Report

Increased Insulin Sensitivity and Obesity Resistance in Mice Lacking the Protein Tyrosine Phosphatase-1B Gene

See allHide authors and affiliations

Science  05 Mar 1999:
Vol. 283, Issue 5407, pp. 1544-1548
DOI: 10.1126/science.283.5407.1544

Abstract

Protein tyrosine phosphatase–1B (PTP-1B) has been implicated in the negative regulation of insulin signaling. Disruption of the mouse homolog of the gene encoding PTP-1B yielded healthy mice that, in the fed state, had blood glucose concentrations that were slightly lower and concentrations of circulating insulin that were one-half those of their PTP-1B+/+ littermates. The enhanced insulin sensitivity of the PTP-1B−/− mice was also evident in glucose and insulin tolerance tests. The PTP-1B−/− mice showed increased phosphorylation of the insulin receptor in liver and muscle tissue after insulin injection in comparison to PTP-1B+/+ mice. On a high-fat diet, the PTP-1B−/− and PTP-1B+/− mice were resistant to weight gain and remained insulin sensitive, whereas the PTP-1B+/+ mice rapidly gained weight and became insulin resistant. These results demonstrate that PTP-1B has a major role in modulating both insulin sensitivity and fuel metabolism, thereby establishing it as a potential therapeutic target in the treatment of type 2 diabetes and obesity.

PTP-1B is implicated in the attenuation of the insulin signal (1). Mice deficient in the heterotrimeric GTP–binding protein subunit Giα 2 exhibit a phenotype of insulin resistance characteristic of type 2 diabetes that correlates with the increased expression of PTP-1B (2). PTP-1B directly interacts with the activated insulin receptor (3), and vanadate, a potent nonselective PTP inhibitor, can function as an insulin mimetic both in vitro and in vivo (4). However, PTPs other than PTP-1B can also dephosphorylate the activated insulin receptor (5). To clarify the role of PTP-1B in insulin action, we generated mice in which the mouse homolog of PTP-1B was disrupted.

The murine gene encoding PTP-1B was cloned from a 129/Sv mouse genomic library and shown to consist of at least nine exons spanning more than 20 kb (6). A targeting vector was designed to delete a portion of the gene that included exon 5 and the tyrosine phosphatase active site in exon 6 and to replace it with the neomycin resistance gene. Two separate embryonic stem cell clones that had undergone homologous recombination and possessed a single integration event were used to microinject Balb/c blastocysts. Chimeric males were mated with wild-type Balb/c females, and heterozygotes from this cross were mated to produce animals homozygous for the PTP-1B mutation (Fig. 1A). Immunoblot analysis of liver microsomes revealed that PTP-1B protein was absent in PTP-1B null mice, and heterozygotes expressed about half the amount of PTP-1B as did wild-type mice (Fig. 1B). PTP-1B−/−, heterozygous, and wild-type littermates were born with the same appearance and with the expected mendelian ratio of 1:2:1. PTP-1B−/−mice grew normally, did not show any significant difference in weight gain as compared to wild-type mice, have lived longer than 1.5 years without any sign of abnormality, and are fertile. Complete necropsies were done on male and female wild-type, heterozygous, and homozygous PTP-1B mutant mice 7 to 8 weeks old, and no gross or histological (brain, liver, muscle, pancreas, and testes) differences were observed.

Figure 1

Gene targeting of the PTP-1B locus. (A) Representative genomic Southern blot analysis of tail DNA digested with Bam HI from a heterozygous cross resulting in wild-type (+/+), heterozygous (+/–), and homozygous (–/–) PTP-1B offspring. (B) PTP-1B immunoblot analysis of liver membrane samples from PTP-1B+/+, PTP-1B+/−, and PTP−/− mice.

If the role of PTP-1B in the insulin signaling pathway is to dephosphorylate the activated insulin receptor, then mice deficient in PTP-1B might have a sustained insulin response because the insulin receptor would remain phosphorylated and hence be activated longer than in PTP-1B+/+ mice. We measured glucose and insulin concentrations in fasted and fed animals (7) (Fig. 2). In the fed state, the PTP-1B−/− mice had a significant 13% reduction in blood glucose concentrations, whereas the heterozygotes had an 8% reduction when compared to wild-type mice (Fig. 2A). The PTP-1B−/−mice had circulating insulin concentrations that were about half those of control fed animals (Fig. 2B). Thus, PTP-1B–deficient mice appeared to be more insulin sensitive, because they maintained lower glucose concentrations with significantly reduced amounts of insulin. In the fasted state, there were no significant differences in concentrations of glucose or insulin.

