ANALYTICAL
BIOCHEMISTRY
204,206209
(1992)
Continuous Spectrophotometric Assay for Restriction Endonucleases Using Synthetic Oligodeoxynucleotides and Based on the ~yperchromic Effect Timothy R. Waters* and Bernard A. Connollyt *Department of Biochemistry (SERC Centre for Molecular Recognition), Southampton, SO9 3TIJ, United Kingdom; TDepartment of Biochemistry NE2 4Hk, United Kingdom
ReceivedDecember
30, 1991
A continuous spectrophotometric assay for the EcoRV restriction endonuclease has been developed. The synthetic self-complementary oligonucleotide d(GACGATATCGTC) (which is double stranded under the assay conditions) is used as the substrate. The EcoRV endonuclease recognizes d(GATATC) sequences cutting between the central T and dA bases. Thus d(GACGATATCGTC) is converted to d(GACGAT) and dCpATCGTC) during catalysis. Both of the hexameric products are single stranded under the assay conditions. The conversion of the dodecameric substrate to the two hexameric products and the concomitant change from doubleto single-stranded DNA is associated with an increase in absorbance at 254 nm due to the hyperchromic effect. This change can be used to monitor column effluents for endonuclease activity and also for K,,, and keat determination under steady-state kinetic conditions. 0 1992 Academic Press, Inc.
Restriction endonucleases are widely used for molecular biology applications. Many have been cloned and overexpressed and are commercially available. Nevertheless many investigators, especially those interested in discovering new endonucleases and those working on the kinetic properties of known endonucleases, prepare the enzymes themselves. Both the purification of restriction endonucleases and the subsequent study of their kinetic properties require an assay for enzyme activity. Assays used during purification usually use plasmids. When a single site is present the closed circle form of the plasmid is converted to the linear form (either directly or via an open circle intermediate). Multiple restriction endonuclease sites on a single plasmid are con204
University of Southampton, Bassett Crescent East, and Genetics, The University, Newcastle-upon-Tyne,
verted to a ladder of products (l-8). Similar plasmid-based assays can be used to study the kinetic properties of purified restriction endonucleases. The enzymology of restriction endonucleases can also be studied using short synthetic oligodeoxynucleotides (9-17). These assays are based either on HPLC or on 5’-32P labeling followed by separation of radioactive substrates and products by denaturing gel electrophoresis or homochromatography. All of the above assays are discontinuous and require sampling of the mixture at discrete time points followed by separation and quantitation of substrates and products (either by HPLC or by gel electrophoresis). As such these assays are time consuming and suffer from the errors inherent in a discontinuous method. These errors arise as the rate is measured at relatively few time points whereas in a continuous assay the entire reaction is observed. A rapid continuous assay for restriction endonucleases would find wide applicability in both enzyme purification and the study of kinetic properties. It is well documented that when double-stranded DNA is heated, melting occurs and the single-stranded form is produced. This is associated with an increase in light absorbance at 254 nm. This is known as the hyperchromic effect and arises because base stacking (associated with diminished absorbance of light) in doublestranded DNA is greater than that in single-stranded DNA (18). All restriction endonucleases require doublestranded DNA as substrates and when short oligodeoxynucleotides are used they must form double strands. However, cleavage of a double-stranded oligodeoxynucleotide (say lo-16 base pairs) often results in shorter products (say 5-8 bases) that are single stranded under assay conditions. Thus it is possible to set up an oligodeoxynucleotide-based assay for restriction enzymes 0003-2697192 $5.00 Copyright Q 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
CONTINUOUS
ASSAY
FOR
RESTRICTION
that converts double- to single-stranded DNA. This should be associated with an increase in absorbance at 254 nm and be amenable to continuous monitoring. In this paper we describe such an assay using the doublestranded dodecamer d(GACGATATCGTC) and the restriction endonuclease EcoRV cleavage site GAT’ATC (19,20). MATERIALS
AND
METHODS
The EcoRV endonuclease was purified from overexpressing Escherichia coli strains as described (19,21,22). The concentration of the endonuclease was determined using an tZsO= 10.4 X lo4 M-’ cm-l (this is for a dimer of subunit molecular weight 29,000 which is the active species) (19). The endonuclease concentration was also determined using Bio-Rad protein assay based on dye binding according to the manufacturer’s instructions. Bovine serum albumin was used as a standard. Both assays gave virtually identical results. Oligodeoxynucleotides were prepared using an Applied Biosystems 381 A DNA synthesiser and phosphoramidite chemistry. The oligodeoxynucleotides were purified and desalted as described (9). The concentrations of d(GACGATATCGTC) and its 5’-phosphorylated derivative were determined using tZs4= 16.6 X lo4 M-’ cm-’ (this is for the double-stranded form which is the true substrate for the endonuclease) (9). The preparation of d(pGACGATATCGTC) was by 5’-phosphorylation of d(GACGATATCGTC) with ATP and polynucleotide kinase as described (9). Spectrophotometric assays were performed using a Kontron Instruments Uvikon 930 spectrophotometer. (a) Comparison of the Endonuclease-Catalyzed of d(GACGATATCGTC) by the Continuous HPLC Assay
Cleavage and an
To a l-cm path length quartz cuvette containing 50 mM Hepes-NaOH, pH 7.5,lOO mM NaCl, 10 mM MgCl,, and 6.8 pM d(GACGATATCGTC) was added EcoRV endonuclease to a concentration of 10 nM. The final volume was 1 ml and the increase in absorbance at 254 nm was monitored over 80 min. The temperature was maintained at 25°C by thermostating the cell compartment. Simultaneously aliquots from an identical assay mixture in an Eppendorf tube were assayed for substrate and products by HPLC as previously described (9).
