Conformational changes of histones and DNA during the thermal denaturation of nucleoprotein

Characterization of the chromosomal protein MC1 from the thermophilic archaebacterium Methanosarcina sp. CHTI 55 and its effect on the thermal stability of DNA

In the deoxyribonucleoprotein complex of Methanosarcina sp. CHTI 55, DNA is associated with two proteins, named MC1 (methanogen chromosomal protein 1) (Mr 10,760) and MC2 (Mr 17,000). Protein MC1, the most abundant of these proteins, is closely related to the Methanosarcina barkeri MS protein MC1. The effect of Methanosarcina sp. CHTI 55 protein MC1 on the thermal stability of DNA has been studied in native deoxyribonucleoprotein complex, as well as in reconstituted complexes, and it has been compared to the effect of E. coli DNA-binding protein II. Both proteins are able to protect DNA against thermal denaturation, but the differences observed in the melting profiles suggest that they interact by different mechanisms. Moreover, our studies indicate that one molecule of protein MC1 protects eight base pairs of DNA.

Differential dissociation of histone tails from core chromatin

The dissociation of the trypsin-sensitive basic tails of the core histones in core chromatin has been followed as a function of [NaCl] using proton NMR spectroscopy. The tails dissociate in a highly cooperative all or none manner over the salt concentration range 0.2-0.6 M, that is, below the salt concentration required to dissociate the complete molecule. Assuming that each basic tail dissociates independently, the total number of salt linkages involved in binding the tails to DNA is 103. This equals the number of basic side chains in the tails of an octamer. The standard free energy of dissociation, delta G degree, in 1 M NaCl at 297 K is 3.6 kcal/mol. Temperature had no effect on the extent of dissociation up to 45 degrees C. However, between 45 and 65 degrees C, where the premelting transition in the core chromatin occurs, the tails dissociated completely. Dissociation of the tails was associated with a conformational transition in the DNA consistent with loss of supercoiling. From this, and the results of a previous study, it can be shown that the structured, trypsin-resistant domain of each core histone octamer makes 100 salt linkages to DNA. Thus, in 10 mM salt, each core octamer makes a total of 203 salt linkages to DNA.

The thermal denaturation of chromatin core particles

The thermal denaturation of chicken erythrocyte core particles has been followed using changes in absorption and circular dichroism. The absorption-denaturation curve is biphasic, with transitions centred at 59 degrees C and 74 degrees C. The first of these transitions is reversible, whereas heating into the second transition produces irreversible changes in DNA and histone secondary structure. After heating to temperatures within the second transition and then cooling, two components are observed in the analytical centrifuge: a slow-sedimenting species which is present in proportions corresponding to the extent of irreversible denaturation and a fast sedimenting species which is present in inverse proportion to the extent of denaturation. The slow-sedimenting component was primarily denatured DNA. The results suggest that thermal denaturation of core particles is a cooperative, 'all-or-none' process.

Thermal melting of chromatin from the pancreas and liver of guinea pigs treated with 1-methyl-1-nitrosourea

The carcinogen 1-methyl-1-nitrosourea (MNU) can cause pancreatic cancer in guinea pigs. We have examined the relative damage produced by MNU treatment on the chromatin from pancreas and liver of these animals. Thermal denaturation of chromatin from guinea pig pancreas and liver was studied following parenteral administration of MNU in several doses. Estimates of single strand breakage were also obtained by examination of the fluorescence of intercalated ethidium bromide. Oligomeric chromatin melted with a main Tm at 78 degrees C, with additional components at 48 degrees C, 55 degrees C and 65 degrees C. Repetitive treatment with MNU at several doses between 20 mg/kg and 70 mg/kg produced destabilization of pancreatic chromatin as shown by a shift from 78 degrees C to lower melting components. The liver by contrast was relatively unaffected. In addition, pancreatic chromatin showed an increase in alkali-induced strand unwinding with MNU treatment, probably due to an increase in single strand breaks, while there was no change in this regard in the liver. The data indicate that the pancreas is more susceptible to damage by MNU than the liver.

