Proc. Nat. Acad. Sci. USA Vol. 72, No. 8, pp. 2895-2899, August 1975

Biochemistry

Location of histones on simian virus 40 DNA (chromatin structure/restriction endonucleases)

BARRY POLISKY AND BRIAN MCCARTHY Department of Biochemistry and Biophysics, University of California, San Francisco, San Francisco, Calif. 94143

Communicated by David M. Prescott, May 1, 1975

The physical location of histone molecules ABSTRACT in a simian virus 40 DNA-histone complex isolated from purified virions was examined using site-specific restriction endonucleases. The complex contains four host histone species but lacks histone Fl. Histones prevent complete cleavage of SV40 DNA by two restriction enzymes, HindIII and EcoRI. From the pattern of DNA fragments resulting from cleavage of the histone-DNA complex by the HindIII endonuclease, which makes six breaks on purified SV40 DNA, we have concluded that histones are randomly arranged on SV40 DNA relative to restriction enzyme cleavage sites. The EcoRI endonuclease, which makes one break in SV40 DNA, was used to determine the degree of physical coverage of the SV40 DNA molecule by histones. We observed that 80% of the EcoRI sites in the complex are accessible to the enzyme while 20% are "closed." This degree of coverage is consistent with the mass ratio of DNA:histone in the complex as revealed by the buoyant density of the formaldehyde-fixed complex. We conclude that the histones in the complex are located randomly on the SV40 genome and cover approximately 20% of the DNA. These results suggest that the histone species F2b, F2al, F2a2, and F3 are bound without regard to nucleotide sequence of SV40 DNA.

Among the remaining unanswered questions concerning chromatin structure is the relationship between histone binding sites and nucleotide sequence in DNA. This question cannot be approached directly in isolated chromatin because of the difficulty in isolating DNA-histone complexes which contain defined nucleotide sequences. Although complexes between histones and various purified DNAs are readily formed in vitro (1-3), their relevance to chromatin structure is uncertain since no criterion for accurate reconstitution exists. A prospective model system for chromatin might involve a small, well-defined DNA molecule interacting with histones in vivo. Such a model is provided by the lytic cycle of the oncogenic simian virus 40 (SV40). Newly synthesized SV40 DNA binds host histones, the synthesis of which is induced by viral infection (4). The SV40 DNA-histone complex is subsequently sequestered in the maturing virion. Therefore, the viral DNA interacts with host histones in the nuclear milieu and is later precluded from nonspecific association with other nuclear components as a consequence of packaging. The SV40-histone complex can be isolated from purified virions by disruption of the virus at pH 10.5 (5). To determine the physical location of histone molecules on the SV40 genome we have used restriction endonucleases as probes. It was assumed that the presence of histone at the restriction enzyme cleavage site would physically prevent the enzyme from cutting the DNA. Coverage of a given cleavage site Abbreviation: SV40, simian virus 40.

2895

can be detected by the appearance of a new DNA fragment consisting of the pair of fragments flanking the covered site. Since the probes used are specific for a very limited segment of the total DNA, rather precise information can be obtained concerning the histone arrangement relative to the cleavage sites. Under ionic conditions in which histone exchange does not occur, histones prevent complete cleavage of SV40 DNA by restriction endonucleases. From the pattern of DNA fragments generated by enzyme treatment of the SV40-histone complex we conclude that histones are arranged without regard to nucleotide sequence. Our results also suggest that 20-30% of complex-associated SV40 DNA is physically covered by histones. MATERIALS AND METHODS Cells and Virus. African green monkey CV-1 cells were infected with SV40 (strain VA 45-54) at a multiplicity of 5 plaque forming units (PFU) per cell. Stock virus was grown from infections at 0.001 PFU per cell. Isotopic labels (Schwarz/Mann) were added at 24 hr post infection. SV40 virions were purified as described by Tan and Sokol (6), except that virus was banded in CsCl rather than KBr density gradients (7). Preparation of SV40 Virion Complex. Virions were disrupted in isotonic Tris-ethanolamine, pH 10.5, and the resulting DNA-protein complex was isolated by velocity sedimentation in a 5-20% neutral sucrose gradient in the SW 27 rotor at 16,000 rpm, 18 hr, 5°. Fractions containing viral DNA were pooled and dialyzed against 0.01 M Tris-HCl, pH 7.2. Treatment with Restriction Endonucleases. EcoRI endonuclease was prepared from E. coli by the method of Greene et al. (8). HindIII endonuclease was prepared by the method of Smith and Wilcox (9), and was a gift of H. Goodman. Samples of complex containing 2 ,ug DNA were treated with an amount of EcoRI sufficient to convert 25 ,ig of purified SV40 DNA to linear form in 30 min at 37'. Reaction mixtures contained 100 mM Tris-HCI, pH 7.5, 50 mM NaCI, and 5 mM MgCl2. Samples were incubated at 37 for 2 hr. The reaction was terminated by the addition of sodium dodecyl sulfate to 1% final concentration. This treatment also removed histones from DNA. HindIII digestion was done in 6.6 mM Tris-HCI, pH 7.5, 60 mM NaCl, 6.6 mM MgCl2 and 1 mM 2-mercaptoethanol at 370 for 4 hr (9, 10). Gel Electrophoresis. DNA samples were analyzed on 2% agarose slabs as described by Greene et al. (8). DNA was photographed after staining with ethidium bromide. Nega-

