Conformation of the HMG17-nucleosome complex

Biochimica et Biophvsica Acta, 783 (1984) 100-104

I00

Elsevier

BBA Report BBA 90033

C O N F O R M A T I O N OF THE H M G I 7 - N U C L E O S O M E C O M P L E X M I T S U O Z A M A *, K A Z U E I M I T A and S A C H I K O I C H I M U R A

Division of Chemistry, National Institute of Radiological Sciences, Chiba 260, Chiba (Japan) (Received March 20th, 1984)

Key words: Nucleosome; High mobility group protein," Circular dichroism," Fluorescent label

The nucleosome core binds more than two molecules of H M G I 7 at low ionic strength (8.9 mM Tris-HCl/8.9 mM boric a c i d / 0 . 2 5 mM Na2EDTA, pH 8.3). Circular dichroism of the complexes showed only minor conformational changes of the nucleosome core DNA on binding of HMG17, with no detectable change in the histone secondary structure. The fluorescence of N-(3-pyrene) maleimide bound to -SH groups at C y s - l l 0 of 1-13 histones in the core particle suggested that the structure of the histone octamer assembly changed little upon binding of HMG17 to the nucleosome. These observations support the idea that even a high level of H M G I 7 binding, e.g., four H M G s per nucleosome, alone, does not open up the core particle.

The structure of the nucleosome core particle has been extensively studied using a number of different techniques, including X-ray crystallography, and its general features are now known with considerable certainty [1]. On the other hand, the mechanism of nucleosome dynamic perturbations, especially of chromatin transcription, is a current subject of intensive investigation. DNA sequences that are capable of being transcribed are sensitive to digestion by DNAase I [2]. This DNAase I sensitivity has been reported to be dependent on the presence of high mobility group ( H M G ) proteins 14 and 17 [3,4]. Two major binding sites were observed for HMG14 or HMG17 on a nucleosome core particle at high ionic strength [5-8]. Neutron scattering [9] and flow birefringence [10] studies on the structure of the HMG14nucleosome core complex have suggested that only minor changes in the nucleosome core structure are induced upon the H M G protein binding and that the complex is in a highly compact state similar to the uncomplexed core itself.

* To whom correspondence should be addressed. 0167-4781/84/$03.00 © 1984 Elsevier Science Publishers B.V.

In this report we describe the effect of HMG17 binding on the structure of nucleosome cores as studied by circular dichroism of the HMG17nucleosome core complex and by the fluorescence of N-(3-pyrene) maleimide molecules which are specifically bound to Cys-ll0 of H3 in the core particle. Nucleosome core particles were prepared from both chicken erythrocytes and calf thymus. Nucleosome core particles from chicken erythrocytes were prepared by micrococcal nuclease digestion of HI- and H5-depleted chromatin as described previously [11], with minor variations. This extraction process which removes H1 and H5 also extracts the H M G proteins. The monomer nucleosome fractions of a 5-20% linear sucrose density gradient ultracentrifugation of the chromatin digests were collected to isolate core particles. These core particles were further purified by preparative polyacrylamide-agarose gel electrophoresis performed at 4 °C in 4% polyacrylamide (acrylamide/bisacrylamide ratio 20 : 1) -0.5% agarose gels, in 0.1 × Tris-borate-EDTA buffer (buffer 1). A Canalco preparative electrophoresis apparatus (Canal Industrial Corp.) was used with

101 an upper column of PD-2/320. Buffer I is 89 mM Tris/89 mM boric acid/2.5 mM Na2EDTA, pH 8.3. Nucleosome core particles from calf thymus were prepared by micrococcal nuclease digestion of the HI-depleted calf thymus chromatin. The isolation and purification of the core particle from the chromatin digests were as described above for chicken erythrocyte nucleosome cores. A sample of fluorescently labeled nucleosome core particles was prepared by combining N-pyrene maleimide with each of the two cysteine residues of the two H3 histones in the chicken erythrocyte core particle, by the method previously reported [12], except that N-pyrene maleimide-labeled nucleosome core samples were further purified by the preparative polyacrylamide-agarose gel electrophoresis as was done for the purification of core particles. The protein HMG17 was prepared from calf thymus by the method described [13,14], with minor modifications. Preparative acid-urea polyacrylamide gel electrophoresis was applied to remove nonhistone proteins of high molecular weights. The purity of HMG17 was checked by SDS-polyacrylamide gel electrophoresis and amino acid analysis, the gel electrophoretic pattern showed no H1 histone band. The concentration of HMG17 was determined by amino acid analysis. The stock solution of HMG17 (1.6.10 -5 M) in redistilled water was stored at - 2 0 o C. Nucleosome cores at concentrations of 0.9-1.1 A260 units/ml in 0.1 × buffer 1 were mixed with various amounts of HMG17 dissolved in the same buffer. HMG17-(N-pyrene maleimide-nucleosome core) complexes were prepared by mixing varying amounts of HMG17 dissolved in 0.1 × buffer 1 with the N-pyrene maleimide-nucleosome core complex at a concentration of A260 = 0.9 in 0.1 × buffer 1. Gel electrophoresis of HMG17-nucleosome and HMG17-(N-pyrene maleimide-nucleosome core) complexes was performed in 0.1 × buffer 1 in 4% polyacrylamide (acrylamide/bisacrylamide ratio 20:1) gels containing 0.5% agarose. The gel was stained with Coomassie blue. Circular dichroism measurements were performed o~a a Jasco J-20 spectropolarimeter at room temperature. Molar extinction coefficient of DNA at 260 nm of 6700 and 6500 were used in calculations of the molar ellipticities, expressed as mol

