Proc. Nati. Acad. Sci. USA Vol. 73, No. 10, pp. 3458-462, October 1976 Biochemistry
Identification of nonhistone chromatin proteins in chromatin subunits (nucleases/phenol-soluble proteins/isoelectrofocusing and sodium dodecyl sulfate/polyacrylamide gel electrophoresis)
C. C. LIEW AND P. K. CHAN Department of Clinical Biochemistry, Banting Institute, University of Toronto, Toronto, Ontario M5G 1LS, Canada
Communicated by Charles H. Best, July 19,1976
Rat liver chromatin was digested by microABSTRACT coccal nuclease. More than 80% of the enzyme-digested chromatin could be recovered after centrifugation. Treatment with sodium deoxycholate and Triton X-100 at concentrations of 0.5% in the final chromatin suspension gave a higher recovery. Chromatin subunits were fractionated on a 5-30% linear sucrose density gradient. Approximately 35% of the chromatin subunits could be recovered from the gradient. Chromatin subunits and their DNA fragments were identified by gel electrophoresis and ultracentrifugation. The presence of nonhistone chromatin proteins (NHCP) in chromatin subunits was demonstrated by the following criteria: (i) Quantitative analysis showed that the mass ratio of histone to NHCP, in the presence or absence of detergents, was 1:0.25 or 1:0.1, respectively. (ii) After the removal of acid-soluble protein from the subunits, it was found that most of the phenol-soluble NHCP were similar to total chromatin NHCP. However, four major fractions of these phenol-soluble NHCP were found to be enriched in the subunits as identified by two-dimensional polyacrylamide gel electrophoresis. (iKi) Experiments using an exchange of isotope-labeled and nonlabeled chromatin showed that NHCP were tightly bound to the chromatin subunits.
the nonhistone chromatin proteins (NHCP) were absent from the llS particles. This report deals with the identification of NHCP in rat liver chromatin subunits. We have found that these subunits are specifically enriched in at least four major fractions of NHCP. MATERIALS AND METHODS All chemicals and organic solvents used in these studies were of reagent grade. Male albino Wistar rats (200 g) were used. Food was removed the night before experiments. [3H]leucine (250 AOi) was given to each rat intraperitoneally (i.p.) for pulse-labeling, and the liver was then removed for the isolation of nuclei. Preparation of Nuclei and Chromatin. Five grams of liver were homogenized in 5 volumes of Medium A (0.25 M sucrose-10 mM Tris-HUl, pH 8.0-3 mM MgCl2-0.1 mM phenylmethylsulfonylfluoride) and centrifuged at 1000 X g for 10 min. The crude nuclear pellet was resuspended in Medium A once and filtered through four layers of nylon. The nuclear pellet was then resuspended in Medium B (0.1% Triton X-100 in Medium A) by gentle homogenization. The Triton-X-100treated nuclear pellet was finally resuspended in Medium C (2.2 M sucrose-10 mM Tris-HCl, pH 8.0-3 mM Mg9l2-0.1 mM phenylmethylsulfonylfluoride) and underlaid with 5 ml of Medium C. Nuclei were isolated by centrifugation at 113,000 X g for 1 hr on the Beckman SW 27 rotor, as described previously (24). Chromatin was prepared essentially by the method of Reeder (25). The chromatin was resuspended in 1 mM Tris-HCI (pH 8.0), 0.1 mM CaC12, and 0.1 mM phenylmethylsulfonylfluoride prior to micrococcal nuclease (Worthington) digestion, as described by Axel (15). Isolation of Chromatin Subunits. Digestion of either nuclei or chromatin into subunits was carried out with 200 units of micrococcal nucleases per g of tissue. After incubation for 10 min at 370, the reaction was arrested by the addition of NaEDTA, pH 7.0, to a final concentration of 2 mM followed by cooling in ice. Sodium deoxycholate and Triton X-100 were each added to a final concentration of 0.5% in suspension prior to the fractionation of chromatin subunits. Lower concentrations (e.g., 0.25%) of detergents failed to give a good recovery. The digested and detergent-treated chromatin subunits were centrifuged at 12,000 X g for 15 min, and the supernatant was layered onto-a 5-30% sucrose density gradient containing 0.2 mM EDTA. Following high-speed centrifugation, fractions containing chromatin subunits were then pooled and precipitated by 10 mM Mg++ as described by Honda et al. (17). Extraction of Chromatin Proteins. The extraction of chromatin proteins was essentially according to the method of Teng et al. (26) except for some modifications as described
In 1969, Murray (1) reported that native or partially dehistonized chromatin could be digested by deoxyribonuclease I and a region of chromatin protected by proteins could be isolated. Clark and Felsenfeld (2) used chemical probes to measure the accessibility of the DNA in chromatin and reported that about half of the DNA could be titrated with divalent cations, polylysine, or histones and could be subjected to staphylococcal nuclease or deoxyribonuclease I digestion. Others also used nucleases as probes to study the interaction of DNA and histones, as well as to isolate an active and inactive chromatin fraction (3, 4). Hewish and Burgoyne (5) demonstrated that a uniform size of chromatin DNA could be obtained by Ca++-Mg++-dependent endonuclease digestion. The electron microscopic observations made by Olins and Olins (6, 7) and Woodcock et al. (8, 9) indicate that chromatin structure resembles beads on a string, greatly substantiating the biochemical investigations (1-5). Recently Noll and others (10-17) demonstrated that the subunit proteins consisted mainly of histones. Kornberg and Thomas (18-20) have proposed a model in which chromatin is constituted as a repeating unit containing eight molecules of histones and about 200 base pairs of DNA. However, these observations on chromatin subunits were mainly concerned with the DNA and histone constituents. The presence of nonhistone proteins in chromatin subunits has not yet been extensively investigated. Lacy and Axel (21) indicated that subunits probably contain acidic protein in amounts analogous to that of intact chromatin. However, Augenlicht and Lipkin (22) and others (16, 17, 23) reported that Abbreviations: NHCP, nonhistone chromatin proteins; Mr, molecular weight.
