T4 Polynucleotide Kinase: Macromolecular Crowding Increases the Efficiency of Reaction at DNA Termini

Transcription

1 ANALYTICAL BIOCHEMISTRY 158, (1986) T4 Polynucleotide Kinase: Macromolecular Crowding Increases the Efficiency of Reaction at DNA Termini BARBARAHARRISON ANDSTEVEN B. ZIMMERMAN Laboratory of Molecular Biology. National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland Received April 25, 1986 The amount of reaction catalyzed by T4 polynucleotide kinase on a variety of its substrates is greatly increased in the presence of polyethylene glycol 8000 (PEG 8000). Both the forward and reverse reactions as well as the exchange reaction can be stimulated. The stimulation is a general effect on T4 polynucleotide kinase reactions involving high molecular weight DNA substrates. The use of PEG 8000 is particularly advantageous for labeling or removing terminal 5 -phosphate groups which are only slowly or incompletely labeled or removed under ordinary conditions, such as those at recessed termini or at nicks in duplex DNA, although the reaction on bluntended or protruding termini is also increased. It is further advantageous for labeling very low Concentrations of substrates Academic PKSS. Inc. KEY WORDS: DNA; DNA modification enzymes; nucleic acid chemistry; DNA sequencing; recombinant technology; kinases. The ease with which a particular class of termini in duplex DNA can be phosphorylated by T4 polynucleotide kinase (EC , reviewed in Ref. ( 1,2)) is affected by the secondary structure at the DNA terminus (3). Termini with protruding 5 -phosphates are generally readily labeled, whereas blunt ends are less easily labeled and termini with recessed 5 -termini or such termini at single-strand breaks in duplex DNA (nicks) are particularly difficult to label. Even partial reaction at recessed 5 -ends or nicks generally requires large amounts of kinase and high ATP concentrations (3-6). We have described a large increase in the exchange reaction of T4 polynucleotide kinase in concentrated polymer solutions (7). This effect allows the ready labeling of 5 -termini of DNA in general, including those at recessed, protruding, or blunt ends, as well as those at nicks. The role of the PEG 8000 was shown Abbreviations used: bp, base pair; Kbp, kilobase pair; PEG, polyethylene gfycol; T4 PNK, T4 polynucleotide kinase. to be a stabilization of the kinase under reaction conditions where the enzyme has a high turnover number. Other high molecular weight polymers besides PEG 8000 caused similar effects whereas related materials of lower molecular weight were relatively ineffective. Such a pattern suggested that the stabilizing effect on kinase activity was caused by macromolecular crowding. We here apply this effect to the practical problems of labeling of DNA termini. The 5 -hydroxyl termini of a polynucleotide substrate can be labeled either by a direct reaction with [Y-~~P]ATP in the forward reaction of the kinase (Reaction 1 of Eq. [ 11) or by the exchange reaction which this enzyme catalyzes in the presence of [y-32p]atp and ADP (Reactions 1 and 2 of Eq. [ 11). 5 -hydroxyl terminus + ATP ti 2 5 -phosphate terminus + ADP Most commonly, the DNA substrates to be labeled are terminated by S-phosphate groups. [l] $3.00 Copyright by Academic Press. Inc. All rights of reproductmn in any form reserved.