nucleotides, [0]p, of calf thymus and chicken erythrocyte nucleosomes, respectively. Fluorescence was measured with a Union Giken FS-401 spectrofluorometer. Solutions of N-pyrene maleimide-nucleosome core complexes always had an absorption of less than 0.01 at the excitation wavelength, 340 nm. HMG14/17-nucleosome core particle complexes have a substantially lower electrophoretic mobility than core particles, providing a convenient binding assay [5-7]. Increasing amounts of HMG17 from calf thymus were mixed with fixed amounts of core particles from the same tissue in 0.1 × buffer 1, and the mixtures were electrophoresed in 4% polyacrylamide-0.5% agarose gels. Fig. 1A shows the gel electrophoresis of HMG17-nucleosome core complexes mixed at ratios of HMG17/nucleosome from 0 to 8. Native nucleosome core particles exhibit a sharp band. With increasing HMG17/nucleosome ratios new sets of bands with lower mobilities appear. At HMG17/nucleosome ratios of 1, 2 and 4, the bands corresponding to nucleosome cores bound to 1, 2 and 4 molecules of HMG17 per particle are dominant. Similar results are obtained with nucleosome cores from chicken erythrocytes complexed with HMG17 from calf thymus, as shown in Fig. lB. These electrophoretic patterns indicate that the nucleosome core can bind more than two molecules of HMG17. Fig. 2 shows the circular dichroism spectra of HMG17-nucleosome core complexes at different HMG17/nucleosome ratios from 0 to 4. It is apparent that the positive circular dichroism signals around 280 nm are depressed further and the magnitude of the negative circular dichroism band at 295 nm increases, with increasing amounts of bound HMG17. The circular dichroism at wavelengths above 250 nm is almost entirely due to the DNA and is strongly affected by its conformational changes. The circular dichroism spectrum of native nucleosome core particles is characterized by the depressed circular dichroism bands at 270-290 nm, in comparison with the bands of free DNA, and a negative band at about 295 nm. This might be accounted for by an increase in the winding angle of the DNA [15,16]. The suppression of the DNA ellipticity on complexing with HMG17 appears to rule out unfolding of the

102

/

,

/

2

-1

iI ×

//

7o

-2

I

II

-3

i -2

[ i -4

L

220

240

260

WAVELENGTH

280

300

320

(nm)

Fig. 1. Electrophoresis on 4% polyacrylamide-0.5% agarose gels s h o w i n g the titration of n u c l e o s o m e core particles with H M G 1 7 at low ionic strength ( 0 . 1 x buffer 1). A. N u c l e o s o m e cores from calf t h y m u s were titrated with H M G 1 7 from the s a m e tissue. M o l a r ratios of H M G 1 7 to n u c l e o s o m e cores: from left to right, 0, 0.5, 0.8, 1, 1.5, 2, 3, 4, 6, 8. B. N u c l e o s o m e cores from chicken erythrocytes were titrated with H M G 1 7 from calf thymus. M o l a r ratios of H M G 1 7 to n u c l e o s o m e cores: from left to right, 0, 0.5, 0.8, 1, 1.5, 2, 3, 4, 6, 8.

