Cloning and nucleotide sequence of the genes coding for the Sau961 ...

.:) 1990 Oxford University Press

Nucleic Acids Research, Vol. 18, No. 16 4659

Cloning and nucleotide sequence of the genes coding for the Sau961 restriction and modification enzymes Laszlo Szilak+, Pal Venetianer and Antal Kiss* Institute of Biochemistry, Biological Research Center, Hungarian Academy of Sciences, PO Box 521, 6701 Szeged, Hungary Received June 29, 1990; Revised and Accepted July 23, 1990

ABSTRACT The genes coding for the GGNCC specific Sau96I restriction and modification enzymes were cloned and expressed in E. coli. The DNA sequence predicts a 430 amino acid protein (Mr: 49,252) for the methyltransferase and a 261 amino acid protein (Mr: 30,486) for the endonuclease. No protein sequence similarity was detected between the Sau96l methyltransferase and endonuclease. The methyltransferase contains the sequence elements characteristic for m5C-methyltransferases. In addition to this, M.Sau961 shows similarity, also in the variable region, with one m5C-methyltransferase (M.Sinl) which has closely related recognition specificity (GGA/TCC). M.Sau961 methylates the internal cytosine within the GGNCC recognition sequence. The Sau961 endonuclease appears to act as a monomer.

INTRODUCTION To understand the molecular mechanism by which type II restriction endonucleases and methyltransferases (MTases) recognize their target sequence, the genes of several restriction enzymes and MTases were cloned and sequenced (ref. in 1) Comparison of the deduced amino acid sequences revealed that endonucleases do not share homology with their cognate methylase (2). Endonucleases seem to have very different primary structures, there is only one case where homology was detected between two enzymes (3). MTases can be classified in three groups according to the methylated base they produce: m5C-, m4C- and m6A-MTases. Analysis of the amino acid sequences revealed that MTases of each type show a varying degree of similarity among themselves (2,4,5,). m5C-MTases are the most uniform, they share extensive similarities in their amino acid sequence (6-18). The homologies detected in m5C-MTases suggest a common building plan and evolutionary relatedness for these enzymes. This holds also for prokaryotic m5C-MTases that are not part of a restriction-modification system (19-24) and even for a eukaryotic MTase (25).

correspondence

should be addressed

no:

X53096

More is known about m5C-MTases than about the other two types. A Pro-Cys dipeptide has been proposed to be part of the

catalytic site (26) and in the multispecific m5C-MTases the domain that is responsible for the sequence specific DNA recognition has been identified: it is in the so-called variable region (27). Although there is no experimental evidence for it, the common architecture suggests that also in monospecific m5C-MTases sequence specificity is determined by the variable region (28). Obviously, the detailed structural analysis and enzymological studies of more enzymes with different specificities is necessary to understand the structural basis of sequence specificity. We started a systematic study of restriction and modification enzymes recognizing GGCC and related sequences (6,7,14,20). One restriction system belonging to this group is Sau96I (recognition sequence: GGNCC, ref.29). This system is especially interesting because the amino acid sequence of the SinI endonuclease and methyltransferase has become available (13). The specificity of the SinI system (GGA/TCC) is closely related to that of Sau96I. We report here the cloning and characterization of the genes coding for the Sau96I endonuclease and methyltransferase. These are the first GGNCC specific restriction endonuclease and MTase for which the amino acid sequence has been determined.

MATERIALS AND METHODS Strains and media Staphylococcus aureus PS 96 (29) was obtained from J.S. Sussenbach. E. coli strains ER1398 (mcrB-) and ER1382 (mcrB+, ref.30) were from E. Raleigh. M13 mpl8 and mp 19 phages were grown in E. coli JM107 (31) or JM107 MA2 (mcrB-, ref.32). Bacteria were grown in LB medium at 37°C. Lambdav,r and phi80, phages (obtained from. J.Brooks) were used to test phage restriction.

Enzymes and chemicals Restriction endonucleases were prepared in this institute or were purchased from New England Biolabs. DNA polymeraseI large fragment was from Vepex (Szeged), BAL31 nuclease from New

'Permanent address: Agricultural Biotechnology Center, POB 170, 2100 G6doll6, Hungary *To whom

EMBL accession

4660 Nucleic Acids Research, Vol. 18, No. 16 England Biolabs and modified T7 DNA polymerase (Sequenase) from United States Biochemical Corp. Deoxyadenosine 5'-alpha(35S)thiotriphosphate was purchased from Amersham.

