Mice lacking dipeptidyl peptidase IV are protected against obesity and insulin resistance Stacey L. Conarello*, Zhihua Li*, John Ronan†, Ranabir Sinha Roy*, Lan Zhu*, Guoqiang Jiang*, Franklin Liu*, John Woods‡, Emanuel Zycband‡, David E. Moller*, Nancy A. Thornberry*, and Bei B. Zhang*§ Departments of *Metabolic Disorders and Molecular Endocrinology, †Comparative Medicine, and ‡Immunology and Inflammation, Merck Research Laboratories, Rahway, NJ 07065 Communicated by Roger H. Unger, University of Texas Southwestern Medical Center, Dallas, TX, March 29, 2003 (received for review December 9, 2002)
T
ype 2 diabetes is a complex metabolic disorder characterized by abnormal insulin secretion caused by impaired  cell function and insulin resistance in target tissues (1, 2). The worldwide prevalence of type 2 diabetes is reaching epidemic proportions, with an expected total of 221 million cases by the year 2010 (3). The single most important risk factor for the pathogenesis of diabetes is obesity and its associated insulin resistance. An effective treatment for diabetes clearly is lacking (4), underscoring the importance of identifying new therapeutic targets for the disease. The underlying causes of obesity are complicated and involve the interaction of environmental and genetic factors that mediate energy intake and expenditure (5–7). Overnutrition in the form of excessive intake of dietary fat contributes greatly to the development of obesity, as indicated by a number of epidemiological studies (8, 9). The molecular mechanisms linking overnutrition and adiposity are not entirely clear; however, recent studies have implicated the involvement of a complex network of hormones derived from peripheral tissues and neuropeptides in the control of body weight (5, 7, 10, 11). Dipeptidyl peptidase IV (DP-IV or CD26) is a multifunctional glycoprotein that contains N-terminal serine dipeptidase activity and is present both in circulation and on the cell surface (12). DP-IV has been implicated in pleiotropic cellular processes involving immune, inflammatory, and endocrine functions (12, 13) and has been shown to cleave several hormones and chemokines in vitro. The best validated in vivo substrates are members of the glucagon family of peptides, including glucagonlike peptide 1 (GLP-1), GLP-2, and glucose-dependent insulinotropic polypeptide (14), which are inactivated by cleavage after their penultimate Ala residue. GLP-1 has multifaceted actions, including glucose-induced stimulation of insulin biosynthesis and secretion, inhibition of glucagon secretion, regulation www.pnas.org兾cgi兾doi兾10.1073兾pnas.0631828100
of gene expression, trophic effects on  cells, inhibition of food intake, and slowing of gastric emptying. These effects contribute to the normalization of elevated blood glucose and the control of satiety and body weight (15–17). In addition, GLP-2 and glucose-dependent insulinotropic polypeptide signaling have been implicated in the regulation of energy balance (18, 19). Previously, investigators have shown that upon glucose challenge, DP-IV⫺兾⫺ mice have improved glucose tolerance, increased insulin, and increased active GLP-1, consistent with the role of DP-IV in the regulation of GLP-1 (20). Here, we investigate the effect of chronic ablation of DP-IV on metabolic control and show that DP-IV⫺兾⫺ mice are resistant to dietinduced obesity and associated insulin resistance. Materials and Methods Experimental Procedures. All animal procedures were performed
in accordance with the guidelines of the Institutional Animal Care and Use Committee.