Fig. 2. Circular d i c h r o i s m spectra of H M G 1 7 - n u c l e o s o m e core complexes. Both H M G 1 7 and n u c l e o s o m e cores are from calf thymus. M o l a r ratios of H M G 1 7 to n u c l e o s o m e cores: - 0; . . . . . . ,1.1; . . . . . . ,2.2; - - - - - , 4.2. The circular d i c h r o i s m s p e c t r u m of free H M G 1 7 ( . . . . ) is also s h o w n for c o m p a r i son. The c o n c e n t r a t i o n of free H M G I 7 is the same as that in the H M G 1 7 - n u c l e o s o m e core complex at an H M G 1 7 / n u c l e o s o m e ratio of 2.2. Solvent: 0.1 x buffer 1.

superhelical conformation of the core DNA. At lower wavelengths below 250 nm both the D N A and the histones contribute to the circular dichroism spectrum of the nucleosome core particle but the D N A contribution is minimal. In the circular dichroism spectra of the core particles complexed with H M G 1 7 the amplitude of the strong negative band at 222 nm, which is indicative of the high a-helix content of the histones in the particle, slightly increases with increasing amounts of H M G 1 7 bound to the core. This increase in the amplitude could simply be explained by the contribution to the ellipticity from bound H M G 1 7 molecules. The circular dichroism spectrum of free H M G 1 7 is also shown in Fig. 2. Thus it seems reasonable to conclude that the secondary structures of both the histone octamer in the nucleosome core particle and the H M G 1 7 protein are unaffected by the interaction between them. Similar circular dichroism results were obtained for H M G 1 7 complexes with nucleosome cores from

chicken erythrocytes. The sequence properties and conformational behavior of H M G 1 7 (i.e., that this protein in free solution has little or no secondary structure over a wide range of p H and' ionic strengths [17]) suggest that the protein interacts in an extended form with chromatin. In order to examine the possibility of any change in the tertiary or quaternary structure of the histone octamer within a core particle on binding of HMG17, we next measured the fluorescence of the reporter molecules, N-pyrene maleimide, introduced into the nucleosome core particle. As shown in Fig. 3, and as has been reported previously [12,18,19], when N-pyrene maleimide molecules are labeled to each of the two cystein residues at position 110 of the two H3 histones of chicken erythrocytes and the nucleosome core particles are reconstituted, excimer fluorescence with a broad peak around 450 nm is observed as well as the fluorescence of pyrene monomers at 375 and 395 rim. If H M G 1 7 binding perturbs the histone oc-

103

350

~00

~50

5e0

5s0

W~VEtZNCTH (.=)

Fig. 3. Fluorescence spectra of HMG17-(N-pyrene maleimidenucleosome core) complexes. Molar ratios of HMG17 to Npyrene maleimide-nucleosomecores: - - , 0; . . . . . . ,2.05; - - - - - , 4.03, Solvent: 0.1 ×buffer 1. Excitation was at 340 nm.

tamer assembly and moves the two sulfhydryl groups of H3 histones apart, it will result in a decrease in the excimer fluorescence and a concomitant increase in the monomer fluorescence [18,19]. Increasing amounts of H M G 1 7 were added to a constant amount of N-pyrene maleimidenucleosome core complexes to make H M G 1 7 - ( N pyrene maleimide-nucleosome core) complexes. HMG17-(N-pyrene maleimide-nucleosome core) complexes gave essentially the same gel electrophoretic patterns and circular dichroism spectra as those shown in Figs. 1B and 2, respectively (data not shown), indicating that the labeling of Npyrene maleimide to the core particle has little effect on the HMG17-nucleosome core interaction. The fluorescence spectra of the complexes are shown in Fig. 3. Only a slight increase in intensity of the entire fluorescence spectrum of N-pyrene maleimide is observed on binding of HMG17, and the increment is larger at high H M G 1 7 / nucleosome ratios of the complexes. The observed change in the fluorescence excludes the possibility of an opening of the histone octamer to move the two sulfhydryl groups of H3 apart and suggests that the octamer remains as compact as that in the H M G - f r e e nucleosome core particle. Similar changes .in the fluorescence spectrum have been observed when N-pyrene maleimide-labled core particles are exposed to relatively low concentrations of organic solvents, e.g., ethanol ( < 40%) t