205
ENDONUCLEASES
TATCGTC) was accurately determined to be 4.8 PM. An excess of EcoRV endonuclease (final concentration, 200 nM) in a small volume was added. The final absorbance after the increase had ceased (about 10 min) was noted. This difference in absorbance represents the hydrolysis of 4.8 nmol of d(GACGATATCGTC) and can be used to evaluate the change per nanomole of substrate. As a check that d(GACGATATCGTC) was completely hydrolyzed an aliquot was assayed by HPLC for substrate and products (9). (c) Rate of Hydrolysis of d(GACGATATCGTC) as a Function of EcoRV Endonuclease Concentration To a l-ml volume in a l-cm path length quartz cuvette containing 50 mM Hepes-NaOH, pH 7.5, 55 mM NaCl, 25 mM MgCl,, and 3 pM d(GACGATATCGTC) was added EcoRV endonuclease to give final enzyme concentrations between 2 and 100 nM. The change in absorbance at 254 nm was noted and used to determine hydrolysis rates. (d) Monitoring of Phosphocellulose Column Eluate during EcoRV Endonuclease Purification To a l-ml volume in a l-cm cuvette containing the buffer and oligodeoxynucleotide given in (c) were added aliquots of fractions collected during the phosphocellulose column purification of the EcoRV endonuclease (19,21). The size of the aliquots were chosen so as to give a reasonable rate of absorbance increase over about 5 min. The amount of endonuclease present in each fraction determined by this assay was compared with the levels present as determined by denaturing SDSPAGE’ electrophoresis (19,23). (e) Determination of the K,,, and k,,, Values for d(pGACGATATCGTC)
(6) Determination of Increase in Absorbance at 254 nm per Nanomole of d(GACGATATCGTC) Hydrolyzed
For accurate kinetic determinations it is best to siliconize the quartz cuvettes prior to use. The cuvettes were rinsed in a 2% solution of dimethyldichlorosilane in l,l,l-trichloroethane (BDH Ltd., Poole, Dorset, UK) for 5 min and then allowed to dry. The cuvettes were rinsed with water. Assays were carried out in either a l-ml (l-cm path length) or a 2-ml (2-cm path length) volume depending on the oligodeoxynucleotide concentration; 50 mM Hepes-NaOH pH 7.5 was shown, 55 mM NaCl, and 25 mM MgCl, was the buffer used and the final EcoRV endonuclease concentration was 5 nM. As-
The absorbance at 254 nm of approximately 5 pM d(GACGATATCGTC) in a l-cm path length cuvette containing the buffer components detailed above was noted. From this the concentration of d(GACGA-
’ Abbreviations used: SDS-PAGE, acrylamide gel electrophoresis; FPLC, raphy.
sodium dodecyl sulfate-polyfast protein liquid chromatog-
206
WATERS 15
,
AND
I 0
= hplc
assay
-6 E e
10
5 3
b s z n
GACGATATCGTC 5 2d(GACGAT)
+ 2d(pATCGTC)
0 0
10
20
30 Time
40
50
60
Imin
FIG.
1. Comparison of the rates of hydrolysis of d(GACGATATCGTC) by the continuous spectrophotometric method (solid line) and by a discontinuous HPLC-based assay (squares). The amount of single-stranded (s/s) product formed is given as a function of time.
says were carried out at 25°C. The d(pGACGATATCGTC) concentrations were 0.28, 0.52, or 0.99 WM (2-ml volume) and 1883.56, or 6.05 PM (l-ml volume). The reactions were monitored over about 30 min and the initial rates determined from the linear portion of the trace (usually the first 10 min). K,,, and IZcatvalues were determined from a plot of [substrate]/velocity versus [substrate] (24,25).
RESULTS
AND
DISCUSSION
Incubation of d(GACGATATCGTC) with the EcoRV endonuclease gives d(GACGAT) and d(pATCGTC). The dodecamer has a 7’, of 53°C (9) and so is double stranded under the assay conditions of 25°C. However the two hexamers are too short to form stable double strands at 25°C and so exist in single-strand form. Thus as shown in Fig. 1 the EcoRV endonuclease converts a double-stranded dodecamer into two single-stranded products. This is associated with an increase in absorbance at 254 nm due to the hyperchromic effect as is also shown in Fig. 1 and this forms the basis of a rapid and continuous assay for the enzyme. Superimposed on the curve for the absorbance change in Fig. 1 are values in which the rate has been determined by a previously described HPLC assay (9). Both methods give the same rate. However, the continuous spectrophotometric assay requires only 80 min assay time. Each time point for the discontinuous HPLC assay requires about 30 min. The spectrophotometric assay is therefore not only more accurate, by virtue of being continuous, but much quicker than the HPLC method.