Animal Studies. The generation of DP-IV⫺兾⫺ mice used in this
study has been described, and the mice were inbred on C57BL兾6 background (20). Male WT (C57BL兾6) and DP-IV⫺兾⫺ mice were housed 10 per cage in a room maintained at constant temperature (25°C or 70°F). Four-week-old mice were fed ad libitum a low- or high-fat diet (LFD or HFD, respectively), with 10% or 45% of Kcal from fat, respectively (Research Diets, New Brunswick, NJ). Individual body weights were measured over a 20-week period. Body composition and hepatic fat content were assessed by a Fat兾Lean Mice Whole Body Magnetic Resonance Analyzer (Bruker, Billerica, MA). For treatment with streptozotocin (STZ), mice were fed a HFD for 8 weeks, injected with STZ i.p. at 85 mg兾kg, and then fed a HFD for an additional 4 weeks. Pair-Feeding Study. Mice were housed individually and permitted to acclimate to this condition for 1 week before food-intake measurements and maintained on a HFD. Food intake amounts for ad libitum fed DP-IV⫺兾⫺ and WT mice were measured daily starting at ⬇5 weeks of age. Pair-feeding was accomplished by measuring the food intake of the ad libitum fed DP-IV⫺兾⫺ mice every 24 h (just before onset of the dark cycle daily). The following day, the individually housed WT mice (pair-fed group) were given the average amount of food consumed by the knockout mice on the previous day. Body weight was measured weekly. Indirect Calorimetry. Male DP-IV⫺兾⫺ and WT mice, 11–12 weeks of
age, were fed a HFD for 6 weeks and investigated in indirect calorimeters. Oxygen consumption (VO2) was determined simulAbbreviations: BAT, brown adipose tissue; DP-IV, dipeptidyl peptidase IV; GLP, glucagonlike peptide; FD, fat diet; HFD, high FD; LFD, low FD; STZ, streptozotocin; SREBP-1C, sterol regulatory element-binding protein-1c; PPAR␣, peroxisome proliferator-activated receptor ␣. §To
whom correspondence should be addressed. E-mail: bei㛭
[email protected].
PNAS 兩 May 27, 2003 兩 vol. 100 兩 no. 11 兩 6825– 6830
PHYSIOLOGY
Dipeptidyl peptidase IV (DP-IV), a member of the prolyl oligopeptidase family of peptidases, is involved in the metabolic inactivation of a glucose-dependent insulinotropic hormone, glucagon-like peptide 1 (GLP-1), and other incretin hormones. Here, we investigated the impact of DP-IV deficiency on body weight control and insulin sensitivity in mice. Whereas WT mice displayed accelerated weight gain and hyperinsulinemia when fed a high-fat diet (HFD), mice lacking the gene encoding DP-IV (DP-IVⴚ兾ⴚ) are refractory to the development of obesity and hyperinsulinemia. Pair-feeding and indirect calorimetry studies indicate that reduced food intake and increased energy expenditure accounted for the resistance to HFD-induced obesity in the DP-IVⴚ兾ⴚ mice. Ablation of DP-IV also is associated with elevated GLP-1 levels and improved metabolic control in these animals, resulting in improved insulin sensitivity, reduced pancreatic islet hypertrophy, and protection against streptozotocin-induced loss of  cell mass and hyperglycemia. Together, these observations suggest that chronic deletion of DP-IV gene has significant impact on body weight control and energy homeostasis, providing validation of DP-IV inhibition as a viable therapeutic option for the treatment of metabolic disorders related to diabetes and obesity.
Fig. 1. Disruption of DP-IV prevents HFD-induced obesity. (A) Body weight gain in WT and DP-IV⫺兾⫺ mice on 10% FD or 45% FD (n ⫽ 17). (B) Histological analyses of epididymal white adipose tissues. (C) Histological analyses of intrascapular BAT. P ⬍ 0.05, comparing WT with DP-IV⫺兾⫺ mice on 45% FD (*) or 10% FD (#).
taneously for multiple animals by indirect calorimetry using an Oxymax System (Columbus Instruments, Columbus, OH). Measurements were taken for a 23-h period, including an 11-h light cycle and a 12-h dark cycle. Data were normalized to body weight. Histological and Immunohistochemical Analysis. Adipose tissue and pancreas were fixed in 10% buffered formaldehyde and embedded in paraffin standard hematoxylin兾eosin procedures or with antiinsulin and antiglucagon antibodies. Blood and Plasma Parameters. Plasma insulin and leptin concen-
trations were measured with ELISA kits (Alpco, Windham, NH). Plasma GLP-1 levels were measured with ELISA kits (Linco Research Immunoassay, St. Charles, MO).
