene,P.J., Betlach,M.C., Heyneker,H.L., Boyer, H.W., Crosa, J.H., Falkow,S. Construction and characterization of new cloning vehicles. A multipurpose cloning system. // Gene. 1977. V. 2. P. 95.
6. Belavin P.A., Dedkov V.S., Degtyarev S. Kh. A simple technique for detection of restriction endonucleases in bacterial colonies. // Applied Biochemistry and Microbiology (Moscow). 1988. V. 24. No. 1. P. 121 – 124 (In Russian).
7. Smith, H.O. and Nathans D. A suggested nomenclature for bacterial host modification and restriction systems and their enzymes. // J. Mol. Biol. 1973. V. 81. P. 419-423.
8. Degtyarev, S.Kh., Belichenko, O.A., Lebedeva, N.A., Dedkov, V.S. and Abdurashitov, M.A. BtrI, a novel restriction endonuclease, recognises the non-palindromic sequence 5′-CACGTC(-3/-3)-3′. // Nucleic Acids Res. 2000. Jun 1; 28(11): e56.