Oligonucleotide cleavages by restriction endonucleases MvaI and EcoRII: a comprehensive study on the influence of structural

Oligonucleotide cleavages by restriction endonucleases MvaI and EcoRII: a comprehensive study on the influence of structural

Biochimica et Biophysica Acta, 1088 (1991) 395-400 395 © 1991 Elsevier Science Publishers B.V. 0167-4781/91/$03.50 ADONIS 016747819100106M BBAEXP 9...

481KB Sizes 0 Downloads 19 Views

Biochimica et Biophysica Acta, 1088 (1991) 395-400


© 1991 Elsevier Science Publishers B.V. 0167-4781/91/$03.50 ADONIS 016747819100106M

BBAEXP 92221

Oligonucleotide cleavage by restriction endonucleases Mva I and EcoRII: a comprehensive study on the influence of structural parameters on the enzyme-substrate interaction E l e n a A . K u b a r e v a i, E l i z a v e t a S. G r o m o v a ~, C l a u s - D i e t m a r P e i n 2 A n t j e K r u g 2. T a t j a n a S. O r e t s k a y a l, D i e t e r C e c h 2 a n d Z o e A . S h a b a r o v a 1 t A.N. Belozersky Laboratory of Molecular Biology and Bioorganic Chemistry at~d Department of Chemistry, Moscow State Universtty. Moscow (U.S.S.R.) and : Department of Chemistry, Ilumboldt Unirers:O, Berhn (F.R G.) (Received 15 October 1990)

Key words: D N A - p r o t e i n interaction; Restriction endonuclease; Synthetic DNA duplex: Flanking ,sequence

To elucidate the mechanism of action of the restriction endonucleases - isoschizomers E c o R l l and M v a i - a study was made of their interaction with a set of synthetic oligonudeotide dt, plexes containing a single 5 " . d ( C C ~ / T G G ) - 3 ' E c o R l l ( M v a l ) recognition site. The substrates had varyin2 length and structure of the nudeotide sequences flanking the recognition site. T i e strueturc of the flanking sequence is important for the cleavage by E c o R l i and M v a i e n z y n ~ s ; there is a structure which was found to speed up the EcoRll and Mval action. ~ d e a v a g e of oligonudeotide duplexes by E c o R i l enzyme does not go to completion. E c o R l l endonudease cleaved extended substrates less efficiently than short ones. Extension of the flanking sequences, with the same nudeotide surrounding of the recognition site, substantially altered the whole kinetic pattern of M v a l hydrolysis. This was not observed with E c o R i i enzyme. The restriction endonudease Mtml distinguished between dA and dT residues in the recognition site, which W~LSreflected in the higher rate of hydrolysis of the dA-containing strand of the quasipalindromic DNA duplex. Introduction

During the last years, we have studied the i:~teraction of the restriction endonucleases E c o R I l and Mt'al whicL are isoschizomers with different synthetic oligonucleotide duplexes containing canonical 5'-d(CCf'/ TGG)-3 ' or modified recognition sites of these enzymes [1-7]. The latter include either base or sugar modified nucleosides. These studies have shown that modifications markedly influence the activity of both enzymes. albeit in a different manner. In addition, notable differences were seen in the cleavage behaviour of both strands in an oligonucleotide duplex. According to our tmdings [3,4,61 and to those of other authors [8-12] it has to be assumed that, conformation of the substrate as well as length and nucleotide sequences of the DNA fragments flanking the recognition site also have a significant influence on the cleavage behaviour of DNA. Prompted by these results we have investigated structural parameters and melting behaviour of synthetic DNA duplexes of various length and sequences Cocrcspondencc: Dr. E.S. Gromova. A,N. Belozersk) Laborato~' of Molecular Biolog3 ar,d Bioorganic chemistry, Mosco~ St,tic University, Moscow W-2"~4, 119899, U.S.S.R.