RNA Preparation and Quantitative RT-PCR. Tissues were extracted with Trizol (Life Technologies, Rockville, MD), and total RNA was isolated from each sample according to the manufacturer’s instructions. Amplification of each target cDNA was performed with TaqMan PCR Reagent Kits in the ABI Prism 7700 Sequence Detection System according to the protocols provided by the manufacturer (Applied Biosystems). The following forward primer兾reverse primer兾probe sets were used for the amplification step: -3 receptor, gggaggcaacctgctggta兾gaagtcacgaacacgttggttatg兾cgcccgcacgccgagact; UCP1 (uncoupling protein), ccttcccgctggacactg兾cctaggacacctttatacctaatggt兾caaagtccgccttcagatccaaggtg; Peroxisome proliferator-activated receptor ␣ (PPAR␣), ttcaatgccttagaactggatgac兾 aatgtagcctatgtttagaaggcca兾ccgatctccacagcaaattatagcagcca; SREBP-1c (sterol regulatory element-binding protein-1c), catcgact acatc cgcttcttg兾ttttgtgtgcacttcgt agggt agcaac cagaagctca. The level of mRNA for each gene was normalized to the level of 18S ribosomal RNA detected in each sample.
DP-IV⫺兾⫺ mice compared with WT was significant by the third week and was maintained throughout the 20-week study period (terminal body weight of 46.0 ⫾ 0.7 g for WT vs. 38.5 ⫾ 1.6 g for DP-IV⫺兾⫺, P ⫽ 0.0002, n ⫽ 17). Additionally, the body weight in LFD-fed DP-IV⫺兾⫺ mice was lower than the WT counterparts. These results indicate that DP-IV⫺兾⫺ mice are protected from HFD-induced accelerated weight gain. As expected, WT mice maintained on the HFD had increased total fat mass and percent fat content relative to the LFD group (Table 1). These parameters were significantly lower for HFDfed DP-IV⫺兾⫺ mice. Lean body mass was comparable for all groups. Histological analysis indicated a significant hypertrophy of fat cells in both epididymal white adipose tissue (Fig. 1B) and intrascapular brown adipose tissue (BAT) (Fig. 1C) after HFD feeding of WT mice. The changes in morphology were most striking in BAT, where the HFD-fed WT mice developed significantly enlarged brown adipocytes resembling metabolically less active white adipocytes. Similar changes were absent or attenuated in HFD-fed DP-IV⫺兾⫺ mice. Consistent with the body composition and histological analyses, DP-IV⫺兾⫺ mice also were resistant to HFD-induced hyperleptinemia (Table 1). These results indicate that HFD-induced adiposity is curtailed in mice lacking the DP-IV gene. Reduced Food Intake and Increased Energy Expenditure in DP-IVⴚ兾ⴚ Mice. Energy intake and consumption are two major components
of energy balance (21). On HFD, the food intake of DP-IV⫺兾⫺ mice was reduced significantly when compared with WT mice (2.2 ⫾ 0.1 vs. 3.0 ⫾ 0.2 g兾d, P ⬍ 0.05, n ⫽ 15). Thus, calculation of feed efficiency (22) indicated that reduction in caloric intake accounted for ⬇70% of reduced weight gain in DP-IV⫺兾⫺ mice on HFD. To more accurately determine the effect of reduced food intake on body weight gain, we performed a pair-feeding experiment (23) to restrict food intake of WT mice compared with that of DP-IV⫺兾⫺ mice. As illustrated in Fig. 2A, pair-fed
Calculations. Data are expressed as means ⫾ SEM. Statistical
analysis was conducted by using Student’s t test or ANOVA. Statistical significance was defined as P ⬍ 0.05. Results
Mice Are Resistant to High Fat-Induced Obesity. Male WT and DP-IV⫺兾⫺ mice (20) were fed a LFD or HFD from 5 to 25 weeks of age. As depicted in Fig. 1A, WT mice fed the HFD predictably exhibited a significant increase in the rate of body weight gain when compared with age- and gender-matched animals fed the LFD. Mice lacking DP-IV, however, weighed significantly less than their WT counterparts maintained on the same diets. This decreased weight gain was most evident across the HFD groups, where a reduced rate of weight gain in DP-IVⴚ兾ⴚ
6826 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0631828100
Table 1. Reduced adiposity and plasma leptin levels in mice lacking the DP-IV gene Diet type WT 10% FD ⫺兾⫺ 10% FD WT 45% FD ⫺兾⫺ 45% FD
Fat mass, g
% Fat content
Lean mass, g
Leptin, ng兾ml
9.1 ⫾ 0.9 9.3 ⫾ 0.8 15.2 ⫾ 0.5* 11.6 ⫾ 1.4**
26.6 ⫾ 2.0 26.9 ⫾ 1.8 36.5 ⫾ 0.9* 30.7 ⫾ 2.6**
19.5 ⫾ 0.5 19.6 ⫾ 0.6 21.1 ⫾ 0.5 20.1 ⫾ 0.8
33.9 ⫾ 4.9 33.8 ⫾ 4.1 62.7 ⫾ 3.3* 41.8 ⫾ 6.5**
WT and DP-IV⫺/⫺ mice were fed with 10% or 45% FD for 20 weeks. Body composition was determined by NMR. *, P ⬍ 0.05, comparing WT mice on 10% vs. 45% FD. **, P ⬍ 0.05, comparing WT on 45% FD with DP-IV⫺/⫺ mice on same diet; n ⫽ 17.
