Structural characterisation of FACT histone chaperone complex

[摘要] Due to the huge amount of DNA and the small space in which this is stored, cells package DNA into condensed chromatin. During DNA replication, repair or transcription, cells need to reorganize chromatin structure in order to gain access to the DNA. To do so they use a wide range of enzyme that can modify chromatin structure such as acetyltransferases, methyltransferases and others. Chromatin remodelling is also accomplished by ATP dependent and independent histone chaperones. One of the ATP independent histone chaperones is FACT (facilitate chromatin transcription), which is a heterocomplex of SSRP1 and Spt16. Studies on FACT showed that it is implicated in DNA replication, transcription and repair. To date little is known about its mechanism of action; but based on recent structural studies of its individual domains, it has been proposed that FACT reorganizes nucleosome by eviction of the heterocomplex H2A/H2B. Nevertheless, how FACT reorganizes nucleosomes remains elusive. Studies conducted in several cancer cell lines have shown that FACT is overexpressed in these cells and that knockdown of FACT reduces cells proliferation suggesting that FACT may be a cancer therapeutic target. Here I report the structure of human Spt16 N-terminal domain, which resembles an aminopeptidase domain lacking a catalytic centre. Sequence conservation and electrostatic surface analyses of this domain reveal some acidic regions that might be associated with histone binding. Indeed, ITC analysis showed that this domain binds both histone H2A/H2B and (H3/H4)2 at mid-low μM affinity. Interestingly, Spt16 N-terminal domain showed a sequential binding event both for G. gallus and X. laevis histone dimer H2A/H2B. On the onset of my project there was no structure for Spt16 middle domain. To gain insight into FACT mechanism of action, I determined the structure of D. rerio Spt16 middle domain. Surprisingly this domain resembles a double PH domain, which is similar to the POB3 middle domain and RTT106 except for having a U-turn motif at its C-terminus. Sequence conservation and electrostatic surface analyses of this domain reveal two possible surfaces for histone binding. Interestingly, ITC analysis showed Spt16 middle domain has weak binding affinity for both histone H2A/H2B and (H3/H4)2 with the latter displaying a double binding event. Moreover, ITC analysis showed that Spt16 middle domain binds histone H2A/H2B via the U-turn motif consistent with the recent published data. Finally, I investigated the solution structure of SSRP1 by using AUC and SAXS analysis. I found that SSRP1 is an elongated homodimer, which assumes an open v conformation. Mutations in PH2 or PH3 domain alone resulted in the formation of monomer suggesting that SSRP1 homodimer may assume an asymmetric conformation. Interestingly, PH2 and PH3 domain mutants displayed weaker histone binding affinity than wild type suggesting that homodimerization plays a role in histone binding.

[发布机构] University:University of Glasgow;Department:Institute of Cancer Sciences

[关键词] QH301 Biology