Genome Integrity Discussion Group June 2017

LEM Domain Proteins Control Adaptation by Modulating a Transcription-Dependent Change in Gene Copy Number

Paolo Colombi, Yale University, Department of Cell Biology

While loss of genome integrity is at the basis of numerous pathologies, including cancer, genome plasticity is necessary to adapt to a changing environment and thus is essential for long-term organismal success. One of the ways in which genomic plasticity manifests is through gene copy number variation (CNV)(Yona, Frumkin, and Pilpel 2015). Historically, the events leading to CNV have been conceptualized through a Darwinian lens as incidental mutations that become fixed in the population because of a fitness advantage. Our work challenges this view; we will present data supporting a targeted mechanism that promotes adaptation to environmental stress by driving site-specific genome instability. We have identified hypervariable protein coding regions in S. cerevisiae by integrating SNP and fragile site distribution across the genome. One such region encodes a cluster of four nearly identical genes (ENA) each coding for a P-type ATPase pump involved in the extrusion of Li+ from the cell; ENA gene expression is induced by growth in high salt concentrations. Interestingly, the load of SNPs near the ENA gene cluster is highly enriched at the 3′ end, which is also an established fragile site; these observations suggest a possible role for ENA gene transcription as a driver of local genome instability. Indeed, a genome stability assay links the fragility of the 3′ end of the ENA gene cluster to both transcription (induced by high salt) and the formation of DNA:RNA hybrids. In chronically high salt conditions, we find that the ENA locus undergoes reproducible gene copy number expansion. Interestingly, we find that the ENA gene cluster is associated with the nuclear periphery in normal growth conditions but moves to the nuclear interior upon exposure of cells to high salt.

To begin to probe if and how CNV of the ENA gene cluster is tied to association with the nuclear periphery, we focused on the LEM domain proteins (Heh1, Heh2); HEH1 and HEH2 genetically interact with genes responsible for transcription-dependent genome instability and Heh1 plays a role in maintaining the stability of the repetitive rDNA (Mekhail et al. 2008). Using an in vitro evolution assay based on prolonged exposure of cells to high concentration of LiCl, we observe that different genotypes (WT, heh1Δ, heh2Δ, heh1Δheh2Δ) adapt at different rates and through distinct pathways, involving CNV and/or point mutations. Taken together, our data suggest the existence of a LEM domain protein-mediated mechanism by which an immediate transcriptional response to a changing environment drives targeted genome instability to promote long-term adaptation through CNV. These results suggest a Lamarckian pathway to promote targeted genome instability in response to a changing environment on which Darwinian selection can act to promote adaptation.

Coauthors: Diane King, Jessica F. Johnston, C. Patrick Lusk, and Megan C. King, Yale University, Department of Cell Biology

References:

Mekhail, Karim, Jan Seebacher, Steven P Gygi, and Danesh Moazed. 2008. "Role for Perinuclear Chromosome Tethering in Maintenance of Genome Stability" 456 (7222). Nature Publishing Group: 667-70. doi:10.1038/nature07460.

Yona, Avihu H, Idan Frumkin, and Yitzhak Pilpel. 2015. "A Relay Race on the Evolutionary Adaptation Spectrum." Cell 163 (3): 549-59. doi:10.1016/j.cell.2015.10.005.

Capturing the Onset of Polycomb Domain Formation
Ozgur Oksuz, NYU Langone School of Medicine (Reinberg lab)

The means by which epigenetically inheritable, repressed chromatin domains are established and maintained for appropriate gene expression in mammals remains largely unknown. Polycomb repressive complex 2 (PRC2) maintains transcriptional inactivity by catalyzing tri-methylation of histone H3 at lysine 27 (H3K27me3) within chromatin. Through the design of a system whereby PRC2-mediated repressive domains can be collapsed and then reconstructed in an inducible fashion in vivo, a two-step mechanism of domain formation became evident. First, PRC2 is recruited to a limited number of "nucleation sites" that are enriched for local CpG islands (CGIs), distinguished from other CGIs by the presence of specific motifs. These nucleation sites lie in close spatial proximity, forming polycomb foci within the nucleus. Second, via binding to its catalytic product, PRC2 spreads both locally and distally via long-range contacts to form H3K27me3-chromatin domains. These mechanistic insights into the establishment and maintenance of polycomb-repressive domains across the genome point to the centrality of the 3D chromatin structure.

Coauthors: Varun Narendra^{1, 2}, Chul-Hwan Lee^{1, 2}, Nicolas Descostes^{1, 2}, Gary LeRoy^{1, 2}, Ramya Raviram³, Kelly Karch⁴, Pedro R. Rocha³, Benjamin A. Garcia⁴, Jane A. Skok³, and Danny Reinberg^{1, 2}.

1. Howard Hughes Medical Institute, New York University School of Medicine.
2. Department of Biochemistry and Molecular Pharmacology, New York University School of Medicine.
3. Department of Pathology, New York University School of Medicine.
4. Epigenetics Program, Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania.

To whom correspondence should be addressed: Danny.Reinberg@nyumc.org