2 308 HARRISON AND ZIMMERMAN To label such substrates by the forward reaction of the kinase requires a preliminary enzymatic removal of the phosphate groups, a step which is not trivial for S-phosphate groups at recessed termini or at nicks. This problem is avoided and labeling procedures are simplified by using the exchange reaction. Berkner and Folk have studied the T4 PNK exchange reaction in detail in the absence of PEG and have optimized conditions for this reaction (6,8). They have suggested the use of the reverse reaction of the enzyme as a means of dephosphorylation of DNA termini (6). Our finding that PEG causes a large stimulation of this otherwise relatively slow reaction should allow a wider use of the kinase as a dephosphorylating agent. Materials MATERIALS AND METHODS SphI, BamHI, PvuII, PstI, and HpaII nucleases and HaeIII nuclease-digest of $X 174RF DNA were purchased from Bethesda Research Laboratories. ADP, ATP, (pdt)ro, HaeIII nuclease, T4 polynucleotide kinase, and T4 DNA ligase were from Pharmacia; Bg/I nuclease was from International Biotechnologies; Escherichia coli alkaline phosphatase (BAPC), pancreatic DNase I, and pancreatic RNase A were from Millipore; PEG 200, PEG 8000, and imidazole were from Baker; [T-~~P]ATP was from New England Nuclear; and dithiothreitol from Calbiochem. pbr322 DNA was digested with various restriction nucleases under conditions described by the suppliers. pbr322 DNA was digested with pancreatic DNase I until the digest contained an average of 0.5 single-strand break/ pbr322 molecule as determined by the level of exchange labeling with excess kinase after heat denaturation of the DNA. E. coli DNA was digested with several levels of pancreatic DNase and the average single-strand length was similarly judged. Digested DNA was reisolated by phenol-chloroform extraction and ethanol precipitation. DNA was denatured by 5 min at 100 C in 10 mm Tris-HCl buffer (ph 8.0)-O. 1 mm EDTA, and quenched in ice water. Partial digests of poly(ra) with micrococcal nuclease were prepared as before (9). pbr322 DNA was either a gift from Dr. M. Gellert or the product of Bethesda Research Laboratories or Pharmacia. The commercial preparations contained small amounts of relatively low molecular weight materials which were highly labeled by T4 PNK. These materials were routinely removed by treatment with boiled pancreatic RNase A (10) and the DNA reisolated by extraction with phenolchloroform and chloroform-i-amyl alcohol followed by ethanol precipitation before use as a substrate. Use of very high specific radioactivity [y- 32P]ATP (ca Ci/mmol) for these reactions has occasionally caused decreased and variable kinase activity when the high specific radioactivity ATP stock was held for more than a few days before use, apparently due to the accumulation of degradation products. This problem, which occurs irrespective of the presence of the PEG 8000 in the assay, is minimized by the use of a lower specific radioactivity stock of [Y-~~P]ATP (e.g., 45 Ci/mmol). Methods Assay of the kinase exchange reaction. Unless otherwise specified, the exchange reaction of T4 PNK was assayed at 37 C for 20 min in polypropylene tubes (Eppendorf) in mixtures (final volume of 10 ~1) containing 50 mm imidazole-hcl buffer (ph 6.4), 4.5 tnm dithiothreitol, 18 mm MgC12, 0.1 mm ADP, 12 PM [T-~~P]ATP (2-20 Ci/mmol), 0.1 pg of the specified DNA, PEG, and other additions as indicated, and 0.1 vol of kinase dilution in a diluent (11) of 50 mm Tris-HCl (ph 7.6)-10 mm fi-mercaptoethanol-0.5 mg/ml of bovine plasma albumin. Kinase dilutions in the above diluent were stable for at least 20 min on ice. The reaction mixtures were assayed for radioactivity migrating at the position of the specified DNA on agarose gels or for acid-precipitable radioactivity, or for both. For the gel assay, the kinase reaction was

