The effect of methylation outside the recognition sequence of restriction endonuclease PvuII on its cleavage efficiency

Gene, 113 (1992) 89-93 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0378-1119/92/$05.00

89

GENE 06367

The effect of methylation outside the recognition sequence of restriction endonuclease Pvull on its cleavage efficiency (Methylated DNA; unmethylated DNA; specific inhibition; preferential cleavage; kinetics of digestion)

Qiang Liu, Xiongwei Chen, Xilin Zhao, Yawen Chen and Defeng Chen Institute.for Molecular Biology, Nankai University. Tianjin 300071 (People's Republic of China) Received by R. Wu: 6 July 1991 Revised/Accepted: 25 November/27 November 1991 Received at publishers: 13 January 1992

SUMMARY This study is to extend our earlier observation that Dam and Dcm methylation outside the Pvull recognition sequence inhibited Pvull cleavage in one of the three PvulI sites of pGEM4Z-ras DNA. In this paper, a new recombinant plasmid DNA, pGEM4-SV4Oori-anti-ras, was constructed which has only two Pvull sites, I and II. The Dam and Dcm-methylated and unmethylated DNAs were produced in Escherichia cob and linearized by Seal. The DNA molecules were digested with different amounts of PvuII. The results show that by comparing the DNA fragment number and intensity of the partial and final products in agarose gel, Pvull site I on the methylated DNA molecule was digested four- to eight-fold more slowly than site II. In the unmethylated plasmid DNA, the two Pvull sites were ,:,leaved at about the same rate. The difference was caused only by methylation of Dam and Dcm sites outside the PvuII recognition sequence. A methylated Dam site immediately adjacent to the less efficiently cut Pvull site I may be responsible for the inhibitory effect. We suggest that a new parameter, involving methylation of sites outside the recognition sequence, be considered in kinetic experiments on cleavage.

INTRODUCTION The type-lI ENase, Pmll, recognizes and cleaves the DNA to produce blunt-end fragments (Gingeras et al., 1981) 5'-CAG/CTG GTC/GAC

Correspondencew: Dr. D. Chen, Institute for Molecular Biology,Nankai University, Tianjin 300071 (P.R. China) Tel. (86-22)31.5992, ext. 419; Fax (86-22)34.4853. Abbreviations: bp, base pair(s); BSA, bovine serum albumin; DTT, dithiothreitol; ENase, restrictionendonuclease;kb, kilobase(s)or 1000bp; MTase methyltransferase; nt, nucleotide(s); ras gene, c-Ha-ras gene; SV4Oori, SV40 origin of DNA replication and early promoter region.

Two groups reported that methylations within its recognition sequence 5 ' - C A G/5"CTC GTS'"C/ GAC

or

5 ' - C A G/4'"CTC GT4'"C/ GAC

block its cleavage activity (Dobritsa and Dobritsa, 1980; Butkus et al., 1987). All previous reports indicate that only site-specific methylation within the recognition sequence can block or inhibit DNA cleavage of an ENase (Nelson and McCleiland, 1989). We have found previously that methylation outside the recognition sequence inhibited cleavage of pGEM4Z-ras DNA molecule which has three PvuII sites. One site was digested about sixteen-fold less efficiently than either of the other two sites by Pvull because of methylation outside the recognition sequence (Chen et al., 1991). In this paper, we constructed a new recombinant plasmid pGEM4-SV4Oori-

90 anti-ras DNA, which has only two Pvull sites, I and II. When this plasmid DNA was cleaved by PvuII, the PvuII site I was cleaved about four- to eight-fold slower than the site II due to methylation outside the recognition sequence.

