Logo of prosciprotein sciencecshl presssubscriptionsetoc alertsthe protein societyjournal home
Protein Sci. 2005 Feb; 14(2): 514–522.
PMCID: PMC2253418

The H2A.Z/H2B dimer is unstable compared to the dimer containing the major H2A isoform


The nucleosome, the basic fundamental repeating unit of chromatin, contains two H2A/H2B dimers and an H3/H4 tetramer. Modulation of the structure and dynamics of the nucleosome is an important regulation mechanism of DNA-based chemistries in the eukaryotic cell, such as transcription and replication. One means of altering the properties of the nucleosome is by incorporation of histone variants. To provide insights into how histone variants may impact the thermodynamics of the nucleosome, the stability of the heterodimer between the H2A.Z variant and H2B was determined by urea-induced denaturation, monitored by far-UV circular dichroism, intrinsic Tyr fluorescence intensity, and anisotropy. In the absence of stabilizing agents, the H2A.Z/H2B dimer is only partially folded. The stabilizing cosolute, trimethylamine-N-oxide (TMAO) was used to promote folding of the unstable heterodimer. The equilibrium stability of the H2A.Z/H2B dimer is compared to that of the H2A/H2B dimer. The equilibrium folding of both histone dimers is highly reversible and best described by a two-state model, with no detectable equilibrium intermediates populated. The free energies of unfolding, in the absence of denaturant, of H2A.Z/H2B and H2A/H2B are 7.3 kcal mol−1 and 15.5 kcal mol−1, respectively, in 1 M TMAO. The H2A.Z/H2B dimer is the least stable histone fold characterized to date, while H2A/H2B appears to be the most stable. It is speculated that this difference in stability may contribute to the different biophysical properties of nucleosomes containing the major H2A and the H2A.Z variant.

Keywords: thermodynamics, chemical denaturation, circular dichroism, fluorescence, histone variant

Within the nucleus of eukaryotic cells, DNA is compacted by the nucleoprotein complex chromatin. The fundamental repeating unit of chromatin is the nucleosome core particle (NCP). The NCP consists of two copies each of the histone proteins, H2A, H2B, H3, and H4, and ~150 bp of DNA, which is wrapped 1.65 times around the central protein octamer. Each of these proteins fold into a structural motif termed the “histone fold,” which contains a long central α helix flanked on each end by a β loop and shorter α helix (Arents and Moudrianakis 1993; Luger et al. 1997a). H2A and H2B heterodimerize in a head-to-tail manner called the “handshake motif” (Fig. 1A). H3 and H4 heterodimerize in a similar structure; two H3/H4 dimers then form the H3/H4 tetramer via a four-helix bundle of the C termini of two H3 proteins.

Figure 1.
(A) Ribbon diagram of the H2A/H2B dimer, derived from the X-ray crystal structure of the core nucleosome (Luger et al. 1997a). H2A is shown in the lighter gray, and H2B is shown in black. The three helices of the canonical histone fold are shown as cylinders. ...

The compaction of DNA afforded by the NCP also restricts the access of cellular machinery to the DNA, and nucleosomal structures are often repressive to the DNA-based chemistries of the cell. During processes such as replication, transcription, repair, and recombination, the structure of chromatin must be altered to provide access to the DNA. Cells have developed several ways in which to modulate chromatin structure and NCP dynamics: (1) posttranslational modifications, (2) ATP-dependent chromatin remodeling complexes, and (3) incorporation of histone variants. Posttranslational modifications, primarily of the histone-tail domains, have been extensively studied (for review, see Strahl and Allis 2000 and Jenuwein and Allis 2001). Modifications such as acetylation, methylation, and phosphorylation alter the properties of chromatin and influence accessibility of the DNA as well as act as signals for the recruitment of nuclear factors. Secondly, ATP-dependent chromatin remodeling complexes utilize the energy from ATP hydrolysis to rearrange both histone–DNA and histone–histone interactions (for review, see Lusser and Kadonaga 2003). Generally, the function of these complexes is to increase NCP mobility, and remodeling may enable the translocation of the histone octamer along the DNA. Finally, the proteins constituting the NCP can be varied by the incorporation of alternate histones.

The structural and functional properties of NCPs can be altered by incorporation of variant histones with biochemical properties that are different from those of the major histone isoform. Variants have been identified for each of the four core histones, but the H2A family contains the most members. The major site of sequence divergence between the H2A family members is in the C-terminal regions of the proteins (Ausio and Abbott 2002). The various H2A isoforms appear to be involved in specific cellular functions. H2A.X, for example, is involved in the response to DNA double-strand breaks, and phosphorylation of a serine residue in the C-terminal tail of the protein is one of the earliest events after induction of DNA double-strand breaks (Rogakou et al. 1998). MacroH2A is thought to play a pivotal role in the inactivation of one copy of the X chromosome, and contains a large non-histone fold C-terminal domain (Costanzi and Pehrson 1998).

This report focuses on the variant H2A.Z, an isoform found in most eukaryotes. It is required for survival in Drosophila melanogaster (van Daal and Elgin 1992), Tetrahymena thermophila (Liu et al. 1996), and mice (Faast et al. 2001). H2A.Z is highly conserved across species, ~90%, but shares only ~60% sequence identity with the major H2A protein (Iouzalen et al. 1996). In D. melanogaster, the region of H2A.Z essential for survival is in the C terminus (Clarkson et al. 1999). Initially, H2A.Z was thought to be involved in transcriptional activation, because in T. thermophila, H2A.Z is found only in the transcriptionally active macronucleus and not in the inactive micronucleus (Stargell et al. 1993). In Saccharomyces cerevisiae, H2A.Z is required for normal induction of the GAL1 and PHO5 genes; the requirement for H2A.Z is enhanced in strains deficient for chromatin remodeling complexes (Santisteban et al. 2000). More recent studies have implicated H2A.Z in the maintenance of pericentrimeric heterochromatin and telomeric heterochromatin (Meneghini et al. 2003; Rangasamy et al. 2003).