Cloning methods S. aureus PS 96 DNA was prepared by the SDS/proteinase K lysis method (33) with the modification that the cells were pretreated with lysostaphin (29). Isolation of plasmids, transformation of E. coli and agarose gel electrophoresis were done by standard procedures (34).

Determination of the nucleotide sequence DNA fragments were cloned in mpl8 and mpl9 phage vectors (35) or in Bluescript plus vector (Stratagene) and sequenced by the chain termination method (36) using deoxyadenosine 5'-alpha(35S)thiotriphosphate and either DNA polymerase I large fragment or Sequenase. To sequence regions that were not readily accessible for sequencing with the universal sequencing primer, unidirectional deletions were prepared by the combined action of exonuclease Ill and mung bean nuclease (37).

Determination of the methylated base The 246 bp HinfI3363 -PstI3609 fragment of pBR322 was isolated from unmethylated (pBR322) and methylated (pSau3) plasmid DNA. The fragments labeled at the Hinfl end were subjected to pyrimidine specific chemical sequencing reactions (38,39) and run on a 6% sequencing gel.

RESULTS AND DISCUSSION Cloning of the sau96IRM genes Transformation of E. coli by the Sau96I-digested S. aureus PS 96 plasmid library yielded several ampicillin resistant clones. Plasmids isolated from these clones were resistant to Sau961 digestion indicating that they carried the Sau96I methyltransferase gene. All plasmids analysed seemed to be identical by restriction enzyme digestions. One plasmid (pSaul, Fig. 1) was chosen for further analysis. pSaul contained an approximately 3.4 kb insert. The E. coli clone containing pSaul did not restrict unmodified phage, suggesting that pSau did not code for the Sau961 endonuclease. Assuming that the Sau96I methyltransferase and endonuclease genes are linked but the insert cloned in pSau did not contain both genes, we tried to reclone the sau96IM gene on an overlapping fragment. The pSaul insert contains a unique Sall site close to one end. Subcloning experiments revealed that this Sall site lies outside of the methyltransferase gene. S. aureus PS 96 DNA was digested with Sall and several other enzymes and hybridized, in Southern experiments, with a probe representing the central part of the pSaul insert. In the SalI-SphI digest a 3.1 kb hybridizing fragment was identified. This fragment was cloned in E. coli ER1398 using pBR322 as vector and the same type of selection which was used to obtain pSaul. The plasmid (pSau3) isolated from this clone was resistant to Sau96I and the clone restricted unmodified phage (restr. ratio: 10-1-10-2). Since this clone did not restrict phage grown on ER1398(pSaul), we concluded that pSau3 coded, in addition to Sau96I

Cloning of the Sau96I methyltransferase gene S. aureus PS 96 DNA was partially digested with MboI and the digested DNA was ligated to pBR322 that had been cleaved by BamHI and dephosphorylated by bacterial alkaline phosphatase. The ligated DNA was transformed into E. coli ER1398. Plasmid DNA isolated from 27,000 ampicillin resistant transformants was digested to completion with Sau961. The digested DNA was used to transform ER1398. Ampicillin resistant colonies were selected. Purification of Sau%I endonuclease Sau96I endonuclease was purified from S. aureus PS 96 by a published procedure (29). This procedure, however, did not give satisfactory results when the E. coli clone containig the cloned Sau961 genes was used as starting material. To purify Sau96I endonuclease from ER1398(pSau5), cells from a 2 1 dense culture were sedimented by centrifugation, then resuspended in PC buffer (10 mM K-phosphate pH 7.5, lo mM 2-mercaptoethanol, 1 mM EDTA) and disrupted by French press. After centrifugation at I00,OOOxg for 1 h, 1 M NaCl was added to the supernatant and it was loaded on a Bio-Gel AO.5m column. Active fractions were dialysed against PC buffer and loaded on a Mono S FPLC column. Sau96I endonuclease was eluted with a 0-1 M NaCl gradient in PC buffer. Determination of molecular weight by gel filtration The molecular weight of Sau961 endonuclease under native conditions was determined by gel filtration on a Superose 12 HRIO/30 (Pharmacia) FPLC column in a buffer containing 50 mM K-phosphate pH 7.0, 100 mM NaCl, 15 mM MgCl2 and 10 mM 2-mercaptoethanol. Sau96I is active in this buffer (unpublished). Bovine serum albumin, ovalbumin and cytochrome c were used as molecular weight standards.