TABLE I The final and partial products of the ScaI-linearized DNA as digested by Pvull Product

Fragment~

Fragment size (bp)

Final

1--11 I I--S

990 1724

S--I

1167

I--II--S S--I--ll

2714 2157

EXPERIMENTAL AND DISCUSSION

(a) Construction of the recombinant plasmid DNA The plasmid pGEM4-SV4Oori-,mti-ra~ DNA was constructed by first inserting the SV40 ori (including early promoter region; 342 bp) into the ,,ector pGEM4 between the Hi, dill and Pvull sites to produce pGEM4-SV4Oori, and then inserting the ras eDNA fragment (775 bp) (Yu et al., 1989; Fasano et al., 1983) in the opposite direction into the pGEM4-SV4Oori between the Pstl and Sinai sites. The map of pGEM4-SV4Oori-anti-ras D N A is shown in Fig. 1.

(b) The cleavage of the methylated and unmethylated DNAs by Pv.ll The entire sequence of the plasmid pGEM4-SV4Oorianti-ras D N A was already known and there are two PvulI sites I and II located at bp 1 and 990, and a Scal site at bp 2714 as shown in Fig. 1. The products of partial and complete digestion of the ScaI-iinearized DNA by Pvull are shown in Table I. The Dam and Dcm-methylated plasmid DNA was isolated from E. coil HB101 (dam +, dcm + ), whereas unmethylated plasmid DNA was isolated from E. coil G M I I 9 (dam-, dcm- ) according to the modified method of alkaline lysis described by Birnboim and Doly (1979), and then the DNAs were linearized by Seal. To test the different cleavage efficiencies of the plasmid DNA by PvulI due to methylation outside the recognition sequence, the Scallinearized methylated and unmethlylated plasmid DNAs were digested with different amounts of Pvull. The results are illustrated in Figs. 2 and 3. In Fig. 2, the Scal-linearized methylated plasmid D N A

Partial

" I and II represent Pvull sites I and II, separately, and S, the Scal site. was cleaved with different amounts of PvulI per #g of DNA. In sample 1, the D N A was digested to the three final products by PvuII, whereas the D N A in samples 2-8 was not completely cleaved by Pvuil. From the D N A band patterns of samples 1-8, we noticed the two facts explained below. First, the intensity of the D N A band of the partial products I--II--S (2714 bp) was far weaker than the partial product S--I--II (2157 bp), even though the larger size of the I--II--S D N A fragment was not taken into consideration. Second, the final product II--S (1724 bp) was about four- to eight-fold more abundant than the final product S--I (1167 bp)in samples with lower amounts of PvuII. The M

0

W

1

23

4

56

~=~ ~

7

8

W I

ScaI-linearized P--~tlal: I--II--S Partial: S--I--II

;

Final: II--S Final: S--I

Final: I--II

.PvuII site I ( 1 )

ras SV40or__~i pGEM4-SV40ori-anti-ras--Sc___~aI (2714)

,

(3881 bp)

Pv___uuII site II ,

(990)

Fig. 1. The map ofplasmid pGEM4-SV4Oori-anti-rasDNA, showingthe positions of two Pruli sites and one Seal site. The position at the PvuII cut site I is at the nt + ! (start point ol the sequence).

Fig. 2. Digestionof Sea ldinearized methylated plasmid (shown in Fig. I) by PvuII. Lane M: ). DNA digested with Hi, dIII; Lane 0, undigested DNA; Lanes 1-8, DNA digested with different amounts of Pvull from 16 units to 0.125 unit per #g DNA. The ScaI.linearized methylated plasmid DNA from E. coli HB lOl[pGEM4-SV4Oori-anti.ras] (dam+, dcm +) was cleaved by Pvull in the buffer(50 mM NaCl/10 mM Tris'HCl/10 mM MgCI2/l mM DTT pH 7.9 at 25°C) provided by NE Biolabs (Beverly, MA), at the final volume of 20 #1. The Pvull (NE Biolabs) was dilated as to obtain 16, 8, 4, 2, 1, 0.5, 0.25 and 0.125 units per #g of DNA for samples 1-8, respectively. The digestion reaction was incubated at 37°C for 2 h, then stopped by adding EDTA to 20 mM final concentration. The DNA samples and 2 DNA/HindIII marker were electrophoresed in an 1.2% agarose gel/TAE buffer (0.04 M Tris.acetate/0.001 M EDTA pH 8.0/0.5 #g ethidium bromide per mi). Electrophoresis was carried out at " -3 volt/cm for 5-6 h. Photographs were taken under short wave ultraviolet illumination.

