Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2005 Oct 25;44(42):13673-82.
doi: 10.1021/bi051333h.

ASF1 Binds to a Heterodimer of Histones H3 and H4: A Two-Step Mechanism for the Assembly of the H3-H4 Heterotetramer on DNA

Free PMC article

ASF1 Binds to a Heterodimer of Histones H3 and H4: A Two-Step Mechanism for the Assembly of the H3-H4 Heterotetramer on DNA

Christine M English et al. Biochemistry. .
Free PMC article


The first step in the formation of the nucleosome is commonly assumed to be the deposition of a histone H3-H4 heterotetramer onto DNA. Antisilencing function 1 (ASF1) is a major histone H3-H4 chaperone that deposits histones H3 and H4 onto DNA. With a goal of understanding the mechanism of deposition of histones H3 and H4 onto DNA, we have determined the stoichiometry of the Asf1-H3-H4 complex. We have established that a single molecule of Asf1 binds to an H3-H4 heterodimer using gel filtration, amino acid, reversed-phase chromatography, and analytical ultracentrifugation analyses. We demonstrate that Asf1 blocks formation of the H3-H4 heterotetramer by a mechanism that likely involves occlusion of the H3-H3 dimerization interface.


Figure 1
Figure 1. Model for histone binding by Asf1
(A) Cartoon diagram of the H3–H4 heterotetramer. Blue depicts the H3 dimer; red depicts two H4 molecules, and cyan depicts the putative Asf1–H3 interacting region at the H3–H3 dimerization interface (amino acids 97–135). The model was derived from the coordinates of the nucleosome structure of PDB entry ikx5 (37). (B) Schematic of possible models for interaction among Asf1, H3, and H4. (i) Model for a 2:2:2 stoichiometric Asf1:H3:H4 ratio, where two Asf1 molecules would bind to the identical interaction surfaces on each of the H3 proteins within the H3–H4 heterotetramer. (ii) Model for a 2:2:2 stoichiometric Asf1:H3:H4 ratio, where interaction between Asf1 and two H3–H4 heterodimers is mediated via a hypothetical Asf1–Asf1 interaction. (iii) Model for a 1:2:2 stoichiometric Asf1:H3:H4 ratio, where interaction between Asf1 and the H3–H4 tetramer is mediated via binding of Asf1 to H3. This model would require that each Asf1 molecule have two identical binding interfaces for H3. (iv) Model for a 1:1:1 stoichiometric Asf1:H3:H4 ratio, where interaction between Asf1 and the H3–H4 heterodimer is mediated via binding of Asf1 to the H3 dimerization interface.
Figure 2
Figure 2. Gel filtration analysis of the recombinant Asf1–H3–H4 complex
(A) Elution profile of the Asf–H3–H4 complex obtained by gel filtration analysis. Recombinant Asf1tr–H3–H4 complex was eluted from a Superdex 75 16/60 HiLoad column, and the resulting trace is shown. The key indicates the identity of the major peaks, as determined by SDS–PAGE analysis of selected fractions. The arrows indicate the positions of the elution peaks of gel filtration standards that were eluted from the same column in the same buffer as the Asf1tr–H3–H4 complex. (B) SDS–PAGE analysis of the purified Asf1tr–H3–H4 complex obtained by gel filtration analysis. Selected fractions from the gel filtration column elution profile shown in panel A were resolved on a 15% polyacrylamide gel and stained with Coomassie blue. The positions of the Asf1, H3, and H4 proteins are indicated.
Figure 3
Figure 3. Recombinant Asftr–H3–H4 complex has nucleosome assembly activity
The ability of nothing (–), Drosophila S190 extract, increasing amounts of purified recombinant Asf1tr–H3–H4 complex (400 ng, 750 ng, or 1 µg), or recombinant Asf1 (400 ng, 750 ng, or 1 µg) to introduce supercoils into relaxed plasmid DNA in a DNA replication-independent supercoiling assay was tested. An image of a Sybr gold stained agarose gel is shown with markers for supercoiled and relaxed species as indicated.