3458
Biochemistry:
Proc. Natl. Acad. Sci. USA 73 (1976)
Liew and Chan
3459
E
0.4
0.2 E
10
30 40 20 FRACTION NUMBER
CY qq:
FIG. 1. Detergent-treated chromatin subunits. The digested chromatin was treated with a mixture of detergents and the supernatants of these chromatin digests were then layered onto a 5-30% linear sucrose density gradient. Centrifugation was then carried out at 82,500 X g for 16 hr. The chromatin absorbance profiles were recorded at 260 nm. The sedimentation rate of chromatin subunits was estimated to be 11 S by using catalase as a marker (G). Catalase activity was estimated by a modification of the method of Chantrenne (33). The high absorbance in fraction A is partly due to the interference by Triton X-100 at 260 nm. Fraction B was re-isolated by sucrose density gradient centrifugation (inset). A = top fraction, B = chromatin subunits.
previously (27). The phenol-solubilized proteins were finally dialyzed against 8 M urea-0.02 M Tris-HCl, pH 8.4-0.02 M glycine-0.1% 2-mercaptoethanol (28). Polyacrylamide Gel Electrophoresis. DNA of nucleasedigested chromatin and its monomers was extracted by the method of Hewish and Burgoyne (5). The DNA was then fractionated by 2.5% polyacrylamide gel electrophoresis as described by Loening (29). The gels were scanned at 260 nm.
Nonhistone chromatin proteins, which were isolated by the phenol extraction method, were subjected to isoelectrofocusing in the first dimension and subsequently fractionated by sodium dodecyl sulfate/polyacrylamide slab gel electrophoresis in the second dimension, as described previously (28, 30). Protein and DNA determinations were carried out by the methods of Lowry et al. (31) and Burton (32), respectively. Ovalbumin [molecular weight (Mr) 45,000] and bovine serum albumin (Mr 67,000) were used as markers. RESULTS Chromatin Subunits. As shown in Fig. 1, the nuclease-digested and detergent-treated chromatin was fractionated into subunits by sucrose density gradient centrifugation. More than 80% of the digested chromatin could be recovered in the detergent-treated chromatin suspension. Similar results could be obtained from nuclei that were subjected to nuclease digestion. Fraction A represents chromatin proteins and nucleotides released from chromatin by nucleases. Fraction B, which sediments at a rate corresponding approximately to 11 S. represents monomer particles of the chromatin (i.e., chromatin subunits). Fraction B was re-isolated by sucrose density gradient cen-
0
1
2
3
4
5
6
cm FIG. 2. Fractionation of DNA fragments from chromatin by gel electrophoresia DNA from nuclease-digested chromatin (A) and the isolated monomer particles (B) was isolated (5) and the DNA from these two sources was fractionated by gel electrophoresis (29). The arrow (1) indicates DNA fragments of monomer particles of chromatin. The broken line represents the blank gels.
trifugation and the result is shown in the inset of Fig. 1. The subunit fraction precipitated by 10 mM MgCl2 was designated as fraction Bm. The subunits of the enzyme-digested chromatin were further identified by their DNA pattern on gels. DNA extracted from either enzyme-digested chromatin or the isolated chromatin subunits was fractionated by gel electrophoresis. As shown in Fig. 2, both fractions contained a major DNA fragment; in particular, fraction B showed only one major peak. These findings indicated that the possible contamination of dimers or oligomers was minimal in fraction B. The isolated chromatin subunits were further identified by electron microscopy (not shown). Composition of Chromatin Subunits. Protein compositions of fractions, A, B, and Bm from the fractionated chromatin
3460
Proc. Nati. Acad. Sci. USA 73 (1976)
Biochemistry: Liew and Chan
Table 1. Protein composition in fractionated chromatin Fraction Nuclei
Chromatin A B Bm
mg/g of tissue 4.6 2.67 0.27 0.92 0.8
Nuclease-digested chromatin was fractionated by a linear sucrose density gradient centrifugation. Fractions were collected as indicated in Fig. 1. The results are expressed as the protein content of each fraction that could be recovered from the starting materials.
were determined and are shown in Table 1. About 60% of the nuclear proteins were recovered with DNA in the isolated chromatin fraction. About 10% of the chromatin proteins were released during nuclease digestion and recovered in the top layer of the sucrose gradient (fraction A). The chromatin subunits (fraction B) isolated from the gradient accounted for more than 35% of the digested chromatin. More than 85% of the chromatin subunits (fraction Bm) could be recovered from fraction B by precipitation with 10 mM Mg++. Following Mg++ precipitation of each fraction, histones were extracted in 0.25 M HCL NHCP were then extracted in phenol. Taking the acid-soluble fraction as equivalent to 1, the mass ratio of NHCP to histone was determined. As shown in Table 2, NHCP constituted a significant part of chromatin proteins in chromatin subunits. The mass ratio of NHCP to histone in fraction Bm was 0.25 I 0.02 in four separate experiments. In fraction A, the mass ratio of NHCP to histone was 4.85, which indicated that the proteins released by nuclease were mainly
A
Mr
t41
67,000-( 45,000-
* 1: *:
-~~~~~~~~~~~~~
'S Ill?-.
B
I ,z
:.:
-Im
:.ftz,,.
67,000L45,OOO.-
* ..
NHCP. Fractionation of NHCP by Two-Dimensional Gel Elec-
trophoresis. NHCP from the top fraction (A), chromatin subunits (B), and whole chromatin (C) were prepared by removal of histone in an attempt to prevent any nonspecific interaction between histone and NHCP. These proteins were fractionated by two-dimensional polyacrylamide gel etectrophoresis. As shown in Fig. 3, most NHCP associated with chromatin subunits (i.e., fraction B) were identical to proteins in whole chromatin, except for several fractions that were increased. A group of proteins that had molecular weights of over 100,000 in the pH range 5.6 to 8.0 was significantly enriched in fraction B (top rectangle) as compared to fraction A. Acidic proteins of fraction B that focused from pH 4.8 to 5.4, with molecular weights of 55,000, 70,000, and 74,000, were significantly reduced as compared to fraction A. Also, proteins in the chromatin subunits having molecular weights of 70,000 and 67,000 at pH 5.8 to 8.2 Table 2. Ratio of histone to nonhistone chromatin protein in chromatin subunits Fraction
Mass ratio
Chromatin
0.62 4.86 0.25 ± 0.02*
A Bm
Chromatin prepared by the method of Reeder (25) has a ratio of DNA to chromatin protein of 1:1.66. Chromatin protein ratios are expressed with histone taken as 1. * The results represent mean + standard deviation of four separate experiments. The mass ratio of histone to NHCP in the chromatin subunits that were obtained from the nuclease digestion without the treatment of detergents was 1:0.10 (Chan and Liew, manuscript in preparation).