[19]. Therefore, the effect, if any, of H M G binding on the structure of the histone octamer in the core particle may be to bring about a lowering of the dielectric constant of the H3-H3 contact regions. Our present results showing that H M G binding does not radically alter the secondary, tertiary and quaternary structure of the nucleosome are in general agreement with those which have been obtained by others by measurements of sedimentation coefficient [7], neutron scattering [9] and flow birefringence [10] of the n u c l e o s o m e - H M G 1 4 / 1 7 complex or by circular dichroism spectra of H M G - b o u n d chromatin [20]. Furthermore, our observations do support the idea that eve a high level of H M G 1 7 binding, e.g., four H M G s per nucleosome, alone, does not open up the core particle. There is a finding, which may relate to these observations, that the presence of an excess of H M G 1 4 / 1 7 does not induce any overall nucleosome unfolding and chromatin unwinding of the adult chicken erythrocyte #-globin gene [211. It has been reported [3,4] that transcriptionally active nucleosornes show enhanced binding affinity for H M G 1 4 and H M G 1 7 compared to bulk nucleosomes and that an interaction with the H M G s makes active nucleosomes especially sensitive to DNAase I. However, some evidence which does not encourage the concept that H M G proteins are the primary agents of the transcriptionally active state of chromatin has been recently presented [22]. There are observations that bulk nucleosomes containing longer than core particle length D N A have a strikingly enhanced affinity for H M G 1 4 / 1 7 [23] and that bulk nucleosomes are also digested more rapidly by DNAase I when they are associated with H M G 1 4 / 1 7 [24]. The problem of how the conformational properties of the nucleosome bound to H M G 1 4 / 1 7 reflect the biological function of the H M G s requires further investigation. This work was supported in part by a Scientific Research G r a n t from the Ministry of Education, Japan. References

1 McGhee, J.D. and Felsenfeld, G. (1980) Annu. Rev. Biochem. 49, 1115-1156

104 2 Weintraub, H. and Groudine, M. (1976) Science 193, 848-856 3 Weisbrod, S. and Weintraub, H. (1979) Proc. Natl. Acad. Sci. USA 76, 630-634 4 Weisbrod, S., Groudine, M. and Weintraub, H. (1980) Cell 19, 289-301 5 Albright, S.C., Wiseman, J.M., Lange, R.A. and Gerrard, W.T. (1980) J. Biol. Chem. 255, 3673-3684 6 Mardian, J.K.W., Paton, A.E., Bunick, G.J. and Olins, D.E. (1980) Science 209, 1534-1536 7 Sandeen, G., Wood, W.I. and Felsenfeld, G. (1980) Nucleic Acids Res. 8, 3757-3778 8 Schr6ter, H. and. Bode, J. (1982) Eur. J. Biochem. 127, 429-436 9 Uberbacher, E.C., Mardian, J.K.W., Rossi, R.M., Olins, D.E. and Bunick, G.J. (1982) Prec. Natl. Acad. Sci. USA 79, 5258-5262 10 Harrington, R.E., Uberbacher, E.C. and Bunick, G.J. (1982) Nucleic Acids Res. 10, 5695-5709 11 lchimura, S., Mita, K. and Zama, M. (1982) Biochemistry 21, 5329-5334 12 Ashikawa, I., Nishimura, Y., Tsuboi, M. and Zama, M. (1982) J. Biechem. 92, 1425-1430 13 Goodwin, G.H., Nicolas, R.H. and Johns, E.W. (1975) Biechim. Biophys. Acta 405, 280-291

14 Sanders, C. (1979) Biochem. Biophys. Res. Commun. 78. 1034-1042 15 Baase, W.A. and Johnson, W.C., Jr. (1979) Nucleic Acids Res. 6, 797-814 16 Chan, A., Kilkuskie, R. and Hanlon, S. (1979) Biochemistry 18, 84-91 17 Abercrombie, B.D., Kneale, G.G., Crane-Robinson, ('., Bradbury, E.M., Goodwin, G.H., Walker, J.M. and Johns, E.W. (1978) Eur. J. Biochem. 84, 173-177 18 Zama, M., Bryan, P.N., Harrington, R.E., Olins, A.L. and Olins, D.E. (1978) Cold Spring Harbor Syrup. Quant. Biol. 42, 31 41 19 Zama, M., Olins, D. E., Prescott, B. and Thomas, G.J., Jr. (1978) Nucleic Acids Res. 5, 3881-3897 20 Sasi, R., HuvOs, P.E. and Fasman, G.D. (1982) J. Biol. Chem. 11448-11454 21 McGhee, J.D., Rau, D.C. and Felsenfeld, G. (1982) Nucleic Acids Res. 10, 2007-2016 22 Seale, R.C., Annunziato, A.T. and Smith, R.D. (1983) Biochemistry 22, 5008-5015 23 Swerdlow, P.S. and Varshavsky, A. (1983) Nucleic Acids Res. 11, 387-401 24 Stein, A. and Townsend, T. (1983) Nucleic Acids Res. 11, 6803-6819