Fig. 1. Plasmid carrying the Sau96I methyltransferase gene is resistant to Sau96I digestion. 1. pSaul undigested 2. pSaul digested with Sau96I 3. pSaul and pBR322 digested with Sau96I

Nucleic Acids Research, Vol. 18, No. 16 4661

A 0

2

:

8

z _

2.0

2.5 (kb)

a ..

.

-

1.5

1.0

0.5

II

U)

I

:

u

OI

IU

Z

X M LJ

~~~~R

M B

~~~Nrul

Hminrd II TCGCGAATTTATGTATT CTAAAGATATATTTrGAAATAGTTrAGTTrArTTAGTTTAGAAAGCrrTTTTAGATTrTAAAT GGTTT GAAACATACTAAGAAAAACTAATAGACTAAAATAGT 9-,

1 15

*

4

GTTTATTATATAATAAACTTTGGATAGGGTT ATG TGT ATA GGA GGT AGT TTAJJAGA TTG AAC AAA GGA AGT ATA ArT GAA AAA ATG AAA T CAA AAT AT AAA ACA CAM ACT GAA TTA GCT GAM * * N K (m) (c) (i) (9) (g) (s) (CLM R L N K G S I K E K N 0 N T Q T I E L A E I

242

(v) AAA ATT OAT ATA TCT AM AGC CA CTA TCA TTC ATG TTT TCA GAT GAA TAT GAG CCT CTT AAA AAG AAT GTI ATT AAG TTG GCA GAT GTT TTA AAA GTA TCT CCT AAT GAC ATT ATA TTA 362 K I D I S K S Q K L S F N F S D E Y E P L K K N V I L A D V L K V S P N D I I L

EcoRI GAC GAA GMA GAC CM ATG CCA ATT AAC TCT GAT TTT AMT AGA TAT GAT TAT AAA TTA GAT GAA TTC ATA GAT GTA AGC AAT GTA AGA MAG AAT AAA GAT TAT AMT GTA TTC GAA ACC TTT 482 D E E D Q N P I N S D F N R I D Y K L D E F I D V S N V R K N K D Y N V F E T

QD

GCC GGG GCT GGT GGA CTA GCA A Q A G G L A EcoR I ATt GAG AAC OAT ATT GAA TtC I E N D I E F

CtA GGC TTA GAA AGC GCT GGC TtG TCA ACT TAT GGA GCA GTT GAA ATT GAT AAA AAT GCC GCA GAA ACT TTG AGA ATt AAT AGO CCA AAA TGO AAA GTT 602 L

G

L

E

S

A (

L

S

T

Y

G

A

V

E

I

D

A

N

K

A

E

T

AtT GCt GAt AAT TtO GAT GAA TTT ATA GAT GAA GAG ATA GAT ATA TTG AGT GGG GGG TAT CCT TGC CAA QO I A D N L D E F I D E E I D I L S G Y Hirndl I I Ttt GCG GAt ACA AGA GGA ACA CtG TtC TAT CCA TAT AGT AAA ATt TTA AGT AAA TTA AAG CCT AAA GCT TTT ATA GCA GAA AAT GTT AGA F A D t R G t L F Y P Y S K I L S K L K P K A F A c R

L

R

N

R

P

K

W

K

V

ACA TTT AGt TAT GCA GGA AM AGA MT GGA F Y A K

(I (D T O(D)

I

(D

(I

722

RN G

GGA TTA GTC AMT CAC GAC GAT GGC AM ACA 842 G

L

V

N

N

D

D

G

K

T

TtA GM GTA ATG TTA AA 0tt t ATT AAG GAAGGt TAT GAO GIT TAC TOG AAT All Tt MAC TCG t10 AAT TAT GAT GTA GCG CAA AAA AGA GMA AGA ATA GTA ATT ATA GGt ATC AGA 962 L E V ML K V F I K E G(O E V Y W N OK )i I L N S W V A YI D V I () E I G I R GAA GAT TTA GTA AM GM CAG AM TAY CCA ITt AGA ttt CCt TTA OCT CA GTA TAC AMA CCt OTA TtO AM GAT GTA CTG AAA GAT GTT CCA AAM AO AMA 0tt ACA GCA tAC AOT GAC 1082 FP V L E D V L E D V P K S K V t A Y S D E D L V K E Q K Y P F R F P L A O V Y

