Kinin B₁ Receptor Deficiency Leads to Leptin Hypersensitivity and Resistance to Obesity

B₁^−/− mice (18) used in the experiments were originated from 10 generations’ backcrossing of an initially mixed genetic background (129/Sv and C57Bl/6) with C57Bl/6 mice (Taconic, Germantown, NY). Therefore, C57Bl/6 animals were used as their controls. The ob/ob-B₁^−/− strain was generated by crossbreeding C57Bl/6 ob^+/− mice (The Jackson Laboratories, Bar Harbor, ME) with B₁^−/− mice. Animals have been obtained from the Universidade Federal de São Paulo (Brazil), from the Max-Delbrück-Center for Molecular Medicine (Berlin-Buch, Germany), and from the Animal House (Toulouse, France). All experiments reported have been conducted as stated in the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Academy Press, Washington, DC, 1996) and approved by a local committee. Animals were maintained on standard mouse chow at 22°C on a 12-h light-dark cycle with ad libitum access to food and tap water. Food consumption and body weight were monitored weekly in individualized animals. In all experiments, 12- to 18-week-old males were used.

Mice were fed either standard diet (10% kcal fat) or high-fat diet (45% kcal fat) (Research Diets, New Brunswick, NJ) for 9 weeks. After the treatment, mice were killed for blood and tissue collection. The sera were separated for leptin and insulin quantification with ELISA kits (R&D Systems [Minneapolis, MN] and Millipore [Billerica, MA], respectively). Tissues explants were weighed and used for adipocyte isolation, or frozen for protein, RNA, or triglycerides extraction.

Total body fat was estimated by dual-energy X-ray absorptiometry using a Hologic QDR 4500 scanner (Hologic, Waltham, MA) as described previously (25).

Isolation of fat cells and calculation of adipocyte volume and number in epididymal fat depots were performed as previously described (26). Briefly, images of isolated adipocytes were acquired from a light microscope fitted with a camera, and the average cellular volume was estimated by measuring the diameter of at least 100 adipocytes per animal using the software ImageJ (http://rsb.info.nih.gov/ij/). Fat cell number was calculated by dividing the adipose tissue triglyceride content by the average adipocyte lipid content, obtained by the multiplication of the average fat cell volume by the triolein density (0.92).

Please refer to the online appendix (available at http://dx.doi.org/10.2337/db07-1508).

In vivo indirect open circuit calorimetry was performed as previously described (26). Briefly, respiratory exchanges and spontaneous activity signals of freely moving mice were computed at 10-s intervals using a computer-assisted data acquisition program (ADDENFI metabolic chamber prototype). Mice were housed individually in the metabolic cages at 10:00 a.m. without food, and the postabsorptive resting energy expenditure (∼basal metabolism) was measured between 4:00 and 6:00 p.m. At 6:00 p.m., we gave a 1-g meal of either control or high-fat diet to the mice and computed the increase in energy expenditure in response to feeding (metabolism and respiratory quotient).

To determine leptin sensitivity, we administered recombinant mouse leptin (40 μg · day⁻¹ · mouse⁻¹ i.p.; R&D Systems) during 4 days (at 5:00 p.m.) and evaluated food intake and/or weight loss. Food intake during 5 days before leptin injection was considered as the basal consumption (100%). An independent group of fasted animals was killed 45 min after the first leptin bolus for hypothalamus collection, protein extraction, and Western blotting quantification of pSTAT3 and STAT3.

ALZET micro-osmotic pumps (Alza, Palo Alto, CA), set to deliver for 28 days either SSR240612 (3 mg · kg⁻¹ body wt · day⁻¹; Sanofi-Aventis, Paris, France) or vehicle (2% DMSO), were subcutaneously implanted in anesthetized C56Bl/6 mice under sterile conditions. Seven days after the implant, mice were submitted to high-fat diet for 3 weeks. Body weight and food intake were assessed weekly in the period of 1 month before and after the implant. At the end of the treatment, pumps were checked for their contents, and animals with filled ones were not considered in the data analysis.

All values were expressed as means ± SE. Statistical analyses were carried out using two-tailed Student's unpaired t test to compare two independent groups or ANOVA followed by Bonferroni's test to compare more than two. Significance was rejected at P > 0.05.

To assess possible contributions of the kinin B₁ receptor to the regulation of satiety and adiposity in mice, we analyzed body weight, energy intake, and adipose mass in standard or high-fat diet–fed B₁^−/− and wild-type animals. Although no differences were found in the body weight (Fig. 1A) and energy intake (Fig. 1B) between wild-type and B₁^−/− mice under control diet, a drastic reduction of fat mass was observed in the knockout animals under this condition (Fig. 1C and D). Lean mass, however, was increased in these mice, accounting for their normal body weight when compared to wild type (Fig. 1C). On the other hand, under high-fat diet, B₁^−/− mice exhibited impaired energy intake (Fig. 1B) and were remarkably refractory to the effect of the diet on body weight gain (Fig. 1A). In addition, total adiposity and fat accumulation by adipocytes were dramatically reduced in B₁^−/− mice under both diets (Fig. 1E and F), whereas the number of fat cells in the gonadal depots was only decreased in B₁^−/− mice under standard diet (Fig. 1G).