CONNOLLY
In order to convert the curve in Fig. 1 to a rate in units of nmol substrate hydrolyzed min-’ mg-’ enzyme it is necessary to determine the change in absorbance at 254 nm due to the cleavage of 1 nmol of substrate. To do this a known amount of d(GACGATATCGTC) was hydrolyzed completely by an excess of the endonuclease and the change in absorbance noted. This experiment is illustrated in Fig. 2, which shows an increase in absorbance at 254 nm of 0.165 for the complete hydrolysis of 4.8 nmol of d(GACGATATCGTC). This corresponds to a change in absorbance of 0.034 per nanomole d(GACGATATCGTC) cleaved. This represents an approximate 20% increase in absorbance for complete conversion of cl(GACGATATCGTC) to single-stranded products. The hyperchromicity observed upon endonuclease cleavage depends on ionic conditions. The ionic strengths of all the buffer solutions used in the assays were identical and the amounts (in terms of both concentrations and volumes) of enzyme and oligodeoxynucleotide added were too small to significantly alter this. Although slightly different conditions were used in assays (a) and (b) as compared to (c), (d) and (e), no differences in hyperchromicities were observed under the two conditions. The assay described here is based on an increase in absorbance at 254 nm and this limits the substrate concentration that can be used. Too low an initial absorbance give poor signal-to-noise and too high approaches the end of the linear response of the spectrophotometer. We have found that initial absorbances of between 0.05 and 1.25 give good results. In a l-ml volume and a l-cm path length these correspond to d(GACGATATCGTC) concentrations of between 0.3 and 7.5 PM. The range of the assay can be extended by
Initial APs4 = 0.800 Final A,,, = 0.965 AA,,,,
’
o’60 -h
0
1 I 1
AA,,, ,
per
hydrolyseq
10
oligonucleotide
= 0.034 20
Time
FIG. 2.
nmol
= 0.165
, 30
/min
Determination of the absorbance change caused per nanomole of d(GACGATATCGTC) completely hydrolyzed by the EcoRV endonuclease; 4.8 nmol of d(GACGATATCGTC) was present initially in a l-ml volume. At the break in the trace (after 1.5 min) an excess of EcoRV endonuclease was added and the reaction allowed to proceed to completion.
CONTINUOUS
ASSAY
FOR
RESTRICTION
207
ENDONUCLEASES
addition the continuous assay has been used to determine Km and kc,, values for oligodeoxynucleotides containing dC and dG analogues. These data will be reported on later and complement our earlier studies using analogues of dA and T (9,lO). It should be noted that both T, values and hyperchromicities have a dependence on oligodeoxynucleotide concentration. However, over the oligomer concentration range used here the variation in hyperchromicity with concentration was extremely low and did not affect K,,, and k,,, determination. 0
20
40
[endonucleasel FIG. 3. Dependence on the rate of TATCGTC) as measured by the continuous say on EcoRV endonuclease concentration.
60
60
100
/nM hydrolysis of d(GACGAspectrophotometric as-
using cells of different path lengths. Cells with path lengths of between 0.25 and 5 cm are available allowing an extension of the starting concentration of d(GACGATATCGTC) by about five times in either direction. Alternatively this range could be extended toward higher substrate concentrations using a higher wavelength, where oligodeoxynucleotides have a lower absorbance. However, at these higher wavelengths changes in absorbance tend to be lower, thus reducing signal/noise. Figure 3 shows that the continuous spectrophotometric assay gives a linear response to EcoRV endonuclease concent,ration over the range 2-100 nM. As an application of the above assay we have monitored the eluate from the phosphocellulose column that forms the first step of the purification of the EcoRV endonuclease (19,Zl). In the absence of a convenient assay this column has previously been monitored by SDS-PAGE. As is shown in Fig. 4 both assays correspond exactly. The continuous assay requires 5 min for each time point, a total time of 50 min. The entire gel electrophoresis procedure (including sample preparation, electrophoresis, staining, and destaining) requires 2-3 h. Finally we have determined the Km and kc,, values for d(pGACGATATCGTC) as shown in Fig. 5. Very good quality data resulting in excellent fits to plots of [substrate]/velocity versus [substrate] can be rapidly obtained. The Km value of 2.92 PM obtained agrees very well with that of 3.8 FM obtained previously using an assay based on 32P labeling and gel electrophoresis (10). A kcat (33.2 min-“) higher than the previous value of 6.9 min-’ is found here. This is due to a change in the final purification matrix from gel filtration to an FPLCbased method. The FPLC method consistently gives more active protein than that obtained previously. In
CONCLUSION A continuous spectrophotometric assay for use with restriction endonucleases has been described and evaluated using the EcoRV enzyme. The advantages of the assay are convenience, speed, and accuracy. A slight disadvantage is the limited range of substrate concen-
10
20
30
Fraction
10 13 18 20
40
50
Number
22 24 28 30 32
Fraction
Number
FIG. 4. Assay by the continuous spectrophotometric fractions eluted from a phosphocelluiose column during endonuclease purification. For comparison an SDS-PAGE the fractions is also given. The position of the endonuclease is arrowed.
method of the EcoRV pattern of on the gel
208
WATERS
0
Time
CONNOLLY
0
a
2
AND
6
6
0
0.5
/min
I -3.0
1.0 Time
I 0.0
I 3.0 [Sol
1.5
2
/min
I 6.0
/pM
FIG. 5.