The X-ray crystal structure of an NCP containing H2A.Z (Suto et al. 2000) reveals an overall structure very similar to that of NCPs containing the major H2A (RMSD <1 Å) (Luger et al. 1997a). The subtle structural differences between two NCPs are in the interface between the H2A/H2B dimer and H3/H4 tetramer and a metal ion binding site in the H2A.Z NCP. These differences appear to have effects within an individual NCP and between nucleosomes in higher-order chromatin structures. A recent report showed that H2A.Z NCPs are more stable than H2A NCPs (Park et al. 2004). Additionally, H2A.Z incorporation facilitates the folding of individual nucleosomal arrays while inhibiting the formation of higher-order interactions between the arrays (Fan et al. 2002).

The properties of the histone oligomers are key determinants of the dynamic properties of the nucleosome which regulate DNA accessibility. To provide biophysical insights into NCP dynamics as influenced by the incorporation of histone variants, we compared the stability of H2B dimers with the major H2A isoform and the H2A.Z variant. Previously, we showed that the H2A/H2B dimer is a moderately stable protein (Gloss and Placek 2002). In this report, we demonstrate that, in contrast, the H2A.Z/H2B dimer is quite unstable.


The histones used in these studies were purified from a recombinant E. coli expression system (see Materials and Methods). Unlike histones purified from eukaryotic sources, the recombinant histones were homogenous in lacking post-translational modifications, as shown by mass spectrometry. The X. laevis genes for the major H2A and H2B isoforms were used for recombinant expression (Luger et al. 1997b); however, the overexpression plasmid for H2A.Z contains the gene from mouse (Suto et al. 2000). This H2A.Z was reconstituted with X. laevis H2B by the methods used for the major H2A (Placek and Gloss 2002). The heterodimer of mouse H2A.Z/Xenopus H2B is the construct that has been used in previous biophysical studies of the effect of H2A.Z on the properties of the nucleosome, including X-ray crystallography (Suto et al. 2000), analytical ultracentrifugation (Fan et al. 2002), and equilibrium assembly and dissociation (Park et al. 2004). Although the histone monomers in this heterodimer are from different organisms, this is not anticipated to have any significant effect on the folding properties of the complex. Mouse and Xenopus H2A.Z are 90% identical overall (Fig. 1B). In the globular, helical, histone-fold domain of ~70 residues, there are six amino acid differences. Five of the substitutions are conservative: two Ser/Thr substitutions; Ser replaced with either Ala or Asn; and an Ala/Gly exchange. The sixth difference is very solvent-accessible, exchanging the Xenopus Leu for a His in mouse. Only one of the sequence differences, a Ser/Thr substitution in the long α2 helix, is significantly excluded from solvent and makes intermonomer contacts.

The stabilities of the variant H2A/H2B dimers were determined from urea-induced unfolding transitions. The extent of native structure as a function of [Urea] was monitored by far-UV circular dichroism (CD) at 222 nm and intrinsic Tyr fluorescence (FL) at 305 nm. Far-UV CD provides a global measure of the secondary structure of the protein. Tyr FL provides information on the tertiary and quaternary structure of the heterodimers. The major X. laevis H2A contains three tyrosine residues at positions 39, 50, and 57. These Tyr residues are largely excluded from solvent in the folded dimer by both intra- and inter-monomer contacts. The most solvent-accessible Tyr at position 39 is not conserved in H2A.Z sequences, and is a Ser in the Xenopus and mouse proteins. The X. laevis H2B monomer contains five Tyr, three of which are also largely excluded from solvent by tertiary and quaternary structure.

Cosolute stabilization of H2A.Z/H2B

In the absence of denaturant, the mean residue ellipticity at 222 nm of the H2A.Z/H2B dimer is 50% less than that of the dimer containing the major H2A isoform (Fig. 2A). Given the very similar structures of the two heterodimers in the context of the nucleosome (Suto et al. 2000), the diminished CD signal of the H2A.Z/H2B dimer suggested that this heterodimer is less well folded than the dimer comprised of the two major isoforms. Comparison of HPLC size-exclusion chromatograms of H2A/H2B and H2A.Z/H2B dimers showed that the H2A.Z-containing species was a mixture of dimeric and monomeric species (Fig. 2B). Glutaraldehyde cross-linking, performed as described by Banks and Gloss (2003), confirmed that the dimer was a heterotypic association of H2A.Z and H2B. Furthermore, the urea-induced unfolding transitions of the two heterodimers confirmed that the H2A.Z/H2B dimer was only partially folded. Much lower urea concentrations were required to unfold the H2A.Z/H2B heterodimer (data not shown). The unfolding transitions exhibited high reversibility, 90%–95%. However, the H2A.Z/H2B unfolding transitions could not be fit to an equilibrium model to determine a free energy of unfolding in the absence of denaturant, ΔG° (H2O), because of a lack of a native baseline to determine the spectral properties of the fully folded native dimer. Therefore, cosolutes were used to stabilize the H2A.Z/H2B dimer.

Figure 2.
(A) TMAO dependence of the Tyr FL anisotropy (•) and far-UV CD at 222 nm (▪) of 5 μM H2A.Z/H2B dimer. The lines are drawn to guide the eye and do not represent fits of the data. (B) HPLC size-exclusion chromatography of the H2A/H2B ...