3 T4 POLYNUCLEOTIDE KINASE 309 stopped by adding 5 ~1 of mixture A (90 mm EDTA- 14% glycerol-o. 14 mg/ml bromphenol blue-2.7% sodium dodecylsulfate) and the samples applied to agarose minigels (0.8% agarose (Sigma Type II, medium EEO) in 89 mm Tris-89 mm boric acid-2.5 mm EDTA). After 90 min at 70 V (Mini-cell, Bio-Rad), the gel was stained 10 min in ethidium bromide in water, destained 20 min in water, and photographed under ultraviolet illumination. The gel was then dried under vacuum (Model SE-540 dryer, Hoefer Scientific) and evaluated by autoradiography. Autoradiograms were quantitated with a Joyce-Loebl recording microdensitometer; lanes with serial dilutions of labeled DNA standards were present on each gel for calibration. For assay of acid-precipitable radioactivity, the kinase reaction was stopped by adding 0.2 ml of 0.36 mg/ml bovine plasma albumin- 10 mm NaP,O,, followed by 0.2 ml of cold 10% trichloroacetic acid. After 5 min at 0 C the tubes were centrifuged for 5 min at 15,000g at 5 C. The pellets were redissolved in 0.1 ml of 0.1 N NaOH, 0.4 ml of cold 5% trichloroacetic acid was added, and the precipitation repeated twice more. Finally, the pellets were dissolved in 0.5 ml of 2 M NH40H, plated in stainless steel planchets, and their radioactivity measured in a low background (0.6 cpm) Geiger counter (Tracerlab). If samples were to be analyzed by both the gel and acid precipitation procedures, the kinase reactions were stopped by addition of 2 ~1 of 0.2 M EDTA and aliquots were removed for those procedures. Assay of the forward reaction of the kinase. Reaction mixtures were as described above for the exchange assay except the DNA was a PstI nuclease digest of pbr322 DNA that had been dephosphorylated by incubation with T4 PNK under conditions where >70% of the 5 -phosphate termini were removed, after which the DNA was reisolated. Samples were assayed by the gel procedure as above. Assay of the reverse reaction of the kinase. Reaction mixtures were as described above for the exchange assay except unlabeled ATP was used and the DNA was a PstI nuclease digest of pbr322 DNA that had been labeled by the exchange reaction and subsequently reisolated. Samples were assayed by the acid precipitation procedure as above. Assay for sedimentability of DNA. Kinase assay mixtures as described above were centrifuged for 5 min at room temperature at 15,OOOg (Eppendorf). The supernatant fluids were cautiously removed, the pellets redispersed in the same volume in 10 mm Tris * HCl buffer (ph 8.0)-O. 1 mm EDTA and an aliquot of each fraction was mixed with mixture A (above) and assayed by the gel procedure described above. P-labeling of DNA standards. HaeIII nuclease-digested 4X 174RF DNA was labeled by exchange in the presence of 6% w/v PEG 8000 under the conditions suggested below. RESULTS Polyethylene Glycol Increases the Labeling of Recessed, Protruding, or Blunt-Ended S-Termini of DNA The effect of addition of PEG 8000 on labeling of protruding or blunt-ended DNA substrates (Fig. 1) is similar to that demonstrated earlier on recessed 5 -termini (7). The range of PEG 8000 concentrations in which the increase in exchange occurs is similar for all of these substrates as well as for two further examples of recessed termini (BglI or QhI nuclease digests of pbr322 DNA) for which data are not shown. For all of these substrates, at concentrations up to 4% w/v PEG 8000, there is no marked stimulation. Then, abruptly, between 4 and 6% PEG 8000, the amount of exchange reaction is increased by loo- to looo-fold over the amount in the absence of PEG. Finally, at still higher PEG 8000 concentrations, the amount of reaction gradually decreases. As a further example of the utility of the effect, in the absence of PEG, even at very high T4 PNK levels only about 20% of the theoretical limit of exchange labeling was attained on the recessed termini formed by PstI nuclease, whereas with 6% w/

