←
Darko Barisic408 posts
Darko Barisic avatar

Darko Barisic

@LabBarisic

Assistant Professor of Cancer Biology @WeillCornell Chromatin & Epigenetics in Cancer.

◷ Joined April 2009▧ 50 Photos and videos
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

Beyond thrilled that the first paper from our lab is out today in @nature! We show that SWI/SNF loss results in gain-of-function for CTCF and cohesin, reshaping the 3D genome, and an unexpected switch between chromatin remodeling complexes. nature.com/articles/s41586-0… 1/n

Darko Barisic avatar
Darko Barisic@LabBarisic·Nov 5

Excited to share our other recent paper in @NatureBiotech from Christina Leslie and our lab. We developed and in vivo validated DynaVelo, a deep learning tool that predicts rescue of genetic loss-of-function mutations. Brilliantly led by @karbalayghareh nature.com/articles/s41587-0…

Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

In summary, SWI/SNF loss can drive a gain of function: cBAF loss enables CTCF/cohesin gain and a switch to PBAF, rewiring the 3D genome and promoting proliferation. This reveals how tumor suppressor loss can lead to a growth advantage for cancer cells. 13/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

Finally, Arid1a loss in mice expands the germinal center dark zone, home to highly proliferative B cells. Deleting one copy of either Ctcf or the PBAF subunit Pbrm1 rescues this phenotype. Two genetic rescues connect the complex switch to cell behavior and phenotype. 12/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

Do we see a parallel in patients? In pan-cancer analyses, patients with alterations in both ARID1A and CTCF have better overall survival than those with ARID1A alterations alone. Also, high CTCF expression is associated with worse survival. 11/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

This is also true in mice. Embryonic fibroblasts proliferate faster after Arid1a loss, and genetically deleting both Ctcf and Arid1a restores normal growth. Interestingly, in our mouse lymphoma model, CTCF-site opening appears before overt disease and persists in tumors. 10/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

Why would cells benefit from CTCF and PBAF gain upon cBAF loss? Faster proliferation. CTCF over-expression alone reproduces the growth advantage of ARID1A loss. Deleting the PBAF subunit ARID2 removes that advantage in ARID1A-mutant cells. 9/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

To test if cBAF and PBAF switch happens on chromatin, we examined their binding with ChIP. In both lymphoma and breast cancer cells, sites that lose ARID1A binding upon ARID1A deletion gain PBAF, showing the complex switch at the same genomic sites. 8/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

Next we asked: What helps CTCF gain when cBAF is lost? Surprisingly, PBAF. CTCF interacts with this distinct SWI/SNF complex that is recruited to sites gaining CTCF. We confirmed this interaction in several cancer cell lines from different tissue-of-origin. 7/n

Post attachment
Darko Barisic avatar
Darko Barisic@LabBarisic·Oct 28

So what did that do to genome folding? A lot. Large TADs split into smaller domains. Boundaries were stronger. Long-range contacts decreased and short-range ones increased. Interestingly, gained CTCF sites picked up more promoter-enhancer contacts and those genes went up. 6/n

Post attachment