Studies on the thermal denaturation of histone-H1-depleted chromatin

The multiphasic thermal denaturation profile of histone-H1-depleted chromatin was studied by using a nucleoprotein preparation which lacked the first high temperature transition at about 72 degrees C. Such a preparation was obtained by heating at 72 degrees C H1-depleted chromatin, the DNA of which was cross-linked with psoralen to ensure a complete renaturation of DNA upon cooling. When this nucleoprotein was redenatured, its melting profile was found to be significantly altered: only one high temperature peak centered at about 82 degrees C was observed in addition to the low temperature transition at about 53 degrees C. The kinetics of digestion of this material with micrococcal nuclease showed a limit digest equal to that found for the 'native' H1-depleted chromatin but the rate of hydrolysis was higher. The monomer particles prepared from this nucleoprotein were found similar to the 'native' monosomes in respect to protein:DNA ratio and size of DNA but showed an altered melting profile: the premelt area was broader, bigger, and centered at lower temperature; the main peak was reduced in size with no change in its melting temperature. On the basis of these data, the following conclusions were drawn: (a) the last two thermal transitions in H1-depleted chromatin most likely reflect the presence within each nucleosome of two regions with different stability of DNA; (b) DNA involved in the first high thermal transition of H1-depleted chromatin belongs to nuclease-accessible DNA, and (c) the main peak in the biphasic melting profile of the monomer particles reflects the denaturation of DNA regions which in the polymer nucleoprotein are involved in the two high temperature transitions.

Thermal transition of core particle is not a two-state process

Thermal transition of core particle which occurs before melting of DNA and can be followed by circular dichroism is not a two-state process; it is the result of two processes which cannot be dissociated in static experiments: unfolding of core particles is immediately followed by their aggregation. It is thus impossible to get thermodynamic parameters of core particle unfolding from its thermal transition monitored by circular dichroism. Thermal denaturation kinetics of core particles gives some information about their stability. Finally core particle structure is more stable in chromatin than in its isolated state.

The interaction of chromatin with alkylating agents. The monofunctional action of bis(2-chloroethyl)methylamine

The reaction of L5178Y lymphoblast cell chromatin with the alkylating agent bis(2-chloroethyl)methylamine has been studied as a function of time, pH and reagent concentration. The reaction with DNA of chromatin from which the proteins were dissociated, as well as with purified calf thymus DNA, was studied in parallel. The extent of alkylation of DNA in intact chromatin was 4--5 times as much as in parallel free DNA samples; up to 4% of nucleotide base pairs were substituted. The extent of monofunctional substitution of the proteins was similar, on a weight basis, to that of DNA. Chromatographic analysis of the depurinated products showed that in chromatin, as in DNA, position N-7 of guanine is the major site of reaction. Up to 25% of the reaction products were guanines cross-linked as bis(2-guanin-7-yl-ethyl)methylamine, indicating a considerable degree of DNA-DNA cross linking. Column analysis shows that up to 40% of the nuclear proteins are cross-linked to DNA at 10 mM bis(2-chloroethyl)methylamine. The increased reactivity of intact chromatin is interpreted in terms of a conformational change in the position of the DNA bases when in the organized nucleohistone complex.

Analysis of chromatin reconstitutiion

The ability of high molecular weight chicken erythrocyte chromatin to spontaneously self-assemble into native-like material, after dissociation by high ionic strength and reassociation by salt gradient dialysis, was critically examined. The native conformational state of the reassembled nucleoprotein complex was regenerated to the extent reflected by circular dichroism spectra and thermally induced helix--coil transition of the nucleoprotein DNA. However, internucleosomal packing of approximately 205 base pairs of DNA per repeating unit, as probed by digestion with micrococcal nuclease, was not regenerated upon reassembly and was replaced by a packing of approximately 160 base pairs per repeating unit. Thus, high molecular weight chromatin containing only lysine-rich histones (H1 and H5) and core histones (H2A, H2B, H3, and H4) is not a true self-assembling system in vitro using the salt gradient dialysis system used herein. Circular dichroism and thermal denaturation studies on core chromatin (lysine-rich histones removed) showed that core histones alone are not capable of reassembling high molecular weight DNA into native-like core particles at low temperature (4 degree C). Reassembly at 21 degree C restored the circular dichroism but not the thermal denaturation properties to those characteristic of undissociated core chromatin. Nonetheless, micrococcal nuclease digestions of both reassembled core chromatin products were identical with undissociated native core chromatin. Ressembly in the presence of the complete complement of histones, followed by removal of the lysine-rich histones, did regenerate the thermal denaturation properties of undissociated native core particles. These results indicated multiple functions of the lysine-rich histones in the in vitro assembly of high molecular weight chromatin.