91 cleavage efficiencies of the Pvull sites I and II are calculated based on the band intensity of the final product I--S divided by a correction factor of 1167 and that of II--S divided by a correction factor of 1724. The correction factors are equivalent to the bp size of the corresponding DNA fragments. Thus, we have demonstrated that the Pvull site I in the methylated DNA is cleaved four- to eight-fold less efficiently than the site II. In Fig. 3, the Scal-linearized unmethylated DNA was digested to the three final products by Pvull in samples 1 and 2, but in samples 3-8 it was partially digested. The DNA band pattern of Fig. 3 differed from that of Fig. 2 in two respects. First, the partial products ofl--II--S (2714 bp) and S--I--II (2157 bp) existed in approximately the same amount, if the different sizes of these two DNA fragments were considered. Second, the bands of the final products of S--I (1167 bp) and II--S (1724 bp) appeared at approximately the same rate as samples 8-3. According to the equations of the cleavage efficiencies mentioned above, we can conclude that the two Pvull sites were digested at about the same efficiency in Fig. 3.

(c) The specific inhibition of Pvull cleavage caused by methylation outside the Pvull recognition sequence Dam and Dcm MTases in E. coli HB101 strain methylate G6mATC and cSmC~GG DNA sequences, respectively (May and Hattman, 1975; Brooks and Roberts, 1982). However, E. coli GMI 1o strain which is dam- and dcm- cannot modify these two nt in DNA mentioned above (Arraj and Marinus, 1983). The plasmid pGEM4-SV4Oorianti-ras DNA has many Dam and Dcm recognition and modification sites as shown in Table II. It is generally accepted that plasmid DNAs in E. coli strain (dam ÷, dcm +) are totally methylated (Dreiseikelmann et al., 1979; Hattman, 1981). We also digested the plasmid DNAs isolated from E. coil GMII9 and E. coil M

0

1

23

4

56

7

8

TABLE II The possible Dam and Dcm-methylated sites on the plasmid sequence Methylated site"' Dam

*4, 228, 1813, 1883, 1899, 1907, 1985, 1997, 210! 2443, 2467, 2507, 2765, 2782, 2818, 3494, 3585.

Dem

100, 115, 292, 336, 880, 897, 1274, 1395, 1408.

The methylated sites on the plasmid sequence are shown starting from the Pvull site I. *4 is the Dam-methylated site nearest to the Pvull site I.

HBI01 transformants with Dpnl, MboI and Apyl. These three enzymes can recognize and cleave the methylated sequence G6mATC modified by the Dam MTase, the unmethylated sequence GATC and the methylated sequence cSmC~-GG modified by the Dcm MTase, respectively. We observed that the plasmid DNA from E. coil GM119 can be digested by Mbol (Fig. 4, lane 2) but not by DpnI (lane 1) and Apyl (lane 3). On the other hand, the plasmid DNA from E. coil HB 101 cannot be cleaved by Mbol (lane 5) but can be digested by Dpnl (lane 4) and Apyl (lane 6). It is important to note that sample 4 was cut to give the 300bp and 224-bp fragments. This is possible only if the two A residues at the nt positions 4 and 5 are methylated and then cut by Dpnl. Thus we conclude that the plasmid DNA

M

0

1

2

3

4

5

6

Nicked 2573 bp Supercoiled

1585 bp

linearized al: I--II--S al: S--I--II : II--S : S--I : I--II

Fig. 3. Digestion ofScal-linearized unmethylated plasmid DNA by Pvuli. The unmethylated DNA was isolated from E. coil GMll9[pGEM4SV4Oori-anti-ras] (dam-, dcm- ). The symbols and other conditions are the same as described in Fig. 2.