C Xj.
67,00045,000
34
3
4
I 6
pH
I
7
I
I
8
9
FIG. 3. Fractionation of NHCP by two-dimensional gel electrophoresis. NHCP (250 ,g per gel), which were dissolved in 8 M urea buffer, were fractionated by two-dimensional gel electrophoresis (28, 30). A and B represent NHCP isolated from fractions A and B (subunits) of the sucrose density gradients and C represents nucleasedigested whole chromatin. Fractions of special interest are enclosed in broken line rectangles.
(long rectangles, center) were strikingly increased. A fraction of proteins in the pH range from 5.0 to 5.5 with molecular weight of 30,000 was identified in fraction A and chromatin (C). The proteins that were enriched in the chromatin subunits were apparently absent in fraction A. Exchange of Protein in Chromatin Subunits. Experiments were designed to examine whether NHCP were dynamically associated with the chromatin subunits.- Rats were pulse-labeled
Biochemistry:
with [3H]leucine for 2 hr. 3H-Labeled chromatin was isolated concurrently with chromatin from a control group that had not received any isotope. Nuclease-digested chromatin from both groups of animals was then fractionated by sucrose density gradient centrifugation. The top layer (fraction A) and the chromatin subunits (fraction B) were each collected from the [3H]leucine-labeled and unlabeled control chromatins. As shown in Fig. 4(I), 3H radioactivity was found in all chromatin fractions The top fraction (A) and chromatin subunits (B) were pooled and used for the isotopic exchange experiments with the nonlabeled fractions of the control group. The top fraction A and fraction B, which contained the radioactivity, were mixed with fraction B and fraction A, respectively, of the control group. Each mixture was then layered onto a 5-30% sucrose density gradient. As shown in Fig. 4(II), it was demonstrated that 3H-labeled NHCP remained in the same fraction and were not interchanged with the control. In another experiment isotope-labeled fraction A was mixed with nonlabeled nuclease-digested chromatin prior to sucrose density gradient centrifugation. As shown in Fig. 4(III), most of the radioactivity remained in the top fraction, and no radioactivity was detected in the chromatin subunits. DISCUSSION Biochemical findings on chromatin composition and functions (1-5, 10-17, 26, 27, 30, 34-39) as well as recent morphological observations involving nuclease digestion of chromatin (6-9) have added considerable insight to our concept of the structure of chromatin. The existence of NHCP in chromatin subunits has not hitherto been established. A few reports (16, 17) have indicated that NHCP are not found in the subunits. In some cases these findings can be explained on the basis that materials (e.g., erythrocytes) used for chromatin subunit preparation had little NHCP to begin with. Augenlicht and Lipkin (22) reported the absence of NHCP in the subunits of human colonic carcinoma cells, but the presence of NHCP cannot be excluded on the basis of their electrophoretic profiles. On the other hand, Lacy and Axel (21) indicated that acidic proteins were probably present in the subunits. It was therefore a matter of some importance to determine whether NHCP are associated with chromatin subunits. Our findings provide clear evidence that NHCP are present in the subunits. An association of NHCP with the chromatin subunits could have resulted from contamination of dimers or oligomers during the isolation procedures. Our results disprove this possibility by the following criteria: (a) the chromatin subunits isolated from the nuclease digest were further purified by sucrose density gradient centrifugation (the purity of these 1IS particles was confirmed by electron microscopy)*; (b) there was only one major DNA fragment from the chromatin subunit fraction after gel electrophoresis; and (c) isotope experiments showed that NHCP released by nuclease digestion of 3H-labeled chromatin were observed not to have interchanged with the NHCP of unlabeled chromatin subunits during re-centrifugation. It was noted that if detergents were omitted from the nuclease-digested suspension the recovery of subunits after centrifugation.was relatively low. The detergents appear to prevent aggregation and maintain the particles in suspension, thus giving a high yield of chromatin subunits. The nuclease-digested chromatin was then fractionated into monomers, dimers, trimers, and oligomers. This methodology is analogous to the treatment of ribosomes (40). However, Smart and Bonner (41) reported that sodium deoxycholate at concentrations greater *
P. K. Chan and C. C.
Proc. Natl. Acad. Sci. USA 73 (1976)
Liew and Chan
Liew, manuscript in preparation.
F! 202p
~~~ so V ~~~~~~~40
0.1
0°.t 0.61
3461
4°
120
*40
0.2 10
20 30 FRACTION NUMBER
40
FIG.' 4. Fractionation of chromatin particles by sucrose density gradients. [3H]Leucine was given to rats and a control group received no isotope. Liver chromatin was then isolated from each group as described in Materials and Methods. Nuclease-digested chromatin was fractionated by linear sucrose density gradient. Fractions A and B identical to Fig. 1 were collected. Mixing of 3H-labeled fraction A
with nonlabeled fraction B or vice versa was carried out and re-cen-
trifugation was performed as above. Solid lines are A260 la I represents chromatin labeled with [3H]leucine and fractionated by sucrose density gradient. II represents in one graph exchanges of 3H-labeled 0) with nonlabeled fraction B or A fraction A (*- -*) or B (0 in two separate experiments. Ill represents the mixture of 3H-labeled fraction A and nuclease-digested nonlabeled chromatin. Radioactivity was determined by dissolving each fraction in 5 ml of Aquasol (New England Nuclear) and counting in the liquid scintillation counter.