Conarello et al.
calorimetry studies. The mice were evaluated after 6 weeks of HFD feeding. The DP-IV⫺兾⫺ mice demonstrated significantly higher rates of oxygen consumption than their WT controls throughout the light and dark cycles, and the increase in the total oxygen consumption in DP-IV⫺兾⫺ mice was highly significant (P ⬍ 0.0001) (Fig. 2 B and C). These results are indicative of increased energy expenditure in mice with ablation of the DP-IV gene. Furthermore, mRNA levels of -3 adrenergic receptor and uncoupling protein 1 in the BAT from HFD-fed DP-IV⫺兾⫺ mice were increased significantly (3- to 4-fold; Fig. 2E). Because both -3 receptor and uncoupling protein 1 play important roles in increasing energy expenditure (24, 25), the increased expression of these messages in DP-IV⫺兾⫺ mice is consistent with elevated metabolic rate in DP-IV⫺兾⫺ mice. The increased energy expenditure, along with the aforementioned reduced food intake, could contribute to the prevention of HFD-induced obesity in these animals. DP-IVⴚ兾ⴚ Mice Are Resistant to Obesity-Induced Insulin Resistance.
Hyperinsulinemia and insulin resistance frequently are associated with obesity (26). Although mice in all of the diet groups maintained euglycemia, WT mice maintained on the HFD exhibited a 4-fold increase in plasma insulin levels compared with similar mice fed the LFD (Fig. 3A). Interestingly, even when comparing animals on the LFD, DP-IV⫺兾⫺ mice exhibited ⬇3-fold-lower insulin levels, indicative of improved insulin sensitivity. Furthermore, histological analysis revealed pancreatic islet hypertrophy only in HFD-fed WT mice (Fig. 3B). In contrast, HFD-fed DPIV⫺兾⫺ animals were protected from hyperinsulinemia and islet hypertrophy. An insulin tolerance test (Fig. 3C) indicated that insulin manifested significantly greater glucose lowering efficacy in DP-IV⫺兾⫺ mice than in WT mice, demonstrating that obesity-linked insulin resistance observed in WT mice was ameliorated in mice lacking the DP-IV gene. The ambient levels of biologically active GLP-1(7–36) were elevated by ⬎6-fold in HFD-fed DP-IV⫺兾⫺ mice when compared with WT mice (6.7 ⫾ 0.3 vs. 0.8 ⫾ 0.3 pM, n ⫽ 10, P ⬍ 0.01). Circulating glucagon levels were reduced significantly by ⬇15– 20% in DP-IV⫺兾⫺ mice when compared with WT mice (77.0 ⫾ 6.0 vs. 105.0 ⫾ 11.5 pg兾ml, P ⬍ 0.05, n ⫽ 10). These data are consistent with the previously reported glucagonostatic effects of GLP-1 (27) and could contribute to improved glucose clearance in the DP-IV⫺兾⫺ mice.
Fig. 2. Energy homeostasis in DP-IV⫺兾⫺ mice. (A) Pair-feeding of WT mice on reduced food intake equal to the amount consumed by DP-IV⫺兾⫺ mice. Shown are body weights of mice in each group [P ⬍ 0.05, comparing WT ad libitum group (#) or WT pair-fed group (*) with DP-IV⫺兾⫺ mice fed ad libitum, n ⫽ 14]. Metabolic rate and oxygen consumption were measured over light and dark phases (B) and expressed as total oxygen consumption over a 23-h period (C). (D) Expression of uncoupling protein 1 and -3 adrenergic receptor in BAT. *, P ⬍ 0.05, n ⫽ 8.