Figure 2

Serum concentrations of glucose and insulin in animals fed ad libitum or fasted overnight. (A) Glucose and (B) insulin concentrations were determined as described (7). Dark bars indicate fed group [(A) and (B),n = 19 to 21]; light bars indicate fasted group [(A),n = 8 to 10; (B), n = 6]. Values are given as means ± SEM. Statistical analysis was done with a two-tailed, unpaired, Student's t test. Compared to the wild type, ∗P = 0.06 and ∗∗P ≤ 0.01.

We examined insulin sensitivity in PTP-1B−/−, PTP-1B+/−, and PTP-1B+/+ mice with oral glucose and intraperitoneal insulin tolerance tests (8). Administration of a bolus of glucose to PTP-1B−/− mice resulted in a more rapid clearance of glucose than was observed in wild-type mice (Fig. 3A). There was a more pronounced hyperglycemia in the wild-type animals at all time points after glucose administration than in PTP-1B–deficient mice. Increased insulin sensitivity was also observed upon injection of insulin (Fig. 3B). Hypoglycemia was evident 30 and 60 min after injection, but by 120 min after injection glucose concentrations were returning to normal values in wild-type mice, whereas the PTP-1B–deficient mice remained hypoglycemic (P < 0.02). The PTP-1B+/− mice did not show altered glucose tolerance as compared to that in wild-type mice, although glucose concentration at 120 min after injection of insulin appeared to be intermediate between that of wild-type and PTP-1B−/− mice (Fig. 3).

Figure 3

Glucose and insulin tolerance tests in PTP-1B+/+ (diamonds), PTP-1B+/− (squares), and PTP-1B−/−(triangles) mice. (A) Glucose tolerance of male mice 10 to 14 weeks old (n = 11). (B) Insulin tolerance of male mice 10 to 14 weeks old (n = 5 to 6). Data are presented as means ± SEM. Statistical analysis was done with a two-tailed, unpaired, Student'st test. Compared to the wild type, ∗P < 0.05 and ∗∗P < 0.02.

Binding of insulin to its receptor results in autophosphorylation of the receptor on tyrosines 1146, 1150, and 1151 in the kinase regulatory domain (9). This causes activation of the insulin receptor tyrosine kinase, which phosphorylates the various insulin receptor substrate (IRS) proteins that propagate the insulin signaling event (9). If PTP-1B dephosphorylates the activated insulin receptor, then the increased insulin sensitivity observed in the PTP-1B−/− mice may be due to increased or prolonged phosphorylation of the receptor (or both). We therefore measured tyrosine phosphorylation of the insulin receptor in liver and muscle tissue (Fig. 4, A and B) after exposure to insulin (10). The kinetics of insulin receptor phosphorylation in the liver were significantly different in PTP-1B−/− mice than those observed in the PTP-1B+/+ mice. The amount of insulin receptor phosphorylation was the same for both PTP-1B−/− and PTP-1B+/+ animals 1 min after injection. However, by 5 min after injection, phosphorylation decreased to about 50% of maximal in the PTP-1B+/+ animals, whereas no reduction was observed in the PTP-1B−/− mice. A greater effect on insulin receptor phosphorylation was observed in muscle of the PTP-1B−/− mice. The absolute amount of receptor phosphorylation was increased about twofold in muscle from PTP-1B−/− mice as compared to that in muscle from PTP-1B+/+ animals (P < 0.05) (Fig. 4B). The amount of insulin receptor phosphorylation in muscle did not significantly change between 2 to 6 min after insulin injection in either PTP-1B−/− or PTP-1B+/+ samples. No difference in the amount of insulin receptor expression was detected between PTP-1B+/+ and PTP-1B−/− mice, as determined by protein immunoblotting (11).

