Cdk4 regulates recruitment of quiescent beta-cells and ductal epithelial progenitors to reconstitute beta-cell mass

Abstract

Insulin-producing pancreatic islet beta cells (beta-cells) are destroyed, severely depleted or functionally impaired in diabetes. Therefore, replacing functional beta-cell mass would advance clinical diabetes management. We have previously demonstrated the importance of Cdk4 in regulating beta-cell mass. Cdk4-deficient mice display beta-cell hypoplasia and develop diabetes, whereas beta-cell hyperplasia is observed in mice expressing an active Cdk4R24C kinase. While beta-cell replication appears to be the primary mechanism responsible for beta-cell mass increase, considerable evidence also supports a contribution from the pancreatic ductal epithelium in generation of new beta-cells. Further, while it is believed that majority of beta-cells are in a state of 'dormancy', it is unclear if and to what extent the quiescent cells can be coaxed to participate in the beta-cell regenerative response. Here, we address these queries using a model of partial pancreatectomy (PX) in Cdk4 mutant mice. To investigate the kinetics of the regeneration process precisely, we performed DNA analog-based lineage-tracing studies followed by mathematical modeling. Within a week after PX, we observed considerable proliferation of islet beta-cells and ductal epithelial cells. Interestingly, the mathematical model showed that recruitment of quiescent cells into the active cell cycle promotes beta-cell mass reconstitution in the Cdk4R24C pancreas. Moreover, within 24-48 hours post-PX, ductal epithelial cells expressing the transcription factor Pdx-1 dramatically increased. We also detected insulin-positive cells in the ductal epithelium along with a significant increase of islet-like cell clusters in the Cdk4R24C pancreas. We conclude that Cdk4 not only promotes beta-cell replication, but also facilitates the activation of beta-cell progenitors in the ductal epithelium. In addition, we show that Cdk4 controls beta-cell mass by recruiting quiescent cells to enter the cell cycle. Comparing the contribution of cell proliferation and islet-like clusters to the total increase in insulin-positive cells suggests a hitherto uncharacterized large non-proliferative contribution.

Figure 1. Enhanced regeneration of β-cell mass in Cdk4 ^R/R mice in response to partial pancreatectomy (PX).

(A) Body weight change and (B) fed-state blood glucose levels were measured before (day 0) and on indicated days after PX on Cdk4 ^WT mice (open squares) and Cdk4 ^R/R mice (closed squares). (C) β-cell mass is increased after PX in both groups. Significant increase of β-cell mass was observed in Cdk4 ^R/R mice (closed bars) compared with Cdk4 ^WT mice (open bars) by 14 days after PX. Data are shown as means and error bars represent S.E. *p<0.05 vs Cdk4 ^WT mice.

Figure 2. Increased islet β-cell and ductal epithelial cell proliferation after PX in Cdk4 ^R/R mice.

(A) Islet β-cell proliferation was significantly increased in Cdk4 ^R/R mice (closed bars) compared with Cdk4 ^WT mice (open bars) by 14 days after PX. The picture on the left shows BrdU⁺ (brown) and insulin⁺ (blue) β-cell (red arrow) and BrdU negative β-cell (yellow arrow). (B) Proliferation of pancreatic ductal epithelial cells was significantly increased within 2 days post-PX in Cdk4 ^R/R mice (closed bars) compared with Cdk4 ^WT mice (open bars). Arrow heads in picture on left indicate BrdU⁺ ductal cells. Data are shown as means and error bars represent S.E. *p<0.05, **p<0.01, and ***p<0.001 vs Cdk4 ^WT mice. Note that BrdU-labeled cells result from a 16-hr-labeling period.

Figure 3. Schematic diagram and protocols of the double-labeling experiment.

