Proc. Nati. Acad. Sci. USA Vol. 88, pp. 10148-10152, November 1991 Biochemistry The nucleosomal core octamer at 3.1 A resolution: A tripartite assembly and a left-handed (//handshake motif/) GINA ARENTS*, RUFUS W. BURLINGAME*t, BI-CHENG WANGt, WARNER E. LOVE§, AND EVANGELOS N. MOUDRIANAKIS*¶l *Department of Biology and §Thomas C. Jenkins Department of Biophysics, The Johns Hopkins University, Baltimore, MD 21218; $Departments of Crystallography and Biological Sciences, The University of Pittsburgh, Pittsburgh, PA 15260; and IDepartment of Biology, University of Athens, Athens, Greece Communicated by Christian B. Anfinsen, August 30, 1991

ABSTRACT The structure of the octameric histone core of differed drastically from those derived from the crystallo- the nucleosome has been determined by x-ray crystaflography to graphic studies of Finch et al. (8) with the nucleosome core a resolution of3.1 A. The histone octamer is a tripartite assembly particle, and from those of Klug et al. (13) on the histone core in which a centrally located (H3-H4)2 tetramer is flanked by two of the nucleosome (hereafter called the MRC model). In this H2A-H2B dimers. It has a complex outer surface; depending on report, we present the results of our crystallographic rede- the perspective, the structure appears as a wedge or as a flat disk. termination of the structure of the histone octamer. It will The disk represents the planar projection of a left-handed pro- become apparent that the "shape and size" issue is now teinaceous superhelix with -28 A pitch. The diameter of the resolved; the outer dimensions of the histone octamer are particle is 65 A and the length is 60 A at its maximum and -10 consistent with the MRC model (length, 55 A; diameter, 70 A) A at its Ium extension; these dimensions are in agreement (13) and not with those reported earlier by us (length, 110 A; with those reported earlier by Klug et al. [Klug, A., Rhodes, D., diameter, 70 A) (11). The shape of the octamer is complex Smith, J., Finch, J. T. & Thomas, J. 0. (1980) Natwre (London) and, depending on the perspective, it can be described as a 287, 509-516]. The folded histone chains are elongated rather wedge or a flat disk (13) as well as a "tripartite assembly with than globular and are assembled in a characteristic "handshake" a central V-shaped (H3-H4)2 tetramer, flanked by two flat- motif. The individual polypeptides share a common central tened balls, the H2A-H2B dimers" (11). The increased length structural element of the helix-oop-helix type, which we name we reported earlier was the result of an apparent rotation of the histone fold. the electron density in the Fourier map about a "special position." This rotation, at the end, resulted in a much In all eukaryotic cells, the nuclear DNA is highly compacted "expanded" tetramer and distally displaced dimers, thus through its association with special , the , generating an octamer with the erroneous length of 110 rather which both neutralize its electrostatic character and provide than 60 A. An extended and detailed account of the crystal- a scaffold for its path. This path must be compatible with the lographic analyses germane to this transformation will be several states oforganization the double helix experiences as published elsewhere (B.-C.W., J. Rose, G.A., and E.N.M., it progresses from interphase chromatin to the metaphase unpublished data). The results of our redetermination of the chromosome and back. At the first level of compaction, the structure of the histone octamer to a resolution of 3.1 A and histones and DNA are organized in repeating units (1) called an R value of 25.5% are presented here at two levels.** First, (2). Within each core nucleosome are two we present solid renderings of the shapes of the histone copies of each of the core histones H2A, H2B, H3, and H4 octamer to facilitate direct comparison with the earlier lower- in the form ofan octameric complex (3). The histone octamer resolution studies (8, 11, 13). Second, we outline salient has been shown by a variety of solution physical chemical recently found features ofthe folding ofthe four polypeptides experiments to be internally organized as a tripartite protein as well as their organization within the H2A-H2B dimer and assembly (4, 5). (H3-H4)2 tetramer subunits of the histone octamer. A de- Over the past decade, the structure ofboth the nucleosome tailed analysis will be presented in a forthcoming article. and the histone octamer