Initially, high salt concentrations, 0.3 M KPi or 1 M KCl, were employed to stabilize the H2A.Z/H2B dimer; these conditions had been shown to greatly stabilize the H2A/H2B dimer (Gloss and Placek 2002). While the addition of salts did stabilize H2A.Z/H2B, as evidenced by requiring higher urea concentrations to unfold the dimer, these conditions did not sufficiently stabilize the dimer such that a native baseline could be discerned. The high-salt conditions also had the additional complication of reducing the reversibility of the unfolding reaction, probably because of aggregation as detected in HPLC size-exclusion chromatograms.

Trimethylamine-N-oxide (TMAO) has been shown to be a potent cosolute for stabilizing partially folded proteins, and extending the native baseline region (Henkels et al. 2001; Mello and Barrick 2003). TMAO is an organic osmolyte that stabilizes proteins through the osmophobic effect because of unfavorable interactions with the peptide backbone (Wang and Bolen 1997; Bolen 2001) and preferential hydration of the native species (Timasheff 1998). The FL and far-UV CD signals of the H2A.Z/H2B dimer increased with increasing TMAO concentrations, reaching a plateau above 3 M TMAO (Fig. 2A). These data suggest that 3–4 M TMAO is required to induce complete folding of the H2A.Z/H2B dimer. HPLC size-exclusion chromatography of the dimer in 1 M TMAO confirmed that the population was mostly dimeric, with no detectable aggregate (Fig. 2B). TMAO does not significantly affect the CD and FL properties of H2A/H2B. However, it is not practical to study the urea-induced unfolding of the histones in such high concentrations of TMAO, because it is not possible to make urea solutions of high enough concentration, in the presence of 3 M TMAO, to completely unfold the histone dimers. Therefore, further studies were performed in 1 M TMAO; the various methods used to determine the native baseline are described below.

Equilibrium stability of H2A.Z/H2B

A data set of 10 urea equilibrium unfolding transitions in 1 M TMAO was collected over a range of dimer concentrations from 2 to 17.3 μM. The data set included unfolding transitions monitored by both far-UV CD and Tyr FL; FL data included both FL intensity and FL anisotropy as a function of [Urea]. Representative transitions are shown in Figure 3, A and B, for far-UV CD and FL intensity, respectively. As expected for a dimeric protein, the unfolding transition is protein concentration-dependent; as the dimer concentration increases, the transition between folded and unfolded species shifts to higher urea concentrations. At the lowest dimer concentrations, no native baseline is observed. The data were globally fitted to a two-state equilibrium model for the unfolding of a dimer to two unfolded monomers, with no populated intermediates (equations 1, 2). In all global fits, the ΔG° (H2O) and m values were treated as global parameters, linked across the entire data set. The well-defined unfolded baselines were treated as local fitted parameters, and not linked between the various titrations in the data set.

Figure 3.
Equilibrium urea unfolding transitions of H2A.Z/H2B in 1 M TMAO. (A) Unfolding monitored by CD at 222 nm: 2 μM dimer, •; 5 μM dimer, ♦; 12 μM dimer, ▪. (B) Unfolding monitored by intrinsic Tyr FL at 305 ...

In the global fitting of the data set, several methods were used to constrain the poorly defined native baselines:

  1. The intercepts of the native baselines (the signal at 0 M urea) were fixed to the values of the spectral signals observed at 3 to 4 M TMAO. The slopes of the native baselines were treated as fitted parameters, either globally linked between titrations monitoring the same spectral signal or treated as local parameters.
  2. The far-UV CD data were normalized to mean residue ellipticity, and the parameters for the native baselines were linked between all of the CD titrations.
  3. The slopes of all native baselines were fixed at 0 for all titrations, and the intercept was treated as a local parameter or linked between titrations monitoring the same spectral signal.
  4. The data set was fit with the slopes and intercepts of the native baselines treated as local adjustable parameters, without any additional constraints to define the native baseline.

The latter method, having the largest number of fitted parameters, and thus the most degrees of freedom, gave the best fit of the data, as assessed by reduced Chi-squared values and the agreement between the fit and the data points. The ΔG° (H2O) and m values for this fit are given in Table 11,, and the results are shown as the solid lines in Figures 3 and 4. The fitted values of ΔG° (H2O) from the various methods of constraining the native baseline ranged from 7.3 to 7.8 kcal mol−1; the m values ranged from 1.0 to 1.3 kcal mol−1M−1. These ranges are within the errors reported in Table 11 for the least constrained fit. Therefore, the ΔG° (H2O) and m values reported in Table 11 for the H2A.Z/H2B dimer are reasonably accurate, despite the poor definition of the native baseline in 1 M TMAO.

Table 1.
Parameters describing the equilibrium stability of the H2A/H2B and H2A.Z/H2B
Figure 4.
Representative Fapp curves for the equilibrium urea unfolding of the H2A/H2B heterodimer variants in 1 M TMAO. CD data were collected at 222 nm, and FL data at 305 nm. (A) H2A.Z/H2B unfolding transitions. 5 μM dimer: CD, ♦; FL intensity, ...

The data from the different spectral probes can be compared by converting the spectral data to Fapp transitions, the apparent fraction of unfolded monomers as a function of [Urea] (equation 1). The far-UV CD and Tyr FL Fapp curves are coincident (Fig. 4A), demonstrating a concerted loss of secondary, tertiary, and quaternary structure. This coincidence suggests that the H2A.Z/H2B dimer unfolds by a two-state equilibrium mechanism. Further support is provided by the agreement between the data and the global fit as a function of dimer concentration (Figs. 3, 4).

