Projects

Role of histone tails in chromatin structure and dynamics

Our mesoscale model with flexible histone tails was applied to analyze chromatin dynamics and conformational states under various salt conditions and the role of each on fiber condensation (Arya et al., 2006; Arya and Schlick, 2006). Dynamics simulations reveal the rapid unfolding of oligonucleosome at low salt, due to strong electrostatic repulsion between linker DNAs, leading to the “beads-on-a-string” model (Arya et al., 2006). At higher salt, oligonucleosomes remain moderately folded due to a balance between the attractive inter-nucleosomal interactions (mediated by the histone tails, especially H3) and repulsive interactions between the linker DNA. Interestingly, the emerging oligonucleosomes are highly flexible and irregular, unlike the static condensed structures portrayed in biology textbooks. Furthermore, tail-mediated fiber/fiber interactions (Fig. 3) emerge as the oligonucleosome chain folds into more compact self interactions (Arya and Schlick, 2006; Arya and Schlick, 2009).

Analyses of the tail positional distributions reveal a broad spread of tail positions consistent with the dynamic and flexible nature of the tails (Arya and Schlick, 2006). The H3 tails, in particular, remain close to the entering and exiting linker DNA and help screen electrostatic repulsion. Detailed energetics and geometric trends in the histone-mediated folded chromatin fibers(Arya and Schlick, 2006) also identify the role of each histone tail: the H4 tails mediate the strongest inter-nucleosomal interactions due to their favorable location on the nucleosome core, especially at high salt (Fig. 3b); the H3 tails interact strongly with the parent linker DNA, which helps screen electrostatic repulsion between the linkers and assist in chromatin folding (Fig. 3f); the H2A and H2B tails mediate considerable fiber/fiber interactions.

chromatin_project_fig1
Fig. 3: Analysis of histone tail interactions in chromatin without linker histones at 0.2 M (□) and 0.01 M (Δ), and “compact” oligonucleosomes at 0.2 M monovalent salt (×).

Representative 48-unit oligonucleosome at 0.2 M salt (a) with boxed regions used to highlight different tail interactions (b-g). In each box (b-g), a cartoon image depicts the interaction plotted as a frequency for the time that tails: mediate internucleosomal interactions (b), attach to parent nucleosomes (c), remain unattached (d), attach to linker DNA not associated with parent nucleosome (e), and attach to linker DNA associated with parent nucleosome (f). Plot (g) provides tail extension lengths. Results are averaged over the two copies of each tail. H2A*: C termini of H2A histones.

Role of linker histones and divalent ions in chromatin compaction and flexibility

The most intriguing part of this work involves the detailed organization of the oligonucleosome and its sensitivity to the presence or absence of the linker histone and to divalent ions (Grigoryev et al., 2009). Divalent ions were modeled to a first order approximation by allowing DNA beads to almost touch one another and reducing the DNA persistence length from 50 to 30 nm (Rouzina and Bloomfield, 1998; Baumann et al., 1997). We analyze the inter-nucleosomal interaction pattern in our irregular zigzag models of oligonucleosomes as obtained from long Monte Carlo simulations by formulating an interaction matrix I that counts in each i,j element the fraction of time (i.e., number of configurations during a large Monte Carlo ensemble) that nucleosomes i and j “interact”. We then plot the inverse of this matrix to dissect the near-neighbor interactions expressed as a high intensity near the matrix diagonal and the fiber/fiber interactions (high intensities in off-diagonal clusters); see Fig. 5 below.

chromatin_projects_fig2
Fig. 4: Chromatin compaction.

The effect of the linker histone: without linker histone and divalent ions (A), chromatin is open and floppy; with linker histone (B), the histone tails mediate many internucleosomal interactions, especially the H4 (green) and H3 tails (blue) which play an important role in chromatin compaction; and with linker histone and divalent ions, the fiber is very compact (C).

chromatin_project_fig3
Fig. 5: Internucleosomal interaction pattern predicted for distinct chromatin 30-nm fiber models.

a-c: Interaction-intensity matrices at 0.15 M monovalent salt without LH and Mg2+ (a), with LH and without Mg2+ (b), and with both LH and Mg2+ (c) show the intensity of the tail-mediated interactions between nucleosome core i and j matrix [I'(i; j)]. d-f: the associated plots decompose the dominant neighboring interactions, I(k), where k is the nucleosome separator. Interactions at k = 1, 2, and 3 nucleosomes are indicated as ±1, ±2, and ±3 correspondingly. g-j: reference internucleosomal interaction patterns corresponding to the three existing models of nucleosome arrangement: (g) solenoid model, (h) two-start zigzag model, and (i) interdigitated nucleosome model compared with that obtained from our Monte Carlo simulations for +LH+Mg chromatin (j).

Without linker histone, the most dominant core/core interactions along the folded oligonucleosome at moderate salt are between each core and its fourth neighbor along the chain (Fig. 4). When the linker histone is considered, the structures are almost twice as compact with interaction matrices exhibiting remarkable increase in fiber/fiber interactions (Fig. 5) mediated through the H2A and H2B tails.

When divalent ions are incorporated, chromatin compacts further (Grigoryev et al., 2009) made possible by a tendency to bend a proportion (e.g., 15%) of the linker DNAs to produce a mostly zigzag fiber accented by solenoid-like features (i.e., bent DNA linkers). This suggestion for a hybrid compact fiber in divalent ion conditions was verified in our work with Dr. Grigoryev who developed a new experimental technique termed EMANIC which uses formaldehyde cross-linking followed by unfolding and EM visualization to capture internucleosomal patterns (Grigoryev et al., 2009). Such an ensemble of interchanging configurations with straight and bent linker DNAs is energetically advantageous since linker DNA bending can minimize repulsion at the fiber axis. Moreover, it merges zigzag and solenoid models for optimal compaction and suggests that such modest linker DNA bending can lead to higher-order structures of chromatin through inter-digitation.

Role of nucleosome repeat length in conjunction with linker histone and counterions in chromatin organization

Different stages of cell cycle and different chromatin species are known to have distinct nucleosome repeat (DNA) lengths (NRLs) (Staynov and Proykova, 2008; Angeles and Franco, 1986). Thus, characterizing the dependence of chromatin organization on NRL in conjunction with linker histone and counterions helps decipher how various levels of chromatin compaction promote or hamper access to genetic material. We studied two biologically relevant NRL values (Fig. 6): 173 bp (short) and 209 bp (moderate), corresponding to 26 and 62 bp DNA linker lengths, respectively. We found that the short-NRL defines arrays that fold into classic zigzag structures, whereas the longer NRL allows heteromorphic fibers with some bending in the linker DNA to fold over the relatively short linker histone to minimize the overall energies and steric clashes (Schlick and Perisic, 2009). Similarly, magnesium ions markedly compact the long NRL chromatin structure but have smaller effects on the short-NRL chromatin. These results suggest that transitions between short and long-NRL chromatin forms have marked structural effects and may help correlate chromatin organization and function during the cell cycle.

chromatin_project_fig5
Fig. 6: Chromatin folds with two NRL values: (top) with Mg2+ but no linker histone, (bottom) with Mg2+ and linker histone.