Determination of the K,,, and k,, for d(pGACGATATCGTC) using the continuous spectrophotometric assay. (A) The rates observed at d(pGACGATATCGTC) concentrations (yM) of: (a) 6.05, (b) 3.56, (c) 1.86. (B) The rates observed at: (d) 0.99, (e) 0.52, (f) 0.28. Note that the volume used in (A) was 1 ml whereas in (B) 2 ml was used. (C) Plot of substrate concentration/velocity versus substrate concentration for K, and kc., determination.
trations that can be used. The assay has proved useful both in the purification of the EcoRV endonuclease and in the evaluation of kinetic constants. Almost all restriction endonucleases use synthetic oligodeoxynucleotides and it is easy to manipulate this system so that a double-stranded substrate is converted to single-stranded products. The continuous assay should therefore be widely applicable and should form a very useful addition to the assay methods already available for restriction enzymes. ACKNOWLEDGMENTS T.R.W. was supported by a UK SERC research studentship. B.A.C. is a research fellow of the Lister Institute of Preventive Medicine. Special thanks go to Mrs. S. Broomfield for preparation of the manuscript.
REFERENCES 1. Modrich, P., and Roberts, R. J. (1982) and Roberts, R. J., Eds.), pp. 109-154, ratory, Cold Spring Harbor, NY.
in Nucleases Cold Spring
(Linn, Harbor
S. M., Labo-
2. Bennett, S. P., and Halford, Cellular Regulation (Horecker, Vol. 30, pp. 57-104, Academic
S. E. (1989) in Current Topics in B. L., and Stadtman, E. R., Eds.), Press, London.
3. Sharp, P. A., Sugden, 12,3055-3063.
Sambrook,
B., and
J. (1973)
Biochemistry
4. 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. 5. Halford, S. E., and Johnson, N. P. (1980) Biochem. J. 191,593604. 6. Rubin, R. A., and Modrich, P. (1978) Nucleic Acids Res. 5, 29912997. 7. Maxwell, 8. Halford, 1777.
A., and Halford, S. E. (1982) Biochem. J. 203, 85-92. S. E., and Goodall, A. J. (1988) Biochemistry 27, 1771-
9. Newman, P. C., Nwosu, V. U., Williams, D. M., Cosstick, Seela, F., and Connolly, B. A. (1990) Biochemistry 29,9891-9901. 10.
Newman, P. C., Nwosu, V. U., Williams, D. M., Cosstick, Seela, F., and Connolly, B. A. (1990) Biochemistry 29,9902-9910.
11.
Mazzarelli, chemistry
12. Fliess, Frank, 3474.
J., Scholtissek, 28,4616-4622.
S., and McLaughlin,
A., Wolfes, H., Rosenthal, A., Schwellus, R., and Pingoud, A. (1986) Nucleic Acids
R., R.,
L. W. (1989)
Bio-
K., Blocker, Res. 14,
3463-
H.,
CONTINUOUS 13. Fliess, A., Wolfes, H., Seela, Acids Res. 16, 11,781-11,793. 14. Brennan, C. A., van Cleve, Biol. Chem. 261,12X-1218. 15. Seela, F., and Kehne, 16. Lesser, D., Kurpiewski, 250, 776-786.
F., and
18.
Pingoud,
M. D., and
and
A. (1988)
Gumport,
A. (1987) Biochemistry M., and Jen-Jacobsen,
17. Aiken, C. R., McLaughlin, L. W., Biol. Chem. 266, 19,070-19,078.
ASSAY
26,
Gumport,
FOR
RESTRICTION
Nucleic
R. I. (1986)
J.
2232-2238. L. (1990) Science R. I. (1991)
20.
Schildkraut, (1984) Gene
21.
Bougueleret, L., Tenchini, (1985) Nucleic Acids Res.
I., Banner, 27, 327-329.
C. D.,
Rhodes,
C. S., and
Parekh,
S.
J., and
Zabeau,
M.
M. L., Botterman,
13, 3822-3839.
22.
Luke, P. A., McCallum, S. A., and Halford, S. E. (1987) in Gene Amplification and Analysis (Chirikian, J. G., Ed.), Vol. 5, pp. 183205, Elsevier, New York.
23.
Laemmli,
24.
Hanes,
25.
Dixon, M., and Webb, E. C. (1979) Longmans Green. London.
J.
Saenger, W. (1984) Principles of Nucleic Acid Structure, pp. 141149, Springer-Verlag, New York. 19. D’Arcy, A., Brown, R. S., Zabeau, M., van Resandt, R. W., and Winkler, F. K. (1985) J. Biol. Chem. 260, 198771996.
209
ENDONUCLEASES
U. K. (1970) C. S. (1932)
Nature
Biochem.