2896

Biochemistry: Polisky and McCarthy

Proc. Nat. Acad. Sci. USA 72 (1975) Fl Fl

2

I

F3 F2B F2A2 F2A1

tt

0

A

t

61

4r..o 0

4-

:E I

2 u U

Top

5

10

15

20

FRACTION

FIG. 1. Sedimentation of SV40 virion complex. SV40 virions were labeled in vivo with a 3H-amino acid hydrolysate (1 gACi/ml) and [14C]thymidine (0.1 IACi/ml) added 24 hr after infection of CV-1 cells. At 72 hr virus was purified from the infected cells as described in Materials and Methods. Virus was disrupted and sedimented in a neutral 5-20% sucrose gradient. Fractions of 1.5 ml were collected and assayed for trichloroacetic-acid-precipitable radioactivity. The arrow marks the position of 20S SV40 supercoiled DNA run in a separate tube.

tives were scanned with a Beckman ACTA CIII spectrophotometer.

Electrophoresis of total virion or complex-associated polypeptides on 15% sodium dodecyl sulfate-acrylamide gel slabs was carried out by the methods of O'Farrell and Gold (11). Gels were electrophoresed for 7 hr at 15 mA. Histones were prepared from HeLa cells by 0.25 N H2SO4 extraction of purified nuclei. CV-1 histones were isolated by 0.25 N HC1 extraction of whole cells as described by Lake et al. (12).

RESULTS

Characterization of the SV40-histone complex Exposure of purified SV40 virions to pH 10.5 results in the production of a nucleoprotein "core" complex. This complex can be separated from SV40 capsid protein by velocity sedimentation. Fig. 1 shows the sedimentation of [3H]thymidine, [14C]leucine-labeled SV40 complex after alkaline treatment. More than 90% of the viral DNA and about 10% of the viral protein sediment together at approximately 30 S. The protein constituents of the complex were analyzed by sodium dodecyl sulfate-acrylamide electrophoresis. Densitometric scans of complex-associated polypeptides and histone from HeLa and CV-1 cells are shown in Fig. 2. Approximately 90% of the protein consists of four polypeptides which co-electrophorese with histones. The remaining 10% of complex-associated protein co-electrophoreses mainly with the viral protein designated VP3 [nomenclature of Huang et al. (5}]. Whereas, the histone species F3, F2a2, F2b, and F2al are all present in the complex, their relative amounts differ from those observed in CV-1 or HeLa chromatin. Histones F2a2 and F2b are reduced relative to their counterparts in chromatin, and F1 is practically undetectable. The same relative stoichiometry is evident when total virus is analyzed on dodecyl sulfate gels, indicating that preparation of the complex does not result in preferential loss of any histone components. It is noteworthy that histone F1 is barely detectable in the virion despite the fact that the CV-1 host cell chromatin contains normal amounts of Fl. To determine the approximate stoichiometry between histone and DNA, isolated complex was treated with formaldehyde and the buoyant density of the fixed complex was measured in an equilibrium CsCl gradient. The complex banded at a density of 1.55 g/ml. Assuming a simple addi-

D FIG. 2. Densitometric scans of complex-associated polypeptides and histones. Samples were electrophoresed on 15% sodium dodecyl sulfate-acrylamide gel slabs, fixed, stained, and scanned as described in Materials and Methods. (A) Polypeptides from purified SV40 virion complex (5 ,ug); (13) acid-soluble polypeptides from purified HeLa nuclei; (C) acid-soluble proteins from total CV-1 cells; (D) photograph of gel scanned to produce tracing A, showing that both F2b and F2a2 are present in the complex, although they were not resolved in the scans. The scale reduction factor for the photograph differs slightly from that of scan A.

tive relationship between the buoyant densities of the DNA and histone components of the complex, the complex consists of about 30% protein and 70% DNA. Similar results have been reported recently for a polyoma virion complex (13).