WT mice gained significantly less weight compared with WT mice fed ad libitum on HFD. However, the pair-fed WT mice still gained significantly more weight when compared with DP-IV⫺兾⫺ mice, suggesting that factors other than food intake also contributed to the resistance in obesity in mice with DPV-IV ablation. To address whether the resistance to weight gain is associated with an increase in energy expenditure, we carried out indirect Conarello et al.
NMR measurement indicated that livers from WT mice maintained on the HFD accumulated significantly higher levels of lipids (expressed as percent hepatic fat content) compared with livers from LFD-fed WT mice (Fig. 4A). In contrast, lipid accumulation in the livers of HFD-fed DP-IV⫺兾⫺ mice was reduced significantly and similar to that observed in WT mice on LFD, suggesting that DP-IV⫺兾⫺ mice are protected against HFD-induced fatty liver formation. Hepatic lipid accumulation is determined by the activities of two opposing pathways, namely, lipid synthesis and fatty acid -oxidation. PPAR␣ is a key transcription factor controlling the expression of a number of genes that are important for hepatic fat oxidation (28, 29). On the other hand, SREBP-1c is a master gene that regulates the expression of lipogenic genes, and increased expression of SREBP-1c is associated with fatty livers in murine models of diabetes (30, 31). In this study, we found that PPAR␣ expression was up-regulated whereas SREBP-1c expression was down-regulated in the livers of HFD-fed DP-IV⫺兾⫺ mice (Fig. 4 B and C). These results are consistent with the notion that reduced hepatic lipid accumulation in DP-IV⫺兾⫺ mice could result from attenuated lipogenesis and increased fat oxidation. PNAS 兩 May 27, 2003 兩 vol. 100 兩 no. 11 兩 6827
PHYSIOLOGY
DP-IVⴚ兾ⴚ Mice Are Resistant to HFD-Induced Hepatic Lipid Accumulation. HFD has been associated with hepatic steatosis (8, 9).
Fig. 4. Deletion of DP-IV confers resistance to high fat-induced hepatic lipid accumulation. (A) Hepatic lipid content in WT and DP-IV⫺兾⫺ mice. (B and C) Expression levels of PPAR␣ and SREBP-1c genes in liver of WT and DP-IV⫺兾⫺ mice (n ⫽ 10). *, P ⬍ 0.05, comparing WT with DP-IV⫺兾⫺ mice on HFD.
Immunohistochemical analysis of pancreas from HFD兾STZtreated WT and DP-IV⫺兾⫺ mice by using antiinsulin and glucagon antibodies revealed a dramatic difference in the ratio of glucagon-positive to insulin-positive cells within individual islets (Fig, 5C). The glucagon兾insulin ratio was 0.48 ⫾ 0.06 for DP-IV⫺兾⫺ mice and 1.62 ⫾ 0.41 for WT mice (P ⫽ 0.02, n ⫽ 15), suggesting a significant loss of  cell mass in WT mice. These observations suggest that in the absence of DP-IV,  cells are resistant to the combined deleterious effects of STZ and HFD. It is possible that the elevated levels of circulating GLP-1 in DP-IV⫺兾⫺ mice contribute to the protective effects on  cells, because GLP-1 exerts trophic effects on  cells in vitro and in vivo (33, 34). Fig. 3. DP-IV⫺兾⫺ mice are protected from high fat-induced insulin resistance. (A) DP-IV⫺兾⫺ mice have reduced circulating plasma insulin levels on 10% and 45% FD for 20 weeks (n ⫽ 17). (B) Histological analyses of pancreatic islets. (C) Glucose levels during insulin tolerance test in WT and DP-IV⫺兾⫺ mice on HFD (n ⫽ 10). Insulin was dosed i.p. at 1 unit兾kg. P ⬍ 0.05, comparing WT with DP-IV⫺兾⫺ mice on HFD (*) or LFD (#).