Equilibrium stability of H2A/H2B in 1 M TMAO

To compare the stability of heterodimers containing the H2A.Z and major H2A isoforms under similar solvent conditions, the stability of the H2A/H2B was determined from equilibrium urea unfolding studies in 1 M TMAO. A data set of 11 unfolding transitions at several dimer concentrations was collected, spanning a dimer concentration range of 1 to 8 μM; the data set contained transitions monitored by both far-UV CD and intrinsic Tyr FL data. The presence of TMAO did not decrease the high reversibility, ≥95%, of the H2A/H2B unfolding reported previously in the absence of TMAO (Gloss and Placek 2002). The data set was globally fit to a two-state equilibrium mechanism, which has been shown previously to adequately describe the equilibrium unfolding in the absence of TMAO (Gloss and Placek 2002). The ΔG° (H2O) and m values were treated as global parameters (Table 11),), linked across the entire data set, while the native and unfolded baselines were treated as local parameters. The coincidence of unfolding transitions monitored by CD and FL data and the agreement between the data and the global fit at various dimer concentrations (Fig. 4B) demonstrate that the unfolding of the H2A/H2B dimer is a two-state process in the presence of 1 M TMAO.

Equilibrium stability of H2A-ΔC/H2B

The region of greatest diversity in sequence, and apparently function, between the variants in the H2A family is in the extended C-terminal tail region of the protein. To assess the contribution of this region to the stability of the H2A/H2B dimer, a C-terminal truncated mutant of the major H2A isoform was generated. The H2A-ΔC mutation terminated the polypeptide after residue 99 by insertion of a stop codon. The C-terminal residues 100–129 of H2A have 78% similarity (54% identity) to those in the slightly shorter tail of H2A.Z (Fig. 1B). The truncation site was chosen to remove all of the C-terminal residues which were in an extended conformation or unresolved in the X-ray crystal structure of the NCP (Luger et al. 1997a). The urea-induced unfolding of the H2A-ΔC/H2B dimer is ≤95% reversible. A data set of nine equilibrium transitions were collected, spanning a range of dimer concentrations from 1 to 20 μM dimer, with unfolding monitored by both CD at 222 nm and FL intensity at 305 nm. The data set was globally fitted to a two-state equilibrium model (equations 1, 2). Representative data and fits are shown in Figure 5, and the globally fitted ΔG° (H2O) and m values are given in Table 11.

Figure 5.
Representative Fapp curves for the equilibrium urea unfolding of the H2A-ΔC/H2B heterodimer. CD data were collected at 222 nm; FL intensity data, at 305 nm. 1 μM dimer: CD, ▴; FL, ▾. 2 μM dimer: CD, ▵; FL, ...


Stability differences between histone dimers

Previous chemical denaturant studies have shown that the H2A/H2B dimer is a relatively stable dimer for its size (Gloss and Placek 2002). For the histones, it is difficult to compare stabilities, as quantified by the ΔG° (H2O) values, between studies using urea and guanidinium chloride (GdmCl); as shown in Table 11,, H2A/H2B exhibits higher stability to GdmCl denaturation than to urea denaturation (Gloss and Placek 2002; Banks and Gloss 2003), apparently because of the ionic nature of the GdmCl salt and the highly basic histone sequence. The stability to GdmCl denaturation of other histone fold-containing dimers has been reported and can be directly compared to that reported for H2A/H2B in GdmCl. Of the eukaryotic NCP heterodimers, the H3/H4 dimer exhibits only 60% of the stability of H2A/H2B (Banks and Gloss 2003). Surprisingly, H2A/H2B exhibits greater stability than even the histone homodimers of thermophilic and hyperthermophilic archae, hMfB (20% greater) and hPyA1 (5% greater) (Topping and Gloss 2004).

In contrast, the H2A.Z/H2B dimer is strikingly unstable, and is the least stable of the histone folds characterized to date. This instability is apparent without quantitative analyses of the chemical denaturation data. The mesophilic archael homodimeric archael histone, hFoB, and the H3-H4 dimer require cosolutes such as TMAO to promote folding to the native state (Banks and Gloss 2003; Topping and Gloss 2004). However, for both of these histone dimers, 1 M TMAO is sufficient to induce complete folding, and yield a native baseline in the presence of denaturants. However, concentrations of TMAO greater than 1 M are required to promote complete folding of the H2A.Z/H2B dimer (Fig. 2A). Fits of the H2A.Z/H2B data indicate that in 1 M TMAO, ≥20% of the monomers are unfolded at concentrations less than 20 μM (Figs. 3, 4A), and the dimer exhibits only 47% of the stability to urea denaturation of the H2A/H2B dimer (Table 11).). In contrast, in 1 M TMAO, 100% of the H3/H4 oligomers are folded, and the dimers require >0.5 M GdmCl to induce unfolding (Banks and Gloss 2003).