4 310 HARRISON AND ZIMMERMAN PEG8000#%: FIG. 1. Effect of PEG 8000 concentration on exchange-labeling of recessed, protruding, or blunt-ended DNA termini by T4 polynucleotide kinase. Kinase exchange assay mixtures, as under Methods, contained &I, BumHI, or PvuII nucleasedigested pbr322 DNA in A, B, or C, respectively, and the indicated concentrations of PEG Kinase concentrations were 5.4,O. 17, or 1.7 units/ml in A. B, or C, respectively. Reaction mixtures were assayed by the gel procedure. The only significant radioactivity visible on the autoradiograms was that of the bands shown above which migrated at the position of linear pbr322 DNA. v PEG 8000, about 70% of that limit was reached-and at much lower enzyme levels (Fig. 2). Generally, product formation in these reactions in 6% w/v PEG 8000 is approximately proportional to time within the limits expected for an exchange reaction. However, with certain substrates, particularly DNA with recessed 5 -termini, there is a reproducible lag of several minutes which is then followed by a reaction that is essentially proportional to time. The origin of the lag is unknown. PEG Does Not Change the Nature of the Kinase Products It was possible that the increased labeling due to PEG was occurring at nicks or sites other than the legitimate 5 -termini of the duplex DNA substrates used above. Accordingly, two types of experiments were used to characterize the 32P-labeled DNA products made by the kinase on recessed, protruding, or bluntended termini in the presence of 6% w/v PEG In the first, recessed termini generated by SphI nuclease digestion of pbr322 DNA were labeled by the PEG-stimulated kinase exchange reaction and the kinase products were ligated by T4 DNA ligase. The substrate was extensively ligated as shown by its altered mobility on agarose gels. The ligated products were then recut with the same restriction nuclease that had been used to form the original S-termini. The result was that the ligated products were completely recut. When the recut sample was further tested, it was found to be sensitive to alkaline phosphatase. These results indicate that the structure of the termini is not altered by the labeling procedure, that the labeled termini are substrates for DNA ligase, and that the label is exactly at the expected point of restriction, and not, for example, at adventitious nicks or at exonucleasetrimmed ends. This type of experiment was also done with similar results using DNA with recessed termini generated by PstI nuclease labeled in the exchange reaction in 6% w/v PEG 8000.

5 T4 POLYNUCLEOTIDE IUNASE 311 1cKl 200 3cY3 400 Em T4 POLYNUCLEOTIDE KINASE, units/ml FIG. 2. Effect of 6% w/v PEG 8000 on the amount of exchange-labeling of recessed DNA termini by large amounts of T4 polynucleotide kinase. Kinase exchange assay mixtures, as under Methods, contained PstI nucleasedigested pbr322 DNA and kinase as indicated. Reaction mixtures were assayed by the acid-precipitation procedure. Results are expressed as percentage ofthe theoretical maximum. A second type of experiment was used to characterize the kinase products on bluntended or protruding termini as well as on these same two examples of recessed termini. In this second type of experiment, DNA termini generated by a given restriction nuclease were also labeled by exchange in 6% w/v PEG 8000 but then the DNA was recut with a second restriction nuclease. The second nuclease was chosen to localize the original S-termini on small DNA fragments which were subsequently resolved on acrylamide gels. In all cases tested, the radioactivity was only present in the two fragments which were expected from the original site of nuclease cleavage, the sum of the two fragments representing 2-6% of the length of pbr322. This second type of experiment as applied to the recessed termini formed by PstI is shown in Fig. 3. Even lo-fold longer exposures failed to show radioactivity other than in the expected fragments. Similar data were also obtained for,sphi fragments redigested by HpaII, for BumHI fragments with protruding termini redigested by H&II, and for PvuII fragments with blunt-ended termini redigested by HpaII nuclease (data not shown). In addition, separate aliquots of all the original labeled samples were also ligated and then recut with the original restriction enzyme. Both the ligation and recutting were quantitative in each case, again indicating that the structures of the original termini were not altered by the kinase reaction. The above experiments indicate the chemical structure of the kinase products