227, J. 26,
680-685. 1406-1421 Enzymes,
3rd
ed., pp. 61-62,
BIOCHEMISTRY
204,206209
(1992)
Continuous Spectrophotometric Assay for Restriction Endonucleases Using Synthetic Oligodeoxynucleotides and Based on the ~yperchromic Effect Timothy R. Waters* and Bernard A. Connollyt *Department of Biochemistry (SERC Centre for Molecular Recognition), Southampton, SO9 3TIJ, United Kingdom; TDepartment of Biochemistry NE2 4Hk, United Kingdom
ReceivedDecember
30, 1991
A continuous spectrophotometric assay for the EcoRV restriction endonuclease has been developed. The synthetic self-complementary oligonucleotide d(GACGATATCGTC) (which is double stranded under the assay conditions) is used as the substrate. The EcoRV endonuclease recognizes d(GATATC) sequences cutting between the central T and dA bases. Thus d(GACGATATCGTC) is converted to d(GACGAT) and dCpATCGTC) during catalysis. Both of the hexameric products are single stranded under the assay conditions. The conversion of the dodecameric substrate to the two hexameric products and the concomitant change from doubleto single-stranded DNA is associated with an increase in absorbance at 254 nm due to the hyperchromic effect. This change can be used to monitor column effluents for endonuclease activity and also for K,,, and keat determination under steady-state kinetic conditions. 0 1992 Academic Press, Inc.
Restriction endonucleases are widely used for molecular biology applications. Many have been cloned and overexpressed and are commercially available. Nevertheless many investigators, especially those interested in discovering new endonucleases and those working on the kinetic properties of known endonucleases, prepare the enzymes themselves. Both the purification of restriction endonucleases and the subsequent study of their kinetic properties require an assay for enzyme activity. Assays used during purification usually use plasmids. When a single site is present the closed circle form of the plasmid is converted to the linear form (either directly or via an open circle intermediate). Multiple restriction endonuclease sites on a single plasmid are con204
University of Southampton, Bassett Crescent East, and Genetics, The University, Newcastle-upon-Tyne,
verted to a ladder of products (l-8). Similar plasmid-based assays can be used to study the kinetic properties of purified restriction endonucleases. The enzymology of restriction endonucleases can also be studied using short synthetic oligodeoxynucleotides (9-17). These assays are based either on HPLC or on 5’-32P labeling followed by separation of radioactive substrates and products by denaturing gel electrophoresis or homochromatography. All of the above assays are discontinuous and require sampling of the mixture at discrete time points followed by separation and quantitation of substrates and products (either by HPLC or by gel electrophoresis). As such these assays are time consuming and suffer from the errors inherent in a discontinuous method. These errors arise as the rate is measured at relatively few time points whereas in a continuous assay the entire reaction is observed. A rapid continuous assay for restriction endonucleases would find wide applicability in both enzyme purification and the study of kinetic properties. It is well documented that when double-stranded DNA is heated, melting occurs and the single-stranded form is produced. This is associated with an increase in light absorbance at 254 nm. This is known as the hyperchromic effect and arises because base stacking (associated with diminished absorbance of light) in doublestranded DNA is greater than that in single-stranded DNA (18). All restriction endonucleases require doublestranded DNA as substrates and when short oligodeoxynucleotides are used they must form double strands. However, cleavage of a double-stranded oligodeoxynucleotide (say lo-16 base pairs) often results in shorter products (say 5-8 bases) that are single stranded under assay conditions. Thus it is possible to set up an oligodeoxynucleotide-based assay for restriction enzymes 0003-2697192 $5.00 Copyright Q 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
CONTINUOUS
ASSAY
FOR
RESTRICTION
that converts double- to single-stranded DNA. This should be associated with an increase in absorbance at 254 nm and be amenable to continuous monitoring. In this paper we describe such an assay using the doublestranded dodecamer d(GACGATATCGTC) and the restriction endonuclease EcoRV cleavage site GAT’ATC (19,20). MATERIALS
AND
METHODS
The EcoRV endonuclease was purified from overexpressing Escherichia coli strains as described (19,21,22). The concentration of the endonuclease was determined using an tZsO= 10.4 X lo4 M-’ cm-l (this is for a dimer of subunit molecular weight 29,000 which is the active species) (19). The endonuclease concentration was also determined using Bio-Rad protein assay based on dye binding according to the manufacturer’s instructions. Bovine serum albumin was used as a standard. Both assays gave virtually identical results. Oligodeoxynucleotides were prepared using an Applied Biosystems 381 A DNA synthesiser and phosphoramidite chemistry. The oligodeoxynucleotides were purified and desalted as described (9). The concentrations of d(GACGATATCGTC) and its 5’-phosphorylated derivative were determined using tZs4= 16.6 X lo4 M-’ cm-’ (this is for the double-stranded form which is the true substrate for the endonuclease) (9). The preparation of d(pGACGATATCGTC) was by 5’-phosphorylation of d(GACGATATCGTC) with ATP and polynucleotide kinase as described (9). Spectrophotometric assays were performed using a Kontron Instruments Uvikon 930 spectrophotometer. (a) Comparison of the Endonuclease-Catalyzed of d(GACGATATCGTC) by the Continuous HPLC Assay
Cleavage and an
To a l-cm path length quartz cuvette containing 50 mM Hepes-NaOH, pH 7.5,lOO mM NaCl, 10 mM MgCl,, and 6.8 pM d(GACGATATCGTC) was added EcoRV endonuclease to a concentration of 10 nM. The final volume was 1 ml and the increase in absorbance at 254 nm was monitored over 80 min. The temperature was maintained at 25°C by thermostating the cell compartment. Simultaneously aliquots from an identical assay mixture in an Eppendorf tube were assayed for substrate and products by HPLC as previously described (9).