than 0.5% may extract some histone from the DNA complex. We limited the detergent concentration to 0.5% and have shown that small amounts of Histone Hi are present only in fraction A. Assuming that the inter-region of the subunit is accessible to nuclease digestion, one would expect the associated proteins to be released after nuclease treatment. We therefore attempted to answer the following questions in regard to the association of NHCP with DNA and histones: (i) are NHCP located only in the inter-region of the subunits? or (ii) are NHCP present in both the chromatin subunits and the inter-regions? We have shown that about 10% of the chromatin proteins were released into the top fraction of the sucrose density gradient after nuclease digestion. More than 85% of the chromatin proteins in this fraction (i.e., fraction A) were NHCP and only histone HI was released in significant amounts during chromatin digestion. The chromatin subunits that were re-isolated from the sucrose density gradients clearly contained both histones and NHCP. The mass ratio of histone to NHCP was 1: 0.25, which indicates that NHCP constitute a significant proportion of the chromatin subunits. However, chromatin subunits that were obtained from the nuclease digestion without the treatment of detergents have a mass ratio of histone to NHCP as 1:0.1.*
3462
Biochemistry: Liew and Chan
We have also determined the composition of chromatin subunits isolated from nuclei that were digested by nuclease. It was found that the amount of NHCP present in the subunits is similar to that in the subunits prepared from nuclease digestion of chromatin. These findings confirmed the report of Lacy and Axel (21) that acidic proteins were probably present in the chromatin subunits. To substantiate the chemical analysis, we have shown that the chromatin proteins extracted from chromatin subunits gave a sodium dodecyl sulfate/polyacrylamide gel electrophoresis pattern similar to that of whole chromatin proteins, except for the histone fraction, which was significantly enriched in the subunits.* These findings clearly establish that NHCP are associated with both the inter-regions and the subunits. We then undertook to examine whether specific NHCP are associated with the subunits or the inter-regions. As shown in Fig. 3, both quantitative and qualitative differences in NHCP could be demonstrated by two-dimensional gel electrophoresis. At least four major and a few minor groups of proteins were recognized as being specifically associated with the subunits. On the other hand, four groups of NHCP were found only in fraction A and were not present in the subunit particles, which suggests that this portion of NHCP was significantly enriched in the inter-regions and could be released by nuclease digestion. Nonhistone chromatin proteins that focused in the neutral range were greatly enriched in the chromatin subunits. Similar results were also obtained from the chromatin subunits even without the treatment of detergents. * In addition, we have found recently that both the chromatin subunits and the top fraction contain different fractions of highly phosphorylated NHCP.* We therefore conclude that there are certain specific proteins associated with the chromatin subunits and other proteins that interact with the inter-region. The support from the Medical Research Council and the Ontario Heart Foundation, Canada, is acknowledged. Comments on the use of detergents from Dr. N. Straus are greatly appreciated. We are thankful for the competent assistance of Mrs. K. L. Yao. 1. Murray, K. (1969) J. Mol. Biol. 39,125-144. 2. Clark, R. J. & Felsenfeld, G. (1971) Nature 229,101-106. 3. Marushige, K. & Bonner, J. (1971) Proc. Natl. Acad. Sd. USA 68, 2941-2944. 4. Mirsky, A. E. (1971) Proc. Nati. Acad. Sci. USA 68, 29452948. 5. Hewish, D. R. & Burgoyne, L. A. (1973) Biochem. Biophys. Res. Commun. 52,504-510. 6. Olins, A. L. & Olins, D. E. (1973) J. Cell Biol. 59, 252a. 7. Olins, A. L. & Olins, D. E. (1974) Sciene 183,330-332. 8. Woodcock, C. L. F. (1973) J. Cell Biol. 59, 368a.
Proc. Nati. Acad. Sci. USA 73 (1976) 9. Woodcock, C. L., Maguire, D. L. & Stanchfield, J. E. (1974) J. Cell Biol. 63, 377a. 10. Noll, M. (1974) Nature 251,249-251. 11. Clark, R. J. & Felsenfeld, G. (1974) Biochemistry 13, 36223628. 12. Sahasrabuddhe, C. G. & Van Holde, K. E. (1974) J. Biol. Chem. 249, 152-156. 13. Spadafora, C. & Geraci, G. (1975) FEBS Lett. 57,79-82. 14. Weintraub, H. (1975) Proc. Natl. Acad. Sci. USA 72, 12121216. 15. Axel, R. (1975) Biochemistry 14,2921-2925. 16. Oudet, P., Gross-Bellard, M. & Chambon, P. (1975) Cell 4, 281-300. 17. Honda, B. M., Baillie, D. L. & Candido, E. P. M. (1975) 1. Biol. Chem. 250, 4643-4647. 18. Kornberg, R. D. (1974) Science 184,868-871. 19. Kornberg, R. D. & Thomas, J. 0. (1974) Science 184, 865868. 20. Thomas, J. 0. & Kornberg, R. D. (1X975) Proc. Natl. Acad. Sci. USA 72,2626-2630. 21. Lacy, E. & Axel, R. (1975) Proc. Natl. Acad. Sci. USA 72, 22. Augenlicht, L. H. & Lipkin, M. (1976) Biochem. Biophys. Res. Commun. 70,540-544. 23. Gottesfeld, J. M., Murphy, R. F. & Bonner, J. (1975) Proc. Natl. Acad. Sci. USA 72, 4404-4408. 24. Liew, C. C., Liu, D. K. & Gornall, A. G. (1972) Endocrinology 90,488-495. 25. Reeder, R. H. (1973) J. Mol. Biol. 80,229-241. 26. Teng, C. S., Teng, C. T. & Allfrey, V. G. (1971) J. Biol. Chem. 246,3597-3609. 27. Suria, D. & Liew, C. C. (1974) Biochem. J. 137, 355-362. 28. Jackowski, G., Suria, D. & Liew, C. C. (1976) Can. J. Biochem. 54, 9-14. 29. Loening, U. E. (1967) Biochem. J. 102,251-257. 30. Suria, D. & Liew, C. C. (1974) Can. J. Blochem. 52, 11431153. 31. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 32. Burton, K. (1956) Biochem. J. 62,315-323. 33. Chantrenne, H. (1955) Biochim. Biophys. Acta 16, 410-417. 34. MacGillivray, A. J. & Rickwood, D. (1974) Eur. J. Biochem. 41, 181-190. 35. Stedman, E. & Stedman, E. (1950) Nature 166, 780-781. 36. Paul, J. & Gilmour, R. S. (1968) J. Mol. Biol. 34,305-316. 37. Stein, G. S., Spelsberg, T. C. & Kleinsmith, L. J. (1974) Science 183,817-824. 38. Kostraba, N. C. & Wang, T. Y. (1973) Exp. Cell Res. 80, 291296. 39. Allfrey, V. G. (1974) in Acidic Proteins of the Nucleus, eds. Cameron, L. L. & Jeter, J. R., Jr. (Academic Press, New York), pp. 1-29. 40. Liew, C. C. & Gornall, A. G. (1973)J. Blol. Chem. 248,977983. 41. Smart, J. E. & Bonner, J. (1971) J. Mol. Biol. 58,651-659.