Treatment of the complex with restriction endonucleases-theoretical considerations Two restriction endonucleases were employed for analysis of histone arrangement on SV40, EcoRI, and HindIII. EcoRI makes a single break in purified SV40 DNA (14, 15), while HindIII makes six breaks (10). It is useful to consider the consequences of various possible arrangements of histones on SV40 DNA with respect to the fragments produced by treatment of the virion complex with restriction endonucleases. If the histones are located at unique sequences with respect to the cleavage sites of EcoRI enzyme, then either all DNA molecules will be cut because the site is "open" or all DNA molecules will remain intact because the site is "closed." If the histones are randomly arranged, then some fraction of DNA molecules will be cut, this fraction reflecting the degree to which the DNA is "covered" with histones. An intermediate case, in which the histones have a degree of preferential affinity for certain sequences, would also generate a fractional yield of cut DNA molecules. In this case the fraction cleaved would not necessarily be related to the physical "coverage" of DNA by histone. Thus, using the EcoRI en-

Proc. Nat. Acad. Sci. USA 72 (1975)

Biochemistry: Polisky and McCarthy

3

FIG. 3. Densitometric scans of SV40 DNA and virion complex after EcoRI endonuclease treatment. SV40 DNA or SV40 virion complex was treated with EcoRI endonuclease as described in Materials and Methods. Digestion products were analyzed on 2% agarose slabs. DNA was stained with ethidium bromide and wells were cut and scanned. (1) SV40 virion complex, no EcoRI treatment; (2) SV40 virion complex following EcoRI treatment; (3) purified SV40 DNA, no EcoRI treatment; (4) purified SV40 DNA following EcoRI treatment. Arrow A indicates the position of SV40 DNA supercoils; arrow B that of the linear SV40 DNA. Electrophoresis was from right to left.

zyme it is possible to eliminate a unique arrangement, but it is not possible to distinguish between random and preferential arrangement. Use of the HindIII enzyme allows a distinction between these possibilities. If the histones are randomly arranged with respect to the six HindIII cleavage sites, then each of a

L)

t,

u

IJMLI. wwwjft-.

-1."

I

a

A B

c

|D FIG. 4. Treatment of SV40 DNA or virion complex DNA with HindIII endonuclease. Samples containing 2 jg of virion complex DNA were treated with increasing amounts of HindIlI endonuclease as described in Materials and Methods. Following incubation for 2 hr at 370, samples were analyzed by agarose gel electrophoresis. Excess enzyme was not found to be inhibitory on purified SV40 DNA. (a) Complex DNA, 1 1Al of enzyme; (b) complex DNA, 5 Ml of enzyme; (c) complex DNA, 10 ul of enzyme; (d) 2 uig of purified SV40 DNA, 1 ,l of enzyme. Major fragments produced by HindIII are labeled A-E. HindIRIF was observed but was too faint to reproduce.