AAA AAA AGA GM GTA AT AMA TTA 0tt CCA CCA GGA GGC TGT TGG GTA GAt TtA CCt GM C AlA OCT AMA GAY TAT ATG 000 AMA AOl TO TAT tCA 00t GOT 0T1 AMA AGA GGA ATG 1202 K K R E V M K L V PFF 0 I A E D Y M G K S W Y S P C W V D L P E D G IK R O N Npel

OCT A

AGA AGO ATA AGt tGG GAC GAO CCt TGT TTG ACG TtA ACA ACA tCt CCT AOT CAA AM CAA ACA GAO R R I S U D E P C L t L t T S P S O E T E K

Att CAA tCG TtC CCt GAt GAA TGO GAA ID

TtA AMT L

S

C

ND

F

P

TTT AAt F N1

D

E

W

tGA

TAt

TtA

E

TT F

S

G

CTt

V

ATT AAG

G

A

Q (Q

R

a

GTA TTA TtA ACT TAA AtA

I

CtA

A

AAt AMT TtO AGT GAO CAA GAT N N L S E O |

AAT TAT CCA N

PF

GCG GCA

GTi GGT TAT GAT TAT GAt G01 V OY D Y D V

TGT GGC TAT GCA GAA C O Y A E

AM TAT

A

GAO E

TtG CAA AGA CAA OTT GAG AAM L 0 R O V E K

D

V

P

ACT

Att

ACT

T

I

I

AAA TTT TTT A K

1

F

AAt CAT N N

V

N

L

A

K

Y

AAt AGA ATT GAT ACA ttT N R I D T F

GTA OGA ACA V

G

T

TTT CAT GAA AAG TTA F H E K L

AMT N

GAT ATT D I

TCO W

GAA TtT AGA AGT E F R S

AAA CCA AGA TtT TCT CAC CCA AGA GTA AGA ATT GCT tCt GGA GAT CM TTT TAT AM ATT OTT ACT GGA GA GAA GAt OCT TtC K P R F S N P R V R I A S O D D F IE I Y V T E E E A A

jD

Hindlll

I

Gt AM G K

AAM CAA CTA OCT L

GGA

A

[j2jJI

A

IL

°

°

[D

j

L

S

L

V

ATC ACt GAt I

t

GAt

D

M

T

F

D

AtT

GGA

AM

ATA

GAO

GGC

I

G

K

I

E

G

S

TAT Y

A

1322

R

CtG L

N

GAO E

Y

GAt 1562 D

AtG G0t ATT 1682

K

GGA AGT G

I) Y

TCT TtA GtA CAT TAT 1442

AAM ATG ACT TT

M 0

I

TAT ACT Y

1802

T

CAA AAA ATA 1922

TTT F O

K

Att GAT GMA AAC ACA AGA I D E N t R

I

GAA 2042

E

MC Att CCt ATA GCA TTA GAT 2162

N

AAA TCA AGC TTA GGA CTA TtT GCA GM CTT TAT CMA CAA GCA AAG GCA MT MC CGA ACA TTA TCA GA GMA K E G S S L G A N R jL A S E E F A E Y Xbl . EcoRV CCT AAA TCT AAT TAC ATT AGT TTC TAG ATGGATGATTTTAACTTTTAACAAAATCATTGCAACGAAGTTATTATTATTGCTACGMATTATATTGTTTTAATCGGCCTTTATTGCGATATC N I S F P Y ** GAT TGG ATA GAA ACG AAA AGG E T K R

AGA GAO TAT GCT AGA

OCT MO GAlGAT AAT ACA CTT ITT GCA GAA ll AM MT AAM CAT AAT ACT ITO ACC GGT ACT CAT ACT A AT A K D D N T rL F A E I K N [ H T H E |t L T G T