Identification of nonhistone chromatin proteins in chromatin subunits (nucleases/phenol-soluble proteins/isoelectrofocusing and sodium dodecyl sulfate/polyacrylamide gel electrophoresis)
C. C. LIEW AND P. K. CHAN Department of Clinical Biochemistry, Banting Institute, University of Toronto, Toronto, Ontario M5G 1LS, Canada
Communicated by Charles H. Best, July 19,1976
Rat liver chromatin was digested by microABSTRACT coccal nuclease. More than 80% of the enzyme-digested chromatin could be recovered after centrifugation. Treatment with sodium deoxycholate and Triton X-100 at concentrations of 0.5% in the final chromatin suspension gave a higher recovery. Chromatin subunits were fractionated on a 5-30% linear sucrose density gradient. Approximately 35% of the chromatin subunits could be recovered from the gradient. Chromatin subunits and their DNA fragments were identified by gel electrophoresis and ultracentrifugation. The presence of nonhistone chromatin proteins (NHCP) in chromatin subunits was demonstrated by the following criteria: (i) Quantitative analysis showed that the mass ratio of histone to NHCP, in the presence or absence of detergents, was 1:0.25 or 1:0.1, respectively. (ii) After the removal of acid-soluble protein from the subunits, it was found that most of the phenol-soluble NHCP were similar to total chromatin NHCP. However, four major fractions of these phenol-soluble NHCP were found to be enriched in the subunits as identified by two-dimensional polyacrylamide gel electrophoresis. (iKi) Experiments using an exchange of isotope-labeled and nonlabeled chromatin showed that NHCP were tightly bound to the chromatin subunits.
the nonhistone chromatin proteins (NHCP) were absent from the llS particles. This report deals with the identification of NHCP in rat liver chromatin subunits. We have found that these subunits are specifically enriched in at least four major fractions of NHCP. MATERIALS AND METHODS All chemicals and organic solvents used in these studies were of reagent grade. Male albino Wistar rats (200 g) were used. Food was removed the night before experiments. [3H]leucine (250 AOi) was given to each rat intraperitoneally (i.p.) for pulse-labeling, and the liver was then removed for the isolation of nuclei. Preparation of Nuclei and Chromatin. Five grams of liver were homogenized in 5 volumes of Medium A (0.25 M sucrose-10 mM Tris-HUl, pH 8.0-3 mM MgCl2-0.1 mM phenylmethylsulfonylfluoride) and centrifuged at 1000 X g for 10 min. The crude nuclear pellet was resuspended in Medium A once and filtered through four layers of nylon. The nuclear pellet was then resuspended in Medium B (0.1% Triton X-100 in Medium A) by gentle homogenization. The Triton-X-100treated nuclear pellet was finally resuspended in Medium C (2.2 M sucrose-10 mM Tris-HCl, pH 8.0-3 mM Mg9l2-0.1 mM phenylmethylsulfonylfluoride) and underlaid with 5 ml of Medium C. Nuclei were isolated by centrifugation at 113,000 X g for 1 hr on the Beckman SW 27 rotor, as described previously (24). Chromatin was prepared essentially by the method of Reeder (25). The chromatin was resuspended in 1 mM Tris-HCI (pH 8.0), 0.1 mM CaC12, and 0.1 mM phenylmethylsulfonylfluoride prior to micrococcal nuclease (Worthington) digestion, as described by Axel (15). Isolation of Chromatin Subunits. Digestion of either nuclei or chromatin into subunits was carried out with 200 units of micrococcal nucleases per g of tissue. After incubation for 10 min at 370, the reaction was arrested by the addition of NaEDTA, pH 7.0, to a final concentration of 2 mM followed by cooling in ice. Sodium deoxycholate and Triton X-100 were each added to a final concentration of 0.5% in suspension prior to the fractionation of chromatin subunits. Lower concentrations (e.g., 0.25%) of detergents failed to give a good recovery. The digested and detergent-treated chromatin subunits were centrifuged at 12,000 X g for 15 min, and the supernatant was layered onto-a 5-30% sucrose density gradient containing 0.2 mM EDTA. Following high-speed centrifugation, fractions containing chromatin subunits were then pooled and precipitated by 10 mM Mg++ as described by Honda et al. (17). Extraction of Chromatin Proteins. The extraction of chromatin proteins was essentially according to the method of Teng et al. (26) except for some modifications as described
In 1969, Murray (1) reported that native or partially dehistonized chromatin could be digested by deoxyribonuclease I and a region of chromatin protected by proteins could be isolated. Clark and Felsenfeld (2) used chemical probes to measure the accessibility of the DNA in chromatin and reported that about half of the DNA could be titrated with divalent cations, polylysine, or histones and could be subjected to staphylococcal nuclease or deoxyribonuclease I digestion. Others also used nucleases as probes to study the interaction of DNA and histones, as well as to isolate an active and inactive chromatin fraction (3, 4). Hewish and Burgoyne (5) demonstrated that a uniform size of chromatin DNA could be obtained by Ca++-Mg++-dependent endonuclease digestion. The electron microscopic observations made by Olins and Olins (6, 7) and Woodcock et al. (8, 9) indicate that chromatin structure resembles beads on a string, greatly substantiating the biochemical investigations (1-5). Recently Noll and others (10-17) demonstrated that the subunit proteins consisted mainly of histones. Kornberg and Thomas (18-20) have proposed a model in which chromatin is constituted as a repeating unit containing eight molecules of histones and about 200 base pairs of DNA. However, these observations on chromatin subunits were mainly concerned with the DNA and histone constituents. The presence of nonhistone proteins in chromatin subunits has not yet been extensively investigated. Lacy and Axel (21) indicated that subunits probably contain acidic protein in amounts analogous to that of intact chromatin. However, Augenlicht and Lipkin (22) and others (16, 17, 23) reported that Abbreviations: NHCP, nonhistone chromatin proteins; Mr, molecular weight.
