sign in sign up
<<
Home , DNA polymerase

Proc. Nat. Acad. Sci. USA Vol. 70, No. 5, pp. 1326-1330, May 1973

Accessibility of DNA in to DNA and RNA Polymerase (chromatin/DNA polymerase/RNA polymerase) BERT SILVERMAN AND ALFRED E. MIRSKY The Rockefeller University, New York, N.Y. 10021 Contributed by Alfred E. Mirsky, February 20, 1973

ABSTRACT The accessibility of DNA in chromatin to enzymatic variability. It will be seen that we have given much both exogenous DNA polymerase and RNA polymerase is slight when compared to isolated DNA. DNA in extracted attention to the technique of scintillation counting, for a chromatin is somewhat more accessible to these mistake in this procedure can lead to erroneous conclusions. than is DNA in the chromatin of isolated nuclei; and the DNA template of chromatin is more accessible to DNA MATERIALS AND METHODS polymerase than to RNA polymerase. In these experiments Materials we have given much attention to the technique of scin- tillation counting, since artifacts arising in this pro- The preparation of nuclei isolated from calf thymus (6) and cedure can lead to erroneous conclusions. liver (7) in 0.25 M sucrose-3 mM CaC12 has been described. DNP (extracted chromatin) was prepared from thymus Since DNA in chromatin is complexed with , ac- nuclei by the method of Paul and Gilmour (5, 8). Highly cessibility to DNA by enzymes and other substances is re- polymerized native calf-thymus DNA was obtained from stricted. The role of histones in restricting access to DNA in Worthington Biochemicals Corp. triphosphates chromatin was first observed by staining with basic dyes. were products of Schwarz/Mann, who also supplied [3H]TTP Staining of the led Flemming (1) in 1882 to coin (specific activity, 18 Ci/mmol) and [3H]UTP (specific activ- the term chromatin to designate the material that takes basic ity, 15 Ci/mmol). Highly purified DNA stains, and Flemming knew that chromatin contains DNA. polymerase was prepared and assayed according to Richard- The combination of DNA with crystal violet was shown by son et al (9). It was supplied by General Biochemicals at a Feulgen (2) in 1913 to be stoichiometric: one of dye specific activity of 5000 units/mg of . RNA poly- for each phosphoric acid group of DNA. Later (1951), in a merase from E. coli B was prepared by the method of Cham- quantitative study of the reaction of isolated chromatin with berlin and Berg (10). It was purchased from General Bio- crystal violet, Mirsky and Ris (3) found that only 8% of the chemicals at 2000 units/mg of protein. Hyamine (T. M. phosphoric acid groups of DNA combined with dye but that Rohm and Haas) was obtained from Nuclear Associates, Inc. removal of histone from the chromatin opened access to DNA so that 30% of the phosphoric acid groups of DNA Experiments with DNA polymerase combined with dye. Blocking of DNA by histones was later (1962) shown by Huang and Bonner (4) in experiments with Incubation. Enzymatic activity was assayed with the follow- RNA polymerase. Recently we have found that only a small ing reaction mixture adapted from Englund (11); 50 mM part of the DNA in chromatin is freely accessible to DNase Tris - HCl buffer (pH 7.4), 5 mM MgCl2, 1 mM 2-mercapto- (5). ethanol, dATP, dCTP, dGTP, and TTP each at 0.04 mM, We report in this paper determinations of the accessibility 10 MACi/ml of [3H]TTP, usually 20 Ag of DNA per ml, and of DNA in chromatin to exogenous DNA polymerase and 4 units of DNA polymerase per ml. Incubations were done at DNA-dependent RNA polymerase. In our experiments we 370 for 30 min with occasional stirring in a minimum volume have studied chromatin in isolated nuclei and also in ex- of 0.75 ml. The reaction was terminated by chilling to 00 and tracted chromatin (deoxyribonucleoprotein, DNP). Most of precipitating the DNA and proteins in 5% trichloracetic the previous work in this field has been on DNP, in which it acid-1% sodium . Incorporation of radio- is possible to replace certain components of chromatin, but activity into an acid-insoluble product was then measured by the purpose of all these investigations is, of course, to under- liquid scintillation counting of triplicate 0.2-ml aliquots from stand chromatin as it is in the nucleus. In the experiments each incubation. Counts per min from control incubations we are now reporting we have compared