205
ENDONUCLEASES
TATCGTC) was accurately determined to be 4.8 PM. An excess of EcoRV endonuclease (final concentration, 200 nM) in a small volume was added. The final absorbance after the increase had ceased (about 10 min) was noted. This difference in absorbance represents the hydrolysis of 4.8 nmol of d(GACGATATCGTC) and can be used to evaluate the change per nanomole of substrate. As a check that d(GACGATATCGTC) was completely hydrolyzed an aliquot was assayed by HPLC for substrate and products (9). (c) Rate of Hydrolysis of d(GACGATATCGTC) as a Function of EcoRV Endonuclease Concentration To a l-ml volume in a l-cm path length quartz cuvette containing 50 mM Hepes-NaOH, pH 7.5, 55 mM NaCl, 25 mM MgCl,, and 3 pM d(GACGATATCGTC) was added EcoRV endonuclease to give final enzyme concentrations between 2 and 100 nM. The change in absorbance at 254 nm was noted and used to determine hydrolysis rates. (d) Monitoring of Phosphocellulose Column Eluate during EcoRV Endonuclease Purification To a l-ml volume in a l-cm cuvette containing the buffer and oligodeoxynucleotide given in (c) were added aliquots of fractions collected during the phosphocellulose column purification of the EcoRV endonuclease (19,21). The size of the aliquots were chosen so as to give a reasonable rate of absorbance increase over about 5 min. The amount of endonuclease present in each fraction determined by this assay was compared with the levels present as determined by denaturing SDSPAGE’ electrophoresis (19,23). (e) Determination of the K,,, and k,,, Values for d(pGACGATATCGTC)
(6) Determination of Increase in Absorbance at 254 nm per Nanomole of d(GACGATATCGTC) Hydrolyzed
For accurate kinetic determinations it is best to siliconize the quartz cuvettes prior to use. The cuvettes were rinsed in a 2% solution of dimethyldichlorosilane in l,l,l-trichloroethane (BDH Ltd., Poole, Dorset, UK) for 5 min and then allowed to dry. The cuvettes were rinsed with water. Assays were carried out in either a l-ml (l-cm path length) or a 2-ml (2-cm path length) volume depending on the oligodeoxynucleotide concentration; 50 mM Hepes-NaOH pH 7.5 was shown, 55 mM NaCl, and 25 mM MgCl, was the buffer used and the final EcoRV endonuclease concentration was 5 nM. As-
The absorbance at 254 nm of approximately 5 pM d(GACGATATCGTC) in a l-cm path length cuvette containing the buffer components detailed above was noted. From this the concentration of d(GACGA-
’ Abbreviations used: SDS-PAGE, acrylamide gel electrophoresis; FPLC, raphy.
sodium dodecyl sulfate-polyfast protein liquid chromatog-
206
WATERS 15
,
AND
I 0
= hplc
assay
-6 E e
10
5 3
b s z n
GACGATATCGTC 5 2d(GACGAT)
+ 2d(pATCGTC)
0 0
10
20
30 Time
40
50
60
Imin
FIG.
1. Comparison of the rates of hydrolysis of d(GACGATATCGTC) by the continuous spectrophotometric method (solid line) and by a discontinuous HPLC-based assay (squares). The amount of single-stranded (s/s) product formed is given as a function of time.
says were carried out at 25°C. The d(pGACGATATCGTC) concentrations were 0.28, 0.52, or 0.99 WM (2-ml volume) and 1883.56, or 6.05 PM (l-ml volume). The reactions were monitored over about 30 min and the initial rates determined from the linear portion of the trace (usually the first 10 min). K,,, and IZcatvalues were determined from a plot of [substrate]/velocity versus [substrate] (24,25).
RESULTS
AND
DISCUSSION
Incubation of d(GACGATATCGTC) with the EcoRV endonuclease gives d(GACGAT) and d(pATCGTC). The dodecamer has a 7’, of 53°C (9) and so is double stranded under the assay conditions of 25°C. However the two hexamers are too short to form stable double strands at 25°C and so exist in single-strand form. Thus as shown in Fig. 1 the EcoRV endonuclease converts a double-stranded dodecamer into two single-stranded products. This is associated with an increase in absorbance at 254 nm due to the hyperchromic effect as is also shown in Fig. 1 and this forms the basis of a rapid and continuous assay for the enzyme. Superimposed on the curve for the absorbance change in Fig. 1 are values in which the rate has been determined by a previously described HPLC assay (9). Both methods give the same rate. However, the continuous spectrophotometric assay requires only 80 min assay time. Each time point for the discontinuous HPLC assay requires about 30 min. The spectrophotometric assay is therefore not only more accurate, by virtue of being continuous, but much quicker than the HPLC method.
