Reduction of hypothalamic protein tyrosine phosphatase improves insulin and leptin resistance in diet-induced obese rats

Abstract

Protein tyrosine phosphatase (PTP1B) has been implicated in the negative regulation of insulin and leptin signaling. PTP1B knockout mice are hypersensitive to insulin and leptin and resistant to obesity when fed a high-fat diet. We investigated the role of hypothalamic PTP1B in the regulation of food intake, insulin and leptin actions and signaling in rats through selective decreases in PTP1B expression in discrete hypothalamic nuclei. We generated a selective, transient reduction in PTP1B by infusion of an antisense oligonucleotide designed to blunt the expression of PTP1B in rat hypothalamic areas surrounding the third ventricle in control and obese rats. The selective decrease in hypothalamic PTP1B resulted in decreased food intake, reduced body weight, reduced adiposity after high-fat feeding, improved leptin and insulin action and signaling in hypothalamus, and may also have a role in the improvement in glucose metabolism in diabetes-induced obese rats.

Figure 1

Immunoblotting and immunohistochemical evaluation of PTP1B distribution and expression in hypothalamus of rats treated with PTP1B-ASO. A, Immunoblotting with anti-PTP1B antibody of whole-tissue extracts from hypothalamus of rats treated with PTP1B-ASO (0, 0.5, 1, 2, 4 nm) for 4 d. B, Immunoblotting with anti-PTP1B antibody of whole-tissue extracts from hypothalamus of control and DIO rats treated with PTP1B-ASO. C, Paraformaldehyde-fixed rat hypothalamic sections (5 μm) were incubated with anti-PTP1B (visualization of immunoreactivity as described in Materials and Methods). D, Imunnoblotting of PTP1B from different areas of the CNS. Data are means ± sem of four independent experiments, i.e. four different cohorts of control rats or DIO rats. *, Control or DIO rats: P < 0.05, PTP1B-ASO vs. PTP1B sense.

Figure 2

Effect of icv PTP1B-ASO on metabolic parameters. A, Twenty-four hour food intake. B, Body weigh. C, Epididymal and retroperitoneal adipose tissue mass. D, serum leptin levels. E, Serum insulin levels in controls and DIO rats treated with PTP1B-ASO or PTP1B-sense or in pair-fed rats. Data are means ± sem. Each group was composed of 10–15 animals. *, Control or DIO rats: P < 0.05, PTP1B-ASO vs. PTP1B-sense; &, P < 0.05 pair-fed vs. PTP1B-ASO; #, P < 0.05 pair-fed vs. PTP1B-sense; **, P < 0.01 PTP1B-ASO vs. PTP1B sense.

Figure 3

Effect of icv PTP1B-ASO on leptin-induced satiety and leptin signaling in hypothalamus of control and DIO rats. A, leptin was icv infused, and 12-h food intake was measured in control and DIO rats treated with PTP1B-ASO or PTP1B-sense or in pair-fed rats. B, Immunoprecipitation (IP) with anti-Jak-2 and immunoblotting (IB) with anti-pY antibodies. C, IP with anti-Jak-2 and IB with anti-PTP1B. D, IP with anti-ObR and IB with anti-pY. E, IP with anti-Stat3 and IB with anti- pY. F, IP with anti-Jak-2 and IB with anti-pY. G, IP with anti-Stat3 and IB with anti-pY. Specific bands were densitometrically quantified. Data are means ± sem of four independent experiments, i.e. four different cohorts of control rats or DIO rats. The comparisons between the groups were performed by using the same time point in the respective group: *, P < 0.01 PTP1B-ASO vs. PTP1B sense; #, P < 0.05 PTP1B-sense vs. PTP1B sense; &, P < 0.05 pair-fed vs. PTP1B-ASO; $, P < 0.05 pair-fed vs. PTP1B-sense.

Figure 4

Effect of icv PTP1B-ASO on insulin-induced satiety and insulin signaling in hypothalamus of control and DIO. A, insulin was icv infused, and 12-h food intake was measured in control and DIO rats treated with PTP1B-ASO or PTP1B-sense or in pair-fed animals. B, Immunoprecipitation (IP) with anti-IR and immunoblotting (IB) with anti-pY antibodies. C, IP with anti-IR and IB with anti-PTP1B. D, IP with anti-IRS-1 and IB with anti-pY. E, IP with anti-IRS-2 and IB with anti-pY. F, IP with anti-IRS-1 and IB with anti- pY. G, IP with anti-IRS-2 and IB with anti-pY. Specific bands were densitometrically quantified. Data are means ± sem of four independent experiments, i.e. four different cohorts of control rats or DIO rats. Control or DIO rats comparing the same time in the respective group: *, P < 0.01 PTP1B-ASO vs. PTP1B sense; #, P < 0.05 PTP1B-sense vs. PTP1B-sense; &, P < 0.05 pair-fed vs. PTP1B-ASO; $, P < 0.05 pair-fed vs. PTP1B-sense.

Figure 5

A, Steady-state glucose infusion rates obtained from averaged rates of 90–120 min of 10% unlabeled glucose infusion during hyperinsulinemic-euglycemic clamp procedures in the control, DIO, and pair-fed rats. B, Glucose transport in skeletal muscle tissue evaluated by 2- deoxyglucose uptake during the last 45 min of the hyperinsulinemic-euglycemic clamp studies. C, Basal and insulin-stimulated rates of hepatic glucose production during the hyperinsulinemic euglycemic clamp procedures in awake rats. Data are means ± sem of four independent experiments, i.e. four different cohorts of control rats or DIO rats. DIO or control rats comparing the same time in the respective group: *, P < 0.05 PTP1B-ASO vs. PTP1B-sense; &, P < 0.05 pair fed vs. PTP1B-ASO; $, P < 0.05 pair-fed vs. PTP1B-sense.

Figure 6

Effect of icv PTP1B-ASO on PTPase and PTP1B activities in control and DIO rats. Total PTPase activity and PTPase activities in immunoprecipitates were assayed by incubation with the pp60c-^src C-terminal phosphoregulatory peptide (TSTEPQpYQPGENL), as described in Materials and Methods. A, Total PTPase activity in hypothalamus. B, PTPase activity in PTP1B immunoprecipitates. C, PTPase activity in IR immunoprecipitate. D, PTPase activity in Jak-2 immunoprecipitates. Data are means ± sem of four independent experiments, i.e. four different cohorts of control rats or DIO rats. Control or DIO rats comparing the same time in the respective group: *, P < 0.05 PTP1B-ASO vs. PTP1B-sense; #, P < 0.05 PTP1B-sense vs. PTP1B-sense; &, P < 0.05 pair-fed vs. PTP1B-ASO; $, P < 0.05 pair-fed vs. PTP1B-sense.