the activities of both containing no DNA template were subtracted from each ex- exogenous DNA and RNA on free DNA and on perimental value. In each case these counts were about 0.5% DNA in both thymus nuclei and extracted chromatin. The of those found with isolated DNA as was found if DNA comparison shows clearly that only a very small fraction of polymerase was omitted from the incubations. the DNA in a nucleus is accessible to either DNA or RNA Scintillation Counting. Initially, we used a generally ac- polymerase; DNA in DNP is slightly more accessible. We cepted procedure that has been described by numerous have used exogenous polymerase in these experiments so that investigators (11). These workers did not vary the concen- availability of the DNA template should not be obscured by tration of DNA template in the reaction mixture. Rather, they used a fixed quantity of template to evaluate procedures Abbreviation: DNP, deoxyribonucleoprotein. concerned with the fractionation and purification of DNA 1326 Downloaded by guest on September 29, 2021 Proc. Nat. Acad. Sci. USA 70 (1973) Accessibility of DNA in Chromatin 1327

polymerase or to study the effects of variables other than template concentration on enzymatic activity. The assay 14 procedure used most frequently in these studies and initially adopted by us is to bring the incubation mixture to a final 12 concentration of 5% trichloroacetic acid-1% sodium pyro- phosphate to precipitate DNA and protein and then to pass 10 the suspension through a glass-fiber filter. The filter is washed 0) to remove acid-soluble radioactivity, dried, and counted to x determine acid-insoluble radioactivity. We used Whatman 6- GF/C glass-fiber filters saturated with 5%0 trichloracetic acid-1% sodium pyrophosphate. After filtration, they were 4- washed with 15-ml aliquots of the trichloroacetic acid- sodium pyrosphosphate solution, 95% , and ether. The filters were air dried, placed in 15 ml of a toluene-based scintillation fluid, and counted. In this procedure more than 0 98% of the radioactivity remained associated with the filter. 20 40 60 80 100 120 140 160 180 200 When we used this assay system to determine the effect of DNA (4esghI) varying the concentration of DNA template in the incubation mixture, we obtained the result illustrated by curve B in FIG. 1. Template activity with DNA polymerase. Isolated Fig. 1. Enzymatic activity reached a maximum at about 20 DNA (A), DNA in DNP (C) extracted from thymus nuclei, and of DNA and decreased at DNA in thymus sucrose nuclei (D) were used in increasing con- ag/ml template fairly sharply centrations as templates for DNA polymerase in a standard reac- higher concentrations of DNA. This result is essentially the tion mixture. After 30 min at 370, the incubations were terminated. same as that obtained by Schwimmer and Bonner (12) using DNA and proteins were precipitated in 5% trichloroacetic acid- a similar assay procedure. Is this maximum characteristic of 1% sodium pyrophosphate, collected on glass-fiber filters, dis- the DNA polymerase reaction or is it an artifact due perhaps solved in hyamine, and counted. In curve (B) isolated DNA to self-absorption by DNA in the scintillation counting was used as the template, but it was not dissolved in hyamine technique? To answer this question, we added increasing before scintillation counting. At a DNA concentration of 400 amounts of nonradioactive DNA to a constant known amount pg/ml (not illustrated), the template activity of isolated DNA of tritiated DNA synthesized with 20 gg/ml of template decreased by about 20% (to 11,100 cpm) while that of nuclei DNA. The counts recovered were inversely proportional to and DNP each had increased by a small amount. the added DNA and mimicked curve B in Fig. 1. Therefore, the apparent decrease in enzymatic activity seen at DNA concentrations greater than 20 yg/ml is an artifact of the with the hyamine counting technique. Bollum has described scintillation counting procedure. Most likely the explanation (16) a procedure in which aliquots of incubation mixtures are is related to self-absorption occurring as increasing amounts of spotted directly onto dry filters, which are then dropped into DNA are collected on the glass filters. a single beaker of trichloroacetic acid. Thus the nucleic acid In order to eliminate scintillation