2897

the six complete fragments would be generated with a yield which is proportional to its length in the genome. Such an expectation is equivalent to stating that the probability of cleavage at all sites is equal. If the histones are preferentially arranged with respect to certain sites, then the yield of certain fragments will be reduced relative to their length in the genome, i.e., the probability of cleavage at some sites is greater than at others. Treatment with EcoRP Samples of the SV40-histone complex and purified SV40 DNA were separately treated with the EcoRI endonuclease. After incubation at 370 for 1 hr the histones were removed by addition of 1% sodium dodecyl sulfate and the DNA was analyzed by agarose gel electrophoresis. Fig. 3 shows that under conditions where 98% of purified SV40 DNA is converted to a linear form, only 80% of the SV40 DNA in the complex is converted to linear fragments. Approximately 2% of SV40 DNA in the complex exists in linear form without EcoRI treatment (Fig. 3) and the data were corrected by this amount. This result indicates that 80% of the EcoRI sites are "open" and 20% "closed" in the SV40-histone complex from virions, and rules out a unique location of histones on the genome. If the histones were randomly arranged this result would imply that about 20% of the SV40 DNA is physically inaccessible to a probe similar in size to EcoRI. Treatment with HindIll The virion complex was treated with increasing amounts of the HindIll restriction enzyme and the digestion products were analyzed by agarose gel electrophoresis. The cleavage pattern observed with the complex (Fig. 4a,b, and c) did not change when the amount of enzyme added was increased 10-fold over that used to completely digest an equivalent amount of SV40 DNA. This result demonstrates that the cleavage pattern of the complex is highly reproducible and is not a consequence of limiting enzyme conditions. Further evidence concerning the latter point is presented in following paragraphs. Fig. 4 shows that treatment of the complex with HindIII enzyme results in the production of the five complete fragments (labeled A-E) observed in a digest of purified SV40 DNA (Fig. 4d). * From densitometric scans of Fig. 4, it was determined that while 95% of purified SV40 DNA was converted to the five fragments only 40% of the DNA of the complex appeared in these regions of the gel. The remaining 60% is found mainly in large DNA fragments. All of these large fragments are observed in a digest of purified SV40 DNA with HindIII under conditions of limiting enzyme (unpublished experiments). Therefore, most DNA molecules present in the complex are cleaved partially by HindIll. Because the partial fragments are numerous and poorly resolved, determining their position on the SV40 DNA map is difficult. However, information on the arrangement of histones is available from quantitative analysis of the relative yields of the five limit fragments (Table 1). The critical point in Table 1 is that the ratio of the observed mass of each limit fragment relative to the mass of total DNA cleaved to limit fragments corresponds closely to the fractional mass of *

Hind III actually produces six fragments from SV40 DNA (10). However, the smallest fragment, comprising 4% of the genome (HindIIIF), was diffuse and difficult to- quantitate accurately. Only the five larger fragments are considered here.

2898

Biochemistry: Polisky and McCarthy

Proc. Nat. Acad. Sci. USA 72 (1975)

each fragment in the genome. This result demonstrates that production of limit fragments from the complex is reduced compared to purified DNA as a result of random deposition of histones relative to restriction enzymes cleavage sites. Were histone molecules located preferentially relative to certain cleavage sites one would expect a selective reduction in the yield of those limit fragments generated as a result of inefficient cleavage at those particular sites. Characteristics of the restriction enzyme-complex

b

a

c

A

C

-

interaction

Since histones are known to exchange under certain ionic conditions (16) it is necessary to demonstrate that no exchange occurs during incubation of the complex with the restriction enzymes. As a probe for exchange, SV40 [32P]DNA was added to the complex prior to addition of the HindIII enzyme. After incubation at 370 the products were analyzed on agarose gels and the pattern of 32P cleavage products was determined by autoradiography. The autoradiogram shown in Fig. 5a indicates complete cleavage of the added 32P-labeled DNA, rather than partial cleavage predicted if histone exchange occurred. A variation of this protocol was used to rule out the possibility that the partial cleavage pattern observed with the SV40 complex is a result of some general inhibition of restriction enzymes by nucleic acid-protein complexes. 32plabeled SV40 DNA was added 2 hr after addition of complex to the HindIII enzyme and incubated for an additional 2 hr. The SV40 [32P]DNA was quantitatively converted to six limit fragments, showing that the enzyme remained active after contact with the complex (Fig. 5b). Digestion of SV40 [32P]DNA alone is shown in Fig. 5c. To determine whether the fragment pattern of tle complex with HindIII was a limit digest, aliquots of enzyme were added to a sample at 6 hr intervals for 24 hr. No difference was detected in the fragment pattern of complex treated for 24 hr compared to that treated for 2 hr. This result indicated that the cleavage pattern is complete. The stability of the cleavage pattern 4espite the addition of fresh enzyme implies the absence of intrastrand histone "sliding" which

E

'MO

F

FIG. 5. Autoradiography of SV40 [32PJDNA cleavage products. To samples containing 1 gg of SV40 virion complex DNA, 0.5 jug SV40 [32PJDNA (5 X 104 cpm) was added either before (a) or 2 hr after (b) the addition of HindIII endonuclease. A separate reaction mix contained 0.5 jtg SV40 [32P]DNA alone (c). Samples (a) and (c) were incubated 2 hr at 37° while sample (b) was incubated 4 hr. Following incubation, samples were analyzed by agarose gel electrophoresis. The gel was dried under vacuum and exposed for 16 hr. The six fragments generated by HindIII digestion are labeled A-F.

would eventually allow all cleavage sites to become accessible to the restriction enzyme.