has been the subject of intensive investigations. They have included small-angle neutron and MATERIALS AND METHODS small-angle x-ray diffraction (6, 7) of solutions of these assemblies, as well as x-ray crystallography of their ordered Structure Determination. The chicken erythrocyte histone states crystallized with the aid of either organic solvents (8, octamer forms large crystals in space group P3221 that 9) or high salts (10, 11). Our earlier crystallographic analysis diffract to better than 3.0-A resolution (10). Only one-half of of the chicken erythrocyte histone octamer described the the octamer-i.e., H2A-H2B-H3-H4-is present in the asym- octamer as having a "tripartite organization, that is, a central metric unit. This requires the molecular and crystallographic (H3-H4)2 tetramer flanked by two H2A-H2B dimers." Its twofold axes to coincide. A heavy atom derivative of the shape was described as an ellipsoid 110 A long and 65-70 A histone octamer crystals with tetrakis-(acetoxymercu- in diameter (11). At that level of analysis, the internal ri)methane was found that gave only one peak in the Patter- organization of the particle was in agreement with our earlier son map (11) at x = 0.322, y = 0.329, z = 0.992 (equivalent biochemical work that demonstrated the tripartite nature of the histone octamer in solution (4) and could accommodate Abbreviation: ISIR, iterated single isomorphous replacement. the histone-DNA contacts suggested by the studies of Mirz- tPresent address: Scripps Clinic and Research Foundation, La Jolla, abekov et al. (12). However, the shape and size of our model CA 92037. "To whom reprint requests should be addressed at: Department of Biology, The Johns Hopkins University, Baltimore, MD 21218. The publication costs of this article were defrayed in part by page charge **The a-carbon coordinates have been deposited in the Protein Data payment. This article must therefore be hereby marked "advertisement" Bank, Chemistry Department, Brookhaven National Laboratory, in accordance with 18 U.S.C. §1734 solely to indicate this fact. Upton, NY 11973 (reference 1HIO). 10148 Downloaded by guest on September 30, 2021 Biochemistry: Arents et al. Proc. Natl. Acad. Sci. USA 88 (1991) 10149 to z = -0.008), and the occupancy of this site was <50o Finally, the asymmetric unit of the new map contains no since the two -SH groups of the octamer are very close to redundant electron density-i.e., there is no partial repeat of each other on either side of the dyad. Thus, only one metal the dimer density near the tetramer. atom can be bonded to one but not to both of the cysteines. Initial Model. The interpretation ofthe map was facilitated Phases calculated by the iterated single isomorphous replace- by the fact that the asymmetric unit contains half of the ment (ISIR) procedure of Wang (15) resulted in the electron (H3-H4)2 tetramer and one H2A-H2B dimer and a single density map shown earlier (11). cysteine as residue 110 ofH3. A simple inspection ofthe map For reasons to be detailed elsewhere (B.-C.W., J. Rose, revealed one contiguous protein domain containing the mer- G.A., and E.N.M., unpublished data), the data were reex- cury of tetrakis-(acetoxymercuri)methane (11) and this was amined and the heavy atom parameters were reevaluated. assigned to the H3-H4 half-tetramer. The other separate but This process resulted in a slight displacement of the heavy closely apposed domain was assigned to the H2A-H2B dimer. metal position to a new site with x = 0.342, y = 0.344, z = It is interesting to note that the density of the dimer domain 0.007, and an improvement of the R(culliS) value for this site in the new map is nearly identical to the density for the from 51.3% to 48.0%o. Further examination of our earlier analogous dimer domain in the previous map. The overall results revealed that the minimum in R value calculated for quality of the present map and clear connectivities between the old site was actually a local minimum. Comparison ofthe secondary structure elements in it allowed for a straightfor- heavy metal difference Patterson (FPH - Fp)2 with theoretical ward tracing of the four chains. Patterson syntheses based on the old and new sites confirmed We have built a model for the histone octamer by using the the correctness ofthe new site; therefore, the old heavy metal program FRODO (16) on an Evans and Sutherland PS-390 and coordinates were abandoned. The ISIR procedure was used MicroVAX II minicomputer. Our