Figure 1.
Modulation of Perk dosage using Perk+/− mice impacted glucose homeostasis.
A. Random fed serum glucose of mice at 15–36 postnatal days. Mice are in mixed genetic background. Shown are means ± SEM (n = 44, 68; *P<0.05). B. Perk mRNA measurement in mouse islets expressed as percentage of Perk+/+ levels. Shown are means ± SEM (n = 29, 26; ***P<0.001). C. Western blot of PERK protein in mouse islets. Left panel showed representative blots. Right panel showed quantification of PERK protein levels in relative to Perk+/+. (n = 5, 5; ***P<0.001). D. Western blot of phosphorylated eIF2α in mouse islets. Left panel showed representative blots. Right panel showed quantification of PERK protein levels in relative to Perk+/+. (n = 5, 4; **P<0.01). E. Immuno-staining of pancreatic section from Perk+/−, Perk+/+, and Perk KO mice using insulin and glucagon antibody.
Figure 2.
Glucose and insulin homeostasis were impacted by modulation of Perk during development.
A. Random fed serum glucose of Perk+/− mice relative to Perk+/+ during postnatal development. Age of mice was shown as indicated, except that mice belonging to P75 group were from P70 to P90. Mice are in C57BL/6 background. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 18, 16 (P5); 16, 17 (P17); 13, 11 (P30); 12, 16 (P50); 17, 18 (P75). *P<0.05, ***P<0.001). B. Random fed serum glucose of Perk+/− mice relative to Perk+/+ at 9–12 month. Mice are in C57BL/6, 129 SvEvTac or mixed background. Shown are means ± SEM. (n for Perk +/−, Perk+/+ = 4, 5 (B6); 23, 20 (129); 5, 3. **P<0.01). C. Random fed serum insulin of Perk+/− mice relative to Perk+/+ during postnatal development. Age of mice was shown as indicated, except that mice belonging to P75 group were from P70 to P90. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 28, 28 (P5); 19, 17 (P17); 25, 18 (P30); 56, 75 (P50); 29, 32 (P75). *P<0.05). D. Glucose tolerance test of P17 mice after 4 hour fasting. Serum glucose was measured at various time points after glucose injection as indicated above. GTT traces are shown as means ± SEM of each time point and quantified by calculating area under the curve (AUC) in relative to Perk+/+ (n for Perk +/−, Perk+/+ = 5, 4). E. Serum insulin collected in P17 mice 30 minutes after injection with glucose or vehicle (n for Perk +/−, Perk+/+ = 5, 4. P = 0.16 for stimulated insulin levels). F. Glucose tolerance test of P50 mice after 16 hour fasting. Experiment and analysis were as in 2D (n for Perk +/−, Perk+/+ = 18, 18, *P<0.05).
Figure 3.
Insulin production was modulated by Perk during development.
A. Total pancreatic insulin content of Perk+/− mice relative to Perk+/+ during postnatal development. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 14, 16 (P5); 8,5 (P17); 18, 16 (P30); 10, 15 (P50); 8, 6 (P180). *P<0.05). B. β-cell volume of Perk+/− mice relative to Perk+/+ during postnatal development. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 3, 4 (P17); 4, 4 (P35); 4, 3 (P50); 3, 3(P75). No statistical significance was observed at any developmental time point). C. Estimated β-cell insulin concentration of Perk+/− mice relative to Perk+/+ during postnatal development. Age of mice was shown as indicated, except that mice belonging to P75 group had ages ranging from P70 to P90. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 7,6 (P5); 9, 11 (P17); 6, 9 (P30); 11, 16 (P50); 9, 8(P75). *P<0.05, **P<0.01). D. Estimated β-cell number of Perk+/− mice relative to Perk+/+ using insulin content in the whole pancreas divided by estimated insulin content per cell.
Figure 4.
P50 Perk+/− mice exhibit higher β-cell number due to elevated β-cell proliferation.
A. Gene expression levels of islets or whole pancreata from P50 Perk+/− mice relative to Perk+/+ control. Shown are means ± SEM. Below the bar graph shows the estimated fold-increase of β-cell number of P50 Perk+/− mice relative to Perk+/+ control using each β-cell specific gene (islet data: n = 3, 4; pancreas data: n = 6, 4. *P<0.05). B. Daily BrdU incorporation rate of β-cells in mice at various ages. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 5, 5 (P5); 3, 4 (P17); 10, 10 (P30); 4, 4 (P50); 3, 3 (P70). *P<0.05).
Figure 5.
Insulin transcription was up-regulated in P17 Perk+/− mice.
A. Gene expression levels of mouse islets relative to Perk+/+. Mice are postnatal 17 days. Shown are means ± SEM. (n for Perk +/−, Perk+/+ = 11,9. *P<0.05). B. Gene expression level of mouse islets relative to Perk+/+. Islets were isolated from mice at various ages as indicated. Shown are means ± SEM. (n for Perk +/−, Perk+/+ = 8, 8 (P35); 3, 4 (P50); 6, 6(1 year). No significant difference was seen for any gene at any age).