( l - I V ) by ultraviolet-spectro~opy and CD, and correlated the data to the interaction of those duplexes with the two iso~hisomeric endonucleases E c o R l l and Mva !. (5"-3")







I *







" Index d (dcoxy) is ormtted. The strands of DNA duplex¢~ are indicated as T or A according 1o the central unit of the recogmtion ~itc. The a~o~-~ r and ; indicate the positions of the s~t~ clea~cd b~ E('oRII and Mval end~mucleas¢~, respectively.

396 The results to be presented in this paper show that nucleotide sequences flanking the recognition site indeed have a pronounced, though distinct, effect on the activity of both enzymes. Materials

and Methods

Restriction endonucleases Mval and EcoRll were commercial preF'_.rations produced by NPO Ferment (Vilnius, U.S.S.R.) and Biolar (Olaine, U.S.S.R.), respectively. MvaI had an activity 60000 units/ml. EcoRll after additional purification had .'m activity of 400 units/ml * Substrates 1, III, V'I and VIII were constructed from oligodeoxyribonucle,,tides synthesized in solution by the triester technique according to Refs. 4. 5, 13. Oligonucleotide cons~auents of duplexes II and IV were synthesized by the phosphoramidate technique on a polymeric support with a Viktoriya-4M automatic synthesizer (Novosibirsk, U.S.S.R.) !141. Oligonucleotides were 5'-phosphorylated and 5'32p-labelled with T4 polynucleotide kinase. The 5'terminal label was introduced alternately in one of the strands of D N A duplexes I - I V and into nonmodified chains of substrates Vll and VIII. The following buher solutions were used for optical measurements and enzymatic hydrolysis: buffer A - 40 mM Tris-HCI (pH 7.6), 50 mM NaCI, 5 mM MgC! 2, 7 m M dithiothreitol; buffer B - 10 m M Tris-ttCl (pH 8.5), 150 mM NaCi, 15 mM MgC! 2, 1 m M dithiothreitol, 100 p g / m l bovine serum albumin and buffer ( - 15 mM sodium citrate (pH 7.25). 150 mM NaCI. The temperature dependence of the ultraviolet eosorption of D N A duplexes were measured on Cary 219 spectrophotometer (Varian, U.S.A.) under monotcnous heating at a rate of 0 . 5 ° C per min in buffer /,. The results of melting were presented in the differentia~ form. The derivatives A,426O/AT- ~ were calculated from the A26 o vs. T profiles as a function of temperature. Oligonucleotide concentrations per nucleotide residue (CN) were determined spectrophotometrically. The molar extinction coefficients {¢26o) of oligonucleotides were taken equal to the sum of ~26o of the component nucleotides. CN were 7.5 • 10 -6 to 2- 10 -3 M. C D spectra were measured on Jasco J-500C spectropolarimeter (Japan) in buffer C, 20°C, CN 6.3 • l0 s to 8 . 4 . 1 0 - s M. The a2P-labeled substrates at concentrations per duplex (Co) of 1.63. I0 -7 to 6 . 6 . 1 0 -7 M were incubated at 2 0 0 C or 3 7 ° C for 1 to 60 rain with 0,4 units of EeoRI1 endonuclease in 10/~! of buffer A, or with 2 to 24 units of Mval endonuclease in 10 pl of buffer B. * The restriction endonuclea~ activity unit was defined as the amount of enzyme able to digest completely l pg of phage X DNA at 37o C inlh.

The reactions were stopped by heating at 9 5 ° C for 2 - 4 rain. The reaction mixtures were analyzed by electrophoresis in 20% polyacrylamide gel with 7 M urea as described [4]. The extent of substrate hydrolysis was determined for each strand separately as the ratio of the radioactivity of the hydrolysis product to the total radioactivity of the product and uncleaved substrate, and kinetic curves were constructed.




