High-throughput Functional Genomics Identifies Regulators of Primary Human Beta Cell Proliferation

Karine Robitaille,Jillian L Rourke,Joanne E Mcbane,Accalia Fu,Stephen Baird,Qiujiang Du,Tatsuya Kin,A.M Jame Shapiro,Robert A Screaton

doi:10.1074/jbc.m115.683912

Abstract

The expansion of cells for regenerative therapy will require the genetic dissection of complex regulatory mechanisms governing the proliferation of non-transformed human cells. Here, we report the development of a high-throughput RNAi screening strategy specifically for use in primary cells and demonstrate that silencing the cell cycle-dependent kinase inhibitors CDKN2C/p18 or CDKN1A/p21 facilitates cell cycle entry of quiescent adult human pancreatic beta cells. This work identifies p18 and p21 as novel targets for promoting proliferation of human beta cells and demonstrates the promise of functional genetic screens for dissecting therapeutically relevant state changes in primary human cells.

Highlights

Discovery-based research using cultured cells, including chemical and functional genomic high-throughput screens (HTS),6 are recognized as powerful, comprehensive tools that identify novel components of signaling pathways [1,2,3]
There is a critical need for approaches that elicit proliferative behavior of mature human beta cells and their progenitors and to establish a working understanding of the underlying mechanisms for precise intervention so as to avoid the risk of unrestricted growth. We address these issues and describe the development and implementation of an HTS approach using RNA interference to analyze proliferation following gene silencing in a mixed population of primary human pancreatic islet cells, and we report that the cell cycle-dependent kinase (CDK) inhibitors CDKN2C/p18
We demonstrate the feasibility of genetic HTS using limited freshly isolated primary human tissue and identify the cell cycle inhibitors p18 and p21 as required for maintenance of adult human beta cell quiescence

Summary

Experimental Procedures

Antibodies/Reagents—Antibodies for Western blotting were as follows: ␤-actin (Sigma, 1:20,000); p53 (Santa Cruz Biotechnology, 1:1000); pRb (Pharmingen, 1:500) and p21 (Pharmingen 1:500); p27 (Cell Signaling, 1:500); PTEN (Santa Cruz Biotechnology, 1:200); pAKT (Cell Signaling, 1:1000) and AKT (Cell Signaling, 1:1000); LKB1 (Santa Cruz Biotechnology, 1:1000), p18 (Cell Signaling 1:200), and HSP90 (Santa Cruz Biotechnology, 1:1000). With the exception of the initial plate coating experiment, islets were always seeded on a PDL-coated plate (described below) to facilitate attachment of dissociated cells. Dissociated islets were seeded onto the coated/ uncoated plates, infected with non-targeting shRNA for 96 h, and fixed for immunofluorescence as described below. GSIS Assay—Human islets were dissociated and seeded at a density of 20,000 cells per well into a 384-well plate in triplicate 96 h before treatment. Lentiviral Infection Optimization—Pilot experiments as outlined in Table 1 showed that the GP-VSVG envelope protein coding vector (pMD2G) provided greater infection efficiency than GP-LCMV (pHCMV) as evaluated by GFP and titer (supplemental Fig. 1 a and Table 1). Virus was packaged in HEK293T for 72 h with pLKO1.1-eGFP plasmid encoding GFP and shRNA (supplemental Fig. 1c) and pMD2G and pCMV8.74 encoding coat and structural genes, respectively. In a direct comparison of crude versus concentrated virus

Body mass index years

Results

EdU score

Discussion