Figure 4
Figure 4. Amino acid analysis indicates equal stoichiometry in the Asf1tr–H3–H4 complex
The amino acid content of recombinant Asf1tr–H3–H4 complex from the gel filtration analysis shown in Figure 1 was determined. The observed ratio of each amino acid in the Asf1tr–H3–H4 complex, relative to alanine (normalized to 1), is represented by the black bars. The error bars indicate the standard deviation of three independent experiments. The expected ratio for a theoretical 1:1:1 Asf1tr–H3–H4 complex is shown with gray bars, while the expected ratio for a theoretical 1:2:2 Asf1tr–H3–H4 complex is shown with white bars. The asterisks indicate the amino acid residues that are significantly different between the 1:1:1 and 1:2:2 complex and can thus be used for determining the stoichiometry of the recombinant Asf1tr–H3–H4 complex.
Figure 5
Figure 5. Reversed-phase HPLC analysis indicates the equal stoichiometry in the Asf1tr–H3–H4 complex
Purified recombinant Asf1tr–H3–H4 complex was analyzed by reversed-phase HPLC at 210 nm and normalized to H4, and the resulting elution profile is shown. The identity of the protein eluting in each peak was determined by mass spectrometry, and is indicated. The ratio of the integrated area units under each peak is 2.3:1.3:1 for the Asf1tr–H3–H4 complex.
Figure 6
Figure 6. Analytical ultracentrifugation analyses demonstrate that the Asf1tr–H3–H4 complex exists as a single 2.9S species in solution
(A) Analysis of the homogeneity of the Asf1tr–H3–H4 sample. van Holde–Weischet analysis of sedimentation velocity analysis of 0.31 mg/mL Asf1tr–H3–H4 complex. Over the majority of the boundary, the sample is characterized by a single value of s20,w of ~3 S. (B) Sedimentation velocity analysis of the Asf1tr–H3–H4 complex. The calculated g(s*) distributions from the sedimentation velocity experiments are shown for increasing loading concentrations of the complex (from 0.074 to 0.64 mg/mL). The smooth curves are the nonlinear least-squares fits of the g(s*) data to a single Gaussian distribution (eq 1). The dotted line corresponds to the sedimentation value of Asf1 alone; s20,w is 1.93 ± 0.02 S, and the molecular mass is 18.7 ± 1.2 kDa. (C) Sedimentation equilibrium analysis of the Asf1tr–H3–H4 complex. The Asf1tr–H3–H4 complex (0.35 mg/mL) was sedimented to equilibrium at 20K (blue circles), 25K (red triangles), and 30K rpm (green squares), at 4 °C. The top panel shows the results of a global, nonlinear least-squares fit of the data to a single-ideal species model (the smooth curves that are drawn through the data). The bottom panel shows the residuals for this fit.
Figure 7
Figure 7. Model for stepwise formation of the H3–H4 heterotetamer on DNA
Old H3–H4 heterotetramers (light blue apple cores) are randomly distributed between the new daughter strands of DNA, while Asf1 and/or CAF-1 deposits new H3–H4 heterodimers onto the newly replicated DNA to form new H3–H4 heterotetramers (red and blue apple cores). Asf1 binds to the H3 dimerization interface of the H3–H4 heterodimer, preventing its dimerization to form an H3–H4 heterotetramer until Asf1 releases the histones. Asf1 may pass the H3–H4 heterodimers to its binding partner CAF-1 to be deposited onto the newly replicated DNA, or Asf1 may deposit the H3–H4 heterodimers directly. CAF-1 is localized to the replication fork via interaction with the replication machinery. Asf1 is also required for replication-independent chromatin disassembly and assembly, otherwise known as histone exchange, which results in the mixing of old and new heterodimers within an H3–H4 heterotetramer.

Similar articles

See all similar articles

Cited by 63 articles

See all "Cited by" articles

Publication types