CCA GAC 0CC ATT TGC TAT TAT OTT AGA ATA AtA GAC ACT AAA AGT CGA P D A I C Y Y V R I I D t K S R

OCT AM

T

EcoRI

AOT

CGA CCA ttt AtC (i P F S I R

CAG GtG GGA TtO TATYAACT AAT AA tAT TTA AOT TTt T N1L S F *jt

GAT

A

ACT

TM

CtA TIC GAA t1t ATT GAA TG T ttG TAT ACA GAO TAT GAA AAA GCG CTT GAA GGA AtT GAT Ttt 1I D F L F E C I E F L Y T E Y E 3 A L E

L

R

CAT CCT GAT GAA C H P D E

tCT GGA GA GTA GOT GCA CAA tAC AGA CAA ATA GOT MC OCT GTA CCt GTA AT CtA OCT AAA TAT ATA

TtA AAA CCA

TGG TtA

AGA TOT

I

AtA I

P

I

A

L

D

AtT tCt ATT AAT TAT N Y I S I

2282 2405

Fig. 2. A) Schematic map of the NruI-EcoRV fragment containing the sau96IRM genes. Horizontal arrows indicate the coding regions. B) Nucleotide sequence of the sau961RM genes and the deduced amino acid sequence of the Sau96I MTase and endonuclease. The putative translational initiation codons are boxed, ShineDalgarno sequences are indicated by dots and translational stop codons by asterisk. The sequence of a seven amino acid N-terminal peptide that would be translated if the ATG triplet at position 147 were used as initiation codon is shown in brackets, with lower case letters. Amino acid residues in the MTase which are most highly conserved in m5C-methyltransferases (4) are circled. Amino acid residues in the nuclease that are part of an intramolecular repeat are boxed. Horizontal arrows mark a sequence which is likely to form a hairpin structure in the putative mRNA. A vertical arrow at nucleotide 194 indicates the endpoint of a deletion cloned in pSau6.

4662 Nucleic Acids Research, Vol. 18, No. 16

-.

_

V.

_

AS

...40 40~~: S

4*

4

_*

Fig. 3. Digestion of lambda phage DNA with Sau96I purified from (lane 1) ER1398(pSau5) and S. aureus PS 96 (lane 2).

methyltransferase, also for Sau96I endonuclease. A 2.4 kb NruIEcoRV fragment of the pSau3 insert was cloned in pBR322 to

give pSau5 (Fig.2.). pSau5 was methylated, ER1398(pSau5) exhibited phage restriction and was shown to produce Sau96I endonuclease (Fig.3), thus the NruI-EcoRV fragment codes for the complete Sau96I system (Fig.2). The level of phage restriction shown by ER1398 containing pSau3 or pSau5 is lower than usually found with cloned restriction-modification systems (e.g. BsuRl, ref.7.). A possible explanation for this is the low level of Sau96I endonuclease found in the clone. It was probably due to this low activity, that in order to get a usable enzyme preparation from ER1398(pSau3 or pSau5) we had to modify the original purification procedure described for Sau96I (29). DNA sequence The nucleotide sequence (2405 nucleotides) of the NruI-EcoRV fragment was determined. Two large open reading frames were found (Fig.2): a longer (ATG147-TGA1458) coding for 437 amino acids, and a shorter (ATG1524-TAG2307) coding for 261 amino acids. Since pSaul contained only the longer ORF, this was assigned to the Sau96I methyltransferase and the shorter one to Sau96I endonuclease. The two genes are tandemly arranged with 63 nucleotides between the coding regions. The ORF coding for M.Sau96I contains, in addition to ATG at 147, another potential start codon: GTG at 168 and several less likely initiation codons further downstream. Starting from the NruI site, BAL31 deletions were prepared to identify the functioning initiation codon. The shortened fragments were cloned in pUC18 in the orientation which would enable the sau96IM gene to be transcribed from the lacZ promoter. The fusion points were sequenced. One of the clones (pSau6), in which the first 194 nucleotides were deleted (Fig.2) and the lacZ and sau96IM genes were fused in different reading frames, did not express Sau96I methyltransferase, suggesting that translation

