Targeted disruption of pancreatic-derived factor (PANDER, FAM3B) impairs pancreatic beta-cell function

Abstract

Objective: Pancreatic-derived factor (PANDER, FAM3B) is a pancreatic islet-specific cytokine-like protein that is secreted from beta-cells upon glucose stimulation. The biological function of PANDER is unknown, and to address this we generated and characterized a PANDER knockout mouse.

Research design and methods: To generate the PANDER knockout mouse, the PANDER gene was disrupted and its expression was inhibited by homologous recombination via replacement of the first two exons, secretion signal peptide and transcriptional start site, with the neomycin gene. PANDER(-/-) mice were then phenotyped by a number of in vitro and in vivo tests to evaluate potential effects on glucose regulation, insulin sensitivity, and beta-cell morphology and function.

Results: Glucose tolerance tests demonstrated significantly higher blood glucose levels in PANDER(-/-) versus wild-type male mice. To identify the mechanism of the glucose intolerance, insulin sensitivity and pancreatic beta-cell function were examined. Hyperinsulinemic-euglycemic clamps and insulin tolerance testing showed similar insulin sensitivity for both the PANDER(-/-) and wild-type mice. The in vivo insulin response following intraperitoneal glucose injection surprisingly produced significantly higher insulin levels in the PANDER(-/-) mice, whereas insulin release was blunted with arginine administration. Islet perifusion and calcium imaging studies showed abnormal responses of the PANDER(-/-) islets to glucose stimulation. In contrast, neither islet architecture nor insulin content was impacted by the loss of PANDER. Interestingly, the elevated insulin levels identified in vivo were attributed to decreased hepatic insulin clearance in the PANDER(-/-) islets. Taken together, these results demonstrated decreased pancreatic beta-cell function in the PANDER(-/-) mouse.

Conclusions: These results support a potential role of PANDER in the pancreatic beta-cell for regulation or facilitation of insulin secretion.

FIG. 1.

Generation and confirmation of the PANDER knockout (KO) mouse. A: Schematic representation of genomic arrangement for mouse PANDER (top), targeting vector (middle), and the disrupted locus (bottom) produced by homologous recombination. The targeting vector was created by cloning a 4.3-kb long arm and 2.9-kb short arm of flanking PANDER genomic sequence surrounding the neomycin (NEO) gene in opposite orientation, intended to disrupt the transcriptional start site (TSS), secretion signal peptide (SP), and exons 1 and 2. Primers 5′-F/5′-R and 3′-F/3′-R were used for confirmation of correct gene targeting, resulting in a 4.3-kb and 3.5-kb product, respectively. Primers 5′-F/3′-R were used for routine PCR genotyping producing a 12.1-kb product for wild-type (WT), 12.1-kb/8.3-kb products for heterozygote (HET), and single 8.3 kb for KO. B: PCR analysis of genomic DNA extracted from transfected ES cells. Primer locations are shown in Fig. 1A. Lane 1: molecular marker; lanes 2 and 3: 4.3-kb arm amplified from two independent correctly targeted ES cell clones; lanes 4 and 5: 3.5-kb arm from correctly targeted ES cell clones; lanes 6 and 7: lack of 5′ and 3′ arm amplification from negative ES clones; and lane 8: 4.3-kb product from artificial template consisting of a plasmid containing the 5′ arm sequence with flanking nontargeted genomic regions. C: PCR confirmation of germline transmission. Genomic DNA was extracted from tail biopsies of mouse offspring with subsequent PCR using the 5′-F/5′-R primers. Lane 1: molecular marker; lanes 2, 3, 5, and 8: mouse offspring negative for germline transmission; and lanes 4, 6, 7: mouse offspring with targeted PANDER deletion. D: Genotyping of mouse offspring. Representative PCR results are shown among WT, HET, and KO mice. E: RT-PCR analysis of RNA isolated from PANDER-positive tissues from WT, HET, and KO mice. F: Western evaluation of protein isolated from pancreatic islets for PANDER (top) and β-actin (bottom) expression from WT, HET, and KO mice.