Figure 6.
New proinsulin synthesis and mature proinsulin level were up-regulated in P17 Perk+/− mice.
A. Newly synthesized proinsulin in mouse islets measured by western blot. Mice are postnatal 17 days. 10 µg/ml puromycin was added 15 minutes before protein samples were harvested. Left panel showed representative blots probed with anti-puromycin and co-localized with proinsulin C-peptide. Right panel showed quantification of blot intensity relative to Perk+/+. (n for Perk +/−, Perk+/+ = 5, 4). B. Total pancreatic proinsulin content of Perk+/− mice relative to Perk+/+ at various postnatal stages. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 8, 8 (P5); 8,7 (P17); 18, 17 (P30). *P<0.05). C. Proinsulin and insulin were measured in the pancreatic samples harvested from P17 mice and proinsulin/insulin ratio was calculated. Shown are means ± SEM (n for Perk +/−, Perk+/+ = 8,7. *P<0.05).
Figure 7.
ER chaperones were impacted in P17 Perk+/− mice.
A–C. Gene expression levels of mouse islets relative to Perk+/+. Mice are postnatal 17 days (figure A, n = 11,9), 35 days (figure B, n = 8, 8), and 50 days (figure C, n = 4,3). Shown are means ± SEM. (*P<0.05, **P<0.01). D. Western blot analysis of islets isolated from P17 mice. Left panel showed representative blots. Right panel showed quantification of blot intensity in relative to Perk+/+. (n for Perk +/−, Perk+/+ = 5, 5. *P<0.05, **P<0.01).
Figure 8.
Perk gene dosage specifically in the β-cells regulates glucose homeostasis and insulin secretion.
A. Random fed serum glucose of liver-specific Perk KO mice (liPerk−/−) or liPerk+/+ at 70 days old. (n = 8, 8. A trend towards high blood glucose was seen in liPerk−/− mice but was not significantly different from wild-type control mice.). B. Random fed serum glucose of pancreatic-specific Perk+/− (pcPerk+/−) or pcPerk+/+ at age of P17-P35. Mice are in C57B/L6 background (n for pcPerk+/−, pcPerk+/+ = 18, 43. **P<0.01). C. Random fed serum glucose of Perk+/+;βPerk mice or wild-type littermates at age of P21-P38. Mice are in C57B/L6 background (n for Perk+/+, Perk+/+;βPerk = 63, 36. *P<0.05). D. Random fed serum insulin of Perk+/+;βPerk mice or wild-type littermates at age of P27-P64. Serum insulin of each mouse was normalized to the average of wild-type littermates. Mice are in C57B/L6 background (n for Perk+/+, Perk+/+;βPerk = 44, 16. ***P<0.001). E. Perk mRNA measurement in INS1 832/13 shPerk β-cells pre-treated with indicated concentration of doxycycline for 24 hours. Perk mRNA level was normalized to the level of non-doxycycline control (n = 4 per dosage point). F. Insulin secretion in response to 2.8 mM or 20 mM glucose in INS1 832/13 shPerk β-cells pre-treated with indicated concentration of doxycycline for 24 hours. Insulin secretion was normalized to total protein and expressed as fold increase in relative to basal insulin secretion (2.8 mM glucose) of non-doxycycline control. Shown are means ± SEM. (n≥4 per treatment. Statistical significance was determined by comparing to no-doxycycline control. * P<0.05, *** P<0.001). G. Gene expression measurement in INS1 832/13 shPerk β-cells pre-treated with indicated concentration of doxycycline for 24 hours. The mRNA level of each gene was normalized to the level of no-doxycycline control (n = 4 per dosage point. Statistical significance was determined by comparing to no-doxycycline control. *P<0.05, **P<0.01).
Figure 9.
Developmental summary and model of PERK-dependent regulation of β-cell development and glucose homeostasis.
A. Genotypic difference of serum glucose and insulin throughout development. Figure 9A was generated based on the data in Fig 2A and Fig 2C. Serum glucose and insulin are inversely correlated during postnatal development of Perk+/− mice. B. Genotypic difference of total pancreatic insulin content (based on data in Fig 3A), insulin content per β-cell (Fig 3C), β-cell number (Fig 3D), and β-cell proliferation rate (Fig 7B). Initially β-cell number is relatively low in Perk+/−, but β-cell and total pancreatic insulin are high. In response to low β-cell number, β-cell proliferation is accelerated between postnatal 17–50 resulting in increased β-cell number in Perk+/−. However, as β-cell number rises, insulin content per β-cell drops resulting in a balance between cell number and insulin concentration in each cell and a return to equivalent levels of total pancreatic insulin content.