Figure 4

Increased and prolonged tyrosine phosphorylation of the insulin receptor in PTP-1B−/− mice. (A) Time course of tyrosine phosphorylation of the insulin receptor (IR) β subunit in the liver after insulin treatment in PTP-1B+/+and PTP-1B−/− mice. Quantification of the insulin receptor β subunit phosphotyrosine content from immunoblots was performed by densitometry. Data are presented by setting the amount of phosphorylation at 1 min to 100 and the subsequent amount after 5 min for the same animal relative to this value. The results are from five PTP-1B−/− and PTP-1B+/+ mice each, from three separate experiments. (B) Phosphorylation of the insulin receptor β subunit in muscle of insulin-treated PTP-1B+/+ and PTP-1B−/− mice. The quantified data from immunoblots (n = 6, from two separate experiments) are presented as arbitrary units. (C) Insulin-stimulated tyrosine phosphorylation of IRS-1 from muscle of PTP-1B+/+ and PTP-1B−/− mice 2 min after injection (n = 3). The quantified data are presented as arbitrary units. Data are means ± SEM. Statistical analysis was done with a two-tailed, unpaired, Student's t test comparing in (A) the 5-min to the 1-min time point value and in (B) the PTP-1B−/− mice 2- and 6-min time point values to the respective values of the PTP-1B+/+ mice (∗P < 0.05).

To confirm that the increased phosphorylation of the insulin receptor in the muscle of insulin-treated PTP-1B−/− mice reflects increased kinase activity, phosphorylation of IRS-1 was also measured (12) (Fig. 4C). Phosphorylation of IRS-1 was increased in insulin-treated muscle from PTP-1B−/− mice in comparison to muscle from wild-type animals (P < 0.05). We have also examined phosphorylation of another receptor tyrosine kinase, the epidermal growth factor receptor, and found no difference between PTP-1B wild-type and deficient animals (11). The increased insulin receptor phosphorylation in muscle and its sustained phosphorylation in the liver probably accounts for the enhanced insulin sensitivity observed in the PTP-1B−/− mice.

The disruption of the PTP-1B gene demonstrates that altering the activity of PTP-1B can modulate insulin signaling in vivo. To determine the effect of the loss of PTP-1B activity on insulin resistance, PTP-1B–deficient, wild-type, and heterozygote littermates were subjected to a diet high in fat (50% of calories from fat) and calories (5286 kcal kg–1; Bioserve, NJ) (13) that normally results in obesity-induced insulin resistance (14). During the 10 weeks the mice were on this diet, male and female wild-type littermates rapidly gained weight, whereas PTP-1B−/− and PTP-1B+/− mice were substantially protected from diet-induced weight gain (Fig. 5). The amount of food consumed by the animals did not differ, which indicates that decreased expression of PTP-1B (heterozygotes have about half the amount of PTP-1B as that in wild-type animals) can influence dietary-induced obesity.

Figure 5

Resistance of PTP-1B null and heterozygous mice to diet-induced obesity. The percent weight gain of male and female wild-type (diamonds), heterozygous (squares), and homozygous (triangles) littermates fed a high-fat diet for 10 weeks is shown. The starting weight (male: +/+, 27.6 ± 1.4 g; +/–, 28.5 ± 1.2 g; and –/–, 26.3 ± 1.2 g; female: +/+, 22.1 ± 0.8 g; +/–, 22.2 ± 0.8 g; and –/–, 21.5 ± 0.8 g) and final weight (male: +/+, 41.4 ± 1.3 g; +/–, 37.2 ± 2.0 g; and –/–, 33.5 ± 1.6 g; female: +/+, 33.3 ± 1.7 g; +/–, 27.3 ± 1.3 g; and –/–, 27.2 ± 1.4 g) of animals put on the high-fat diet are indicated. The final weight was significantly different (P< 0.05) for PTP-1B null and heterozygous mice as compared to wild-type mice, except for male wild-type mice versus heterozygotes (P = 0.1).