DP-IVⴚ兾ⴚ Mice Are Resistant to Streptozotocin-Induced Hyperglycemia. Impaired insulin secretion because of insufficient  cell func-
tion is a key feature of type 2 diabetes (1). In this study, mice were fed HFD for 8 weeks, followed by injection of a moderate dose of STZ to induce hyperglycemia (32). A time-course study indicated that blood glucose levels were similar in both groups (WT and DP-IV⫺兾⫺) during the first 2 weeks after STZ injection. At 3–4 weeks post-STZ injection, however, WT mice developed hyperglycemia, whereas DP-IV⫺兾⫺ mice maintained euglycemia (Fig. 5A). Circulating insulin levels were significantly higher in DP-IV⫺兾⫺ mice than in the WT mice (Fig. 5B), providing a potential explanation for the maintenance of euglycemia in the former. 6828 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0631828100
Discussion We have demonstrated that mice lacking the DP-IV gene are refractory to the development of obesity and adiposity induced by HFD, resulting from reduced energy intake and concomitant increase in energy expenditure. Ablation of DP-IV also is associated with improved metabolic control in these animals, resulting in improved insulin sensitivity, reduced pancreatic islet hypertrophy, and protection against STZ-induced loss of  cell mass and hyperglycemia. In addition, DP-IV deficiency either directly or indirectly induces a signal that activates the PPAR␣ pathway and down-regulates SREBP-1 expression, thereby increasing lipid oxidation, reducing lipogenesis, and preventing hepatic steatosis that typically is associated with high-fat feeding. Despite previous in vitro studies proposing a role for CD26兾 DP-IV in T cell activation, the knockout mice develop normally with no obvious defects in immune function (20). Furthermore, we have performed comparative histological studies on ⬎40 organs from DP-IV⫺兾⫺ and WT mice and have not observed any remarkable differences (data not shown), suggesting that the Conarello et al.
Fig. 5. DP-IV⫺兾⫺ mice are resistant to STZ-induced hyperglycemia and loss of  cell mass. (A) Glucose levels in WT and DP-IV⫺兾⫺ mice after STZ injection (n ⫽ 8 –12). (B) Plasma insulin levels at 3 weeks post-STZ injection (n ⫽ 8 –12). (C) Immunohistochemical analyses of islets from DP-IV⫺兾⫺ (Upper) and WT (Lower) mice. Insulin-positive cells are shown in green, and glucagon-positive cells are shown in red. *, P ⬍ 0.05, comparing WT with DP-IV⫺兾⫺ mice.
deficiency in DP-IV is well tolerated in mice. Consistent with this finding, Fischer rats that are genetically deficient in DP-IV are also normal phenotypically and exhibit improved HFD-induced insulin resistance (35). GLP-1 is a DP-IV substrate that is of prominent medical interest for the treatment of diabetes (15–17). DP-IV converts the active GLP-17–36-amide to inactive GLP-19 –36-amide (36– 38). Circulating GLP-1 levels were elevated in the knockout mice in this study and could contribute to the reduced food intake and decreased weight gain in these animals, consistent with the proposed role of this incretin in gastric emptying and central satiety control (15–17). However, because DP-IV degrades most peptides containing a penultimate Pro or Ala residue at the amino terminus (12–14), it is anticipated that several such regulatory peptides would be stabilized in the knockout animals. Hence, one cannot exclude the possibility that elevated levels of some of the unknown substrates may impact the physiology of DP-IV⫺兾⫺ mice and contribute to the improved metabolic phenotype. Indeed, disruption of GLP-1 signaling by deletion of the GLP-1 receptor in mice is not associated with the development of obesity (39), possibly because of compensatory upregulation of other incretin hormones, such as glucosedependent insulinotropic polypeptide. Therefore, the resistance to obesity observed in DP-IV⫺兾⫺ mice likely is due to effects extending beyond the stabilization of bioactive GLP-1. Further investigations into the substrate specificity of DP-IV should lead to identification of novel peptides that are present at higher levels in DP-IV⫺兾⫺ mice, providing valuable information for unraveling the mechanisms of body weight control. Pathways that control insulin secretion and regulate  cell mass are integral to the pathology of diabetes mellitus (40). It has been shown that when lipids accumulate in nonadipose tissues during overnutrition, fatty acids trigger deleterious signals that cause apoptosis of lipid-laden cells, such as  cells and cardiomyocytes (41, 42). In the pathogenesis of diabetes in the ZDF ( fa兾fa) rat, there are biphasic  cell responses to lipid overload. The initial phase involves  cell hypertrophy and hyperplasia
We thank Dr. D. Marguet and the Institut National de la Sante´ et de la Recherche Me´dicale for providing the CD26兾DP-IV⫺兾⫺ mice. We acknowledge Raynald Bergeron, Deborah Szalkowski, Beverly Shelton, Leslie Balogh, and Xioalan Shen for expert technical assistance and helpful discussions. S.L.C. is supported by a Merck兾Columbia University Postdoctoral Fellowship in Comparative Medicine.