676 544 377 341 300 258 224

bp bp bp bp bp bp bp

Fig. 4. Digestion patterns by Dpnl, Mbol and Apyl of the plasmid DNAs produced in E. coil GMII9 and HBI01. The DNA samples were fractionated on an !.5% agarose gel. Lane M, 2 DNA digested with EcoRI and HindIII; lane 0, undigested plasmid; lanes 1-3, plasmid from E. coil GM 119 (see Fig. 3); lanes 4-6, the plasmid DNA from E. coli HB 101 (see Fig. 2); lanes 1 and 4, digested by Dpnl; lanes 2 and 5, digested by Mbol; lanes 3 and 6, digested by Apyl.

92 TABLE I11 Two Pvull sites in pGEM4-SV4Oori-anti-rasDNA, their adjacent sequences and the methylated site nearest to the Pvull site I Pvull site

The nt sequence and the methylated site~

I

5'-AGT GCG CTG ACC ATC CAG/CTG *AT C CAG AAC CAT TTT TCA CGC GAC TGG TAG GTC/GAC TA* G GTC TTG GTA AAA 5'-CTG ACA CAC ATT CCA CAG/CTG CAT GAC TGT GTG TAA GGT GTC/GAC GTA

TAA TGA ATC GGC ATT ACT TAC CCG

CAG/CTG is the recognition and cleavage site of Pvull (underlined). " Sequence_GTC/GAC *A is the Dam.methylated site nearest to the Pvull site I.

from E. coli HBI01 was completely methylated in the sequences G6mATC and CsmC~GG, but the plasmid D N A from E. coli G M 119 was not modified at these two sites. Table III shows the sequences surrounding the two PvulI sites and the nearest methylated site to the Pvull site I on the plasmid sequence. Preferential cleavage means that an ENase cleaves some recognition sites within a given D N A molecule more easily than other sites. This phenomenon was previously explained to be due to the adjacent nt sequence or the proximity to the termini of a linear D N A molecule (Armstrong and Bauer, 1982; 1983; Yoo and Agarwal, 1980; Alves et al., 1984). The influences of the adjacent sequence were believed to have contributed to local structural changes of the DNA double-helices. Alternatively, an ENase may span a cegion on DNA larger than its recognition sequence. Our results show that Pvull did digest the methylated plasmid about four- to eight-fold less efficiently at Pvull site I than at PvulI site I1, On the other hand, the two Pvull sites on the unmethylated plasmid D N A were cleaved at about the same rate, Thus, the difference between the methylated and unmethylated DNA can be attributed only to the methylation outside the Pvull recognition sequence. We also examined the digestion reactions under other conditions such as using the Pvull specific buffer supplied by Promega (Madison, WI) (6 mM Tris.HCl pH 7.4 at 37 °C/50 mM NaCI/6 mM MgCI2/6 mM 2-mercaptoethanol/0.1 mg BSA per ml), using a lower amount of Pvull but with a longer incubation time, using a lower temperature such as 25°C, or using Pvull samples from different companies, i.e., BRL Life Technologies (Gaithersburg, MD) and Promega. However, there was no apparent difference under different conditions (data not shown). From Tables II and Ill, we can see many Dam and Dcm-modified sites on the methylated plasmid D N A prepared from E. coli HB 101. However, only one Dam site is near the Pvull site I. Thus, we suggest that the methylation immediately adjacent to the Pvull site I may be responsible for the less efficient cleavage of this site, even though

the effect(s) of the other methylated site(s) could not be excluded completely.

