Faculty Profile

Ed Luk Ph.D

Assistant Professor
Department of Biochemistry and Cell Biology

410 Life Sciences Building
Stony Brook University
Stony Brook, NY 11794-5215
Office telephone: 631-632-1903

E-mail: ed.luk@stonybrook.edu

Earlier work and future studies


  image1
  Figure 1: Crystal structure of the H2A- and H2A.Z-containing nucleosomes solved by the laboratories of Timothy Richmond and Karolin Luger [2,3].
   


Our recent work has focused on a chromatin remodeling enzyme, called SWR1, which is responsible for depositing the histone variant H2A.Z at promoters. Most nucleosomes are made up of a histone core consisting of two copies of each of the major histones H2A, H2B, H3 and H4 (Figure 1, left). However, in a minor fraction of nucleosomes (~5-10%), H2A is replaced with the H2A.Z variant (Figure 1, right). While H2A and H2A.Z share ~60% sequence identity, unique residues are located throughout these polypeptides. Nucleosomes containing H2A.Z mark gene promoters and flank a nucleosome-depleted region, which is typically found upstream of the transcription start site (Figure 2) [4,5]. H2A.Z nucleosomes are less stable in vivo. It has been proposed that they help poise genes for transcription by facilitating the transaction of transcription factors on promoter DNA [6,7].

 
Figure 2: Organization of nucleosomes in and around a typical yeast promoter   Figure 3: Histone replacement reaction catalyzed by SWR1

 

  image1
 

Figure 4: Mechanism of SWR1-mediated histone replacement


How H2A.Z is deposited at gene promoters is of great interest to us. Using an in vitro approach, we previously demonstrated that SWR1 deposits H2A.Z by replacing the two H2A-H2B dimers in a nucleosome with two H2A.Z-H2B dimers in a stepwise manner (Figure 3), generating heterotypic H2A/H2A.Z nucleosomes (A/Z) as intermediate and homotypic H2A.Z/H2A.Z nucleosomes (Z/Z) as final product [8]. This replacement reaction is driven by ATP hydrolysis and is unidirectional (i.e. replacing H2A with H2A.Z, but not vice versa) [8].

The specificity of the reaction tells us that SWR1 is able to distinguish the unique structural features of H2A and H2A.Z. Indeed, we found that the ATPase motor of SWR1 becomes hyperstimulated only when SWR1 simultaneously binds to H2A nucleosomes and H2A.Z-H2B dimer (Figure 4) [8].

These mechanistic insights are supported by our in vivo data. When we isolated the A/Z and Z/Z nucleosomes from yeast cells and “mapped” the nucleosomal DNA by microarray analysis, we found that both species are present at gene promoters (Figure 5) [8]. But somewhat unexpectedly, for any given promoter, homotypic H2A nucleosomes (A/A) are also frequently present (Figure 5) [8].

 
image1
Figure 6: Proposed model
Figure 5: Nucleosome density profiles of the Z/Z, A/Z, and A/A nucleosomes. ‘All’ shows the profile before the nucleosomal species were fractionated. Arrows indicate the position of transcription start sites.


These observations, together with data from other labs that showed histones turn over more quickly at promoters [9,10], lead us to speculate that there is an opposing pathway that involves the eviction of the labile Z/Z nucleosome and reassembly of the A/A nucleosome. We propose that the resulting “histone cycle” is what is needed to open up the underlying promoter sequence to facilitate the assembly of the transcription machinery (Figure 6). We are currently testing this model and identifying the genes that are involved in this pathway.

In yeast and humans, H2A.Z nucleosomes are also found at genomic sites other than promoters, e.g. DNA repair sites, telomere boundaries, insulators, enhancers, and intron-exon junctions. It is conceivable that the cycling of histone may also occur at these genomic sites to regulate the binding of an array of chromatin factors. Therefore, we are also investigating whether SWR1 and H2A.Z have functional importance beyond transcription, especially in DNA repair processes.

The outcome of our research will likely have a broader impact than simply understanding transcriptional control or DNA repair in yeast. The mammalian H2A.Z and SWR1 homologs are known to play important roles in the regulation of tumor suppressor and developmental genes [11,12]. The technologies and principles developed for the yeast can potentially be applied to mammalian systems and generate new insights into the molecular basis of these processes in humans. Students in our laboratory will learn how to use fluorescence-based in vitro assays, genomic technologies, and yeast genetics to study questions fundamental to the understanding of chromosome biology.


Top of Page