CONNOLLY
In order to convert the curve in Fig. 1 to a rate in units of nmol substrate hydrolyzed min-’ mg-’ enzyme it is necessary to determine the change in absorbance at 254 nm due to the cleavage of 1 nmol of substrate. To do this a known amount of d(GACGATATCGTC) was hydrolyzed completely by an excess of the endonuclease and the change in absorbance noted. This experiment is illustrated in Fig. 2, which shows an increase in absorbance at 254 nm of 0.165 for the complete hydrolysis of 4.8 nmol of d(GACGATATCGTC). This corresponds to a change in absorbance of 0.034 per nanomole d(GACGATATCGTC) cleaved. This represents an approximate 20% increase in absorbance for complete conversion of cl(GACGATATCGTC) to single-stranded products. The hyperchromicity observed upon endonuclease cleavage depends on ionic conditions. The ionic strengths of all the buffer solutions used in the assays were identical and the amounts (in terms of both concentrations and volumes) of enzyme and oligodeoxynucleotide added were too small to significantly alter this. Although slightly different conditions were used in assays (a) and (b) as compared to (c), (d) and (e), no differences in hyperchromicities were observed under the two conditions. The assay described here is based on an increase in absorbance at 254 nm and this limits the substrate concentration that can be used. Too low an initial absorbance give poor signal-to-noise and too high approaches the end of the linear response of the spectrophotometer. We have found that initial absorbances of between 0.05 and 1.25 give good results. In a l-ml volume and a l-cm path length these correspond to d(GACGATATCGTC) concentrations of between 0.3 and 7.5 PM. The range of the assay can be extended by
Initial APs4 = 0.800 Final A,,, = 0.965 AA,,,,
’
o’60 -h
0
1 I 1
AA,,, ,
per
hydrolyseq
10
oligonucleotide
= 0.034 20
Time
FIG. 2.
nmol
= 0.165
, 30
/min
Determination of the absorbance change caused per nanomole of d(GACGATATCGTC) completely hydrolyzed by the EcoRV endonuclease; 4.8 nmol of d(GACGATATCGTC) was present initially in a l-ml volume. At the break in the trace (after 1.5 min) an excess of EcoRV endonuclease was added and the reaction allowed to proceed to completion.
CONTINUOUS
ASSAY
FOR
RESTRICTION
207
ENDONUCLEASES
addition the continuous assay has been used to determine Km and kc,, values for oligodeoxynucleotides containing dC and dG analogues. These data will be reported on later and complement our earlier studies using analogues of dA and T (9,lO). It should be noted that both T, values and hyperchromicities have a dependence on oligodeoxynucleotide concentration. However, over the oligomer concentration range used here the variation in hyperchromicity with concentration was extremely low and did not affect K,,, and k,,, determination. 0
20
40
[endonucleasel FIG. 3. Dependence on the rate of TATCGTC) as measured by the continuous say on EcoRV endonuclease concentration.
60
60
100
/nM hydrolysis of d(GACGAspectrophotometric as-
using cells of different path lengths. Cells with path lengths of between 0.25 and 5 cm are available allowing an extension of the starting concentration of d(GACGATATCGTC) by about five times in either direction. Alternatively this range could be extended toward higher substrate concentrations using a higher wavelength, where oligodeoxynucleotides have a lower absorbance. However, at these higher wavelengths changes in absorbance tend to be lower, thus reducing signal/noise. Figure 3 shows that the continuous spectrophotometric assay gives a linear response to EcoRV endonuclease concent,ration over the range 2-100 nM. As an application of the above assay we have monitored the eluate from the phosphocellulose column that forms the first step of the purification of the EcoRV endonuclease (19,Zl). In the absence of a convenient assay this column has previously been monitored by SDS-PAGE. As is shown in Fig. 4 both assays correspond exactly. The continuous assay requires 5 min for each time point, a total time of 50 min. The entire gel electrophoresis procedure (including sample preparation, electrophoresis, staining, and destaining) requires 2-3 h. Finally we have determined the Km and kc,, values for d(pGACGATATCGTC) as shown in Fig. 5. Very good quality data resulting in excellent fits to plots of [substrate]/velocity versus [substrate] can be rapidly obtained. The Km value of 2.92 PM obtained agrees very well with that of 3.8 FM obtained previously using an assay based on 32P labeling and gel electrophoresis (10). A kcat (33.2 min-“) higher than the previous value of 6.9 min-’ is found here. This is due to a change in the final purification matrix from gel filtration to an FPLCbased method. The FPLC method consistently gives more active protein than that obtained previously. In
CONCLUSION A continuous spectrophotometric assay for use with restriction endonucleases has been described and evaluated using the EcoRV enzyme. The advantages of the assay are convenience, speed, and accuracy. A slight disadvantage is the limited range of substrate concen-
10
20
30
Fraction
10 13 18 20
40
50
Number
22 24 28 30 32
Fraction
Number
FIG. 4. Assay by the continuous spectrophotometric fractions eluted from a phosphocelluiose column during endonuclease purification. For comparison an SDS-PAGE the fractions is also given. The position of the endonuclease is arrowed.
method of the EcoRV pattern of on the gel
208
WATERS
0
Time
CONNOLLY
0
a
2
AND
6
6
0
0.5
/min
I -3.0
1.0 Time
I 0.0
I 3.0 [Sol
1.5
2
/min
I 6.0
/pM
FIG. 5.