initial model contained once again to calculate new phases for the original data by side-chain atoms for approximately two-thirds of the H2A- using the coordinates of the new heavy atom site, and a new H2B residues, while all others were built in as alanines. A electron density map was calculated. single round of refinement by simulated annealing using the By comparing the two Fourier maps, it is clear that the program XPLOR (17) reduced the crystallographic R value molecular boundary found in the old map is larger than that from 48% to 29% for the 10- to 3.5-A data set. An improved found in the new map. The old boundary contains the same map was calculated by using the SIGMAA program of Read fundamental domains of electron density present in the new (18) to combine the model phases with the ISIR phases. map, as well as partial repeats of one of those domains. In Another round ofrebuilding, refitting, and refining using both addition, there exists a 1200 rotational displacement of the XPLOR and the restrained least-squares refinement program major domains between the old and new maps; it was this PROFFT (19-21) was initiated by using the 10- to 3.1-A data rotation about the special position that generated the false set-i.e., 13,542 reflections with F. > 2 X o(Fo). The image in the old map and this resulted in an elongated shape resulting model (no water molecules included) has good for the histone octamer (B.-C.W., J. Rose, G.A., and overall stereochemistry (rms deviations from ideal bond E.N.M., unpublished). We constructed our molecular model lengths, 0.016 A; rms deviations from ideal bond angle within the new map for the following reasons. In this appli- lengths, 0.039 A), and 4-q values that are reasonable for a cation of the ISIR procedure, we used modified heavy atom partially refined model (22)-i.e., <5% fall in energetically coordinates that had refined to a better R value and the unfavorable regions. The R value for 2644 nonhydrogen phases generated thereby had a better figure of merit (new atoms with individual thermal factors is 25.5% (thermal (m) = 0.75; former (m) = 0.65); consequently, this map shows variabilities between main chain atoms weighted to 1.5 A2). improved connectivities between major secondary structure The difference Fourier map suggests that significant improve- elements as well as improved side-chain densities (Fig. 1). ments to the model can be made in two areas: (i) by refitting the extended chain regions between major helices, and (ii) by extending the model building into the areas of weak electron density distal to the current carboxyl end of H2A and amino end of H2B. RESULTS Secondary Structure. Our current model accounts for =70% ofthe histone octamer, and ofthese residues =65% are found in helices (Fig. 2). This is slightly in excess of the 62% assigned to structured residues by the earlier studies of Bradbury and associates (23-25). Within the interpreted part 00 H2A F - - DMOO _ o Omo0OO

H2B ------Wom oooo

H3 ------00000

H4 F- - - g M o--m

1 20 40 60 80 100 120 140 Residue Number FIG. 2. Secondary structure of the four core histone chains. a-Helices are represented by coils, extended chains are represented FIG. 1. Stereo photograph demonstrating the quality of the ISIR by solid lines, and unseen residues are represented by dotted lines. map. The long helix of H2A is shown, with some density below it Boldface segments of each chain participate in the common histone attributable to H2B. fold. Downloaded by guest on September 30, 2021 10150 Biochemistry: Arents et al. Proc. Natl. Acad. Sci. USA 88 (1991)

FIG. 3. Three orthogonal views of the histone octamer. (a) View showing the tripartite nature of the histone octamer, looking down the molecular twofold axis with the superhelical axis running hori- zontally from left to right. We refer to this as the front view. (b) Protein wedge as it appears by looking down at a plane containing the twofold and the superhelical axes, with the twofold axis running from top to bottom. The apex of the wedge is formed exclusively by the tetramer, while the H2A-H2B dimers form the lobes ofthe wedge. (c) View showing the histone octamer as a disk, looking down into the superhelical axis with the twofold axis horizontal. Protrusions from the curved surface are due to the termini of H2A, H2B, and H3. Surfaces were calculated from a-carbon positions at twice the van der Waals radius. The H2A-H2B dimers are dark blue and the tetramer is white.