FIG. 2.

Metabolic evaluation of PANDER^−/− mice. A: Intraperitoneal GTT performed by injecting mice with glucose at 2 g/kg and measuring serum glucose concentration at the indicated time points (n = 14). B: AUC calculated from glycemic levels measured during the course of the GTT. C: Intraperitoneal insulin tolerance test performed by injecting mice with insulin at 0.75 units/kg and measuring glucose concentration at indicated time points (n = 8). D: Mean steady-state glucose infusion rate (GIR) during hyperinsulinemic-euglycemic clamp (n = 8). E: Measured insulin levels during GTT (n = 14). F: Intraperitoneal arginine tolerance test. Arginine was injected at 2 g/kg (n = 8–9). For all experiments above, male mice ∼4–6 months old were evaluated. Values are means ± SE. *P < 0.05. **P < 0.01 by Student t test. n.s., not significant; KO, knockout; WT, wild-type.

FIG. 3.

HGP during hyperinsulinemic-euglycemic clamp is impaired in PANDER^−/− mice. HPG during the hyperinsulinemic-euglycemic clamps was determined by subtracting the glucose infusion rate from the whole-body glucose appearance. Male mice ∼5–7 months old were evaluated (n = 8). A: Basal HGP and production at the end of the hyperinsulinemic-euglycemic clamps. B: HGP suppression. Values are means ± SE. *P < 0.05 by Student t test. KO, knockout; WT, wild-type.

FIG. 4.

Normal islet architecture and gene expression in PANDER^−/− mice. Pancreatic sections from 4- to 6-month-old male mice were evaluated via immunofluorescence for detection of insulin (green) and glucagon (red). DAPI nuclear staining is shown in blue. A: Representative micrographs of insulin- and glucagon-labeled pancreatic sections from PANDER WT (left panel) and PANDER^−/− mice (right panel), respectively. B: Gene expression analysis via RT-PCR in isolated islets of PANDER^−/− and WT mice. Various genes involved in GSIS were evaluated in 5- to 7-month-old male mice (n = 3). Values are means ± SE. P = not significant for all genes as determined by Student t test. KO, knockout. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 5.

Islet perifusions demonstrate impaired insulin secretion in PANDER^−/− mice. A: Pancreatic islets were isolated and evaluated from male mice ∼5–7 months old. Islets were isolated and incubated with increasing concentrations of glucose of 0 mmol/l (G0), 2 mmol/l (G2), 5 mmol/l (G5), 10 mmol/l (G10), and 30 mmol/l (G30) with a terminal step of 30 mmol/l KCl (KCl30) at indicated times. Time at which glucose condition was increased is shown above and noted on the x-axis. Effluent fractions were collected and insulin was measured by radioimmunoassay (n = 6–7). B: AUC calculated for insulin secretion from the G0 to the G30 condition. C: AUC determined for the insulin secretion during the KCl condition. D: After conclusion of the perifusion experiment, islets were collected and lysed with acid alcohol, and subsequent measurements of insulin concentrations were taken via radioimmunoassay. Values are means ± SE. *P < 0.05 by Student t test.

FIG. 6.

Abnormal calcium response to glucose stimulation in PANDER^−/− islets. Intracellular calcium concentration was determined via fura-2 fluorescence imaging during a glucose perifusion on isolated PANDER^−/− and WT islets with stimulation from a glucose ramp and KCl (n = 3). Time of glucose increase is shown above and on the x-axis. Representative calcium plots are shown. Values are means ± SE.

FIG. 7.

Decreased insulin clearance in PANDER^−/− mice. Intraperitoneal GTT performed by injecting mice with glucose at 2 g/kg and measuring serum insulin and C-peptide concentrations at the indicated time points. Male mice ∼5–7 months old were evaluated (n = 3–6). Insulin (A) and C-peptide (B) serum concentrations were determined with commercially available ELISA (ALPCO). C: C-peptide–to–insulin ratio based on AUC calculated from glycemic levels of insulin and C-peptide measured during the course of the GTT. Values are means ± SE. *P < 0.05 by Student t test.