Effect of tetramethylammonium ions on conformational changes of DNA in the premelting temperature range

The reversible conformational change of DNAs and polydeoxyribonucleotides occurring before melting was followed by circular dichroism. deltatheta/deltaT, the rate of change of ellipticity theta with temperature, was used mainly as a measure of this premelting phenomenon. If sodium ions were replaced by tetramethylammonium ions deltatheta/deltaT decreased for poly (dA) poly (dT) and poly (dA.dT) poly (dT.dA), but increased for poly (dG.dC) poly (dC.dG). DNAs of different base composition showed no more premelting (deltatheta/deltaT approximately 0) even at low molarities of TMACl provided the Na/TMA ratio was very small. For all cases studied the theta values at 0 degrees C and at a given ionic strength were smaller in NaCl than in TMACl. When studying the series of ammonium ions from NH4+ to (C2H5)4N+, the deltatheta/deltaT values first decreased, going through zero with TMA+ ions, and then increased again. A tentative and qualitative explanation of our results can be given: (a) Hydration of the polymers increases in presence of TMA ions and their average stability decreases; locally, however, (AT) pairs are preferentially stabilized by TMA ions owing to a specific interaction at the level of O2 of thymine. (b) In order to explain the different behaviour of (AT) polymers and DNA, it is assumed that only the B structure is able to accommodate TMA ions in the small groove of the double stranded helix.

Alteration in nucleosome structure induced by thermal denaturation

Mononucleosomes prepared from goose erythrocyte nuclei exhibited limited heterogeneity with respect to number of electrophoretic components, histones and DNA composition. The components differ slightly in ionic strength induced self-association. Thermal denaturation of each component gave only two dominant, highly cooperative, melting transitions, T" and T"'. Urea and trypsin were used to establish the differential lability of these two transitions. Comparison of the morphologies of the mononucleosomes at various stages throughout the melting profile indicated that the 13.3 +/- 1.5 nm diameter mononucleosomes start to disrupt only in the latter half of transition T" and do not unfold until after reaching T"'. The resultant, open ended (17.4 +/- 2.2 nm diameter) toroids are still largely negatively staining and much more uniform in shape if fixed simultaneously with gluteraldehyde.

Superstructure and CD spectrum as probes of chromatin integrity

Two types of chromatin were extracted from the same stock of rat liver nuclei by a short exposure to micrococcal nuclease and by shearing respectively. These two materials which are identical in their protein/DNA content and by the presence of the five histones, were compared by means of circular dichroism and electron microscopy. Under the electron microscope and in absence of any divalent cation a superstructure of the unfixed chromatin fiber can be viewed only with native material but is no more present in sheared one. The increase of CD signal at 280 nm (from 2000 to about 4000 cm2 deg.dmole-1) in the case of sheared chromatin is not related to the loss of superstructure but to the structural changes of DNA inside the nucleosomal core which are always produced by shearing. These two correlated observations offer new sensitive probes of the integrity of any native or reconstituted chromatin.

Thermal denaturation of nucleosomal core particles

Thermal denaturation of very homogeneous preparations of core particles from chicken erythrocyte chromatin is studied by several techniques. The change in absorbance, which is very closely paralleled by changes in heat capacity, which is very closely paralleled by changes in heat capacity, is a biphasic process with inflexions at 60 degrees C and 74 degrees C. In contrast, isolated DNA of the same length denatures in a single transition around 44 degrees C. Monitoring the circular dichroism of the cores during thermal denaturation reveals biphasic changes in the secondary structure of the DNA, preceding the base unstacking by 10 degrees C in the first and 3 degrees C in the second phase. However, measurable alterations in the secondary structure of the histones are confined to the second phase with a melting temperature at 71 degrees C. Increase in the ionic strength of the buffer from 1 mM to 10 mM leads to almost monophasic melting curves as measured by absorbance and CD, while not causing any measurable conformational changes at room temperature. The melting of core particles is interpreted as a denaturation of about 40 base pairs in the first phase, followed by a massive breakdown of the native structure of a tight histone-DNA complex, which frees the remaining 100 base pairs for unstacking.

Oestradiol-receptor complexes in subnuclear fractions of rat uterine tissue

The subnuclear distribution of 3H-oestradiol-receptor complexes was studied in uterine tissue of ovariectomized adult rats. Nuclei were sonically disrupted and 8 different subnuclear fractions were isolated by discontinuous sucrose density gradient centrifugation. 3H-Oestradiol-receptor complexes, measured by hydroxylapatite column chromatography, were localized in a light chromatin fraction as well as in a heavy chromatin fraction. Using the hydroxylapatite chromatography technique it was possible to demonstrate three classes of oestradiol-receptor complexes which differ in affinity for the chromatin. Oestradiol-receptor complexes with a high affinity for the chromatin were predominantly localized in the heavy chromatin fraction, whereas complexes with a lower affinity for their acceptor sites were present in the lighter chromatin fraction.