3458
Biochemistry:
Proc. Natl. Acad. Sci. USA 73 (1976)
Liew and Chan
3459
E
0.4
0.2 E
10
30 40 20 FRACTION NUMBER
CY qq:
FIG. 1. Detergent-treated chromatin subunits. The digested chromatin was treated with a mixture of detergents and the supernatants of these chromatin digests were then layered onto a 5-30% linear sucrose density gradient. Centrifugation was then carried out at 82,500 X g for 16 hr. The chromatin absorbance profiles were recorded at 260 nm. The sedimentation rate of chromatin subunits was estimated to be 11 S by using catalase as a marker (G). Catalase activity was estimated by a modification of the method of Chantrenne (33). The high absorbance in fraction A is partly due to the interference by Triton X-100 at 260 nm. Fraction B was re-isolated by sucrose density gradient centrifugation (inset). A = top fraction, B = chromatin subunits.
previously (27). The phenol-solubilized proteins were finally dialyzed against 8 M urea-0.02 M Tris-HCl, pH 8.4-0.02 M glycine-0.1% 2-mercaptoethanol (28). Polyacrylamide Gel Electrophoresis. DNA of nucleasedigested chromatin and its monomers was extracted by the method of Hewish and Burgoyne (5). The DNA was then fractionated by 2.5% polyacrylamide gel electrophoresis as described by Loening (29). The gels were scanned at 260 nm.
Nonhistone chromatin proteins, which were isolated by the phenol extraction method, were subjected to isoelectrofocusing in the first dimension and subsequently fractionated by sodium dodecyl sulfate/polyacrylamide slab gel electrophoresis in the second dimension, as described previously (28, 30). Protein and DNA determinations were carried out by the methods of Lowry et al. (31) and Burton (32), respectively. Ovalbumin [molecular weight (Mr) 45,000] and bovine serum albumin (Mr 67,000) were used as markers. RESULTS Chromatin Subunits. As shown in Fig. 1, the nuclease-digested and detergent-treated chromatin was fractionated into subunits by sucrose density gradient centrifugation. More than 80% of the digested chromatin could be recovered in the detergent-treated chromatin suspension. Similar results could be obtained from nuclei that were subjected to nuclease digestion. Fraction A represents chromatin proteins and nucleotides released from chromatin by nucleases. Fraction B, which sediments at a rate corresponding approximately to 11 S. represents monomer particles of the chromatin (i.e., chromatin subunits). Fraction B was re-isolated by sucrose density gradient cen-
0
1
2
3
4
5
6
cm FIG. 2. Fractionation of DNA fragments from chromatin by gel electrophoresia DNA from nuclease-digested chromatin (A) and the isolated monomer particles (B) was isolated (5) and the DNA from these two sources was fractionated by gel electrophoresis (29). The arrow (1) indicates DNA fragments of monomer particles of chromatin. The broken line represents the blank gels.
trifugation and the result is shown in the inset of Fig. 1. The subunit fraction precipitated by 10 mM MgCl2 was designated as fraction Bm. The subunits of the enzyme-digested chromatin were further identified by their DNA pattern on gels. DNA extracted from either enzyme-digested chromatin or the isolated chromatin subunits was fractionated by gel electrophoresis. As shown in Fig. 2, both fractions contained a major DNA fragment; in particular, fraction B showed only one major peak. These findings indicated that the possible contamination of dimers or oligomers was minimal in fraction B. The isolated chromatin subunits were further identified by electron microscopy (not shown). Composition of Chromatin Subunits. Protein compositions of fractions, A, B, and Bm from the fractionated chromatin
3460
Proc. Nati. Acad. Sci. USA 73 (1976)
Biochemistry: Liew and Chan
Table 1. Protein composition in fractionated chromatin Fraction Nuclei
Chromatin A B Bm
mg/g of tissue 4.6 2.67 0.27 0.92 0.8
Nuclease-digested chromatin was fractionated by a linear sucrose density gradient centrifugation. Fractions were collected as indicated in Fig. 1. The results are expressed as the protein content of each fraction that could be recovered from the starting materials.
were determined and are shown in Table 1. About 60% of the nuclear proteins were recovered with DNA in the isolated chromatin fraction. About 10% of the chromatin proteins were released during nuclease digestion and recovered in the top layer of the sucrose gradient (fraction A). The chromatin subunits (fraction B) isolated from the gradient accounted for more than 35% of the digested chromatin. More than 85% of the chromatin subunits (fraction Bm) could be recovered from fraction B by precipitation with 10 mM Mg++. Following Mg++ precipitation of each fraction, histones were extracted in 0.25 M HCL NHCP were then extracted in phenol. Taking the acid-soluble fraction as equivalent to 1, the mass ratio of NHCP to histone was determined. As shown in Table 2, NHCP constituted a significant part of chromatin proteins in chromatin subunits. The mass ratio of NHCP to histone in fraction Bm was 0.25 I 0.02 in four separate experiments. In fraction A, the mass ratio of NHCP to histone was 4.85, which indicated that the proteins released by nuclease were mainly
A
Mr
t41
67,000-( 45,000-
* 1: *:
-~~~~~~~~~~~~~
'S Ill?-.
B
I ,z
:.:
-Im
:.ftz,,.
67,000L45,OOO.-
* ..
NHCP. Fractionation of NHCP by Two-Dimensional Gel Elec-
trophoresis. NHCP from the top fraction (A), chromatin subunits (B), and whole chromatin (C) were prepared by removal of histone in an attempt to prevent any nonspecific interaction between histone and NHCP. These proteins were fractionated by two-dimensional polyacrylamide gel etectrophoresis. As shown in Fig. 3, most NHCP associated with chromatin subunits (i.e., fraction B) were identical to proteins in whole chromatin, except for several fractions that were increased. A group of proteins that had molecular weights of over 100,000 in the pH range 5.6 to 8.0 was significantly enriched in fraction B (top rectangle) as compared to fraction A. Acidic proteins of fraction B that focused from pH 4.8 to 5.4, with molecular weights of 55,000, 70,000, and 74,000, were significantly reduced as compared to fraction A. Also, proteins in the chromatin subunits having molecular weights of 70,000 and 67,000 at pH 5.8 to 8.2 Table 2. Ratio of histone to nonhistone chromatin protein in chromatin subunits Fraction
Mass ratio
Chromatin
0.62 4.86 0.25 ± 0.02*
A Bm
Chromatin prepared by the method of Reeder (25) has a ratio of DNA to chromatin protein of 1:1.66. Chromatin protein ratios are expressed with histone taken as 1. * The results represent mean + standard deviation of four separate experiments. The mass ratio of histone to NHCP in the chromatin subunits that were obtained from the nuclease digestion without the treatment of detergents was 1:0.10 (Chan and Liew, manuscript in preparation).