is unaffected by use of PEG. The overall conformation of large duplex DNA can, however, be reversibly altered in media containing PEG and Mg +. Both the labeled DNA product and the DNA substrate in reaction mixtures containing 6% w/v PEG 8000 and 18 mm Mg*+ are readily sedimentable (see Methods). Apparently the DNA undergoes the psi transition discovered by Lerman in which DNA tends to become highly collapsed in the presence of concentrated polymers and Mg2+ or high monovalent salt concentrations ( 12,13). There is a good correlation between the PEG 8000 concentrations which cause the abrupt stimulation in kinase exchange rate and those which cause the DNA to become sedimentable. This sedimentability provides a simple means for isolation of the product and for its separation from the PEG 8000 (see Comments on the Use of PEG-Stimulated Kinase Reactions, below). The DNA is no longer readily sedimentable after removal of the PEG or when the Mg2+ is complexed with EDTA. PEG Increases Exchange Labeling at Nicks in DNA Pancreatic DNase I was used to introduce nicks, i.e., 5 -phosphate terminated singlestrand breaks, into either pbr322 DNA (ca. 0.5 nick/molecule of 4400 bp) or into E. coli DNA (ca. 1 nick/200 bp). The addition of 8 to 10% w/v PEG 8000 increased the amount of exchange labeling by a given amount of kinase on either nicked substrate from lo- to loo-fold, the extent depending on the assay period and kinase level chosen. As expected, the radioactivity in these labeled products could be converted by T4 DNA ligase to a form resistant to phosphomonoesterase action

6 312 HARRISON AND ZIMMERMAN WELLS- ) Hae Illnuclease, units: FIG. 3. Localization of 32P in a DNA with recessed S-termini after exchange-labeling in PEG 8000 solution. A PsfI nuclease digest of pbr322 was labeled under kinase exchange conditions, as under Methods, using 170 units/ml of T4 PNK in the presence of 6% w/v PEG The DNA was reisolated by phenolchloroform extraction and ethanol precipitation. Aliquots (50 ng DNA in 10 ~1 total volume) were digested with the indicated amounts of Hue111 nuclease and applied to a 12% acrylamide-7 M urea gel (18) from which this autoradiogram was prepared. Aliquots of a Hoe111 nuclease digest of +X174 (differing ninefold in amount of DNA, lanes 6 and 7) serve as molecular weight standards and also illustrate the relatively diminished kinase labeling in 6% w/v PEG 8000 of lower molecular weight DNA (see text). (assayed as Ref. ( 14)). The PEG 8000 concentration-dependence (Fig. 4) is qualitatively similar to that for its effect on the other types of termini described above, but maximal labeling of nicks occurs at a slightly higher PEG 8000 concentration than is optimal for the other types of termini described above. PEG Efects on Exchange Labeling of 5 - terrnini of Single-Stranded Polynucleotides or ofhp)lo The effect of PEG on T4 PNK exchange labeling of single-stranded DNA was tested on a heat-denatured partial DNase I digest of E. coli DNA (average chain length = 1800 residues). A stimulation of ca. 1 O-fold was caused by 9% w/v PEG The PEG 8000 concentration-dependence was similar to that shown for nicks in Fig. 4. The reaction was poorly proportional to time and enzyme so that the changes in apparent rates are only estimates. The exchange reaction on (pdtho was proportional to time for ca. 5 min at 37 C and was unaffected by the presence of PEG 8000 (0.3 PM decamer in 0 to 12% PEG 8000; assayed on 12% acrylamide-7 M urea gels). The

7 T4 POLYNUCLEOTIDE KINASE 313 WELLS- ; NICKED CIRCULAR DNA - LINEAR DNA - PEGSOOO,%:O S FIG. 4. Effect of PEG 8000 concentration on exchange labeling of nicked DNA. Kinase exchange assay mixtures, as under Methods, containing DNase I-nicked pbr322 DNA, 54 units/ml of T4 PNK, and PEG 8000 as indicated were assayed by the gel procedure. The linear DNA band had <) of the intensity of the nicked DNA band by ethidium bromide staining of this gel. rates of labeling of S-hydroxyl terminated oligo(ra) fractions (chain lengths from about 30 to 300 residues) using the forward reaction of the kinase were also unchanged in 6% w/v PEG 