td) Conclusions (I) Methylation outside the recognition sequence caused specific inhibition of Pvull cleavage in the methylated pGEM4.SV40-ori-anti-ras DNA, in which the Pvull site I was digested four- to eight-fold less efficiently than the site II. (2) We propose that methylation outside the recognition sequence may specifically inhibit the cleavage activity of an ENase, thus it may be necessary to consider it as a new factor in kinetic experiments. REFERENCES Aires, J., Pingoud, A., Haupt, W., Langowski, J., Peters, F., Maass, G. and Wolff,C.: The influence of sequence adjacent to the recognition site on the cleavage of oligodeoxynucleotides by the EcoRl endonu¢lease. Eur. J. Biochem, 140 (1984) 83-92. Armstrong, K.A. and Bauer, W.R.: Preferential site-dependent cleavage by restriction endonuclease Pstl. Nucleic Acids Res. 10 (1982) 9931007. Armstrong, K.A. and Bauer, W.R.: Site-dependent cleavage of pBR322 DNA by restriction endonuclease Hinfl. Nucleic Acids Res. 11 (1983) 4109-4126. Arraj, J.A. and Marinus, M.G.: Phenotypic reversal in dam mutants of Escherichiacoil K-12 by a recombinant plasmid containing the dam gene. J. Bacteriol. 153 (1983) 562-%5. Birnboim, H.C. and Doly, J.: A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7 (1979) 1513-1523. Brooks, J.E. and Roberts, R.J.: Modification profiles of bacterial genomes. Nucleic Acids Res. 10 (1982) 913-934. Butkus, V., Klimasauskas, S., Petrauskiene, L., Maneliene, Z., Lebionka, A. and Janulaitis, A.: Interaction ofA/ui, Cfr61 and Pvull restrictionmodification enzymes with substrates containing either N 4methylcytosine or 5-methylcytosine. Biochim. Biophys, Acta 909 (1987) 201-207. Chen, D., Liu, Q., Chen, X., Zhao, X. and Chen, Y.: The inhibition of restriction endonuclease Pvull cleavage activity by methylation outside its recognition sequence. Nucleic Acids Res. 19 (1991) 57035705. ÷

93 Dobritsa, A.P. and Dobritsa, S.V.: DNA protection with the DNA methylase M.Bbvl from Bacillus bre~,is tar. GB against cleavage by the restriction endonuclease Pstl and Pvull. Gene 10 (1980) 105112. Dreiseikeimann, B., Eichenlaub, R. and Wackernagei, W.: The effect of differential methylation by Escherichia coli of plasmid DNA and phage T7 and ,;~DNA on the cleavage by restriction endonucleaseMbol from Moraxeila bin,is. Biochim. Biophys. Acta 562 (1979)418-428. Fasano, O., Taparowsky, E., Fiddes, J., Wigler, M. and Goldfarb, M.: Sequence and structure of the coden region of the human H-ras-I gene from T24 bladder carcinoma cells. J. Mol. Appl. Genet. 2 (1983) 173-180. Gingeras, T.R., Greenough, L., Schildkraut, I. and Roberts, R.J.: Two new restriction endonucleases from Proteus vulgaris. Nucleic Acids Res. 9 (1981) 4525-4536.

Hattman, S.: DNA methylation. In: Boyer, P.D. (Ed.), The Enzymes, Vol. XIV, Part A. Academic Press, New York, 1981, pp. 517-548. May, M.S. and Hattman, S.: Analysis ofbacteriophage deoxyribonucleic acid sequence methylated by host- and R-factor-controlled enzymes. J. Bacteriol. 123 (1975) 768-770. Nelson, M. and McCleiland, M.: Effect of site-specific methylation on DNA modification methyltransferases and restriction endonucleascs. Nucleic Acids Res. 17 (1989) Sup. r389-r415. Yoo, O.J. and Agarwal, K.L.: Cleavage of single strand oligonucleotides and bacteriophage ~pX174 DNA by Mspl endonuclease. J. Biol. Chem. 255 (1980) 10559-10562. Yu, Z., Chen, D., Black, R.J., Blake, K., Ts'o, P. O.P., Miller, P. and Chang, E.H.: Sequence sp :ific inhibition of in-vitro translation of mutant or normal ras P21. J. Exp. Pathol. 4 (1989) 97-106.