Fig. 4. M.Sau96I methylates the internal C within the GGNCC recognition sequence. 1. Unmethylated DNA-C reaction 2. Methylated DNA-C reaction 3. Methylated DNA-C + T reaction The absence of a band in the methylated DNA indicates the methylated base within the Sau96I site at 3410 in the pBR322 sequence. Part of the sequence is shown at right.

starts upstream from this deletion end point. Of the two potential initiation codons located upstream of position 194, GTG168 is preceded by a much stronger Shine-Dalgarno sequence than ATG147 (16.9 kcal/mol vs. 7.7 kcal/mol, Fig.2, ref. 40). Gram positives are characterized by strong ribosomal binding sites (41) thus we think that GTG168 is the real translational start codon of the Sau96I methyltransferase gene. The reading frame between GTG168 and TGA1458 determines a protein of 430 amino acids (Mr: 49,252). The calculated molecular weight of the endonuclease is 30,486. The sau96IRM genes are highly AT-rich. The coding regions are 67.6% and 70.1 % A+T for the MTase and the endonuclease, respectively. There is no Sau96I recognition site in the sequenced region (2405 bp).

Determination of the methylated base Plasmids containing the sau96IM gene transform the mcrB+ E. coli strain ER1382 several orders of magnitude poorer than the isogenic mcrB- ER1398. Since the mcrB system recognizes GmC or PumC (30), the restriction shown by the mcrB+ strain suggests that M.Sau96I methylates the internal cytosine (or in a less likely case both). The nature of the Sau96I methylation reaction was further investigated by the chemical sequencing method. A fragment of pBR322 DNA isolated from modified and unmodified plasmid was subjected to pyrimidine specific chemical sequencing reactions. The missing band in the modified

Nucleic Acids Research, Vol. 18, No. 16 4663

Sinl Sau961

VIII 255 CJlirerviiicSrdgSRVPFLQPTHSEKGEYGLPKWITLRETITNLKNITHEHVLFPEKRLKYYRLLKEGQYW 254 qkreriviigired-LVKEQKYPFRFPLAQVYKPVLKDVLKDVPKSKVTAYSDKKREVMKLVPPGGCW x x xxx xx

Sinl Sau961

*x

x

x

x

x

Ix 327 KHLPEDLUKEALGKSFFLGGGKTGFLRRVAWDRPSPTLVTHPAMPATDLAHPDELrptsvqeykviq 321 VDLPEQIAKDYMGKSWYSGGGKRGMARRISWDEPCLTLTTSPSQKQIERCHPDETrpfsireyariq xx

x

** **** xx xxx*

*****

***

*

E*****

**

393 387

**

Fig. 5. Amino acid sequence in the variable region of the Sau96I and SinI MTases. Residues in the variable region are typed in upper case and conserved blocks VIII and IX forming the bounderies of the variable regions (16) are typed in lower case. Identical amino acids are marked with asterisk.

DNA indicates that M.Sau96I methylates the inner C in the recognition sequence (Fig.4) Since N4m-cytosine, unlike C5mcytosine, is known to react with hydrazine (43) this observation also suggests that M.Sau96I produces C5m-cytosine.

Comparison of amino acid sequences No amino acid sequence similarity was found between the Sau96I endonuclease and methyltransferase. This seems to be a common feature of type II restriction and modification enzymes and probably reflects the different molecular mechanism by which they recognize their target sequence. M.Sau96I contains all sequence motifs characteristic for m5C-methyltransferases (4,Fig.2). In addition to this, its variable region exhibits striking similarity to the variable region of the SinI methyltransferase (Fig.5). The variable region of multispecific m5Cmethyltransferases is responsible for the sequence specific DNA recognition (27). The common general architecture shared by the multispecific and monospecific enzymes and the identification of a conserved 'core' structure in the variable region of multispecific and monospecific enzymes led to the suggestion that the variable region determines sequence specificity also in monospecific enzymes (28). The similarity shown by the variable regions of M.Sau96I and M.SinI is even more significant if we consider that these enzymes are in taxonomically unrelated bacteria: Staphylococcus, a Gram positive and Salmonella, a Gram negative. It seems likely that the similarity of the putative target recognizing domains of M.Sau96I and M.SinI is not a chance similarity, rather it reflects the use of common structural elements for the recognition of closely related targets. In contrast to the MTases, there is no similarity between R.Sau96I and R.SinI. A search for intramolecular homologies found a pattern of a few amino acids in the N-terminal half of the Sau96I endonuclease which are repeated in the C-terminal half (Fig.2.). The observation of similar twofold and fourfold repeats in a few other type II restriction enzymes led to the assumption that these enzymes evolved by gene duplication (42). The intramolecular repeat detected in R.Sau96I gives further support to this idea. Such intramolecular duplication, however,is not evident at the level of primary structure in all type II restriction enzymes. There are few data available for the subunit structure of type II endonucleases. Some enzymes, e.g. BsuRI (44) act as monomers, others, e.g. EcoRI (45) as dimers. It was interesting to know what subunit structure Sau96I has. We found that, when tested by gel filtration under native conditions, it eluted from the column between ovalbumin (Mr: 43,000) and cytochrome c (Mr:

12,500) indicating that it consists of a single subunit (not shown). It was suggested (42) that it is the symmetrical structure of BsuRI, indicated by the fourfold repeat in the amino acid sequence, that enables this enzyme to act as a monomer. Based on the available data, we think it is possible that type II endonucleases that act as monomers have retained more of the original intramolecular symmetry and, unlike enzymes that evolved as dimers, are characterized by intramolecular repeats evident also at the level of the primary structure. Further data will be needed to see whether such correlation really exists.

ACKNOWLEDGEMENTS We thank Dr.J.S.Sussenbach, E.A.Raleigh, J.Brooks and R.Blumenthal for strains and E.Csorba and E.Komocsin for the technical assistance.

REFERENCES 1. Lunnen,K.D., Barsomian,J.M., Camp,R.R., Card,C.O., Chen,S.-Z., Croft,R., Looney,M.C., Meda,M.M., Moran,L.S., Nwankwo,D.O., Slatko,B.E., Van Cott,E.M. and Wilson,G.G. (1988) Gene, 74, 25-32. 2. Chandrasegaran,S. and Smith,H.O. (1988) in Structure and Expression. Vol.1: From Proteins to Ribosomes. Eds. Sarma,R.H. and Sarma,M.H, Adenine Press, New York, pp. 149-156. 3. Stephenson,F.H., Ballard,B.T., Boyer,H.W., Rosenberg,J.M. and Greene,P.J. (1989) Gene, 85, 1-13. 4. Posfai,J., Bhagwat,A.S., P6sfai,G. and Roberts,R.J. (1989) Nucl. Acids Res., 17, 2421-2435. 5. Klimasauskas,S., Timinskas,A., Menkevicius,S., Butkiene,D., Butkus,V. and Janulaitis,A. (1989) Nucl. Acids Res., 17, 9823-9832. 6. Posfai,G., Kiss,A., Erdei,S., P6sfai,J. and Venetianer,P. (1983) J. Mol. Biol., 170, 597-610. 7. Kiss,A., Posfai,G., Keller,C.C., Venetianer,P. and Roberts,R.J. (1985) Nucl. Acids Res., 13, 6403-6421. 8. Sznyter,L.A., Slatko,B., Moran,L., O'Donnell,K.H. and Brooks,J.E. (1987) Nucl. Acids Res., 15, 8249-8266. 9. Caserta,M., Zacharias,W., Nwankwo,D., Wilson,G.G. and Wells,R.D. (1987) J. Biol. Chem., 262, 4770-4777. 10. Som,S., Bhagwat,A.S., and Friedman,S. (1987) Nucl. Acids Res., 15, 313-331. 11. Slatko,B.E., Croft,R., Moran,L.S. and Wilson, G.G. (1988) Gene, 74, 45-50. 12. Sullivan,K.M. and Saunders,J.R. (1988) Nucl. Acids Res., 16,4369-4387. 13. Karreman,C. and de Waard,A. (1988) J. Bacteriol., 170, 2527-2532. 14. Kupper,D., Zhou,J.-G., Venetianer,P. and Kiss,A. (1989) Nucl. Acids Res., 17, 1077-1088 15. Lin,P.M., Lee,C.H. and Roberts,R.J. (1989) Nucl. Acids Res., 17, 3001-3011. 16. Card,C.O., Wilson,G.G., Weule,K., Hasapes,J., Kiss,A. and Roberts,R.J. (1990) Nucl. Acids Res., 18, 1377-1383.