8000 but were inhibited at higher PEG concentrations. Effects of Reaction Components on the Exchange Reaction in 6% w/v PEG 8000 Berkner and Folk optimized the T4 PNK exchange reaction in the absence of PEG (6,8). We initially compared activity changes due to PEG on recessed DNA termini both under their conditions and under those recommended by Maniatis et al. ( 15). Since the PEG effects were at least as large in the assay system of Berkner and Folk, that system was used for further testing. In the experiments of this section, we have surveyed the effects of reaction components on the exchange-labeling of a PstI nuclease digest of pbr322 DNA in 6% w/v PEG As a result of this survey, we made two minor modifications to the Berkner and Folk conditions. First, gelatin has been omitted from their formulation with an improvement in reproducibility in media containing PEG and no significant change in the amount of labeling. Second, 50 mm imidazole-hcl buffer (ph 6.4) replaces 25 tytm imidazole- HCl buffer (ph 6.6). ph and ionic conditions. The exchange reaction in the absence of PEG is optimal between ph 6 and 7 (6,16) and is very sensitive to ionic strength (6,8). The response of the kinase to changes in ph or ionic strength in media containing 6% w/v PEG 8000 is generally similar to that in the absence of PEG on the same substrate. The rate of T4 PNK exchangelabeling in imidazole-hcl buffers in 6% w/v PEG 8000 media was the same at ph 6.1,6.4, and 6.7, and somewhat decreased at ph 7.1. At higher buffer concentrations, the reaction in 6% w/v PEG 8000 was inhibited: the rate was the same in 25 or 50 mm imidazole-hcl buffer, ph 6.4, but decreased ca. two- and fourfold at 75 and 100 IrtM buffer, respectively. Additions of NaCl caused effects comparable to those of the same molar concentrations of buffer. The rate in 6% w/v PEG 8000 is greatly influenced by the Mg*+ concentration. At 12 tnm MgC12, there is cf the rate at 15, 18, or 24 tnm MgC12. The DNA is not sedimentable (see above) in the 12 InM MgC12 medium, where there is less reaction, but is fully sedimentable at 15, 18, or 24 tnm MgC12.

8 314 HARRISON AND ZIMMERMAN ADP and A TP concentrations. Omission of ADP decreased the exchange labeling in 6% w/v PEG 8000 by > 1 O-fold, while 50 pm ADP resulted in a rate significantly lower than the rate at the usual assay concentration of 100 pm ADP. Varying the ATP concentration between 6 and 24 pm gave little change in rate. DNA molecular weight. In 6% w/v PEG 8000, the rate of exchange labeling of short DNA duplexes by low levels of T4 PNK is less than that on longer duplexes. For example, the HaeIII nuclease digest of 4X 174RF DNA used as a size standard in Fig. 3 was labeled under the suggested conditions (see below); it contains fragments of sizes from 1353 to 72 base pairs. The ends of those fragments from 1353 to 3 10 base pairs were uniformly labeled, whereas the ends of those of shorter length showed decreasing labeling with decreasing chain length. Differential rates of labeling by T4 PNK were also noted in the absence of PEG, but the effects were much smaller. Small DNA fragments may not be subject to increased labeling in PEG simply because they are too small to undergo the psi transition mentioned above. Consistent with this suggestion, the 125-bp duplex fragment in a Hind111 nuclease digest of XDNA was not readily sedimentable (see above) and did not become highly labeled whereas larger fragments (20.56 kbp) were both highly labeled and sedimentable. Comments on the Use of PEG-Stimulated Kinase Reactions In addition to PEG, reaction mixtures should contain 50 mm imidazole-hcl buffer (ph 6.4), 4.5 mm dithiothreitol, 18 mm MgQ, 0.1 I nm ADP, 12 pm ATP, DNA, 0.1 vol of kinase dilution (in a diluent of 50 mm Tris-HCl (ph 7.6), 10 mm,f3-mercaptoethanol, and 0.5 mg/ml of bovine plasma albumin), and water. Reactions are incubated at 37 C. The concentration of PEG is of critical importance. In all of the dozens of PEG-stimulated kinase reactions which we have examined, there has been an abrupt transition be- tween low and high kinase rates within a narrow range of