counting artifacts that is precipitated within the filter rather than being precipitated can occur using the "standard" procedure that we originally in a test tube and then collected on the filter. The filters are adopted, we used hyamine to dissolve the nucleic acid col- washed, dried, and counted as in other procedures. When we lected on the glass-fiber filters. After the filters were washed used this technique to repeat experiments varying the con- and dried as previously described, each filter was placed in a centrations of template DNA, we found we could reproduce scintillation vial containing 0.5 mnl of hyamine and gently and corroborate the results obtained with the hyamine pro- agitated in a 700 water bath for 30 min or kept overnight at cedure (curve A, Fig. 1). Apparently counting efficiency is 380. 15 ml of scintillation fluid was then added, and the constant over a wide range of DNA concentrations when the filters were counted as before. In contrast to the previous DNA is precipitated within the glass-fiber filters. Unfortu- finding, essentially all of the radioactivity was found in the nately, when nuclei are used as the source of template DNA scintillation fluid. When DNA in concentrations up to 400 they tend to wash off or out of the filters. Thus Bollum's ug/ml was precipitated in 5% trichloroacetic acid collected technique was not suitable for most of our experiments. on filters, washed, dried, and dissolved in hyamine, it did not If, instead of isolated DNA, nuclei or DNP serve as the quench the counts in an added known amount of tritiated template for DNA polymerase, hyamine is still required to toluene. Thus all samples assayed in this manner are counted dissolve the template before scintillation counting. Omission at the same efficiency, about 20%0, irrespective of the amount of hyamine leads to results that are lower than those obtained of DNA used as template. when the nuclei are dissolved. However, even at very high When we repeated DNA polymerase experiments, varying concentrations, nuclei do not act as quenching agents when the concentration of DNA template, and used hyamine to dissolved in hyamine and mixed with a known amount of dissolve the DNA, the results were entirely different from tritiated toluene. Thus quenching by either protein or DNA those obtained when the radioactivity remained immobilized is not a factor in this counting procedure. on the glass filters. A plateau of enzymatic acitivity was Template Activity of DNA in Chromatin. Curves C and D in reached at about 100 jtg/ml (Fig. 1, curve A). This is compa- Fig. 1 illustrate the effectiveness of intact nuclei and DNP rable to results reported by some other investigators (13-15). when used as the source of template DNA for DNA poly- Another assay procedure was used in experiments with merase. Neither template demonstrates a maximum or even a isolated DNA and found to corroborate the results obtained well-defined plateau value of enzymatic activity as its con- Downloaded by guest on September 29, 2021 1328 : Silverman and Mirsky Proc. Nat. Acad. Sci. USA 70 (1978) centration is increased. Both are rising slowly even at 200 berlin and Berg (10): 40 mM Tris HCl buffer (pH 7.9), sg/ml of DNA, and both are only slightly accessible to DNA 0.15 M KCl, 1 mM MnCl2, 4 mM MgCl2, 10mM 2-mercap- polymerase when compared to isolated DNA. At 100 /g/ml toethanol, ATP, CTP, GTP, and UTP each at 0.12 mM, 10 of DNA, the beginning of the plateau found with isolated jACi/ml of [3H]UTP, usually 20 ug of DNA per ml, and 4 DNA, the template activity of DNP is about 7%, while that units of RNA polymerase per ml. Incubations were done at of nuclei is 4%, of the value found with DNA. The corre- 370 for 20 min with occasional stirring in a minimum volume sponding values at 200 ,ug/ml of DNA are 14% for DNP and of 0.75 ml. The reaction was terminated and assayed as 9% for nuclei. Additional experiments showed that nuclei described for DNA polymerase. isolated in 0.01 M citric acid exhibited essentially the same template activity as those prepared in isotonic sucrose. Scintillation Counting. The problems of assaying the in- An interesting and significant addition to these data was corporation of a radioactive nucleotide into an acid-insoluble made when liver nuclei replaced thymus nuclei as the source product are essentially