C Xj.
67,00045,000
34
3
4
I 6
pH
I
7
I
I
8
9
FIG. 3. Fractionation of NHCP by two-dimensional gel electrophoresis. NHCP (250 ,g per gel), which were dissolved in 8 M urea buffer, were fractionated by two-dimensional gel electrophoresis (28, 30). A and B represent NHCP isolated from fractions A and B (subunits) of the sucrose density gradients and C represents nucleasedigested whole chromatin. Fractions of special interest are enclosed in broken line rectangles.
(long rectangles, center) were strikingly increased. A fraction of proteins in the pH range from 5.0 to 5.5 with molecular weight of 30,000 was identified in fraction A and chromatin (C). The proteins that were enriched in the chromatin subunits were apparently absent in fraction A. Exchange of Protein in Chromatin Subunits. Experiments were designed to examine whether NHCP were dynamically associated with the chromatin subunits.- Rats were pulse-labeled
Biochemistry:
with [3H]leucine for 2 hr. 3H-Labeled chromatin was isolated concurrently with chromatin from a control group that had not received any isotope. Nuclease-digested chromatin from both groups of animals was then fractionated by sucrose density gradient centrifugation. The top layer (fraction A) and the chromatin subunits (fraction B) were each collected from the [3H]leucine-labeled and unlabeled control chromatins. As shown in Fig. 4(I), 3H radioactivity was found in all chromatin fractions The top fraction (A) and chromatin subunits (B) were pooled and used for the isotopic exchange experiments with the nonlabeled fractions of the control group. The top fraction A and fraction B, which contained the radioactivity, were mixed with fraction B and fraction A, respectively, of the control group. Each mixture was then layered onto a 5-30% sucrose density gradient. As shown in Fig. 4(II), it was demonstrated that 3H-labeled NHCP remained in the same fraction and were not interchanged with the control. In another experiment isotope-labeled fraction A was mixed with nonlabeled nuclease-digested chromatin prior to sucrose density gradient centrifugation. As shown in Fig. 4(III), most of the radioactivity remained in the top fraction, and no radioactivity was detected in the chromatin subunits. DISCUSSION Biochemical findings on chromatin composition and functions (1-5, 10-17, 26, 27, 30, 34-39) as well as recent morphological observations involving nuclease digestion of chromatin (6-9) have added considerable insight to our concept of the structure of chromatin. The existence of NHCP in chromatin subunits has not hitherto been established. A few reports (16, 17) have indicated that NHCP are not found in the subunits. In some cases these findings can be explained on the basis that materials (e.g., erythrocytes) used for chromatin subunit preparation had little NHCP to begin with. Augenlicht and Lipkin (22) reported the absence of NHCP in the subunits of human colonic carcinoma cells, but the presence of NHCP cannot be excluded on the basis of their electrophoretic profiles. On the other hand, Lacy and Axel (21) indicated that acidic proteins were probably present in the subunits. It was therefore a matter of some importance to determine whether NHCP are associated with chromatin subunits. Our findings provide clear evidence that NHCP are present in the subunits. An association of NHCP with the chromatin subunits could have resulted from contamination of dimers or oligomers during the isolation procedures. Our results disprove this possibility by the following criteria: (a) the chromatin subunits isolated from the nuclease digest were further purified by sucrose density gradient centrifugation (the purity of these 1IS particles was confirmed by electron microscopy)*; (b) there was only one major DNA fragment from the chromatin subunit fraction after gel electrophoresis; and (c) isotope experiments showed that NHCP released by nuclease digestion of 3H-labeled chromatin were observed not to have interchanged with the NHCP of unlabeled chromatin subunits during re-centrifugation. It was noted that if detergents were omitted from the nuclease-digested suspension the recovery of subunits after centrifugation.was relatively low. The detergents appear to prevent aggregation and maintain the particles in suspension, thus giving a high yield of chromatin subunits. The nuclease-digested chromatin was then fractionated into monomers, dimers, trimers, and oligomers. This methodology is analogous to the treatment of ribosomes (40). However, Smart and Bonner (41) reported that sodium deoxycholate at concentrations greater *
P. K. Chan and C. C.
Proc. Natl. Acad. Sci. USA 73 (1976)
Liew and Chan
Liew, manuscript in preparation.
F! 202p
~~~ so V ~~~~~~~40
0.1
0°.t 0.61
3461
4°
120
*40
0.2 10
20 30 FRACTION NUMBER
40
FIG.' 4. Fractionation of chromatin particles by sucrose density gradients. [3H]Leucine was given to rats and a control group received no isotope. Liver chromatin was then isolated from each group as described in Materials and Methods. Nuclease-digested chromatin was fractionated by linear sucrose density gradient. Fractions A and B identical to Fig. 1 were collected. Mixing of 3H-labeled fraction A
with nonlabeled fraction B or vice versa was carried out and re-cen-
trifugation was performed as above. Solid lines are A260 la I represents chromatin labeled with [3H]leucine and fractionated by sucrose density gradient. II represents in one graph exchanges of 3H-labeled 0) with nonlabeled fraction B or A fraction A (*- -*) or B (0 in two separate experiments. Ill represents the mixture of 3H-labeled fraction A and nuclease-digested nonlabeled chromatin. Radioactivity was determined by dissolving each fraction in 5 ml of Aquasol (New England Nuclear) and counting in the liquid scintillation counter.