Urea-induced structural changes in chromatin obtained by sedimentation

Sedimentation coefficients have been determined for fractionated preparations of whole and stripped (depleted of very lysine-rich histones and non-histone proteins) chicken erythrocyte chromatin fragments in 0-10 M urea. Significant differences in urea effects are observed between these preparations; differences which can be interpreted structurally by use of Kirkwood's dynamical theory of the translational frictional coefficient. This type of analysis implies that urea-induced chain-swelling in stripped chromatin is due largely to the urea effect upon the constituent nu-bodies, whereas the much larger swelling observed in whole chromatin appears to involve also the effect of urea upon the region between adjacent nu-bodies.

Structural transition in chromatin induced by ions in solution

Structural transition in chromatin was measured as a function of counter ions in solution (NaCl or MgCl(2)) and of histones bound on the DNA. The addition of counter ions to aqueous solutions of chromatin, partially dehistonized chromatin, and DNA caused a drastic reduction in viscosity and a significant increase in sedimentation coefficient. Transitions occurred primarily at about 2 x 10(-3) M NaCl and 1 x 10(-5) M MgCl(2) and are interpreted as a change in structure of chromatin induced by tight binding of cations (Na(+) or Mg(++)) to DNA, either free or bound by histones, and is an intrinsic property of DNA rather than of the type of histone bound. At a given ionic condition, removal of histone H1 from chromatin had only a minor effect on the hydrodynamic properties of chromatin while removal of other histones caused a drastic change in these properties. An increase in the sedimentation coefficient of DNA was observed also for protamine. DNA complexes wherein the bound protein contains only unordered coil rather than the alpha-helices found in histones.

Conformational states of chromatin nu bodies induced by urea

Monomer chromatin nu bodies (nu1) from chicken erythrocyte nuclei were exposed to 0-10 M urea plus 0.2 mM EDTA (PH 7). Alterations in nu1 conformation were examined using hydrodynamic methods (i.e., S, eta, and (formula: see text)), thermal denaturation, circular dichroism, reactivity of histone thiol groups to N-ethyl maleimide, and electron microscopy. The two domains of a nu body (i.e., the DNA-rich shell and the protein-rich core) aeared to respond differently to the destabilizing effects of increasing urea; DNA conformation and stability exhibited noncooperative changes; the core protein structure revealed cooperative destabilization between 4 and 7 M urea. Companion studies on the conformation of the inner histone "heterotypic tetramer" also revealed cooperative destabilization with increasing urea concentration.

Interaction of DNA with poly(L-Lys-L-Ala-Gly) and poly(L-Lys-L-Ala-L-Pro). Circular dichroism and thermal denaturation studies

Complexes of DNA with polypeptides composed of Lys, Ala, and Gly in both a sequential order, poly(L-lysine-L-alanine-glycine), and a statistical distribution, poly(L-lysine36-L-alanine28-glycine), were prepared using gradient dialysis. These polypeptide-DNA complexes were studied using ultraviolet absorption (UV) and circular dichroism (CD) to probe the conformation, binding, and melting behavior of DNA in the complex. Complexes with the sequential polypeptide showed no structural change in the DNA; however, the complexes with the random polypeptide yield CD spectra similar to phi DNA [Maniatis, T., Venable, Jr., J.S., and Lerman, L.S. (1974), J. Mol. Biol. 84, 37]. A second sequential polypeptide, poly(L-Lys-L-Ala-L-Pro)n, -DNA complex was also studied. It was found to exhibit pronounced structural changes as a function of ionic strength and poly-peptide-DNA ratio, more similar to the random sequence that the ordered sequence of the Lys, Ala, Gly polymer. Thus the importance of the composition and amino acid sequence in polypeptides which bind to DNA, even in such simple systems, is demonstrated. Evidence from thermal denaturation, employing simultaneous monitoring of CD and UV changes, supports a model in which specific polypeptides cause condensation of the DNA in the complex into an asymmetric tertiary structure. The relevance of these model systems to chromatin is discussed.

