Paul Kaufman

Full Directory Information

Research Interests

We study how eukaryotic cells assemble chromosomes. We are especially interested in the mechanisms of assembly of chromatin during DNA replication, and how the specialized chromatin structures at heterochromatic and centromeric loci are formed. We study these processes using biochemical, biophysical, genetic and cell biological techniques.

Current Projects

Multiple nucleosome assembly proteins function in vivo.
Histone octamers, the fundamental repeat structures of chromatin, are assembled onto nascent DNA soon after passage of a DNA replication fork in vivo. In all eukaryotic cells, two evolutionarily conserved protein complexes perform the first step in chromatin assembly, deposition of histones H3 and H4 onto DNA. One such factor is a three-subunit protein complex termed CAF-I (Chromatin Assembly Factor-I). CAF-I is essential for progression through S phase in human cells and for normal assembly of heterochromatin and centromeric chromatin in budding yeast.

In addition to CAF-I, the Hir proteins are required for histone H3/H4 deposition. We have recently purified a four-subunit Hir protein complex, and we are currently exploring its biochemical activities. Additionally, we discovered that Hir proteins directly bind Asf1, a ubiquitous eukaryotic histone-binding and deposition protein that mediates nucleosome formation in vitro and is required for genome stability in vivo. In addition to its direct interaction with histones and CAF-I, Asf1 also binds proteins involved in chromatin silencing, transcription, chromatin remodeling and DNA repair, making it a central player in many chromatin-related processes.

Biochemical and structural analysis of nucleosome assembly proteins.
CAF-I is recruited to replicated DNA via direct interaction with the DNA polymerase processivity factor PCNA, and mutations in PCNA in yeast and Drosophila lead to defects in heterochromatic gene silencing. We demonstrated that Cac1 contains a canonical PCNA-binding motif required for interaction with PCNA. Surprisingly, we found that mutation of this motif does not destroy CAF-I activity, but instead causes it to be dependent on the Asf1/Hir proteins in vivo and in vitro. Additionally, we discovered direct physical interaction between Asf1 and the Cac2 subunit of CAF-I, consistent with the mechanistic coupling between these factors. We propose that Asf1/Hir proteins functionally assist CAF-I via this direct interaction, and that this assistance is critical for nucleosome formation when Cac1-PCNA interactions are impaired.

To facilitate precise tests of this and other hypotheses regarding Asf1 function, we have recently solved the crystal structure of the highly conserved globular domain of Asf1, which mediates all functions of the full-length protein. In collaboration with the laboratory of James Berger at UC Berkeley, we determined the crystal structure of the Asf1 core domain to 1.5 Å resolution, revealing a compact immunoglobulin-like fold. The surface of Asf1 displays a conserved hydrophobic groove flanked on one side by an area of strong electronegative surface potential. These regions represent potential binding sites for histones and other interacting proteins. The structural model also allowed us to interpret recent mutagenesis studies of the human Asf1/Hir protein interaction and to functionally define the region of Asf1 responsible for Hir-dependent telomeric silencing in budding yeast. We are currently performing a detailed structure-function analysis of Asf1 to gain insight into how this protein interacts with other histone deposition and DNA replication proteins.

Genome stability in yeast: Kinetochore structure and function require chromatin assembly proteins.
Yeast cells lacking both CAF-I and Hir proteins experience a 30-45 minute delay in cell cycle progression after DNA synthesis but prior to anaphase. This G2/M delay is partially relieved by deletion of spindle assembly checkpoint genes, suggesting improper kinetochore function in these mutants. Consistent with this idea, these mutants display greatly elevated rates of chromosome missegregation and genetic synergies with kinetochore mutants. Using immunolocalization and chromatin immunoprecipitation assays, we have demonstrated that a subset of CAF-I and Hir proteins in the cell are stably localized to centromeric chromatin. Furthermore, the nucleosomal structure at centromeres is perturbed in the absence of CAF-I and Hir proteins. We are currently initiating genomic approaches to determine the other, unknown sites of CAF-I proteins localization in the cell.