of the map the density for most side chains is excellent (Fig. the perspective of viewing this assembly, three distinct shapes 1), with the expected exception of truncated density for the can be perceived and the overall mass distribution can be side chains of 30 polar residues. Since the poorly imaged and described as tripartite or tetrapartite. The tripartite image is most missing portions of the electron density are not expected to clearly seen when looking straight into the dyad axis as it enters contribute additional secondary structure, we calculate that the tetramer apex; we referto this as the front view (Fig. 3a). The -45% of the total octamer residues are found in helices, in centrally located larger mass, which from this perspective ap- agreement with biochemical studies (refs. 26 and 27 and pears laterally biconcave and resembles a left-handed propeller, references therein). Of the remaining 35% of the residues in is the (H3-H4)2 tetramer. The two smaller masses, one on each the model, 10%o are in P-strands, mostly found in intermo- side ofthe tetramer, are the H2A-H2B dimers. The surface ofthe lecular contacts. The remaining 25% of the model is in a octameris traversed by several grooves and ridges, which appear random-coil configuration. to follow the path of a left-handed superhelix. The axis of this Description ofthe Octamer StrUCture. The histone octamer is a superhelix is perpendicular to the twofold axis and runs horizon- tripartite protein assembly consisting of two H2A-H2B dimers tally from left to right in Fig. 3a. A perspective orthogonal to the flanking one centrally located (H3-H4)2 tetramer. Depending on earlier view and with the molecular dyad running from top to bottom reveals the octamer as a tripartite wedge (Fig. 3b). When viewed down the superhelical axis (Fig. 3c), the octamer resem- I bles a disk =65 A in diameter. A closer examination ofthis view reveals that the disk really represents the planar projection of a tetrapartite, left-handed, proteinaceous molecular superhelix (Fig. 4) formed by the spiraling ofthe fourdomains (H2A-H2B)l/

3 4

FIG. 4. Left-handed protein superhelix. The first dimer is at the upper left and sits behind the tetramer, which curves forward at the FIG. 5. Stereo pair of the H2A-H2B and H3-H4 dimer domains bottom left, crosses the center, and curves back touching the second within the histone octamer, viewed approximately down the super- dimer at the upper right. The dimers are dark blue; the first helical axis so as to optimize visualization of the four chains and the half-tetramer is gray; the second half-tetramer is white. In this view, histone fold. For reasons of clarity, only one of each histone pair is the histone octamer can be perceived as tetrapartite. Black line shown. The amino end of each chain in the model is marked by an represents the path of the left-handed protein superhelix. arrow. Downloaded by guest on September 30, 2021 Biochemistry: Arents et al. Proc. Natl. Acad. Sci. USA 88 (1991) 10151 (H3-H4)1/(H3-H4)2/(H2A-H2B)2 along a central axis. The order Finch et al. (8) and Klug et al. (13), the 16-A-resolution of these four histone domains on the protein superhelix is in neutron diffraction study of Bentley et al. (30), and the agreement with that proposed by Klug et al. (13). Viewed this 7-A-resolution x-ray diffraction study ofRichmond et al. (31). way, the protein masses line the inside of an imaginary cylinder These reports, as well as our current work, show the protein with a diameter of 65 A and occupy one and two-thirds turns portion of the nucleosome to be a wedge-like structure with about its axis. This results in a protein trapezoid or wedge with a persistently curving outer surface that resembles a helical a thin side (-10 A) at the tip and a thick side (-60 A) at the lobes ramp. Most of the mass in this protein assembly is concen- of the wedge. trated within four major segments, a feature also clearly Tetramer Structure. The tetramer contains two elements of noted in the earlier work of Bentley et al. (30). We have identical volume, each consisting of one H3 and one H4 explained how the tripartite and wedge views of the octamer molecule. Each H3-H4 dimer (or half-tetramer) resembles a are alternate perspectives of this assembly. crescent-shaped sector, like a sector of an orange, but with The main area of difference between our interpretation of one blunt end and one pointed end. The pointed ends of the the present histone octamer map and the model presented by two sectors establish strong contacts as they overlap at the Richmond et al. (31) is a result of the difference in resolution twofold axis and form the tetramer. Each sector is rotated