Assembly of DNA with histones and nonhistone chromosomal proteins in vitro

We have examined, by protein binding assays, thermal denaturation, and circular dichroism, the possible effects of histones on nonhistone chromosomal protein (NHCP) interactions with DNA. For these studies, we have fractionated mouse Krebs II chromosomal proteins into three discrete fractions: Mo, 5 M urea-soluble NHCP; M1, 5 M urea-1 M NaCl-soluble NHCP from 5 M urea-extracted chromatin; and M3, 5 M urea-3 M NaCl-soluble chromosomal proteins from 5 M urea-1 M NaCl-extracted chromatin. These fractions contain heterogeneous populations of NHCP, and were found to differentially affect histone binding to DNA by methods of reconstitution, or by direct binding of M0, M1, or M3 to urea-salt reconstituted DNA with histones. M0 was found to exert a unique effect on the thermal denaturation and circular dichroic spectra of DNA-histone complexes. M0 from Krebs II chromatin was also found to complete for DNA sites in the presence of M0 from mouse liver chromatin. In addition, in 5 M urea, pH 8.0, histone binding to DNA reached saturation at 1.85 mg/mg of DNA, higher than the in vivo ratio of 1.00 mg/mg of DNA. Saturation of histone binding to DNA occurred only in the presence of 5 M urea, resulting in a reduction of nonspecific histone-histone interactions on DNA.

Circular dichroism as a probe of DNA structure inside reconstituted nucleohistones

Reconstituted nucleohistones were obtained by mixing in given conditions acid extracted histones and eukaryotic DNA. The histone/DNA ratio (w/w) was in the range 0.35 - 0.95. With the four histones (H2A2B) we have been able to obtain subunits (nucleosomes or upsilon-bodies). The variation of cirsular dichroism signal with temperature at 280 nm was measured to follow structural changes of the DNA inside the complex. The true change of ellipticity (see article) of histone-bound DNA regions, is similar for reconstituted nucleohistone and H1-depleted chromatin, and is therefore a physical probe of the presence of nucleosomes.

Chromatin and nucleosome structure

Chromatin nucleosomes (mononucleosomes through pentanucleosomes) have been isolated by staphylococcal nuclease digestion of calf thymus nuclei. The peak value ellipticity is the same for all oligomers, 1900 deg cm2, mol-1 at 280-nm, 23 degrees C. The dh280/dT vs T show a progressive increase in Tm of the main thermal band (73.5 degrees C, monomer; 79 degrees C, pentamer). Very small amounts of free DNA can be observed in the melting profiles, and shoulders at 60 degrees C and 93 degrees C appear and increase in magnitude as the particle size increases. The magnitude of the change, delta[theta]280, increases with oligomer size. This pattern could result from an initial unfolding of an asymmetric assembly of nucleosomes (polynucleosome superhelix) in addition to the denaturation of the internal nucleosome structure, and a subsequent or simultaneous denaturation of the double strand DNA. The extent of this unfolding appears to depend upon the size of the oligomer and therefore implies interactions between asymmetrically assembled neighboring nucleosomes.

Thermal denaturation of sheared and unsheared chromatin by absorption and circular dichroism measurements

Thermal denaturation of chromatin is observed by simultaneously monitoring absorption and circular dichroism at 276 nm as functions of temperature. Either observation indicates that sheared chromatins shows less thermal stability than native chromatin. The temperature-dependent ellipticities at 276 nm of these chromatins show features not seen in the absorption curves: the ellipticity of unsheared chromatin increases with temperature, while this increase is abolished or greatly reduced in the same chromatin after shearing. After its first thermal transition (prior to the helix-coli transition) the unsheared chromatin achieves the same ellipticity as sheared chromatin.

Studies on the structure and function of chick-oviduct chromatin. 1. Fractionation by ECTHAM-cellulose chromatography and physico-chemical characterization