Fitting of chemical denaturation data provides the stability, quantified by the ΔG° (H2O) parameter, and also the m value, which describes the steepness of the unfolding transition, i.e., the sensitivity to the denaturant concentration (equation 2). For the relatively stable dimers, H2A/H2B, and the archael histones, hMfB and hPyA1, the m value correlates quite well with that expected from the change in solvent-accessible surface area between the unfolded monomers and native dimer (Myers et al. 1995). However, for the unstable H2A.Z/H2B and H3/H4 dimers, and to a lesser extent, the archael hFoB homodimer, the m values are significantly less than expected for proteins of their size (Banks and Gloss 2003; Topping and Gloss 2004). This suggests that a low m value may be a common feature of poorly folded histones, particularly in TMAO. Even the m value of H2A/H2B is affected by TMAO, decreasing from 6.4 to 5.4 kcal mol−1M−1 in 0 and 1 M TMAO, respectively (Banks and Gloss 2003). Lower than expected m values can also be an indication of the presence of an equilibrium intermediate, for example (Spudich and Marqusee 2000). However, the folding of H2A.Z/H2B appears to be a two-state process based on (1) the coincidence of transitions monitoring multiple spectroscopic probes (Fig. 4A), and (2) the quality of the global fits to a two-state model as a function of protein concentration (Figs. 3, 4A). The poorly defined native baseline of H2A.Z/H2B makes it difficult to differentiate between the quality of fits to a two-state model and a model with additional states, and thus additional fitting parameters.

The regions required for the function of H2A.Z are the C-terminal residues 92–102, the C-terminal α-helix beyond the canonical histone fold, and residues 106–119, the distal end of the extended C-terminal tail (Fig. 1; Clarkson et al. 1999). To ascertain the effect of this region on the stability of H2A/H2B dimer, the C-terminal residues 100–129 were truncated in the mutant H2A-ΔC/H2B. The presence of the C-terminal tail does confer some additional stability to the H2A/H2B dimer, 1.6 kcal mol−1 (Table 11;; Fig. 5). The destabilization resulting from removal of the C-terminal tail is greater than that observed for removing the 15 amino acids of the H2A N-terminal tail, 0.9 kcal mol−1 (Placek and Gloss 2002). However, the truncation of the functionally important C-terminal tail is not nearly as destabilizing as the difference in stability between dimers containing H2A and H2A.Z, 8.2 kcal mol−1 (Table 11).). Therefore, the sequence differences that encode the stability differences between the two histone heterodimers are most likely to be within the globular, helical histone fold region. In support of this, a recent report has demonstrated that removal of the C-terminal tail of H2A does not significantly affect the assembly and apparent stability of nucleosome core particles (Bao et al. 2004).

Sequence differences that may impact the relative stability of H2A/H2B and H2A.Z/H2B dimers

The relative instability of H2A.Z/H2B occurs despite ~70% identity in the helical, globular regions of the two H2A sequences (between residues 16 and 97) (Fig. 1B) and very similar structures, in the context of the nucleosome. The backbone structures of these regions of H2A.Z/H2B and H2A/H2B dimers are superimposable, except for the loop between α1 and α2 in H2A, which is one residue longer in H2A.Z. While there is sequence divergence in the extended N- and C-terminal tails—regions that can modulate dimer stability, the histone fold domain is the major determinant of dimer stability. There are three clusters of sequence differences between H2A and H2A.Z that may, individually or in combination, encode the stability difference between the heterodimers formed by these proteins: (1) the loop region between α1 and α2 of the histone fold (X.l.A residues 37–41, also known as L1); (2) the N terminus of α2, particularly the Pro to Ala substitution at X.l.A residue 48 and the completely buried Leu to Ser or Thr change at X.l.A. residue 51; and (3) the C terminus of α2 and the loop connecting to α3 (X.l.A residues 69–78). The α1 to α2 loop is one residue shorter in the major H2A structure, and also contains a Tyr that is absent in H2A.Z; one face of Tyr residue is packed against the protein surface through hydrophobic interactions. It has been suggested that the L1:L1 interactions between H2A.Z monomers in the nucleosome may be responsible for the greater stability of H2A.Z-containing NCPs (Suto et al. 2000; Park et al. 2004). The substitution of a Leu for an Asn (C terminus of α2, X.l.A. residue 73) removes the hydrogen bond C-capping interaction of the Asn. The clusters of mutations in the helix termini and loop regions connecting the shorter helices to the long central helix of the histone fold may serve to alter the packing and interactions between the three helices of the histone fold, as well as how the H2A and H2A.Z monomers interact with the H2B monomer. Further mutational studies will be required to identify the exact determinants of the stability difference between H2A- and H2A.Z-containing heterodimers.

Implications for nucleosome function

The nucleosome core particle is a highly dynamic structure that must be restructured to allow access to the DNA during cellular processes such as replication, transcription, and repair. The incorporation of histone variants is one means to alter the dynamics of the NCP and chromatin structure. Histone variants introduce new recognition sites for the recruitment of other nuclear proteins that facilitate the downstream actions. Additionally, the biophysical properties of the histone variant may directly alter the stability and dynamics of the NCP. A recent report has demonstrated that incorporation of H2A.Z/H2B dimer does affect NCP stability (Park et al. 2004). NCPs containing H2A.Z/H2B dimers are more stable to the NaCl-induced dissociation of the H2A/H2B dimers than H2A-containing NCPs. The instability of the H2A.Z/H2B dimer reported in this paper provides new insights into NCP stability.