Ph)~lco-chemical properttes of the substrates Synthetic oligonucleotide duplexes l - I V were characterized by melting curves and C D spectra (Fig. 1, 2). The data for duplex 1 have been reported previously by Kuznetsova et al. [1~] The diffcreutial melting curves for substrates 1-1II are characteristic for a cooperative process (Fig. 1 and Ref. 13). The melting points under the conditions of EcoRll hydrolysis (buffer A) are 5 6 ° C for duplex 1, 6 1 ° C for duplex II and 7 3 ° C for duplex III, respectively. The latter melting point is close to that of natural DNA. The melting behaviour of duplex I in 0.15 M NaCi, 0.015 M sodium citrate buffer (pH 7.25) [13] traditionally used for optical measurements is essentially the same as in buffer A. The C D spectrum of duplex III (Fig. 2) in the region of 220-320 nm has a conservative shape with a crossover near 260 nm. It is similar to that of natural B-DNA. The C D of shorter duplex II shows some differences from the C D of duplex I11 but it is also looks like the C D of B-form. So one may conclude that the~e substrates maintain B-like conformation. The C D spectrum of duple;: I (Fig. 2 and Ref. 13) looks different



< x










8o t °C

Fig. I. Differential melting cur~es of DNA duplexes I (C N 3.4, 10 ~ ML II (C~ 2.1.10- ~ M) and Ill (C~. 7.5-10-6 M) in buffet A.

397 6


a£ -




V ~8o

~ o .~,r~ ,



Fig. 2. C D spectra o f D N A duplexes ! ( C N 7.9.10 -~ M L I t ( C ~ 8 . 4 - 1 0 "~ M ) a n d I H ( C ~ 6.3.10 - ~ M ) in b u f f e r C, 20 ° C.


from that of II and 111. It has some of the features of A-DNA, making it probable that duplex 1 assumes such an A-like conformation. There are some examples of A-like form in the case of short DNA duplexes containing like duolex l •..G-G...

• o •C-C...

clusters [15-17]. However, we cannot exclude the possibility, that the unusual CD spectrum of I could be a consequence of the specific sequence of the DNA, give, the short length. In the c~se of substrate IV it wa~ of interest whether the anticipated structure was correct. Under the conditions of the digestion with EcoRll, namely at a low concentration of the oligonucleotide and a low ionic strength, substrate IV additionally to the proposed structure may exist as homoassociates of type V and VI, respectively. AATGCCTGGCATT =o.o== •°=oo°

AAT6CCAGGCATT .•.••o • .....



The existence of hairpin structures cannot be excluded either. The melting curves for this substrate in buffer A (Fig. 3) in the range of a concentration from 10 3 M to 10 -~ M show that the melting process is monophasic and that the Tm increases with increasing oligonucleotide concentration. In accordance with data from the literature [18,19] these results exclude the existence of intramolecular hairpin structures under the above conditions. It seems especially unlikely, that under the higher ionic strength conditions of buffer B (used for Mval endonuclease digestion) hairpin structures are formed. The homoassociates V and VI are actually not formed. It follows from the temperature dependence of the ultraviolet-absorption of oligonucleotide AATGCCTGGCATI" as well as that of AATGCCAGGCATI'. The differential melting curves (Fig. 4) are characterized by two transitions. It is expected [19,20] that the low

t 0 Fig. 3. D i f f e r e n t i a l m e l t i n g c u ~ ' e s o f D N A d u p l e x IV in b u f f e r A: C N 5.1 - 10 M (1), 4 . 4 . 1 0 * M (2), 2,1-10 - 3 M (3). Inset, plot o f T,,: 1 vs. In Cr, for d u p l e x IV

temperature transition corresponds to the melting of DNA duplex V (curve a) or V! (curve b). The melting points remarkably appear below the Tm of duplex IV at the corresponding concentration (Figs. 3, 4). The second broad peak of the differential melting curve (Fig, 4) corresponds to the melting of hairpin structures preferably formed under these conditions. The melting curves suggest that duplex IV exists mainly in the anticipated structure under the conditions of the enzyme reactions (buffer A as well as buffer B). Homoassociates of type V and VI, respectively, as well as hairpin structures are not formed.