Chick oviduct chromatin was separated into a ribonucleoprotein fraction and two chromatin fractions (early and late eluting). We utilized a gentle procedure in which moderately hydrated chromatin was subjected to chromatography on a weak ionic-exchange resin (ECTHAM-cellulose) eluted with a combined pH-salt gradient. Chemical analysis of the early (fraction I) and late (fraction II) eluting fractions revealed that their histones were identical and their nonhistone proteins were markedly different. Control experiments showed that these differences were not due to protein rearrangements during chromatin preparation and/or fractionation. The physical properties of fraction I and II differed in certain aspects. The aggregation response of fraction I to increasing concentrations of monovalent cations was five times lower than that of fraction II but the aggregation response to divalent cations was identical. Thermal denaturation assays of DNAs isolated from fractions I and II revealed identical derivative profiles of hyperchromicity vs temperature, thereby indicating similar base composition in the two fractions. Circular dichroism, spectra of the purified DNAs isolated from both fractions showed identical B-type conformations. However, DNA renaturation kinetics analyzed by computer technique indicated that fraction I DNA contained less than half the amount of highly repetitive sequences as compared to either unfractionated chromatin or fraction II. Circular dichroism spectra of fraction I and II chromatins (at room temperature) showed significant differences in a wavelength region were only DNA is optically active (i.e. 255-320 nm). These results indicated that the DNA complexed to proteins in fraction II assumed a more C-type conformation than the DNA in fraction I. The differences in the circular dichroism spectra could not be accounted for by differences in the RNAs or protein chromophores contained in fraction I and fraction II. When the circular dichroism spectra of fraction I and II were recorded at 55 degrees C, the differences between the two fractions were abolished. These results were interpreted to mean that the differences in the DNA conformations found in fractions I and II were due to the differences in their nonhistone proteins. These proteins were effective in maintaining DNA conformation differences only when they were in their native form but not when heated to 55 degree C. Comparison of the sedimentation coefficients of fractions I and II with their calculated molecular weights suggested a more extended structure in fraction I as compared to a more compact structure in fraction II. Only small differences were observed between fraction I and fraction II with respect to either buoyant density analysis in a metrizamide gradient or in the number of phosphate charges accessible to polylysine.

Interactions between arginine-rich histones and deoxyribonucleic acids. I. Thermal denaturation

Physical properties of histone-DNA complexes very often depend upon the method of complex formation. In an attempt to make the studies of histone-DNA interactions more relevant to biological systems, results from thermal denaturation of native chromatin were used as references for determining how closely a given histone-DNA complex approaches its native state in chromatin. In the case of arginine-rich histones H3 (III or f3) and H4 (IV or f2a1), four methods were used for making complexes with calf thymus DNA: (A) NaCl gradient dialysis with urea; (B) NaCl gradient dialysis without urea; (C) direct mixing in 2.5 x 10(-4) EDTA, pH 8.0; and (D) direct mixing in 0.01 M sodium phosphate, pH 7.0. It was observed that a complex made by direct mixing in phosphate (method D) is closer to the native than is one made by direct mixing in EDTA (method C) than the one made by gradient dialysis with urea (method A) or without urea (method B). Regardless of the method used for complex formation, no substantial differences were observed between complexes with histone H3 dimer with disulfide bond(s) and a reduced H3 without disulfide bond, implying that perhaps a dimer with or without disulfide bond is a natural fundamental subunit in our experimental conditions. When the method of direct mixing in EDTA is used, the melting properties of the complexes vary only slightly with any one of the following H3 histones: from calf thymus, H3 without disulfide bond, H3 dimer, and H3 oligomer with disulfide bonds, also, from duck erythrocyte, H3 monomer and dimer. The complexes formed between DNA and a mixture of H3 and H4 by method D have melting properties similar to those of native chromatin. Since an equimolar mixture of histone H3 and H4 in 0.01 M phosphate, pH 7.0, was shown to form a tetramer (D'Anna, J.A., and Isenberg, I. (1974), Biochem. Biophys. Res. Commun. 61, 343), our results suggest that, a tetramer of H3 and H4, likely to be (H3)2(H4)2, formed from one H3 dimer and one H4 dimer, can bind DNA in a manner similar to that in native chromatin.

Interactions between arginine-rich histones and deoxyribonucleic acids. II. Circular dichroism