Taylor, S. I. (1999) Cell 97, 9–12. Saltiel, A. R. & Kahn, C. R. (2001) Nature 414, 799–806. Zimmet, P., Alberti, K. G. & Shaw, J. (2001) Nature 414, 782–787. Kopelman, P. G. (2000) Nature 404, 635–643. Friedman, J. M. & Halaas, J. L. (1998) Nature 395, 763–770. Barsh, G. S., Farooqi, I. S. & O’Rahilly, S. (2000) Nature 404, 644–651. Spiegelman, B. M. & Flier, J. S. (2001) Cell 104, 531–543. Golay, A. & Bobbioni, E. (1997) Int. J. Obes. Relat. Metab. Disord. 21, Suppl. 3, S2–S11.
9. Bray, G. A. & Popkin, B. M. (1998) Am. J. Clin. Nutr. 68, 1157–1173. 10. Schwartz, M. W., Woods, S. C., Porte, D., Jr., Seeley, R. J. & Baskin, D. G. (2000) Nature 404, 661–671. 11. Palmiter, R. D., Erickson, J. C., Hollopeter, G., Baraban, S. C. & Schwartz, M. W. (1998) Recent Prog. Horm. Res. 53, 163–199. 12. De Meester, I., Korom, S., Van Damme, J. & Scharpe, S. (1999) Immunol. Today 20, 367–375. 13. Hegen, M., Kameoka, J., Dong, R. P., Morimoto, C. & Schlossman, S. F. (1997) Adv. Exp. Med. Biol. 421, 109–116.
Conarello et al.
PNAS 兩 May 27, 2003 兩 vol. 100 兩 no. 11 兩 6829
PHYSIOLOGY
1. 2. 3. 4. 5. 6. 7. 8.
along with increased insulin secretion to counter the development of insulin resistance resulting from obesity. As this prediabetic stage of obesity progresses, the lipid-laden  cells undergo apoptosis with an accompanying decline in insulin production to below the levels required to compensate for insulin resistance and hyperglycemia ensues. In this study, we show that WT mice develop insulin resistance and islet hypertrophy when maintained on a HFD. When HFD-induced insulin resistance was combined with partial islet damage elicited by a moderate dose of STZ, WT mice exhibited loss of insulin-positive  cells, decreased circulating insulin levels, and hyperglycemia. These defects were absent or attenuated in DP-IV⫺兾⫺ mice. These data suggest that inhibition of DP-IV may present an opportunity for preventing or delaying  cell loss and promoting islet regeneration in the treatment of type 2 diabetes. Over the past 10 years, GLP-1 has been explored as a therapeutic agent with demonstrated efficacy in lowering glucose in diabetic patients (43–45). However, the usefulness of native GLP-1 in the treatment of diabetes is hampered by the very short half-life of the hormone because of inactivation by DP-IV. Therefore, strategies for GLP-1-based treatment of diabetes have focused on injections of GLP-1 derivatives that are resistant to DP-IV or stabilization of endogenous GLP-1 through the use of DP-IV inhibitors. Such approaches have led to promising agents with glucose-lowering efficacy in diabetic patients (46, 47). The data presented in this study provide strong evidence that chronic ablation of DP-IV activity may have added benefits in metabolic control extending beyond glucose homeostasis per se and may serve as a unique intervention for treating both diabetes and obesity.
14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.