of the two studies and concerns the location of the individual away from the twofold axis by -15°; therefore, the entire histone molecules within the octamer. At the original 22- and tetramer resembles a shallow horseshoe that has been par- subsequent 7-A-resolution structures of the MRC group, it tially twisted open. As measured along the axis of the was not possible to assign side-chain densities corresponding molecular superhelix, the twofold axis, and their mutual to the histone sequences and thus directly identify the perpendicular, the entire tetramer measures about 46 x 52 x individual histone chains. To assign locations for the histone 66 A. When viewed down the axis of the protein superhelix, molecules within the nucleosome structure, Klug et al. (13) the tetramer occupies =270° of the cylinder. A large central used information from the histone-DNA cross-linking stud- cavity is bounded by the interior curvature ofthe tetramer on ies of Mirzabekov et al. (cf. ref. 12) and from several one face and by the dimers on the other faces. The openings histone-histone cross-linking studies (cf. ref. 13). To this they from the solvent into this cavity are roughly circular, with an applied the straightforward minimum assumption that the approximate diameter of 8 A. The current estimate for the central portion of each histone chain (excluding the highly volume of the cavity is -5500 A3. At our present resolution charged chain termini), because of its small size, consisted of and level of refinement, we find traces of electron density in a single (roughly globular) domain. At 3.1-A resolution, we one portion of the cavity that is near the start of the ordered can now trace the four unique polypeptide backbones as well portion of H2B; therefore, we expect that in our final model as identify the side-chain densities and place the histone some of the volume of the cavity may be filled by additional sequences. None of the four core histones is residues. compacted into a single globular domain. Instead, each chain Dimer Structure. The H2A-H2B dimer has some similarity is folded in a rather elongated fashion; upon assembly into in shape and volume to the half-tetramer domain. When their physiological subunits, the domains of the folded poly- viewed in the coordinate system described above, the dimer peptides interdigitate extensively rather than each chain measures approximately 31 x 50 x 44 A. The outside of the occupying a unique and contiguous segment on the surface of dimer is not as smoothly curved as the half-tetramer; it can the octamer. This arrangement generates the potential for be better described as a somewhat flattened and elongated several noncontiguous contacts between each of the four ball rather than as a crescent-shaped sector. In the protein polypeptides and the DNA helix as it winds its path around superhelix of the octamer, each dimer occupies -160°. the octamer. Within both the H3-H4 (half-tetramer) and the H2A-H2B In addition, certain features of our model correlate well dimer domains, the pairwise association of the folded histone with several physical and biochemical studies. Martinson and chains follow a characteristic "handshake" motif; that is, coworkers (32) showed that a cross-link between Pro-26 of rather than assembling like the globular domains of the a and H2A and Tyr-40 of H2B could be induced by UV irradiation /3 chains of the hemoglobin dimer, which have small local of dimers. We find that these two residues participate in one contacts, the histone chains, by clasping each other, develop of the numerous contacts between H2A and H2B. Martinson an extensive molecular contact interface (Fig. 5). Also, et al. (33) have also found that three cross-links can be formed within both the H2A-H2B and the H3-H4 domains, the in intact octamers between the last 18 residues of H4 and the individual polypeptide chains are folded in a somewhat carboxyl-terminal half of H2B. The octamer model contains similar manner, most noticeably in the central portion of each two different types ofcontacts for H4 and H2B, both ofwhich chain. The common motif, which we call the histone fold, include the residues identified above; the carboxyl terminus consists ofa long central helix flanked on either side by a loop of H4 is located near the dimer-dimer interface, and, accord- segment and a shorter helix (Figs. 2 and 5). This structural ingly, the carboxyl-terminal one-third of H4 is able to form similarity suggests a common evolutionary origin for the four contacts with both H2B molecules. core histones, a finding supported by a mild primary se- The percentage and location ofordered protein found in the quence homology among these four chains (ref. 28 and new map also correlate well with the findings ofBradbury and references therein). The separation of the amino- and car- coworkers (23-25), who reported that residues 25-95 ofH2A, boxyl-terminal regions by a long (24-28 residues) helix is 37-114 of H2B, 42-110 of H3, and 33-102 of H4 are ordered. reminiscent of the shape of the troponin family of structures, Our model includes these residues and extends the structured but the histone loops are shorter than those found in EF portion of the octamer by 66 more residues (Fig. 2). hands (cf. ref. 29) and do not contain the necessary amino Conclusions. The octameric histone core ofthe nucleosome acid residues for binding calcium. The greatest differences in is a tripartite protein assembly that, depending on the per- the folding of the ordered portions of the histone chains are spective of its viewing, appears more or less as a flat disk or found at their amino and carboxyl ends. a wedge. As noted by Baldwin et al. (6), Finch et al. (8), and Klug et al. (13), the histone octamer serves as a protein "spool" around which 140 base pairs of right-handed DNA DISCUSSION is wrapped in the form of a left-handed DNA superhelix Correlations. The overall size and shape of the 3.1-A- (DNA-SH) (inner diameter, -70 A; length, -55 A). In the resolution octamer structure we describe here is entirely present study, we have also resolved this core protein spool consistent with the results of the 22-A-resolution study of as a left-handed protein superhelix (Pr-SH) (outer diameter, Downloaded by guest on September 30, 2021 10152 Biochemistry: Arents et al. Proc. Natl. Acad. Sci. USA 88 (1991) -65 A; extension along the superhelical axis, -60 A). The also thank Dr. David Schmickel for providing some of the crystals outer surface of this fairly evenly curved Pr-SH has regularly used in this study and Dr. Andreas Baxevanis for numerous critical spaced ridges and valleys that define a strong left-handed discussions, his help with the illustrations, and his unrestrained path of -28-A pitch, very suggestive of the path of the devotion to this project. We also acknowledge the support of the DNA-SH in the nucleosome. The left-handed superhelical Supercomputing Laboratory of the National Cancer Institute (Fred- spool of protein is formed by the ordered spiraling assembly erick, MD). This work was supported by a grant from the National of one H2A-H2B dimer to one side (left, with the front as Institutes of Health (GM-33495 to E.N.M.). defined earlier in Fig. 3a) of one (H3-H4)2 tetramer and a 1. Kornberg, R. D. (1974) Science 184, 868-871. second to the other side (right) of the tetramer. It should be 2. Oudet, P., Gross-Bellard, M. & Chambon, P. (1975) Cell 4, noted that the area of the dimer-tetramer contact interface is 281-300. much more extensive than the interface between the two 3. Thomas, J. 0. & Kornberg, R. D. (1975) Proc. Natl. Acad. Sci. H3-H4 half-tetramers. However, the inter-dimer-tetramer USA 72, 2626-2630. interface is more open and potentially accessible to solvent 4. Eickbush, T. H. & Moudrianakis, E. N. (1978) Biochemistry than the intra-tetramer interface. Equilibrium binding and 17, 4955-4964. calorimetric studies have already established that this exten- 5. Benedict, R. C., Moudrianakis, E. N. & Ackers, G. K. (1984) Biochemistry 23, 1214-1218. sive dimer-tetramer interface is by far the preferred surface 6. Baldwin, J. P., Boseley, P. G., Bradbury, E. M. & Ibel, K. for the first octamer disassembly step (4, 5) and thus is (1975) Nature (London) 253, 245-249. energetically less stable than the smaller intra-tetramer in- 7. Perez-Grau, L., Bordas, J. & Koch, M. H. J. (1984) Nucleic terface. The intersection of each of the two dimer-tetramer Acids Res. 12, 2987-2996. interfaces with the surface of the octamer define an "en- 8. Finch, J. T., Lutter, L. C., Rhodes, D., Brown, R. S., Rush- trance perimeter" of potential solvent channels leading into ton, B., Levitt, M. & Klug, A. (1977) Nature (London) 269, the inside of the octamer. This is consistent with our earlier 29-36. proposal of these interfaces as sites for the regulation of 9. Finch, J. T., Brown, R. S., Rhodes, D., Richmond, T., Rush- chromatin compaction-decompaction processes (4). ton, B., Lutter, L. C. & Klug, A. (1981) J. Mol. Biol. 145, 757-769. The overall structural features of the histone octamer lead 10. Burlingame, R. W., Love, W. E. & Moudrianakis, E. N. (1984) to interesting suggestions about possible selective pressures Science 223, 413-414. in the evolution of deoxynucleoproteins. We have proposed 11. Burlingame, R. W., Love, W. E., Wang, B. C., Hamlin, R., earlier (34) that the process of chromatin assembly can be Xuong, N.