Circular dichroism (CD) was used to investigate the conformations of arginine-rich histones, H3 (III or f3) and H4 (IV or f2a1), and DNA in the complexes prepared by four different methods: (A) NaCl gradient dialysis with urea; (B) NaCl gradient dialysis without urea; (C) direct mixing in 2.5 x 10(-4) M EDTA, pH 8.0; and (D) direct mixing in 0.01 M sodium phosphate, pH 7.0. Using the CD spectrum of native chromatin as a criterion to judge the closeness of a complex to its native state, it was observed that a complex made by direct mixing at low ionic strength (methods C and D) is better than the ones made by NaCl gradient dialysis with or without urea (methods A and B). It is explained as a result of lack of ordered secondary structures in histones due to the presence of urea in method A or due to nonspecific aggregation in NaCl without urea (method B). Compared with all the earlier reports in literature on the CD of histone-DNA complexes, the CD spectra of arginine-rich histone-DNA complexes prepared by methods C and D are closest to that of native chromatin both in shape and in amplitude. These results imply (a) that arginine-rich histones play an important role in maintaining the conformation of chromatin and (b) that the binding of these two histones to DNA prepared by methods C and D are close to that in native chromatin. Noticeable variation in conformation of free and bound histone and histone-bound DNA has also been observed in histone H3 with one or two cysteine residues, and in reduced or oxidized state even when the complexes were prepared and examined in the same condition. CD spectra of arginine-rich histones in 0.01 M phosphates, pH 7.0, indicate the presence of alpha-helix which could be responsible for a favorable binding of the less basic regions of these histones to DNA under this condition as demonstrated by thermal denaturation (Yu, S. .S, Li H. J., and Shih, T. Y. (1976), Bio-chemistry, the preceding paper in this issue). To preserve or generate alpha-helical structures in histones seems to be a critical step in reconstituting good histone-DNA complexes.

A model for chromatin structure

A model for chromatin structure is presented. (a) Each of four histone species, H2A (IIbl or f2a2), H2B (IIb2 or f2b), H3 (III or f3) and H4 (IV or f2al) can form a parallel dimer. (b) These dimers can form two tetramers, (H2A)2(H2b)2 and (H3)2(H4)2. (C) These two tetramers bind a segment of DNA and condense it into a "C" segments. (d) The adjacent segments, termed extended or "E" segments, are bound by histone H1 (I or fl) for the major fraction of chromatin; the other "E" regions can be either bound by non-histone proteins or free of protein binding. (e) The binding of histones causes a structural distortion of the DNA which, depending upon the external conditions, may generate the formation of either an open structure with a heterogeneous and non-uniform supercoil or a compact structure with a string of beads. The model is supported by experimental data on histone-histone interaction, histone-DNA interaction and histone subunit-DNA interaction.

Comparison between histones FV and F2a2 of chicken erythrocyte. II. Interaction with homologous DNA

The conformation and stability of artificial complexes between chicken erythrocyte DNA and homologous histones FV and F2a2 was studied by circular dichroism (CD) and thermal denaturation followed by both absorbance and CD measurements. The complexes are made after a stepwise potassium fluoride gradient dialysis without urea and studied at low ionic strength (10-minus 3 M). 1) No structural changes of the DNA can be detected up to r equals 0.2 with FV and r equals 0.6 for F2a2. With FV at higher values of r the CD spectrum is altered, indicating the organization of DNA and histones in some kind of aggregate. 2) The conformation of histone molecules inside the complexes is not related to the ionic strength of the medium but to an effective ionic environment close to 0.1 M. This ionic strength would also correspond to the melting temperature of histone-covered DNA. 3) From the analysis of the absorbance melting profile the length of DNA covered with an histone molecule can be estimated. A good agreement is found between the negative charge of this piece of DNA and the net positive charge of the histone. 4) Since the CD transition at 227 nm occurs before the second absorbance transition at 280 nm, the DNA is stabilized no longer by native histone but partially or fully denatured histones. The helical regions of the histone molecule are not involved in the binding process, which appears to be almost purely coulombian and most likely related to some structural fit between the pattern of negative charges in the DNA helix and that of positive charges along the peptide chain.

Comparison between histones FV and F2a2 of chicken erythrocyte. I. Structure, stability and conformation of the free proteins

Purified chicken erythrocyte histones FV and F2a2 were studied by means of circular dichroism as a function of ionic strength and temperature. The percentage of alpha-helical regions was calculated by comparison with reference spectra obtained with four standard proteins of known tertiary structure. Maximal alpha-helical organization, reached in high ionic strength, was estimated to 14% and 23% for FV and F2a2 respectively. We have compared our experimental determinations of the secondary structure of F2a2 with predictions made from amino-acid sequence according to Fasman's rules. When instability induced by the presence of charged residues close together is taken into account, a good agreement is found between predicted and observed values. The thermal denaturation of FV is cooperative and, unlike F2a2, seems to obey a two-state transition. The classical Arrhenius plot is linear, which indicates that the heat capacity is the same in both the native and the denatured state. Such a behaviour is typical of an expanded configuration of FV even in the "native" state.