than 0.5% may extract some histone from the DNA complex. We limited the detergent concentration to 0.5% and have shown that small amounts of Histone Hi are present only in fraction A. Assuming that the inter-region of the subunit is accessible to nuclease digestion, one would expect the associated proteins to be released after nuclease treatment. We therefore attempted to answer the following questions in regard to the association of NHCP with DNA and histones: (i) are NHCP located only in the inter-region of the subunits? or (ii) are NHCP present in both the chromatin subunits and the inter-regions? We have shown that about 10% of the chromatin proteins were released into the top fraction of the sucrose density gradient after nuclease digestion. More than 85% of the chromatin proteins in this fraction (i.e., fraction A) were NHCP and only histone HI was released in significant amounts during chromatin digestion. The chromatin subunits that were re-isolated from the sucrose density gradients clearly contained both histones and NHCP. The mass ratio of histone to NHCP was 1: 0.25, which indicates that NHCP constitute a significant proportion of the chromatin subunits. However, chromatin subunits that were obtained from the nuclease digestion without the treatment of detergents have a mass ratio of histone to NHCP as 1:0.1.*
3462
Biochemistry: Liew and Chan
We have also determined the composition of chromatin subunits isolated from nuclei that were digested by nuclease. It was found that the amount of NHCP present in the subunits is similar to that in the subunits prepared from nuclease digestion of chromatin. These findings confirmed the report of Lacy and Axel (21) that acidic proteins were probably present in the chromatin subunits. To substantiate the chemical analysis, we have shown that the chromatin proteins extracted from chromatin subunits gave a sodium dodecyl sulfate/polyacrylamide gel electrophoresis pattern similar to that of whole chromatin proteins, except for the histone fraction, which was significantly enriched in the subunits.* These findings clearly establish that NHCP are associated with both the inter-regions and the subunits. We then undertook to examine whether specific NHCP are associated with the subunits or the inter-regions. As shown in Fig. 3, both quantitative and qualitative differences in NHCP could be demonstrated by two-dimensional gel electrophoresis. At least four major and a few minor groups of proteins were recognized as being specifically associated with the subunits. On the other hand, four groups of NHCP were found only in fraction A and were not present in the subunit particles, which suggests that this portion of NHCP was significantly enriched in the inter-regions and could be released by nuclease digestion. Nonhistone chromatin proteins that focused in the neutral range were greatly enriched in the chromatin subunits. Similar results were also obtained from the chromatin subunits even without the treatment of detergents. * In addition, we have found recently that both the chromatin subunits and the top fraction contain different fractions of highly phosphorylated NHCP.* We therefore conclude that there are certain specific proteins associated with the chromatin subunits and other proteins that interact with the inter-region. The support from the Medical Research Council and the Ontario Heart Foundation, Canada, is acknowledged. Comments on the use of detergents from Dr. N. Straus are greatly appreciated. We are thankful for the competent assistance of Mrs. K. L. Yao. 1. Murray, K. (1969) J. Mol. Biol. 39,125-144. 2. Clark, R. J. & Felsenfeld, G. (1971) Nature 229,101-106. 3. Marushige, K. & Bonner, J. (1971) Proc. Natl. Acad. Sd. USA 68, 2941-2944. 4. Mirsky, A. E. (1971) Proc. Nati. Acad. Sci. USA 68, 29452948. 5. Hewish, D. R. & Burgoyne, L. A. (1973) Biochem. Biophys. Res. Commun. 52,504-510. 6. Olins, A. L. & Olins, D. E. (1973) J. Cell Biol. 59, 252a. 7. Olins, A. L. & Olins, D. E. (1974) Sciene 183,330-332. 8. Woodcock, C. L. F. (1973) J. Cell Biol. 59, 368a.
Proc. Nati. Acad. Sci. USA 73 (1976) 9. Woodcock, C. L., Maguire, D. L. & Stanchfield, J. E. (1974) J. Cell Biol. 63, 377a. 10. Noll, M. (1974) Nature 251,249-251. 11. Clark, R. J. & Felsenfeld, G. (1974) Biochemistry 13, 36223628. 12. Sahasrabuddhe, C. G. & Van Holde, K. E. (1974) J. Biol. Chem. 249, 152-156. 13. Spadafora, C. & Geraci, G. (1975) FEBS Lett. 57,79-82. 14. Weintraub, H. (1975) Proc. Natl. Acad. Sci. USA 72, 12121216. 15. Axel, R. (1975) Biochemistry 14,2921-2925. 16. Oudet, P., Gross-Bellard, M. & Chambon, P. (1975) Cell 4, 281-300. 17. Honda, B. M., Baillie, D. L. & Candido, E. P. M. (1975) 1. Biol. Chem. 250, 4643-4647. 18. Kornberg, R. D. (1974) Science 184,868-871. 19. Kornberg, R. D. & Thomas, J. 0. (1974) Science 184, 865868. 20. Thomas, J. 0. & Kornberg, R. D. (1X975) Proc. Natl. Acad. Sci. USA 72,2626-2630. 21. Lacy, E. & Axel, R. (1975) Proc. Natl. Acad. Sci. USA 72, 22. Augenlicht, L. H. & Lipkin, M. (1976) Biochem. Biophys. Res. Commun. 70,540-544. 23. Gottesfeld, J. M., Murphy, R. F. & Bonner, J. (1975) Proc. Natl. Acad. Sci. USA 72, 4404-4408. 24. Liew, C. C., Liu, D. K. & Gornall, A. G. (1972) Endocrinology 90,488-495. 25. Reeder, R. H. (1973) J. Mol. Biol. 80,229-241. 26. Teng, C. S., Teng, C. T. & Allfrey, V. G. (1971) J. Biol. Chem. 246,3597-3609. 27. Suria, D. & Liew, C. C. (1974) Biochem. J. 137, 355-362. 28. Jackowski, G., Suria, D. & Liew, C. C. (1976) Can. J. Biochem. 54, 9-14. 29. Loening, U. E. (1967) Biochem. J. 102,251-257. 30. Suria, D. & Liew, C. C. (1974) Can. J. Blochem. 52, 11431153. 31. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193,265-275. 32. Burton, K. (1956) Biochem. J. 62,315-323. 33. Chantrenne, H. (1955) Biochim. Biophys. Acta 16, 410-417. 34. MacGillivray, A. J. & Rickwood, D. (1974) Eur. J. Biochem. 41, 181-190. 35. Stedman, E. & Stedman, E. (1950) Nature 166, 780-781. 36. Paul, J. & Gilmour, R. S. (1968) J. Mol. Biol. 34,305-316. 37. Stein, G. S., Spelsberg, T. C. & Kleinsmith, L. J. (1974) Science 183,817-824. 38. Kostraba, N. C. & Wang, T. Y. (1973) Exp. Cell Res. 80, 291296. 39. Allfrey, V. G. (1974) in Acidic Proteins of the Nucleus, eds. Cameron, L. L. & Jeter, J. R., Jr. (Academic Press, New York), pp. 1-29. 40. Liew, C. C. & Gornall, A. G. (1973)J. Blol. Chem. 248,977983. 41. Smart, J. E. & Bonner, J. (1971) J. Mol. Biol. 58,651-659.