Mentlein, R. (1999) Regul. Pept. 85, 9–24. Kieffer, T. J. & Habener, J. F. (1999) Endocr. Rev. 20, 876–913. Drucker, D. J. (1998) Diabetes 47, 159–169. Holst, J. J. (1999) Curr. Med. Chem. 6, 1005–1017. Tang-Christensen, M., Larsen, P. J., Thulesen, J., Romer, J. & Vrang, N. (2000) Nat. Med. 6, 802–807. Miyawaki, K., Yamada, Y., Ban, N., Ihara, Y., Tsukiyama, K., Zhou, H., Fujimoto, S., Oku, A., Tsuda, K., Toyokuni, S., et al. (2002) Nat. Med. 8, 738–742. Marguet, D., Baggio, L., Kobayashi, T., Bernard, A. M., Pierres, M., Nielsen, P. F., Ribel, U., Watanabe, T., Drucker, D. J. & Wagtmann, N. (2000) Proc. Natl. Acad. Sci. USA 97, 6874–6879. Coleman, D. L. (1982) Diabetologia 22, 205–211. Parekh, P. I., Petro, A. E., Tiller, J. M., Feinglos, M. N. & Surwit, R. S. (1998) Metabolism 47, 1089–1096. Ste. Marie, L., Miura, G. I., Marsh, D. J., Yagaloff, K. & Palmiter, R. D. (2000) Proc. Natl. Acad. Sci. USA 97, 12339–12344. Lowell, B. B. & Flier, J. S. (1997) Annu. Rev. Med. 48, 307–316. Argyropoulos, G. & Harper, M. E. (2002) J. Appl. Physiol. 92, 2187–2198. Kahn, B. B. & Flier, J. S. (2000) J. Clin. Invest. 106, 473–481. Komatsu, R., Matsuyama, T., Namba, M., Watanabe, N., Itoh, H., Kono, N. & Tarui, S. (1989) Diabetes 38, 902–905. Kersten, S., Seydoux, J., Peters, J. M., Gonzalez, F. J., Desvergne, B. & Wahli, W. (1999) J. Clin. Invest. 103, 1489–1498. Leone, T. C., Weinheimer, C. J. & Kelly, D. P. (1999) Proc. Natl. Acad. Sci. USA 96, 7473–7478. Shimomura, I., Hammer, R. E., Richardson, J. A., Ikemoto, S., Bashmakov, Y., Goldstein, J. L. & Brown, M. S. (1998) Genes Dev. 12, 3182–3194. Brown, M. S. & Goldstein, J. L. (1998) Nutr. Rev. 56, S1–S3, S54–S75.
6830 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0631828100
32. Luo, J., Quan, J., Tsai, J., Hobensack, C. K., Sullivan, C., Hector, R. & Reaven, G. M. (1998) Metabolism 47, 663–668. 33. Susini, S., Roche, E., Prentki, M. & Schlegel, W. (1998) FASEB J. 12, 1173–1182. 34. Wang, Y., Perfetti, R., Greig, N. H., Holloway, H. W., DeOre, K. A., Montrose-Rafizadeh, C., Elahi, D. & Egan, J. M. (1997) J. Clin. Invest. 99, 2883–2889. 35. Yasuda, N., Nagakura, T., Yamazaki, K., Inoue, T. & Tanaka, I. (2002) Life Sci. 227–238. 36. Mentlein, R., Gallwitz, B. & Schmidt, W. E. (1993) Eur. J. Biochem. 214, 829–835. 37. Kieffer, T. J., McIntosh, C. H. & Pederson, R. A. (1995) Endocrinology 136, 3585–3596. 38. Deacon, C. F., Johnsen, A. H. & Holst, J. J. (1995) J. Clin. Endocrinol. Metab. 80, 952–957. 39. Scrocchi, L. A., Brown, T. J., MaClusky, N., Brubaker, P. L., Auerbach, A. B., Joyner, A. L. & Drucker, D. J. (1996) Nat. Med. 2, 1254–1258. 40. Bell, G. I. & Polonsky, K. S. (2001) Nature 414, 788–791. 41. Unger, R. H. (2002) Annu. Rev. Med. 53, 319–336. 42. Unger, R. H. & Orci, L. (2000) Int. J. Obes. Relat. Metab. Disord. 24, Suppl. 4, S28–S32. 43. Gutniak, M., Orskov, C., Holst, J. J., Ahren, B. & Efendic, S. (1992) N. Engl. J. Med. 326, 1316–1322. 44. Nathan, D. M., Schreiber, E., Fogel, H., Mojsov, S. & Habener, J. F. (1992) Diabetes Care 15, 270–276. 45. Nauck, M. A., Holst, J. J. & Willms, B. (1997) Horm. Metab. Res. 29, 411–416. 46. Holst, J. J. & Deacon, C. F. (1998) Diabetes 47, 1663–1670. 47. Ahren, B., Simonsson, E., Larsson, H., Landin-Olsson, M., Torgeirsson, H., Jansson, P. A., Sandqvist, M., Bavenholm, P., Efendic, S., Eriksson, J. W., et al. (2002) Diabetes Care 25, 869–875.
Conarello et al.