The association of the H2A/H2B dimers with the NCP has been shown to be a dynamic process. Several studies have demonstrated that the H2A/H2B dimer can readily exchange between nucleosomes both in vivo and in vitro (Louters and Chalkley 1984, 1985; Kimura and Cook 2001). Furthermore, transcriptionally active chromatin is often depleted in H2A/H2B dimers (Louters and Chalkley 1985; Jackson 1990), and it has been shown that RNA pol II can displace H2A/H2B dimers during transcription (Kireeva et al. 2002). Therefore, there is an important equilibrium between the mature, fully assembled NCP and partially unfolded NCPs in which the H2A/H2B dimers are less tightly bound or completely dissociated. Histone variants or histone posttranslational modifications can impact this equilibrium by altering the free energy of the mature NCP or by altering the free energy of the dissociated dimer (Gloss and Placek 2002). Therefore, the stability of the free dimer will impact this equilibrium, and the state of nucleosome assembly may be intimately linked to histone stability. A more stable H2A/H2B dimer should favor a more unfolded, dissociated state of the NCP. We have previously shown that the highly basic N-terminal tails destabilize the H2A/H2B dimer by electrostatic repulsion (Gloss and Placek 2002; Placek and Gloss 2002). The charge state of the tails is modulated by posttranslational modifications such as Ser phosphorylation and Lys acetylation, which should stabilize the free H2A/H2B dimer, and hyperacetylation correlates with transcriptional activity and depletion of the H2A/H2B dimer. In this report, we have shown that the free H2A.Z/H2B dimer is unstable, and this instability correlates with a shift of the equilibrium toward a more stable, assembled NCP, making the dimers less likely to dissociate under conditions that lead to dissociation of the H2A/H2B dimer (Park et al. 2004).

Materials and methods


Ultrapure urea was purchased from ICN Biomedicals. TMAO (Sigma) was dissolved in H2O and deionized in batch mode with AG 11A8 mixed-bed resin (BioRad). The TMAO concentration was determined from its refractive index as described (Bolen 2001).

Production of recombinant histones

Recombinant H2A, H2A-ΔC, and H2B were overexpressed and purified as described previously (Gloss and Placek 2002). The C-terminal truncation of H2A was generated by introducing a stop codon after the codon for Arg99, using extralong PCR as described previously (Placek and Gloss 2002). H2A.Z was overexpressed and purified by the same methods as the other histones, using a pET vector containing the mouse H2A.Z gene under the control of a T7 promoter (Suto et al. 2000). The H2A.Z and H2A-ΔC monomers were reconstituted with the H2B monomer to form the native heterodimer and further purified as described previously (Placek and Gloss 2002). The native molecular weights of the histone oligomers were determined by HPLC-size-exclusion chromatography. The proteins were chromatographed at room temperature on a Phenomenex BioSep-SEC-S 3000 column. Elution of the H2A-H2B dimers was compared to that of protein standards over the same HPLC column: bovine serum albumin, ovalbumin, chymotrypsinogen, and ribonuclease A.

Urea equilibrium unfolding titrations

The standard buffer conditions were 20 mM potassium phosphate, 0.2 M KCl, 0.1 mM EDTA (pH 7.2), 25°C. CD and FL data were collected on an AVIV 202SF spectrophotometer and an AVIV Model ATF-105/305 differential/ratio spectrofluorometer, respectively. Both instruments were interfaced with Hamilton Model 500 automated titrators. Automated CD unfolding titrations were monitored at 222 nm in a 1-cm cell. At the highest H2A.Z monomer concentrations, sample absorbance was too high to use the 1-cm cell necessary for the automated titrations. Therefore, the unfolding transitions were monitored by preparing individual samples at each [Urea] and collecting wavelength spectra between 260 and 210 nm in a 0.2-cm cell. The spectra were analyzed by singular value decomposition (SVD) (Henry and Hofrichter 1992; Ionescu et al. 2000). SVD analyses have two advantages: (1) better signal-to-noise relative to data collected at a single wavelength, and (2) the ability to analyze data at multiple wavelengths to either enhance the detection of or confirm the absence of equilibrium intermediates. Automated FL titration data were collected using an excitation wavelength of 280 nm and emission wavelength of 305 nm, with 2-nm bandwidths. The equilibration time for each titration was 3–5 min; this time interval was significantly longer than the time required to complete the slowest kinetic step in the folding and unfolding reactions of both histone dimers. The reversibility of the urea-induced unfolding reactions of the heterodimers was quantified by comparing the CD and FL spectra of matched, individual samples prepared from folded protein stocks and urea-unfolded stocks; samples were prepared with final urea concentrations that spanned the unfolding transitions to demonstrate a lack of hysteresis in the folding/unfolding transitions.

Data analysis

The equilibrium unfolding titrations were fitted to a two-state model for a dimeric system (described in detail by Gittelman and Matthews 1990). For a two-state dimeric system, Fapp (the apparent fraction unfolded monomer) is related to the equilibrium constant, Keq, and total monomer concentration, PT, as well as the observed spectral properties, by the following equation:

equation M1

where Yi is the CD signal measured at [Urea], and YN and YU are the spectral properties of the folded and unfolded species. A linear extrapolation between the free energy of unfolding, ΔG°, and the urea concentration was used (Pace 1986):

equation M2

where ΔG° (H2O) is the free energy of unfolding in the absence of denaturant at a standard state of 1 M dimer, and the m value reflects the sensitivity of the transition to urea concentration. Data collected with different probes and at varied monomer concentrations were fitted globally with the program Savuka 5.1 (Zitzewitz et al. 1995; Gualfetti et al. 1999). In global fits, the ΔG° (H2O) and m values were linked across the multiple data sets; the native and unfolded baselines were treated as either local parameters or linked across selected data sets.


This work was supported by NSF grant MCB-9983831 to L.M.G. B.J.P. was partially supported by NIH Biotechnology training grant GM08336-13. The pET overexpression vectors for the H2A.Z histone were kindly provided by Karolin Luger (Colorado State University) and David J. Tremethick (Australian National University).