DISCUSSION complex of SV40 DNA and four histones from CV-1 cells isolated from purified virions the histones seem to be arranged randomly, without regard to nucleotide sequence. Further, by determining the accessibility of the EcoRI site [consisting of six nucleotide pairs (17)] to the EcoRI restriction endonuclease, we conclude that approximately 20% of the SV40 genome is covered with histone. This degree of In

a

Table 1. Densitometric analysis of HindIII cleavage of SV40 virion complex DNA

Exp. no.

(X) Total mass* (mg)

125.6 129.7 126.5 128.3 120.4 135.6 128.8 127.8 Average Observed average mass/average Y Measured fractional length § 1 2 3 4 5 6 7

(Y)

Observed mass (mg)

Total fragment masst (mg)

Y/X

A

BCt

D

E

52.4 53.1 51.3 50.5 46.8 56.1 54.5 52.1

0.417 0.409 0.405 0.393 0.388 0.414 0.423 0.407

17.5 19.0 17.6 17.0 17.2 20.3

21.8 21.9 21.1 22.2 19.4 22.7 22.6 21.7 0.416 0.425

6.5 6.2 7.3 6.3 5.9 6.5 7.0 6.5 0.125 0.105

6.0 5.2 4.8 4.9 4.2 4.5 4.6 4.8 0.092 0.085

19.9 18.3 0.351 0.340

Scans of seven separate HindHI digests of SV40 virion complex DNA were analyzed. Representative digests are shown in Fig. 4a, b, and c. Peaks representing partially digested fragments and limit fragments A, BC, D, and E were cut out and weighed. * Total mass is the sum of all limit and partial fragment masses. t Total fragment mass is the sum of the limit fragment masses. Complete densitometric resolution of limit fragments B and C was not achieved. Data are presented for the double peak corresponding to the sum of B and C. § Measured fractional length is the fractional length of the SV40 genome represented by each limit fragment. Data are taken from Danna etal. (10).

Biochemistry: Polisky and McCarthy coverage is in good agreement with the protein:DNA ratio of the complex as measured by its buoyant density in CsCl. To determine whether the HindIII data were compatible with 20% coverage, a computer program was written which fit the observed HindIII total fragment yield and individual fragment yields to various coverage models assuming a random distribution of histone on DNA. Approximately 30% coverage provided a best fit for the HindIII data. We consider this theoretical result to be in reasonable agreement with that obtained directly with EcoRI and suggest that intrinsic differences in affinity between HindIII and EcoRI for the SV40-histone complex do not influence the outcome of these experiments. Recent electron microscopic observations of swollen nuclei have revealed a subunit structure in chromatin fibers, consisting of spherical particles about 70 A in diameter connected by thin strands about 15 A wide (18). Essentially similar structures have been seen in preparations of the SV40histone complex isolated from virions or in the replicating complex of SV40 DNA isolated from infected cells (19). Thus, with regard to histone composition and morphology it appears that the SV40 virion complex shares some fundamental characteristics with chromatin. Since the implication of these results is that histones are bound without regard to DNA nucleotide sequence in vivo, it is necessary to consider possible preparational artifacts and alternative explanations of the data. With regard to the former, the reduced amounts of histones F1, F2a2, and F2b in the complex relative to chromatin seem to result from selective packaging rather than selective degradation during preparation of the complex, since the virus and complex contain identical complements of histones. Furthermore, control experiments with purified chromatin show that all histone species are stable to treatment at pH 10.5. It is conceivable that the protection against endonuclease digestion is due to the placement of non-histone protein components of the complex. While we cannot rule out this possibility absolutely, three observations make it unlikely; (1) about 90% of complex-associated protein is histone; (2) addition of histone to purified SV40 DNA in vitro results in a complex which gives a restriction enzyme cleavage pattern similar to the virion complex (unpublished experiments); (3) a portion of the 10% non-histone complex-associated protein is known to be located at a unique site on SV40 DNA (20). These results strongly suggest that the histone components are responsible for the observed cleavage patterns. Finally, although we have ruled out the possibility of inter- or intrastrand histone exchange during incubation with the restriction enzymes, we cannot exclude intrastrand histone rearrangement during preparation of the complex. Huang et al. (5) have presented evidence that interstrand histone exchange does not occur during complex preparation. Why are histones packaged with SV40 DNA? One obvious possibility is that they function in the condensation of DNA required for packaging. However, it is interesting that the histone probably responsible for the condensation of the