Influence of the nucleotide sequenc¢:~flanking the recognition site on the cleavage of syntl~etic subsfrates by the EcoRlI and Mvai endonucleases Undoubtedly. the interaction of restriction endonuclease with the DNA double helix is mainly directed to the recognition site. Nevertheless. the nucleotide sequences adjacent to the recognition site have a pro nounced eff,~t on the efficiency of double-strand cleavage [21,22]. Furthermore. the length of the flanking


<] it ¸ {

Fig, 4





D d f e r e n t i a t m e l t i n g curves o f A A T C r C C ' i - G C ~ C A T T (a) a n d AATGCCAGGCATT (b) tn b u f f e r A. ( ' ~ 5 . 1 0 ~ M.

398 o

o,0 I / 5 _



- ¸

/ . . -

Ill ~ . , ' "












_ 10

Time (rain) Fig. 5. Cleavage of I.)NA duplexes i ~ . A), l! ( o , e ) and I l l (E~,l ) by E c o R I I , 0.4 units (a) and M v a l , 2 units (b) restriction endonucleases, 37 o C. - T-strand, - . . . . . A-strand, C o 3.5. t0 - 7 M .

sequence becomes significant for an undisturbed formation of the enzyme-substrate complex. Thus, it has been shown that Ecokll endonuclease interacts with the nucleotide sequences adjacent to both ends of the recognition site and, in particular, no digestion was obtained for substrates in which the recognition site is flanked on one side by two base pairs [1]. As shown in Fig. 5a. the 14-mer oligonucleotide duplexes I and 11 were effectively digested by EcoRll endonuclease indicating that 4-5 base pairs adjacent to the recognition site are sufficient for the enzymatic reaction. The same conclusion can be drawn for the digestion by Mual endonuclease (Fig. 5b). Oligonucleotide duplexes 1 and ,I in which the recognition site is flanked by fragments of the same length but different sequences are cleaved with different rates by EcoRll or by Mval endonuclease (Fig. 5). It is clearly shown that cleavage of the duplex 1 proceeds appreciably faster than that of duplex II both by EcoRll


and Meal endonucleases. The preference of I as substrate correlates with a different CD-spectrum and probably with A-like conformation of this duplex. In duplexes II and llI the recognition site is flanked by fragments of the same sequence in its vicinity but of different length. Our findings indicate that such elongation of the flanking sequences may lead to a significant change of the kinetics of Meal endonuclease digestion. As shown in Fig. 5b, during the whole time of hydrolysis the A-strand of the 14-mer oligonucleotide duplex II is digested with a higher efficiency than the corresponding T-strand. In the case of 30-mer duplex 111, the A-strand is cleaved with a higher initial rate than the T-strand, but, after a short time the cleavage rate of the A-strand decreases significantly, whereas the T-strand is digested effectively. After 1 h of hydrolysis the cleavage of A-strand is not more than 60% and a nicked duplex with the structure of VII is formed in addition to the products of the complete digestion.


,, . . . . . , . . . . . . , , ° , , o , , . . . . . . . . C-T'A-C-G-A-C-G-G-T-T-G-G--A~C-C-G-A-G-A-T-C-G-A-A-G-T-A-T-G


G-A-T-G-C-T-G-C-C-A-A-C-C~T__G_G_C_T_C_T_A_E_C_T_T_C_A_T_A_C . . . . . .


. . . . . . . . . . . . . . . . . .






C-T-A-C-G-A-C-G-G-T-T-G-G-ApC-C-G-A-G-A-T-C-G-A-A-G-T-A-T-G Vlll


,00 80






lime Cmin) Fig. 6. _ . e a v a g : of D N A duplexes II1 (1. T-strand; 2, A-strand), VII (3, A-strand) and V i I I 14. T-strand) by 24 units of Mt,al endonuclease, 2 0 ° C , C o 1.63.10 ~7 M.