the same whether RNA or DNA of DNA in the assay system. The template activities of these polymerase is used. Thus the procedure most commonly nuclei were indistinguishable from each other in spite of the described by other investigators (11) for assaying RNA relatively high content of nonhistone protein in liver chro- polymerase activity is identical to that which has been most matin (3, 17). often used for the DNA polymerase system: the nucleic acid It should also be noted that when DNP was used as the is precipitated in 5% trichloroacetic acid, collected on a source of template DNA, it was entirely insoluble during the filter, washed to remove acid-soluble radioactivity, dried, and incubation with DNA polymerase. At concentrations above counted. As was the case with DNA polymerase, most other 80 ;Ag of DNA per ml some of the DNP was obviously in- investigators did not vary the concentration of template in soluble in the incubation medium. To find out how much DNA, the reaction mixture. Therefore we repeated the experiments if any, was in solution, we centrifuged incubation mixtures first done using DNA polymerase with RNA polymerase and containing different samples of DNP, hydrolyzed any DNA obtained the same general results (Fig. 2) as were presented 'in. the supernatent with 0.5 M perchloric acid, and measured in Fig. 1. When the nucleic acid was collected on glass-fiber the absorbance at 265 nm. As determined by this procedure, filters, a maximum was obtained at about 25 mg/ml of DNA less than 1% of the DNA in DNP was in solution during the (curve B, Fig. 2) and template activity decreased at higher incubation with the . Almost certainly this is also concentrations of DNA. However if the nucleic acid was dis- true in experiments conducted by others, since only a small solved from the filter with hyamine, the plateau of template amount of divalent cation (18), here 5 mM, is necessary to activity shown in curve A of Fig. 2 was observed. Thus the precipitate DNP. Furthermore if DNP is precipitated in same experiments and arguments are relevant for both DNA 0.14 M NaCl it appears more "clumped" than when pre- and RNA polymerase assays: the procedure where nucleic cipitated by the divalent cation in the incubation mixture, acid remains immobilized on the filters is subject to an arti- yet its template activity is only slightly reduced. Of course fact of scintillation counting as the concentration of DNA is the chromatin within isolated nuclei is insoluble and remains increased. However, the use of hyamine to dissolve the nu- so during our experiments. If the nuclei are broken mechan- cleic acid eliminates this artifact. There have been several ically, the template activity of the preparation is about 80% investigators (20-22) who have varied the concentration of of the value found with isolated nuclei. This is strong evidence DNA template in the RNA polymerase reaction, dissolved that the nuclear membrane is not limiting the availability of the nucleic acids and proteins, and then proceeded with the DNA polymerase to the chromatin of isolated nuclei. Thus scintillation counting. Their results are essentially the same the observation that the DNA in chromatin is only slightly as reported in Fig. 2, curve A. Furthermore Bollum's tech- accessible to DNA polymerase, when compared to isolated nique (16), to which we have already referred, also corrobo- DNA, is not due to the enzyme failing to penetrate into the rates the results of Fig. 2, curve A. Unfortunately, this tech- nucleus. nique cannot be used if nuclei replace isolated DNA in the And as was the case with DNA when The results that have just been presented were obtained incubation. polymerase, with the standard assay system described under Incubation. nuclei or DNP are used as templates for RNA polymerase, Varying the concentration of nucleotide triphosphates, the hyamine must be used to dissolve the template before scin- incubation time, and the concentration of enzyme did not tillation counting. alter the main conclusion drawn from Fig. 1: the DNA of Template Activity of DNA in Chromatin. Curves C and D in either nuclei or DNP, as compared to isolated DNA, is only Fig. 2 illustrate the effectiveness of nuclei and DNP when slightly accessible to DNA polymerase. Incubation times of used as the source of template DNA for RNA polymerase. 