Proc. Nat. Acad. Sci. USA 72 (1975)

2899

complex of DNA and the other four histones, F1 (21, 22), is almost completely absent from the virion relative to the other histones. A second possibility is that histone binding to SV40 DNA occurs as a trivial consequence of the induction of cell DNA and histone synthesis by SV40 and plays no essential role in the infectious cycle. According to this view, the histone composition of the virion complex, namely large amounts of F3 and F2al, lesser amounts of F2b and F2b2 and trace amounts of F1, would reflect simply the tightness of binding of these various histones to DNA in vivo. A variety of ionic dissociating agents remove histones in the order F1, then F2b and F2a2, then F2al and F3 together, as the concentration of the dissociating agent is raised (23). The reduced amounts of F1, F2a2 and F2b might occur as a consequence of competition with capsid proteins for DNA binding during packaging. B.P. is a Fellow of the Leukemia Society of America. This research was supported by a grant from the National Science Foundation, GB35704. We thank Hugo Martinez for many helpful discussions. 1. Garrett, R. A. (1968) J. Mol. Biol. 38,249-250. 2. Olins, D. E. (1969) J. Mol. Biol. 43,439-460. 3. Rubin, R. & Mondrianakis, G. (1972) J. Mol. Biol. 67, 361374. 4. Winocour, E. & Robbins, E. (1970) Virology 40,307-315. 5. Huang, E. S., Estes, M. K. & Pagano, J. S. (1972) J. Virol. 9, 923-929. 6. Tan, K. B. & Sokol, F. (1974) J. Gen. Virol. 25,37-51. 7. Roblin, R., Harle, E. & Dulbecco, R. (1971) Virology 45, 555-566. 8. Greene, P., Betlach, M., Boyer, H. & Goodman, H. (1900) in Methods in Molecular Biology, ed. Wickner, R. B. (Marcel Dekker, New York), Vol. 7, pp. 87-111. 9. Smith, H. & Wilcox, K. (1970) J. Mol. Biol. 51,379-391. 10. Danna, K., Sack, G. & Nathans, D. (1973) J. Mol. Biol. 78,

363-376. 11. O'Farrell, P. & Gold, L. (1973) J. Biol. Chem. 248, 55025511. 12. Lake, R., Barban, S. & Salzman, N. (1973) Biochem. Biophys. Res. Commun. 54,640-647. 13. McMillen, J. & Consigli, R. (1974) J. Virol. 14, 1326-1336. 14. Mulder, C. & Delius, H. (1972) Proc. Nat. Acad. Sci. USA 69,

3215-3219. 15. Morrow, J. & Berg, P. (1972) Proc. Nat. Acad. Sci. USA 69,

3365-3369. 16. Clark, R. & Felsenfeld, G. (1971) Nature New Biol. 229, 101-106. 17. Hedgpeth, J., Goodman, H. & Boyer, H. (1972) Proc. Nat. Acad. Sci. USA 69,3448-3452. 18. Olins, A. & Olins, D. (1974) Science 183,330-32. 19. Griffith, J. (1975) Science 187, 1202-1203. 20. Griffith, J., Dieckmann, M. & Berg, P. (1975) J. Virol. 15, 167-172. 21. Bradbury, E., Inglis, R. & Mathews, H. (1974) Nature 247, 257-261. 22. Bradbury, E., Inglis, R., Mathews, H. & Langan, T. (1974) Nature 249,553-556. 23. Ohlenbusch, H., Olivera, B., Tuan, D. & Davidson, N. (1967)

J. Mol. Biol. 25,299-315.

Location of histones on simian virus 40 DNA.

Proc. Nat. Acad. Sci. USA Vol. 72, No. 8, pp. 2895-2899, August 1975 Biochemistry Location of histones on simian virus 40 DNA (chromatin structure/r...
1MB Sizes 0 Downloads 0 Views