  • CD, circular dichroism
  • FL, fluorescence
  • Fapp, apparent fraction of unfolded monomer
  • ΔG° (H2O), the free energy of unfolding in the absence of denaturant
  • GdmCl, guanidinium chloride
  • H2A, major Xenopus laevis isoform
  • H2A.Z, H2A variant encoded by the mouse gene
  • H2A-ΔC, C-terminal tail truncation, after Arg99, of the Xenopus laevis major H2A isoform
  • KPi, potassium phosphate m-value, parameter describing the sensitivity of the unfolding transition to [Urea]
  • MRE, mean residue ellipticity
  • NCP, nucleosome core particle
  • SVD, singular value decomposition
  • TMAO, trimethylamine-N-oxide


Article published online ahead of print. Article and publication date are at http://www.proteinscience.org/cgi/doi/10.1110/ps.041026405.


  • Arents, G. and Moudrianakis, E.N. 1993. Topography of the histone octamer surface: Repeating structural motifs utilized in the docking of nucleosomal DNA. Proc. Natl. Acad. Sci. 90 10489–10493. [PMC free article] [PubMed]
  • Ausio, J. and Abbott, D.W. 2002. The many tales of a tail: Carboxyl-terminal tail heterogeneity specializes histone H2A variants for defined chromatin function. Biochemistry 41 5945–5949. [PubMed]
  • Banks, D.D. and Gloss, L.M. 2003. Equilibrium folding of the core histones: The H3-H4 tetramer is less stable than the H2A-H2B dimer. Biochemistry 42 6827–6839. [PubMed]
  • Bao, Y., Konesky, K., Park, Y.J., Rosu, S., Dyer, P.N., Rangasamy, D., Tremethick, D.J., Laybourn, P.J., and Luger, K. 2004. Nucleosomes containing the histone variant H2A.Bbd organize only 118 base pairs of DNA. EMBO J. 23 3314–3324. [PMC free article] [PubMed]
  • Beechem, J.M. 1992. Global analysis of biochemical and biophysical data. Methods Enzymol. 210 37–54. [PubMed]
  • Bolen, D.W. 2001. Protein stabilization by naturally occurring osmolytes. Methods Mol. Biol. 168 17–36. [PubMed]
  • Clarkson, M.J., Wells, J.R., Gibson, F., Saint, R., and Tremethick, D.J. 1999. Regions of variant histone His2AvD required for Drosophila development. Nature 399 694–697. [PubMed]
  • Costanzi, C. and Pehrson, J.R. 1998. Histone macroH2A1 is concentrated in the inactive X chromosome of female mammals. Nature 393 599–601. [PubMed]
  • Faast, R., Thonglairoam, V., Schulz, T.C., Beall, J., Wells, J.R., Taylor, H., Matthaei, K., Rathjen, P.D., Tremethick, D.J., and Lyons, I. 2001. Histone variant H2A.Z is required for early mammalian development. Curr. Biol. 11 1183–1187. [PubMed]
  • Fan, J.Y., Gordon, F., Luger, K., Hansen, J.C., and Tremethick, D.J. 2002. The essential histone variant H2A.Z regulates the equilibrium between different chromatin conformational states. Nat. Struct. Biol. 9 172–176. [PubMed]
  • Gittelman, M.S. and Matthews, C.R. 1990. Folding and stability of trp aporepressor from Escherichia coli. Biochemistry 29 7011–7020. [PubMed]
  • Gloss, L.M. and Placek, B.J. 2002. The effect of salts on the stability of the H2A-H2B histone dimer. Biochemistry 41 14951–14959. [PubMed]
  • Gualfetti, P.J., Bilsel, O., and Matthews, C.R. 1999. The progressive development of structure and stability during the equilibrium folding of the α subunit of tryptophan synthase from Escherichia coli. Protein Sci. 8 1623–1635. [PMC free article] [PubMed]
  • Henkels, C.H., Kurz, J.C., Fierke, C.A., and Oas, T.G. 2001. Linked folding and anion binding of the Bacillus subtilis ribonuclease P protein. Biochemistry 40 2777–2789. [PubMed]
  • Henry, E.R. and Hofrichter, J. 1992. Singular value decomposition: Application of analysis of experimental data. Methods Enzymol. 210 129–192.
  • Ionescu, R.M., Smith, V.F., O’Neill Jr., J.C., and Matthews, C.R. 2000. Multistate equilibrium unfolding of Escherichia coli dihydrofolate reductase: Thermodynamic and spectroscopic description of the native, intermediate, and unfolded ensembles. Biochemistry 39 9540–9550. [PubMed]
  • Iouzalen, N., Moreau, J., and Mechali, M. 1996. H2A.ZI, a new variant histone expressed during Xenopus early development exhibits several distinct features from the core histone H2A. Nucleic Acids Res. 24 3947–3952. [PMC free article] [PubMed]
  • Jackson, V. 1990. In vivo studies on the dynamics of histone-DNA interaction: Evidence for nucleosome dissolution during replication and transcription and a low level of dissolution independent of both. Biochemistry 29 719–731. [PubMed]
  • Jenuwein, T. and Allis, C.D. 2001. Translating the histone code. Science 293 1074–1080. [PubMed]
  • Kimura, H. and Cook, P.R. 2001. Kinetics of core histones in living human cells: Little exchange of H3 and H4 and some rapid exchange of H2B. J. Cell Biol. 153 1341–1353. [PMC free article] [PubMed]
  • Kireeva, M.L., Walter, W., Tchernajenko, V., Bondarenko, V., Kashlev, M., and Studitsky, V.M. 2002. Nucleosome remodeling induced by RNA polymerase II. Loss of the H2A/H2B dimer during transcription. Mol. Cell 9 541–552. [PubMed]