The study of digestion kihetics cY the chemically synthesized duplexes VII and VIII is taken as indicative for the unusual cleavage of duplex II1. It turned out that the cleavage rates of the intact strands in VII and VIil are different (Fig. 6). While a nick in T-strand of VII leads to a significant decrease of the cleavage rate, the corresponding nick in A-strand of VIII increased the efficiency of the digestion. From the above results it was concluded that Mval endonuclease -leaves its site in the following manner. Al the beginmqg of the reaction, the A-strand of duplex ii! is hydrolyzed much faster resulting in an accumulation of an intermediate with a nicked A-strand (equal to VIII). This intermediate is then cleaved more effectively in the T-strand than the intact duplex II1 (see Fig. 6). Therefore, during the course of the reaction the cleavage rate of the T-strand continuously increases. At the same time, the intermediate nicked in the T-strand (equal to VII) inhibits further cleavage of the A-strand (Fig. 6). As a result, the cleavage of the A-strand of Iii slows down. In contrast to this, there is no fundamental difference in the kinetics of digestion of oligonucleotide duplexes 11 and I!I by EcoRil endonuclease. As can be seen in Fig. 5a, in both cases the T-strand was cleaved with a higher rate than the A-strand. The rate of hydrolysis of each strand of the shorter duplex is faster than that of the corresponding strand of the longer one. Our results are in line with recently published findings 123,24 ! that extended D N A of the phages T3 or T7 containing only one or three EcoRll sites are not digested. These D N A

can be made t:'coRll.sensitive bv co-incubation ~ l h :~ second su~eptible DNA. "lhe following hypothesis ~a~ presented to explain thts interesting results. The activation of EcoRll endonuclease reqaires the simultaneous interaction of the enzyme with two recognition sites [23]. The probability of su:h an event decreases upon increasing the content of 'non-site' containing DNA, as in 30-membered substrate III. It is interesting in this connection that EcoRll endonuclease proved to bind to a 21-23 base pair fragment of concatemer D N A duplexes containing EcoRll sites repeated every 9 base pairs [25,261. There are two EcoRll sites in such a fragment. As one can see from Figs. 5a and 7 the EcoRII catalyzed cleavage reactions do not go to completion. One of the explanations of this may be that as it was shown previously [1] nicking A- or T-strand effectively prevents cleavage in the opposite strand. Also, the inactivation of the EcoRll endonuclease during the time of the hydrolysis is possible. However. it was shown, that the decrease of the cleavage as the result of the enzyme inactivation does not exceed 10-15%.

Studies of strand-dependent cleavage by EcoRll and Mt~al endonucleases As seen from Fig. 5a the T- and A-strand of duplexes I - I I ! are cleaved with different rates by EcoRll endonuclease. The initial ra,e of hydrolysis of the T-strand is at an average 1.5-times higher compared to that of the corresponding A-strand. We argue that the varying degree of symmetry beyond the EcoRll site may influence the rate of cleavage of D N A by EcoRll endonuclease. Indeed, for endonucleases which recognize symmetrical sites, it has been reported that differences in the rate of cleavage are caused by the influence of the flanking sites [8,9,27,28]. However, for EcoRll endonuclease the differences in the cleavage rate of both strands may also result from the derivation from complete symmetry, of the recognition site owing to the degenerac~ of the /tr4 80

i I

6~ t

,,/ i -I ~%__

f 1

f / /

IL- 2

_ _.D--- -

- --

i /.--'S.--" -





7O ._-x





Time (min)

Fig. 7. Cleavage of D N A duplex IV at 3 7 ~ C by EcoRil endonuclease,

0.4 units, C D 6.6-10 -7 M 0,2) and Meal endonudease, 2.4 units. C"o 1.63-10- ~ M (3.4). - - - - T-~trand. - . . . . . A-strand,