15-120 min did not change the template activity of chromatin Both templates are only very slightly accessible to the en- relative to isolated DNA. This was true when thymus nuclei zyme. Increases in template activity are not observed when or DNP or liver nuclei were used as the source of template the concentration of DNA is increased. DNA in nuclei ex- DNA. Concentrations of DNA polymerase up to 5-times hibits about 2% of the template activity of isolated DNA, greater than that used in Fig. 1 also did not affect the relative while -the figure for DNP is about 4%. Liver nuclei (not il- template activity of DNA in chromatin. Therefore, the ac- lustrated) assay at about 1% relative to isolated DNA. Thus cessibility to DNA polymerase observed in these experiments thymus and liver nuclei have essentially the same accessibility is a reflection of the structure of the chromatin template, not toward RNA polymerase, as was the case in experiments with of parameters of the assay system used to measure it. DNA polymerase. Mechanically broken thymus nuclei have the same template activity as unbroken nuclei. Thus, as was Experiments with RNA polymerase true with DNA polymerase, the limited accessibility of DNA Incubation. Enzymatic activity was assayed with the follow- in these nuclei is not due to the enzyme failing to penetrate ing incubation mixture, which is modeled after that of Cham- into the nucleus. It should also be noted that when DNA Downloaded by guest on September 29, 2021 Proc. Nat. Acad. Sci. USA 7o (1973) Accessibility of DNA in Chromatin 1329

in DNP is used as a template, it is completely insoluble during the incubation with RNA polymerase. '4 A In order to eliminate the possibility that RNase activity i2 was degrading the RNA synthesized in these experiments, the virtual absence of such activity was demonstrated in the 10 following manner: Tritiated RNA was synthesized using

nuclei depleted of lysine-rich histone. (This increased the Q8

template activity of the nuclei, as will be reported in a future x paper.) The nuclei were centrifuged and the supernatent was E~~~~~~~~~ assayed for [3H]RNA before and after an incubation with added intact thymus or liver sucrose nuclei (40 jig of DNA 4- per ml). After 20 min at 370, 99% of the counts in the super- M. v natent representing ['HIRNA were still acid-insoluble, and 2 thus the RNA had not been hydrolyzed by RNase activity. As was found for DNA polymerase, when enzyme con- centration, incubation time, and the concentration of nu- 20 40 60 80 100 120 140 160 180 200 cleotide triphosphates were varied, they did not affect the DNA (/.g/ml) conclusions drawn from Fig. 2. The accessibility of DNA in FIG. 2. Template activity with RNA polymerase. Isolated nuclei to RNA polymerase, relative to isolated DNA, did not DNA (A), DNA in DNP (C) extracted from thymus nuclei, and change when more than three times the enzyme concentra- DNA in thymus nuclei (D) were used in increasing concentrations tion was used. It should be noted that even when higher as the templates for RNA polymerase. The reaction mixture was template activities were obtained, as when histones were ex- that described under Incubation. After 20 min at 370, the nucleic tracted from nuclei, the template activity relative to isolated acids and proteins were precipitated in 5% trichloroacetic acid- DNA was independent of the concentration of enzyme. Ob- 1% sodium pyrophosphate, collected on glass filters, dissolved in viously time was not a significant factor at the extremely low hyamine, and counted. The template used in curve (B) was template activity of intact nuclei. However when we used isolated DNA, but it was not dissolved in hyamine before scin- nuclei whose template activity was con- tillation counting. At a DNA concentration of 400 ,ug/ml (not histone-depleted illustrated), the template activity of isolated DNA, nuclei and siderably higher, the relative template activity of these DNP each remained constant. nuclei did not change with incubations ranging from 15-60 mM. observations on the marked restriction of template activity in transcription are, of course, in accord with the results of DISCUSSION DNA-RNA hybridization experiments that have shown that We have just described observations on the effects of exog- only a small part of the DNA in a nucleus is