-H. & Moudrianakis, E. N. (1985) Science 228, viewed as a two-step event; first, local histone-DNA inter- 546-553. actions result in the local dehydration of the double helix, 12. Mirzabekov, A. D., Shick, V. V., Belyavsky, A. V. & which in turn leads to the development of a local left-handed Bavykin, S. G. (1978) Proc. Natl. Acad. Sci. USA 75, 4184- curvature or bending of the DNA axis. Progressive and 4188. sequential addition of more histone assemblies along the 13. Klug, A., Rhodes, D., Smith, J., Finch, J. T. & Thomas, J. 0. (1980) Nature (London) 287, 509-516. length of the double helix will compound this initially local 14. Ferrin, T. E., Huang, C. C., Jarvis, L. E. & Langridge, R. curvature and extend it to a contiguous curvature, which (1988) J. Mol. Graphics 6, 13-21. eventually gives rise to a long-range left-handed DNA super- 15. Wang, B. C. (1984) Acta Crystallogr. A 40, C12. helix. This nucleoprotein superhelix is further stabilized and 16. Jones, T. A. (1978) J. Appl. Crystallogr. 11, 268-272. simultaneously punctuated as a "string of beads" by means 17. Brunger, A. T., Kuryan, J. & Karplus, M. (1987) Science 235, of the short-range histone-histone interactions within the 458-460. core octamer. Thus, the histone masses, by virtue of their 18. Read, R. J. (1986) Acta Crystallogr. A 42, 140-149. association with this DNA superhelix, are expected to dis- 19. Hendrickson, W. A. (1985) Methods Enzymol. 115, 252-270. tribute themselves in a protein superhelix of the same sense. 20. Finzel, B. (1987) J. Appl. Crystallogr. 20, 53-55. 21. Sheriff, S. (1987) J. Appl. Crystallogr. 20, 55-57. In our earlier model studies of the compaction of pure DNA 22. Ramachandran, G. N., Ramakrishnan, C. & Sasisehkaran, V. in the absence of any protein, we observed that reduction of (1963) J. Mol. Biol. 7, 95-99. the water activity by simple chemical means led to the 23. Bradbury, E. M. & Rattle, H. W. E. (1972) Eur. J. Biochem. formation of left-handed DNA supercoils with dimensions 27, 270-281. compatible with those found in chromatin (35). We have 24. Bradbury, E. M., Cary, P. D., Crane-Robinson, C. & Rattle, suggested that the histones must have been selected through H. W. E. (1973) Ann. N. Y. Acad. Sci. 222, 266-288. evolution to fit within those dimensions characteristic to the 25. Moss, T., Cary, P. D., Abercrombie, B. D., Crane-Robinson, DNA supercoil and to facilitate DNA bending and supercoil C. & Bradbury, E. M. (1976) Eur. J. Biochem. 71, 337-350. formation by reducing the local water their 26. Godfrey, J. E., Baxevanis, A. D. & Moudrianakis, E. N. activity upon (1990) Biochemistry 29, 965-972. association with the double helix. Reciprocally, histone- 27. Beaudette, N. V., Fulmer, A. W., Okabayashi, H. & Fasman, histone interactions would have been optimized for the G. D. (1981) Biochemistry 20, 6526-6535. degree of specificity and favorable energetics to harmonize 28. Brown, A. P. (1983) J. Theor. Biol. 104, 401-416. with the DNA superhelix. We propose that the selective 29. Strynadka, N. C. J. & James, M. N. G. (1989) Annu. Rev. pressure of such interactions is responsible for the evolu- Biochem. 58, 951-998. tionary stability of the individual subunits of the histone 30. Bentley, G. A., Lewit-Bentley, A., Finch, J. T., Podjarny, octamer and for the strength and stringency of the next level A. D. & Roth, M. (1984) J. Mol. Biol. 176, 55-75. of interactions of these subunits to form the histone octamer. 31. Richmond, T. J., Finch, J. T., Rushton, B., Rhodes, D. & In other words, the selective pressure in histone evolution Klug, A. (1984) Nature (London) 311, 532-537. 32. Callaway, J. E., DeLange, R. J. & Martinson, H. G. (1985) must derive from the extensive interfacial contacts between Biochemistry 24, 2686-2692. the histones in the octamer on one hand and the outer surface 33. Martinson, H. G., True, R., Lau, C. K. & Mehrabian, M. of the histone superhelix with the inside of the DNA super- (1979) Biochemistry 18, 1075-1082. helix on the other. 34. Moudrianakis, E. N., Anderson, P. L., Eickbush, T. H., Longfellow, D. E., Pantazis, P. & Rubin, R. L. (1977) in The We are grateful to Drs. Mario Amzel and Wayne Hendrickson for Molecular Biology of the Mammalian Genetic Apparatus, ed. their critical reading of the manuscript and many helpful suggestions. Ts'o, P. 0. P. (Elsevier North-Holland, Amsterdam), Vol. 1, We thank Dr. Ernesto Freire for allowing us access to the IRIS pp. 301-321. graphics station and use of the MIDAS (14) software in the Biocalo- 35. Eickbush, T. H. & Moudrianakis, E. N. (1978) Cell 13, 295- rimetry Center and Dr. John Rose for computational assistance. We 306. Downloaded by guest on September 30, 2021