central base pair ( T - A or A - T ) . In order 1o clarify w h e t h e r differences in the c o m p o s i t i o n of the flanking regions lead 1o differences in the cleavage rate of the Tand A-strand in duplexes l - I l l , we have c o n s t r u c t e d q u a s i p a l i n d r o m i c d u p l e x IV c o n t a i n i n g identical nucleotide sequences on both sides o f the recognition site. EcoRli and Mval digestion was c o m p a r e d . It is s h o w n (Fig. 7) that there is no preferential cleavage o f the T- or A - s t r a n d o f 1V in the case o f E c o R I l digestion. Therefore, we c o n c l u d e that differences in the cleavage rates o f both s t r a n d s of d u p l e x e s l - I l l are exclusively caused by the influence o f the regions adj a c e n t to the EcoRIl recognition site, probably, because o f their interaction with the enzyme. Likewise, differences in the cleavage rates of the two s t r a n d s are o b t a i n e d for the Mual digestion. The As t r a n d o f duplexes !1 a n d !11 is initially cleaved faster than the T-strand (Fig. 5b). in contrast, the initial cleavage rates o f both s t r a n d s o f duplex I are nearly the same. The hydrolysis of the T- and A - s t r a n d o f the q u a s i p a l i n d t o m i c duplex IV by Mval e n d o n u c l e a s e proceeds with different rates (Fig. 7). T h e results o b t a i n e d d e m o n s t r a t e that differences in the cleavage o f b o t h s t r a n d s o f duplexes I - I V by Mval, unlike EcoRll, are due mainly to two factors. First, the s t r a n d d e p e n d e n c e of the rate o f reaction most likely reflects the structure differences o f sequences next to the Mval site. Second, the cleavage by Mval is strictly influenced by the a s y m metric p r i m a r y structure o f the recognition site due to the degeneracy o f the central base pair. In the case o f Mval e n d o n u c l e a s e the scissile p h o s p h o d i e s t e r b o n d is situated next to the central base pair within the recognition site. The fact that the b o n d s b e t w e e n d C a n d d A in substrate IV are cleaved faster than those b e t w e e n d C and d T suggests that the nature o f the nucleoside residue adjoining the cleavage site may influence the efficiency o f hydrolysis o f the p h o s p h o d i e s t e r bond. T h e differences in the cleavage rate of both s t r a n d s of sub.,,irate IV by Mval obviously result from the nonequivalence of scissile b o n d s within the T- a n d A - s t r a n d . Thus, Mval e n d o n u c l e a s e may take a d v a n t a g e ,.,f the d A - over tiT-residue within the recognition site. It should be noted that in the case of EcoRI! e n d o n u c l e a s e the scissile p h o s p h o d i e s t e r b o n d s in V,oth s t r a n d s o f the substrate IV are equivalent. Acknowledgement We m a n k Olga V. Petrauskiene for her help in ( ' D spectra m e a s u r e m e n t s and for c o n t r i b u t i o n in the enzyme inactivation experintents. References