transcribed. enous polymerases on the template activity of DNA in chromatin. How are these observations to be related to the DNA replication template activities of DNA with respect to replication and Although but a small fraction of the DNA in a nucleus is transcription in nuclei when these activities depend upon the transcribed, all of the DNA of a nucleus is replicated in each endogenous polymerases? First, transcription: autoradio- . According to a report by Schwimmer and Bonner graphs of RNA synthesis in calf-thymus nuclei (23) and (12), this difference between replication and transcription Chironomus polytene chromosomes (24) show that only a can be accounted for by what appears to be the far greater very small part of the DNA is transcribed-in the former accessibility of DNA in chromatin to DNA polymerase than mainly in the relatively small amount of diffuse chromatin, to RNA polymerase. Schwimmer and Bonner (12) reported in the latter mainly in the puffs. It is not generally supposed that "although nucleohistone is totally ineffective in the that the limited transcription in such cases is due to lack of support of DNA-dependent RNA synthesis ..... nucleo- RNA polymerase, but this possibility has not been excluded. histone is nearly as effective in support of DNA synthesis as This is a possibility that should be considered, for RNA is deproteinized DNA" at a high concentration of the DNA polymerase can have an exceedingly short turnover time template. In a recent review (28) the situation was put this (25). In experiments on the effect of exogenous RNA poly- way: "Native chromatin, even if inactive in support of RNA merase on transcription in extracted chromatin, numerous synthesis by RNA polymerase, is active in support of DNA reports (26) have shown that the meager RNA synthesis is synthesis by DNA polymerase." In another review (29) it due to inaccessibility of the DNA template. The experiments was concluded that "Any model for the effect of histones on using exogenous polymerase described in the present report DNA replication must account for the findings of Schwimmer show that also for chromatin in nuclei only a very small part and Bonner that native chromatin (with histones intact) can of the DNA template is accessible. In these experiments the be used for DNA synthesis by DNA polymerase." results were the same for thymus and liver nuclei. Since the In a recent paper (30), we referred to the experiments of ratio of nonhistone (or acidic) protein to histone is much Schwimmer and Bonner because we had found that DNase greater in liver than in thymus nuclei (3, 17), it does not ap- in sufficient concentration can in the course of time digest pear that the over-all nonhistone protein is a significant practically all the DNA in an isolated nucleus (5). It seemed factor in template restriction. Application of exogenous to us noteworthy that although DNA in chromatin is only RNA polymerase to isolated polytene chromosomes shows, slightly accessible to RNA polymerase, our observations and according to a recent report (27), that in this material too, those of Schwimmer and Bonner (12) indicated that the inaccessibility of the template restricts transcription. The DNA of chromatin is far more accessible to DNase and DNA Downloaded by guest on September 29, 2021 1330 Biochemistry: Silverman and Mirsky Proc. Nat. Acad. Sci. USA 70 (1973)

polymerase. When we repeated their experiments, which pre- 3. Mirsky, A. E. & Ris, H. (1951) J. Gen. Physiol. 34, 475-492. sented no difficulty, we were at first satisfied with their con- 4. Huang, R. C. & Bonner, J. (1962) Proc. Nat. Acad. Sci. USA 48, 1216-1222. clusion. This can be understood by inspection of curves B and 5. Mirsky, A. E. (1971) Proc. Nat. Acad. Sci. USA 68, 2945- C, Fig. 1; as the template concentration increases, the tem- 2948. plate activity of chromatin approaches that of isolated DNA 6. Allfrey, V. G., Mirsky, A. E. & Osawa, S. (1957) J. Gen. as represented by curve B. The approach is due to the fact Physiol. 40, 451-490. that curve B falls rapidly with increasing template concentra- 7. Chauveau, J., Moul6, Y. & Rouiller, C. (1956) Exp. Cell Res. 11, 317-321. tion. This fall in curve B (compared with curve A) depends, as 8. Paul, J. & Gilmour, R. S. (1968) J. Mol. Bwl. 34,305-316. we have shown in this paper, on an artifact of scintillation 9. Richardson, C. C., Schildkraut, C. L., Aposhian, H. V. & counting. It can now be seen that because of this artifact Kornberg, A. (1964) J. Biol. Chem. 239, 222-232. Schwimmer and Bonner reached an erroneous conclusion. 