4664 Nucleic Acids Research, Vol. 18, No. 16 17. Walter,J., Noyer-Weidner,M. and Trautner,T.A. (1990) EMBO J., 9, 1007-1013. 18. Karreman,C. and De Waard,A. (1990) J. Bacteriol., 172, 266-272. 19. Buhk,H.-J., Behrens,B., Tailor,R., Wilke,K., Prada,J.J., Gunthert,U., Noyer-Weidner,M., Jentsch,S. and Trautner,T.A. (1984) Gene, 29, 51-61. 20. P6sfai,G., Baldauf,F., Erdei,S., Posfai,J., Venetianer,P. and Kiss,A. (1984) Nucl. Acids Res., 12, 9039-9049. 21. Tran-Betcke,A., Behrens,B., Noyer-Weidner,M. and Trautner,T.A. (1986) Gene, 42, 89-96. 22. Behrens,B., Noyer-Weidner,M., Pawlek,B., Lauster,R., Balganesh,T.S. and Trautner,T.A. (1987) EMBO J., 6, 1137-1142. 23. Hanck,T., Gerwin,N. and Fritz,H.-J. (1989) Nucl. Acids Res., 17, 5844. 24. Renbaum,P., Abrahamove,D., Fainsod,A., Wilson,G.G., Rottem,S. and Razin,A. (1990) Nucl. Acids Res., 18, 1145-1152. 25. Bestor,T., Laudano,A., Mattaliano,R. and Ingram,V. (1988) J. Mol. Biol., 203, 971-983. 26. Wu,J.C. and Santi,D.V. (1987) J.Biol.Chem., 262, 4778-4786. 27. Balganesh,T.S., Reiners,L., Lauster,R., Noyer-Weidner,M., Wilke,K. and Trautner,T.A. (1987) EMBO J., 6, 3543-3549. 28. Lauster,R., Trautner,T.A. and Noyer-Weidner,M. (1989) J. Mol. Biol., 206, 305-312. 29. Sussenbach,J.S., Steenbergh,P.H., Rost,J.A., van Leeuwen,W.J. and van Embden,J.D.A. (1978) Nucl. Acids Res., 5, 1153-1163. 30. Raleigh,E.A. and Wilson,G. (1986) Proc. Natl. Acad. Sci. USA 83, 9070-9074. 31. Yanisch-Perron,C., Vieira,J. and Messing,J. (1985) Gene, 33, 103-119. 32. Blumenthal,R.M., Gregory,S.A. and Cooperider,J.S. (1985) J. Bacteriol., 164, 501-509. 33. Silhavy,T.J., Berman,M.L. and Enquist,L.W. (1984) Experiments with Gene Fusions. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 34. Maniatis,T., Fritsch,E.F. and Sambrook,J. (1982) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 35. Norrander,J., Kempe,T. and Messing,J. (1983) Gene, 26, 101-106. 36. Sanger,F., Nicklen,S. and Coulson,A.R. (1977) Proc. Nadl. Acad. Sci. USA, 74, 5463-5467. 37. Henikoff,S. (1984) Gene, 28, 351-358. 38. Maxam,A.M. and Gilbert,W. (1980) Methods Enzymol., 65, 499-560. 39. Ohmori,H., Tomizawa,J. and Maxam,A.M. (1978) Nucl. Acids Res., 5, 1479-1485. 40. Tinoco,I., Borer,P.N., Dengler,B., Levine,M.D., Uhlenbeck,O.C., Crothers,D.M. and Gralla,J. (1973) Nature New Biol., 246, 40-41. 41. McLaughlin,J.R., Murray,C.L. and Rabinowitz,J.C. (1981) J. Biol. Chem., 256, 11283-11291. 42. Lauster,R. (1989) J. Mol. Biol., 206, 313-321. 43. Butkus,V., Klimasauskas,S., Kersulyte,D., Vaitkevicius,D., Lebionka,A. and Janulaitis,A. (1985) Nucl. Acids Res., 13, 5727-5746. 44. Bron,S. and Horz,W. (1980) Methods Enzymol., 65, 112-132. 45. McClarin,J.A., Frederick,C.A., Wang,B.-C., Greene,P., Boyer,H.W., Grable,J. and Rosenberg,J.M. (1986) Science, 234, 1526-1541.