(A) Schematic description of cell proliferation under the double-labeling experiment. Two initial populations of quiescent (black) and active (white) cells are subjected to two pulse (CldU and IdU) and one chase periods. The numbers of once- and twice-proliferated cells are represented by and (during the CldU period); and (during the IdU period); and and (during the chase period). The number of purple twice-proliferated (once in CldU and once in IdU) cells are denoted with . Note that some labeled cells divide during the chase period and decrease their labeling intensity. They are described with slashed green and red cells in the chase period. (B) Three protocols of the double-labeling experiment with two pulse periods and following chase period.

Figure 4. Proliferation of β-cells and ductal epithelial cells of Cdk4 ^WT and Cdk4 ^R/R mice after pancreatectomy.

Changes of quiescent, active, and proliferated cell numbers (,, and ) normalized by initial active cell number; proliferation rate of active cells; and ratio of proliferated cells among entire cell population are plotted.

Figure 5. Pdx-1 expression in pancreatic ductal epithelial cells from Cdk4 ^R/R mice after pancreatectomy.

(A) A representative picture of Pdx-1 expression observed within the cytokeratin (CK)-positive pancreatic ductal epithelium from Cdk4 ^R/R mice. Arrows show Pdx-1⁺:CK⁺ cells. Blue color identifies DAPI-stained nuclei. (B) Percentage of Pdx-1⁺ ductal epithelial cells was significantly increased by 2 days post-PX in Cdk4 ^R/R mice (closed bars) compared with Cdk4 ^WT mice (open bars). Data are shown as means from four mice of each genotype per group and error bars represent S.E. **p<0.01, and ***p<0.001 vs Cdk4 ^WT mice.

Figure 6. Increased number of islet-like cell clusters (ICCs) in pancreatectomized Cdk4 ^R/R mice.

(A) The number of ICCs that contain 5 or less insulin⁺ cells was counted. Left picture shows DAPI staining and the pancreatic duct (D) is identified by white stippled line. Right picture shows the mature islet (I) and an ICC (yellow arrow) identified upon immunofluorescence staining with insulin antibodies. (B) Number of ICCs was significantly increased before and after PX in Cdk4 ^R/R mice (closed bars) compared with Cdk4 ^WT mice (open bars). Values are the results obtained from four mice per group at each time point and expressed as mean number of ICCs per section. Error bars represent S.E. *p<0.05, **p<0.01, and ***p<0.001 vs Cdk4 ^WT mice.

Figure 7. Proposed model of Cdk4-p16^Ink4a-regulated regeneration of β-cell mass via effects on the islets and the pancreatic ductal epithelium.

(A) Like all cells, β-cells and pancreatic ductal epithelial cells also traverse the cell cycle. A significant fraction of these cells are in the G0 state referred to as quiescence. Our data suggest that Cdk4 promotes cell cycle re-entry of quiescent cells. In addition, Cdk4 enhances cell cycle progression via stimulating G1 to S-phase transition. The p16^Ink4a-Cdk4 pathway shown is operational during recruitment of quiescent cells into the cell cycle and to promote the G1 to S-phase cycling of cells already in cycle. These events are suppressed by p16^Ink4a and promoted by Cdk4 which inactivates RB thereby releasing E2F to drive S-phase progression. (B) Cdk4 appears to target the islets and the pancreatic ductal epithelium during a regenerative response to pancreatectomy and the proposed model outlines the events orchestrated by Cdk4. Islets are comprised of mature differentiated β-cells and a small fraction of these β-cells undergo proliferation in response to regeneration. Cdk4 promotes limited replication of pre-existing β-cells. In addition, we propose that majority of β-cells are quiescent. In response to pancreatectomy, Cdk4 recruits the quiescent cells into an active cell cycle, thereby increasing the pool of actively-dividing cells within the islet. Cdk4 appears to elicit a slightly different program in ducts in response to pancreatectomy. The duct is mainly comprised of cytokeratin-positive (CK+) epithelial cells, with rare proliferating cells, a limited number of cells that express the early pancreatic development marker, Pdx-1, and rare duct-associated ICCs. In response to pancreatectomy, the proliferation activity, Pdx-1 expression, and numbers of ICCs are increased by the cell cycle progression in addition to recruitment of quiescent ductal cells into the active cell cycle.