Regulation of Human Pancreatic β-cell Proliferation by Localization, Interaction, and Transcription of Endogenously Expressed Cell Cycle Molecules

Abstract

Diabetes mellitus was the seventh leading cause of death in the United States in 2012, and it affects approximately 9.3% of the American population. The two most common forms of diabetes, Type 1 and Type 2, are characterized by reduced pancreatic β-cell mass. To develop novel therapeutic approaches for diabetes, I investigated molecular alternations in localization, function, and transcriptional regulation of endogenously expressed cell cycle molecules in non-proliferating human β-cells. In recent reports, our laboratory demonstrated that overexpression of the early G1-phase cell cycle molecules CDK4, CDK6, and CCND1 results in their nuclear translocation from the cytoplasm with resultant activation of cell cycle progression. This suggests that (1) an unknown mechanism may constrain CDK4/6 and CCND1 in the cytoplasm and prevent cell cycle activation in human pancreatic β-cells, (2) CDK4/6 and CCND1 may have important biological functions in the cytoplasm, and (3) activation of CCND1 transcription alone may lead to human β-cell proliferation.

I first identified cytoplasmic CDK4 and CDK6 partner proteins using quantitative mass spectrometry of CDK4 and CDK6 immunoprecipitates in human β-cells. Among those identified, I focused on p18^INK4c and TMG2 to investigate if they regulate CDK4 and CDK6 localization and proliferative activity in human pancreatic β-cells. Interaction of CDK4/6 with p18^INK4c and TMG2 was confirmed with co-immunoprecipitation and in situ proximity ligation assay (isPLA). Unfortunately, overexpression or silencing of p18^INK4c or TGM2, which was confirmed by immunoblot and immunofluorescence, did not affect localization of CDK4 and CDK6 and as a result did not alter the proliferation of human β-cells as determined by Ki67 and BrdU labeling. I also used the Gene Ontology database to identify potential biological functions of the CDK4 and CDK6 interacting proteome. I demonstrated that CDK4 and CDK6 are not significantly involved in ER stress, apoptosis, or trafficking in human islets, although these biological functions were predicted by the Gene Ontology pathway analysis. Interestingly, long-term in vivo treatment of palbociclib, vi a selective inhibitor of CDK4 and CDK6, resulted in ER stress in β-cells of young rats. Lastly, I assessed transcription factors that may regulate CCND1 expression. Among many, we chose to study LEF1 as it is known to increase CCND1 transcription in cell lines via the Wnt//β-catenin pathway. In human islets, LEF1 overexpression increased expression of Wnt target genes but did not affect CCND1 transcription. I demonstrated that while YY1, a transcription factor for LEF1 and CCND1, increases CCND1 mRNA levels, it has no effect on proliferation in human islets. This suggests that an increase in CCND1 expression may not be sufficient to induce human β-cell proliferation.

These studies provides deeper insights into novel physiological functions of cytoplasmic cell cycle molecules in non-proliferating human pancreatic β-cells, and may translate into therapeutics for human β-cell regeneration in diabetes.