10. Chamberlin, M. & Berg, P. (1962) Proc. Nat. Acad. Sci. USA 48, 81-94. When the curves (C and D, Fig. 1) for DNA replication in 11. Englund, P. T. (1971) in Procedures in Nucleic Acid Re- chromatin are compared with the true curve (A) for replica- search, eds. Cantoni, G. L. & Davies, D. R. (Harper and tion of isolated DNA, it is clear that the DNA template in Row, New York), Vol. 2, pp. 864-874. chromatin is only slightly accessible to DNA polymerase. 12. Schwimmer, S. & Bonner, J. (1965) Biochim. Biophys. Acta 108, 67-72. The template is, however, significantly more accessible to 13. Howk, R. & Wang, T. Y. (1969) Arch. Biochem. Biophys. DNA polymerase than it is to RNA polymerase (Fig. 2). 133, 238-246. For both polymerases the accessibility of DNA in thymus 14. Kraemer, R. J. & Coffey, D. S. (1970) Biochim. Biophys. nuclei is about the same as it is in liver nuclei, despite the Acta 224, 553-567. relatively large amount of nonhistone protein in liver chro- 15. Haines, M. E., Wickremasinghe, R. G. & Johnston, I. R. (1972) Eur. J. Biochem. 31, 119-129. matin (3, 17). 16. Bollum, F. J. (1968) in Methods in Enzymology, eds. Gross- It should be noted that the accessibility of DNA in chro- man, L. & Moldave, K. (Academic Press, New York), Vol. matin to DNase is also the same in liver nuclei as it is in thymus 12B, pp. 169-173. nuclei (30). The marked difference between the polymerases 17. Dingman, C. W. & Sporn, M. B. (1964) J. Bil. Chem. 239, in on a 3483-3492. and DNase their action chromatin is that the latter at 18. Huiskamp, W. (1901) Z. Physiol. Chem. 32, 145-196. high concentration of enzyme and in the course of time finally 19. Burgess, R. R. & Travers, A. A. (1971) in Procedures in digests practically all of the DNA (5). This result is probably Nucleic Acid Research, eds. Cantoni, G. L. & Davies, D. R made possible by a change in structure of chromatin due to (Harper and Row, New York), Vol. 2, pp. 851-863. the decomposition of DNA. 20. Furth, J. J., Hurwitz, J. & Anders, M. (1962) J. Biol. Chem. 237, 2611-2619. If, as we have found, DNA in the chromatin of interphase 21. Chambon, P., Karon, H., Ramuz, M. & Mandel, P. (1968) nuclei is only slightly accessible to exogenous DNA poly- Biochim. Biophys. Acta 157, 520-531. merase, it seems likely that there is a change in chromatin at 22. Kamiyama, M. & Wang, T. Y. (1971) Biochim. Biophys. the time of replication. Two recent reports indicate that there Acta 228, 563-576. 23. Littau, V. C., Allfrey, V. G., Frenster, J. H. & Mirsky, A. E. is a change in chromatin at the time of replication: Pederson (1964) Proc. Nat. Acad. Sci. USA 52, 93-100. (31) found that such chromatin was more accessible to diges- 24. Pelling, C. (1964) Chromosoma 15, 71-122. tion by DNase; and Tan and Lerner (32) found that single- 25. Yu, F. L. & Feigelson, P. (1972) Proc. Nat. Acad. Sci. USA stranded DNA was not detected immunologically in the 69, 2833-2837. Biochim. nuclei of lymphocytes during phase but was 26. Spelsberg, T. C. & Hnilica, L. S. (1971) Biophys. multiplying G, Acta 228, 202-211. "abundantly present" in phase S. It may well be that this 27. Robert, M. (1971) Chronwsoma 36, 1-33. single-stranded DNA is accessible to a DNA polymerase 28. Elgin, S. C. R., Froehner, S. C., Smart, J. E. & Bonner, J. present in the phase S nucleus. (1971) in Advances in Cell and Molecular Biology, ed. Du Praw, E. J. (Academic Press, New York), Vol. 1, pp. 2-57. B.S. is a postdoctoral fellow of the U.S. Public Health Service 29. De Lange, R. J. & Smith, E. L. (1971) Annu. Rev. Biochem. (GM 35130). We thank Miss Ellen Hein for her assistance. 40, 279-314. This work was supported in part by a grant from the New York 30. Mirsky, A. E. & Silverman, B. (1972) Proc. Nat. Acad. Sci. Heart Association. USA 69, 2115-2119. 1. Flemming, W. (1882) in Zellsubstanz, Kern und Zelltheilung 31. Pederson, T. (1972) Proc. Nat. Acad. Sci. USA 69, 2224- (Vogel, Leipzig), pp. 92, 129, 374, 375. 2228. 2. Feulgen, R. (1913) Z. Physiol. Chem. 84, 309-328. 32. Tan, E. M. & Lerner, R. A. (1972) J. Mol. Biol. 68, 107-114. Downloaded by guest on September 29, 2021