We examined the effect of the high-fat diet on insulin sensitivity in these animals. Glucose and insulin concentrations in the serum of fasting animals were measured, and glucose and insulin tolerance tests were done on all groups of animals [values for males are presented; females had essentially the same response (11)]. In wild-type mice, the high-fat diet produced a 30% increase (P < 0.05) in fasting blood glucose concentrations and a threefold increase in serum insulin concentrations (Table 1) as compared to wild-type mice on a normal diet. In contrast, the PTP-1B−/− animals maintained glucose and insulin concentrations while on the high-fat diet, which were not significantly different from those in animals on a normal diet (Table 1). PTP-1B heterozygotes on a high-fat diet showed increased fasting concentrations of circulating insulin but had fasting glucose concentrations that were not significantly different from those in animals on a normal diet (Table 1). PTP-1B−/− mice also had enhanced insulin sensitivity as compared to their wild-type littermates in both glucose and insulin tolerance tests (Fig. 6, A and B). The difference in insulin sensitivity between the PTP-1B−/− and PTP-1B+/+ mice became more augmented on the high-fat diet because of the obesity-induced insulin resistance of the wild-type mice. Obesity-induced insulin resistance results in a reduction in insulin receptor phosphorylation and hence in insulin signaling (14). Examination of insulin-stimulated receptor phosphorylation in mice on the high-fat diet revealed that there was a much greater difference in the amount of insulin receptor phosphorylation between the PTP-1B wild-type and deficient animals (Fig. 6C). This increased difference appears to be due to both a reduction in the amount of insulin-stimulated receptor phosphorylation in wild-type mice and a slight increase in the amount of insulin-stimulated receptor phosphorylation in the PTP-1B−/− mice. The PTP-1B heterozygotes on the high-fat diet also appeared to maintain a better response to insulin-stimulated receptor phosphorylation than did wild-type animals (Fig. 6).

Figure 6

Insulin tolerance of mice lacking PTP-1B and of wild-type mice on a high-fat diet. (A) Glucose and (B) insulin tolerance tests of male mice (n= 7 to 8) on a high-fat diet. (C) Insulin-stimulated insulin receptor tyrosine phosphorylation in muscle of fat-fed mice. The quantified data from immunoblots (PTP-1B–/– and PTP-1B+/+, n = 5; and PTP-1B+/−, n = 3; from two separate experiments) are presented as arbitrary units. Data are means ± SEM. Statistical analysis was done with a two-tailed, unpaired, Student's t test (P < 0.05).

Table 1

Fasting glucose, triglyceride, and insulin levels of male wild-type, heterozygote, and PT-1B−/− littermates fed a normal or high-fat diet. Values are given as the means ± SEM. Statistical analysis was done with a two-tailed, unpaired, Student'st test. ND, not determined.

View this table:

The reason for the obesity resistance observed in the PTP-1B−/− mice is unclear at this time, but analysis of triglyceride concentrations indicates that fat metabolism has been affected in these animals. The PTP-1B–/– mice on either diet had significantly lower triglyceride concentrations than did wild-type and heterozygous mice (Table 1). We also examined insulin-stimulated receptor phosphorylation in adipose tissue and found no significant difference between wild-type and PTP-1B-deficient animals on either diet (11). Thus, PTP-1B–deficient mice appear to show tissue-specific insulin sensitivity; muscle and liver have an enhanced insulin sensitivity, whereas adipose tissue remains unchanged relative to wild-type mice. The effects that were observed in adipose tissue could be the result of some compensatory mechanism such as the up-regulation of some other PTP [although no difference was observed in the amount of PTP activity between wild-type and PTP-1B−/− tissue extracts (11)] or could possibly be due to the overall enhanced insulin sensitivity of the PTP-1B–deficient animal.

The data presented identify PTP-1B as having a major role in the insulin signaling pathway. What this function is remains to be clarified, but the simplest explanation would be that it dephosphorylates the activated insulin receptor. Recently, the disruption of the leukocyte antigen–related (LAR) PTP, which has also been suggested to affect the insulin signaling cascade, has been described. The targeted mutagenesis of LAR produced mice with impaired mammary gland development (15) but with blood glucose concentrations within the normal range (16). In contrast, the LAR-deficient mice generated by insertional mutagenesis had body weights that were half those of control mice and were insulin resistant (17). The reason for the difference in phenotype between these two LAR-deficient mice strains is unknown.

We have shown that the loss of PTP-1B activity causes enhanced insulin sensitivity and resistance to weight gain in mice. These results make PTP-1B a potential therapeutic target for the treatment of type 2 diabetes and obesity.

  • * To whom correspondence should be addressed. E-mail: brian_kennedy{at}merck.com

REFERENCES AND NOTES

View Abstract

Stay Connected to Science

Navigate This Article