Determination of the K,,, and k,, for d(pGACGATATCGTC) using the continuous spectrophotometric assay. (A) The rates observed at d(pGACGATATCGTC) concentrations (yM) of: (a) 6.05, (b) 3.56, (c) 1.86. (B) The rates observed at: (d) 0.99, (e) 0.52, (f) 0.28. Note that the volume used in (A) was 1 ml whereas in (B) 2 ml was used. (C) Plot of substrate concentration/velocity versus substrate concentration for K, and kc., determination.
trations that can be used. The assay has proved useful both in the purification of the EcoRV endonuclease and in the evaluation of kinetic constants. Almost all restriction endonucleases use synthetic oligodeoxynucleotides and it is easy to manipulate this system so that a double-stranded substrate is converted to single-stranded products. The continuous assay should therefore be widely applicable and should form a very useful addition to the assay methods already available for restriction enzymes. ACKNOWLEDGMENTS T.R.W. was supported by a UK SERC research studentship. B.A.C. is a research fellow of the Lister Institute of Preventive Medicine. Special thanks go to Mrs. S. Broomfield for preparation of the manuscript.
REFERENCES 1. Modrich, P., and Roberts, R. J. (1982) and Roberts, R. J., Eds.), pp. 109-154, ratory, Cold Spring Harbor, NY.
in Nucleases Cold Spring
(Linn, Harbor
S. M., Labo-
2. Bennett, S. P., and Halford, Cellular Regulation (Horecker, Vol. 30, pp. 57-104, Academic
S. E. (1989) in Current Topics in B. L., and Stadtman, E. R., Eds.), Press, London.
3. Sharp, P. A., Sugden, 12,3055-3063.
Sambrook,
B., and
J. (1973)
Biochemistry
4. 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. 5. Halford, S. E., and Johnson, N. P. (1980) Biochem. J. 191,593604. 6. Rubin, R. A., and Modrich, P. (1978) Nucleic Acids Res. 5, 29912997. 7. Maxwell, 8. Halford, 1777.
A., and Halford, S. E. (1982) Biochem. J. 203, 85-92. S. E., and Goodall, A. J. (1988) Biochemistry 27, 1771-
9. Newman, P. C., Nwosu, V. U., Williams, D. M., Cosstick, Seela, F., and Connolly, B. A. (1990) Biochemistry 29,9891-9901. 10.
Newman, P. C., Nwosu, V. U., Williams, D. M., Cosstick, Seela, F., and Connolly, B. A. (1990) Biochemistry 29,9902-9910.
11.
Mazzarelli, chemistry
12. Fliess, Frank, 3474.
J., Scholtissek, 28,4616-4622.
S., and McLaughlin,
A., Wolfes, H., Rosenthal, A., Schwellus, R., and Pingoud, A. (1986) Nucleic Acids
R., R.,
L. W. (1989)
Bio-
K., Blocker, Res. 14,
3463-
H.,
CONTINUOUS 13. Fliess, A., Wolfes, H., Seela, Acids Res. 16, 11,781-11,793. 14. Brennan, C. A., van Cleve, Biol. Chem. 261,12X-1218. 15. Seela, F., and Kehne, 16. Lesser, D., Kurpiewski, 250, 776-786.
F., and
18.
Pingoud,
M. D., and
and
A. (1988)
Gumport,
A. (1987) Biochemistry M., and Jen-Jacobsen,
17. Aiken, C. R., McLaughlin, L. W., Biol. Chem. 266, 19,070-19,078.
ASSAY
26,
Gumport,
FOR
RESTRICTION
Nucleic
R. I. (1986)
J.
2232-2238. L. (1990) Science R. I. (1991)
20.
Schildkraut, (1984) Gene
21.
Bougueleret, L., Tenchini, (1985) Nucleic Acids Res.
I., Banner, 27, 327-329.
C. D.,
Rhodes,
C. S., and
Parekh,
S.
J., and
Zabeau,
M.
M. L., Botterman,
13, 3822-3839.
22.
Luke, P. A., McCallum, S. A., and Halford, S. E. (1987) in Gene Amplification and Analysis (Chirikian, J. G., Ed.), Vol. 5, pp. 183205, Elsevier, New York.
23.
Laemmli,
24.
Hanes,
25.
Dixon, M., and Webb, E. C. (1979) Longmans Green. London.
J.
Saenger, W. (1984) Principles of Nucleic Acid Structure, pp. 141149, Springer-Verlag, New York. 19. D’Arcy, A., Brown, R. S., Zabeau, M., van Resandt, R. W., and Winkler, F. K. (1985) J. Biol. Chem. 260, 198771996.
209
ENDONUCLEASES
U. K. (1970) C. S. (1932)
Nature
Biochem.
227, J. 26,
680-685. 1406-1421 Enzymes,
3rd
ed., pp. 61-62,