Properties of chromosomal proteins of human leukemic cells

Purified chromatin isolated from lymphocytic cells derived from patients with acute leukemia, or other lymphoproliferative disorders has been compared with chromatin isolated from normal human lymphocytic cells by gel electrophoresis and differential gradient ultracentrifugation. Thermal denaturation studies showed higher Tm values for chromatin from leukemic cells, as compared to that of lymphocytic cells from normal donors or patients with infectious mononucleosis, reflecting the diverse complexity of these chromatins with respect to their varying chemical compositions. There are significant differences in the ratios of DNA:RNA:protein, as well as in the ratios of chromatin-associated histone and non-histone proteins; although chromatin-associated histones were more homogeneous than were the non-histone proteins, as adjudged by amino acid analyses and acrylamide gel electrophoresis. These differences in chromatin structure may relate to the differences in gene expression characteristic of these lymphocytic cells. The chromosomal acidic proteins isolated from the purified chromatin of human leukemic cells greatly stimulated the template activity of the chromatin in in vitro RNA synthesis. The non-histone proteins selectively interact with chromatins and influence the RNA polymerase reactions, indicating that there is selective tissue specificity of non-histone proteins.

Interaction of histone f2al fragments with deoxyribonucleic acid. Circular dichroism and thermal denaturation studies

The glycine-arginine-rich histone, f2al (IV) (102 amino acids), from calf thymus was cleaved at residue 84 with cyanogen bromide. Complexes containing homologous DNA and each f2al fragment were reconstituted by means of Gdn-HC1 gradient dialysis. The circular dichroic (CD) spectra of these complexes were all examined in 0.14 M NaC1. The CD spectra of the DNA-f2al fragment complexes did not differ appreciably from that of DNA alone in the wavelength region above 240 nm. However, intact f2al-DNA complexes yield CD spectra which differ significantly (enhanced, blue-shifted, 273-nm band) from that of native DNA (Shih and Fasman, 1971). The small C-terminal fragment (85-102) was bound weakly to DNA under the conditions used. However, the large basic N-terminal fragment (1-83) was bound as well to DNA as was whole f2al, but produced no CD distortion. The conformation of the N-terminal fragment, unlike intact f2al, was not changed upon increasing the ionic strength to 0.14 M NaF. These results complement previous studies on f2al and its N-terminal CNBr fragment (Ziccardi and Schumaker, 1973). Thermal denaturation of the complexes in 2.5 X 10(-4) M EDTA was monitored simultaneously by changes in the absorption and CD spectra. All complexes showed a thermal transition at 45 degrees (Tml), attributable to the melting of free, double-stranded DNA. In addition, f2al-DNA and N fragment-DNA complexes displayed melting phenomena at 88 and 78 degrees (Tm2), respectively, caused by the denaturation of the histone-bound DNA. This difference in Tm2 constitutes further evidence that loss of the 18-amino-acid carboxyl end segment of f2al prohibits the unique type of interaction which occurs between DNA and the intact histone.

Studies on interaction between histone V (f2c) and deoxyribonucleic acids

Histone V (2fc) from chick erythroctes was used in the study of its interaction with DNA from various sources. Complexes between this histone and DNA were formed using the procedure of continuous NaCl gradient dialysis in urea. Two physical methods, namely thermal denaturation and circular dichroism (CD), were used as analytical tools. Thermal denaturation of nucleohistone V with chick or calf thymus DNA shows three melting bands: band I at 45-50 degrees corresponds to free base pairs; band II at 75-79 degrees, and band III at 90-93 degrees correspond to histone-bound base pairs. In histone-bound regions, there are 1.5 amino acid residues/nucleotide in nucleohistone V. In contrast, a value between 2.9 and 3.3 was determined for nucleohistone I (fl) (H. J. Li (1973), Biopolymers 12, 287). Similar melting properties have been observed for histone V complexed with bacterial DNA from Micrococcus luteus. Histone V binding to DNA induces a slight transition from a B-type CD spectrum to a C-type spectrum. Trypsin treatment of nucleohistone V reduces melting band III much more effectively than band II. Such a treatment also restores DNA to B conformation in the free state. Reduction of the melting bands of nucleohistone V by polylysine binding follows the order of I greater than II greater than III, accompanied by the increase of a new band at 100 degrees. When two bacterial DNAs of varied A + T (adenine + thymine) content simultaneously compete for the binding of histone V, the more (A " T)-rich DNA is selectively favored. Under experimental conditions described here, Clostridium perfringens DNA with 69% A + T is bound by histone V in preference to chicken DNA with 56% A + T although the latter has natural sequences for histone V binding.