  • Kraulis, P.J. 1991. MOLSCRIPT: A program to produce both detailed and schematic plots of protein structures. J. Applied Crystallogr. 24 946–950.
  • Liu, X., Li, B., and Gorovsky, M.A. 1996. Essential and nonessential histone H2A variants in Tetrahymena thermophila. Mol. Cell Biol. 16 4305–4311. [PMC free article] [PubMed]
  • Louters, L. and Chalkley, R. 1984. In vitro exchange of nucleosomal histones H2a and H2b. Biochemistry 23 547–552. [PubMed]
  • ———. 1985. Exchange of histones H1, H2A, and H2B in vivo. Biochemistry 24 3080–3085. [PubMed]
  • Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F., and Richmond, T.J. 1997a. Crystal structure of the nucleosome core particle at 2.8 Å resolution. Nature 389 251–260. [PubMed]
  • Luger, K., Rechsteiner, T.J., Flaus, A.J., Waye, M.M., and Richmond, T.J. 1997b. Characterization of nucleosome core particles containing histone proteins made in bacteria. J. Mol. Biol. 272 301–311. [PubMed]
  • Lusser, A. and Kadonaga, J.T. 2003. Chromatin remodeling by ATP-dependent molecular machines. Bioessays 25 1192–1200. [PubMed]
  • Mello, C.C. and Barrick, D. 2003. Measuring the stability of partly folded proteins using TMAO. Protein Sci. 12 1522–1529. [PMC free article] [PubMed]
  • Meneghini, M.D., Wu, M., and Madhani, H.D. 2003. Conserved histone variant H2A.Z protects euchromatin from the ectopic spread of silent heterochromatin. Cell 112 725–736. [PubMed]
  • Myers, J.K., Pace, C.N., and Scholtz, J.M. 1995. Denaturant m values and heat capacity changes: Relation to changes in accessible surface areas of protein folding. Protein Sci. 4 2138–2148. [PMC free article] [PubMed]
  • Pace, C.N. 1986. Determination and analysis of urea and guanidine hydrochloride denaturation curves. Methods Enzymol. 131 266–280. [PubMed]
  • Park, Y.J., Dyer, P.N., Tremethick, D.J., and Luger, K. 2004. A new fluorescence resonance energy transfer approach demonstrates that the histone variant H2AZ stabilizes the histone octamer within the nucleosome. J. Biol. Chem. 279 24274–24282. [PubMed]
  • Placek, B.J. and Gloss, L.M. 2002. The N-terminal tails of the H2A-H2B histones affect dimer structure and stability. Biochemistry 41 14960–14968. [PubMed]
  • Rangasamy, D., Berven, L., Ridgway, P., and Tremethick, D.J. 2003. Pericentric heterochromatin becomes enriched with H2A.Z during early mammalian development. EMBO J. 22 1599–1607. [PMC free article] [PubMed]
  • Rogakou, E.P., Pilch, D.R., Orr, A.H., Ivanova, V.S., and Bonner, W.M. 1998. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J. Biol. Chem. 273 5858–5868. [PubMed]
  • Santisteban, M.S., Kalashnikova, T., and Smith, M.M. 2000. Histone H2A.Z regulates transcription and is partially redundant with nucleosome remodeling complexes. Cell 103 411–422. [PubMed]
  • Spudich, G. and Marqusee, S. 2000. A change in the apparent m value reveals a populated intermediate under equilibrium conditions in Escherichia coli ribonuclease HI. Biochemistry 39 11677–11683. [PubMed]
  • Stargell, L.A., Bowen, J., Dadd, C.A., Dedon, P.C., Davis, M., Cook, R.G., Allis, C.D., and Gorovsky, M.A. 1993. Temporal and spatial association of histone H2A variant hv1 with transcriptionally competent chromatin during nuclear development in Tetrahymena thermophila. Genes & Dev. 7 2641–2651. [PubMed]
  • Strahl, B.D. and Allis, C.D. 2000. The language of covalent histone modifications. Nature 403 41–45. [PubMed]
  • Suto, R.K., Clarkson, M.J., Tremethick, D.J., and Luger, K. 2000. Crystal structure of a nucleosome core particle containing the variant histone H2A.Z. Nat. Struct. Biol. 7 1121–1124. [PubMed]
  • Timasheff, S.N. 1998. Control of protein stability and reactions by weakly interacting cosolvents: The simplicity of the complicated. Adv. Protein Chem. 51 355–432. [PubMed]
  • Topping, T.B. and Gloss, L.M. 2004. Stability and folding mechanism of mesophilic, thermophilic and hyperthermophilic archael histones: The importance of folding intermediates. J. Mol. Biol. 342 247–260. [PubMed]
  • van Daal, A. and Elgin, S.C. 1992. A histone variant, H2AvD, is essential in Drosophila melanogaster. Mol. Biol. Cell 3 593–602. [PMC free article] [PubMed]
  • Wang, A. and Bolen, D.W. 1997. A naturally occurring protective system in urea-rich cells: Mechanism of osmolyte protection of proteins against urea denaturation. Biochemistry 36 9101–9108. [PubMed]
  • Zitzewitz, J.A., Bilsel, O., Luo, J., Jones, B.E., and Matthews, C.R. 1995. Probing the folding mechanism of a leucine zipper peptide by stopped-flow circular dichroism spectroscopy. Biochemistry 34 12812–12819. [PubMed]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.
  • Taxonomy
    Taxonomy records associated with the current articles through taxonomic information on related molecular database records (Nucleotide, Protein, Gene, SNP, Structure).
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...