1 Yolov, A.A., Gromova, E.S., Kubareva, EA.. Potapov, VK+ and Shabarova. ZA. (1985) Nucleic A~tds Res. 13, 8969-8991.

/olov. A.A., Vinogradova, M.N, Gromova, E.S.. Rosenlhal, A., (e, h+ !).. Veiko, V.P., Metelev. V.G., Kosykh. V.G., Buryanov, JaA Bacv, ArA. and Shabarova. ZA. (1985) Nucleic Acids Res. 13, 8983 8998, Vinogradova, M.N, Gromova, E,S, Gryaznova, O.1., Isagulyants, M.G.. Kuznetsova, S.A., Kosykh, V,G. and Shabarova, Z.A. (1987) Bioorg, Khimia (Russ.) 13, 1194-1204. Kubareva, E.A. Gromova. E.S.. Oretskaya. T.S. and Shabarova. Z.A, (1987) Bioorg, Khimia (Russ.) 13, 1205-1211. Gromova, E.S., Kubareva, E.A., Pcin, C.-D., Oretskaya, T.S., Shabarova, Z.A., Cech, D. and Prokofiev, M.A. (1987) Doklady Akad. Nauk SSSR 295, 1493-1497. Pcin, C.-D., Cech, D., Gromova, E.S., Orezkaya, T.S., Shabarova, Z.A. and Kubareva, E.A. (1987) Nucleic Acids Res., Symp. Series No 18, 225-228. Kubareva, E.A., Pein, C.-D., Gromova, E.S., Kuznezova. S.A., Tashlitzki. V.N., Cech, D. and Shaharova, Z.A+ (1988) Eur. J Biochem. 175. 615-618. 8 Aires, J., Pingoud, A., Haupt, A.. Langowski, J., Peters, F., Maass, G. and Wolff. C. (1984) Eur. J. Biochem, 140, 83-92, 9 Wolfes. H. Fliess, A. and Pingoud, A. (1985) Eur. J. Biochem. 150. 105-110. 10 Soslau, G., Parker, J. and Nelson, J.W. (1986) Nucleic Acids Res. 14. 7237-7252. II Ohtsuka, E., Ishino, Y., Ibaraki, K. and Ikehara, H. (1984) Eur. J. Biochem. 139, 447-450. 12 Nazarenko, I.A., Gorbunov, Yu.A. and Malygn, E.G, (1987) Bioorg, Khimia (Russ.) 13, 928-933. 13 Kuznetsova. S.A., Kubareva, E.A.. Oretskaya, T.S.. Dolinnaya, N.G., Krynetskaya, N.F., Gromova, E.S., Shabarova, Z.A. and Cech, D. (1987) Biopolimery ! Kletka (Russ.) 3, 283-289. 14 Volkov, E.M., Romanova, E.A., Krug, A., Oretskaya, T.S., Potapov, V.K. and Shabarova, Z.A. (1988) Bioorg. Khimia (Russ.) 14, 1034-1039. 15 Shakked, Z. and Rabinovich, D. (1983) J. Mol. Biol. 166, 183-201. 16 Benevides, J.M., Wang. A.H.-J.. Rich, A,, Kyogoku, Y., Van der Marel, GA., Van Boom, J,H. and Thomas, G.J., Jr. (1986) Biochemi,.,tlj 25.41-50. 17 Frederick, C.A., Quigley. G.J., Teng, M.-K., Coil, M., Van der Marel, G.A., Van Boom, J.H., Rich, A. and Wang, A.H.-J. (1989) Eur. J. Biochem. 181,295-307. 18 Marky, L.A., Blumenfeld, K.S., Kozlowski, S. and Breslauer, K.J. (1983) Biopolymcrs 22, 1247-1257. 19 Xodo, L.E, Manzini, G., Quadrifoglio, F., Van der Marel, G.A. and Van Boont, J.H+ (1986) Nucleic Acids Res. 14, 5389-5398. 20 Summers, M.F., Byrd, R.A., Gallo, K.A., Samson, C.J., Zon, G. and Egan, W. (1986) Nucleic Acids Res. 13, 63"'~-6386. 21 Greene, P.J., Poonian, M.S., Nussbaum, A.L., Tobias, L., Garfin, D.E., Boyer, H.W. and Goodman, H.M. (1975) J. Mol. Biol. 99, 237-261. 22 McClarin, J.A.. Frederick, C.A.. Wang, B.C., Greene, P., Boyer, HW,. Grable. J. and Rosenberg, J.M. (198,5) Science 234., 15261541.

23 Kruger, D.H., Barcak, C.J., Reutcr, M. and Smith, II.O. (1988) Nucleic Acids Res. 16, 3997-4008. 24 Pcin, C.-D., Reuter, M., Cceh D. and Kruger, D.I-I.{1989) FEBS Lett. 245, 141--144. 25 Yolov, A.A., Gromova, E.S. and Shabarova, Z.A. (1985) Mol. Biol. Rep. lO, 173-176. 26 inogradova, M.N., Gromova, E.S., Kosykh, V.G.. Buryarmv. Ja.l. and Shabarova, Z.A. (1990) Mol. Biologia {Russ.) 24, 847-850.

21 Amstrung, K.A. and Bauer, W.R. (1982) Nucleic Acids Res. 10, 993-1007. 28 Amstrong, K.A. and Bauer, W.R. (1983) Nucleic Acids Res. 11, 4109-4126.