Genome stability in human cells: surveillance of nucleosome assembly during S phase.
To investigate the contribution of CAF-I to chromatin formation in human cells, we first designed dominant-negative inhibitors based on our mapping of the CAF-I subunit interaction domains. Specifically, we focused on a C-terminal fragment of the human CAF-I large subunit (termed p150C) that binds the middle CAF-I subunit but not PCNA. First, we confirmed that p150C inhibits nucleosome assembly but not DNA synthesis in vitro. In collaboration with Dr. Peter Adams, Fox Chase Cancer Center, we discovered that transient expression of p150C causes a dramatic S phase delay in human tissue culture cells. This delay is very similar to that observed upon overexpression of the human Hir protein homolog, HIRA. Specifically, histone H2AX becomes phosphorylated, and p53 becomes stabilized and phosphorylated on serine 15. Importantly, the cell cycle delay caused by p150C or HIRA overexpression requires the presence of either the related ATR or ATM checkpoint kinases. These data indicated that perturbation of nucleosome assembly results in DNA damage recognized by the ATM/ATR kinases. We hypothesize that this results from instability of DNA replication forks when rapid nucleosome assembly is absent. To pursue studies of the role of CAF-I in genome stability, we have generated inducible siRNA cell lines that destroy CAF-I in a regulated manner. These cells will allow us to arrest large uniform populations for further physiological and biochemical analysis.

Selected Publications

Histone deposition proteins: links between the DNA replication machinery and epigenetic gene silencing. [Franco AA and Kaufman PD. (2004) Cold Spring Harbor Symposia on Quantitative Biology, in press]

Structure and function of the conserved core of histone deposition protein Asf1. [Daganzo SM, Erzberger JP, Lam WM, Skordalakes E, Zhang R, Franco AA, Brill SJ, Adams PD, Berger JM, and Kaufman PD. (2003) Current Biology 13: 2148-2158]

The budding yeast silencing protein Sir1 is a functional component of centromeric chromatin. [Sharp JA, Krawitz DC, Gardner KA, Fox CA, and Kaufman PD. (2003) Genes Dev. 17: 2356-61]

Inhibition of S-phase chromatin assembly causes DNA damage, activation of the S-phase checkpoint and S-phase arrest. [X. Ye, A.A. Franco, H. Santos, P.D. Kaufman, and P.D. Adams (2003) Mol. Cell 11: 341-351]

Sas4 and Sas5 are required for the histone acetyltransferase activity of Sas2 in the SAS complex. [A. Sutton, W-J. Shia, D. Band, P.D. Kaufman, S. Osada, J.L. Workman, and R. Sternglanz (2003) J. Biol. Chem. 278:16887-16892]

Chromatin proteins are determinants of centromere function. [J.A. Sharp and P.D. Kaufman (2003) Current Topics in Microbiology and Immunology 274:23-52 ]

Defects in SPT16 or POB3 (yFACT) Cause Dependence on the Hir/Hpc Pathway: Accessing DNA May Degrade Chromatin Structure. [T. Formosa, S. Ruone, M.D. Adams, A.E. Olsen, P. Eriksson, Y. Yu, A.R. Rhoades, P.D. Kaufman, and D.J. Stillman. (2002) Genetics 162: 1557-1571]

Chromatin Assembly Factor-I and Hir proteins contribute to building functional kinetochores in Saccharomyces cerevisiae. [J.A. Sharp, A.A. Franco, M.A. Osley and P.D. Kaufman (2002) Genes Dev. 16, 85-100]

Chromatin Assembly Factor-I mutants defective for PCNA binding require Asf1/Hir proteins for silencing. [D.C. Krawitz, T. Kama and P.D. Kaufman (2002) Mol. Cell. Biol. 22, 614-625]

Yeast Histone Deposition Protein Asf1p Requires Hir Proteins and PCNA for Heterochromatic Silencing. [J.A. Sharp, E.T. Fouts, D.C. Krawitz and P.D. Kaufman (2001) Current Biology 11, 463-473]

Last Updated 2004-09-02