PEG concentrations. That range varies slightly with the particular DNA substrate and can be affected by changes in Mg2+ or salt concentrations. It may therefore be useful to test several PEG 8000 levels (e.g., 6, 8, and 10% w/v) for a given application. As noted elsewhere (7), the PEG effect on the exchange reaction is accompanied by generally similar effects on both the forward and reverse reactions of the kinase. No attempt has been made to optimize the conditions for the forward or reverse reactions. However, the above conditions support very high rates of the forward reaction (ADP omitted) or the reverse reaction (ATP omitted) in the presence of 6% w/v PEG The standard DNA concentration in these studies is arbitrarily fixed at 10 pg/ml, but a large PEG effect on exchange labeling of PstI fragments also occurred at DNA concentrations as low as 0.03 pg/ml. The size of the polynucleotide substrate can be important. As noted above, the phosphorylation of duplex DNA of < ca. 300 bp in length is less affected by the presence of PEG. DNA can be readily isolated from the PEG 8000 by extraction of the PEG with CHC13 (17) or simply by centrifuging the reaction mixtures, since, as described above, the DNA is readily sedimentable under conditions which cause a stimulation of the kinase. Finally, it may be noted that low levels of PEG 8000 have been innocuous in a number of experimental situations and may not need to be removed. DISCUSSION The PEG-dependent stimulation of the kinase reactions appears to be a general effect on DNA substrates that are of sufficient size. Kinase reactions on RNA-DNA hybrids or on double-stranded RNA might also be similarly affected. For certain substrates of T4 PNK such as single-stranded DNA or duplex DNA with protruding 5 -termini, there may be little practical advantage to introducing PEG into the reaction mixtures. These sub-

9 T4 POLYNUCLEOTIDE KINASE 315 strates can be phosphorylated at high rates in the absence of crowding, particularly at relatively high DNA and kinase concentrations. However, for substrates such as those with recessed S-termini, or for blunt-ended or nicked substrates, the stabilization by PEG can provide very large increases in the amounts of phosphorylation and dephosphorylation that occur in incubations of conventional duration. Even for those substrates which are phosphorylated at high rates, the addition of PEG may allow more efficient reaction at very low substrate concentrations. ACKNOWLEDGMENTS We thank Martin Gellert and Gary Felsenfeld for their comments and Betty Canning for her expert assistance with the manuscript. REFERENCES I. Kleppe, K., and Lillehaug, J. R. (1979) in Advances in Enzymology (Meister, A., ed.), Vol. 48, pp , Wiley, New York. 2. Richardson, C. C. (1981) in The Enzymes (Boyer, P. D.. ed.), Vol. 14A, pp Academic Press, New York. 3. Lillehaug, J. R., Kleppe, R. K., and Kleppe, K. (1976) Biochemistry 15, Weiss, B.. Live, T. R., and Richardson, C. C. (1968) J. Biol. Chem. 243, Chaconas, G., van de Sande, J. H., and Church, R. B. (1975) B&hem. Biophys. Res. Commun. 66, Berkner, K. L., and Folk, W. R. (1977) J. Biol. Chem. 252, Harrison, B., and Zimmerman, S. B. (1986) Nucleic Acids Res. 14, 1863-l Berkner, K. L., and Folk, W. R. ( 1979) J. Biol. Chem. 254, Harrison, B., and Zimmerman, S. B. (1984) Nucleic Acids Rex 12, I. 10. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual, p. 45 1, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 11. Richardson, C. C. (1965) Proc. Nacl. Acad. Sci. USA 54, Lerman, L. S. (197 1) Proc. Nat/. Acad. Sci. USA 68, Auer, C. (1978) Vanderbilt University, Thesis. 14. Zimmerman, S. B., and Levin, C. J. (1975) J. Biol. Chem. 250, Maniatis, T., Fritsch, E. F., and Sambrook. J. (1982) Molecular Cloning: A Laboratory Manual, p. 127, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 16. Van de Sande, J. H., Kleppe. K., and Khorana. H. G. (1973) Biochemistry 12, Rudin, L.. and Albertsson, P.-A. (1967) Biochim. Biophys. Acta 134, Maniatis, T.. Jeffrey, A., and van de Sande, H. (1975) Biochemistry 14,

10 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具