Obesity is normally associated with insulin resistance and glucose intolerance. Thus, we investigated these parameters in wild-type and B₁^−/− mice under standard or high-fat diet. In agreement, B₁^−/− mice presented improved glucose tolerance and reduced insulin levels in both diets in comparison with the wild type (online appendix), confirming our previous data showing increased insulin sensitivity in these mice (21).

Next, we measured serum leptin levels (Fig. 2A) and the expression of leptin mRNA (Fig. 2B) and protein (Fig. 2C) in the epididymal white adipose tissue of wild-type and B₁^−/− mice. In all cases, we observed a significant decrease in leptin content in B₁^−/− mice when compared with wild-type mice. Notably, high-fat diet did not lead to an increase of leptin levels in B₁^−/− mice.

Taken together, these results demonstrate that B₁^−/− mice have reduced adiposity and are strongly resistant to diet-induced hyperleptinemia and body weight gain.

Analysis of feed efficiency (body weight gain/energy consumed) (Fig. 3A) revealed that inhibition of weight gain observed in B₁^−/− mice under high-fat diet could not be totally explained by decreased energy intake (Fig. 1B). Therefore, we quantified the components of total energy expenditure under normal and high-fat feeding in B₁^−/− and wild-type mice. No differences in resting energy expenditure, increase in energy expenditure in response to feeding, or differences in respiratory quotient (respiratory quotient, VCO₂/VO₂) in the postabsorptive state (4 h before meal and 8–12 h after meal) were observed between B₁^−/− and wild-type mice (Table 1). Under standard diet, respiratory quotient levels after feeding were >1.0 in both groups, suggesting that de novo lipogenesis in response to carbohydrate ingestion is occurring in these animals at a similar extent. In contrast, under high-fat diet, B₁^−/− mice exhibited a meal-induced increase in respiratory quotient that lasted 4 h less than in wild-type mice (Fig. 3B and C), indicating that postprandial inhibition of lipid oxidation is significantly lower in these animals. In addition, B₁^−/− mice under high-fat regimen expended significantly more energy with activity (+46.6%, P < 0.002; Table 1). Together, these data suggest that B₁^−/− mice can resist high-fat diet–induced weight gain in part due to increased energy expenditure with spontaneous activity and decreased inhibition of lipid oxidation in response to a fat meal.

Together, hypoleptinemia, protection against high-fat diet–induced hyperleptinemia, elevated energy expenditure, and reduced food intake under high-fat diet suggest increased leptin responsiveness in B₁^−/− mice. To test this hypothesis, we administered exogenous leptin to B₁^−/− and wild-type mice and quantified daily food intake and phosphorylation of STAT3, a downstream component of the Ob-Rb signaling pathway, in the hypothalamus of these animals. A stronger inhibitory effect of exogenous leptin on food intake in B₁^−/− mice was observed when compared with wild-type mice (Fig. 4A). Furthermore, the levels of phosphorylated STAT3 after leptin injection were significantly higher in the hypothalamus of B₁^−/− in comparison with wild-type mice (Fig. 4B). In addition, we measured the expression of Ob-Rb mRNA in hypothalamus of B₁^−/− and wild-type mice and found upregulation of this transcript in B₁^−/− (Fig. 4C). Moreover, the expression of SOCS3, a potent inhibitor of the Ob-Rb signaling, was decreased in hypothalamus of B₁^−/− mice (Fig. 4D). These data suggest that Ob-Rb signaling is facilitated in B₁^−/− mice, which could contribute to the inhibition of food intake and enhanced energy expenditure in these animals under high-fat diet. Another aspect of leptin sensitivity, the passage of leptin through the blood-brain barrier, was addressed in B₁^−/− and wild-type mice by measuring brain accumulation of intravenously injected [¹²⁵I]-labeled leptin. Although B₁^−/− mice had a tendency to an increased leptin passage to the brain, no significant differences were observed in relation to the wild-type mice (data not shown).

We also assessed kinin B₁ receptor mRNA expression in the hypothalamus of wild-type mice after 9 weeks of high-fat regimen. Interestingly, we found higher levels of this mRNA in high-fat diet–fed mice, in comparison with animals under standard diet (Fig. 4E). Taken together, these results indicate that the kinin B₁ receptor may participate in the mechanism of leptin resistance in obesity.

To estimate the importance of leptin hypersensitivity to the phenotype of resistance to obesity in B₁^−/− mice, we generated mice lacking both B₁ receptor and leptin (ob/ob-B₁^−/−). In comparison with ob/ob mice, ob/ob-B₁^−/− had no differences in body weight (Fig. 5A) and exhibited no significant morphological alterations, except for mildly heavier heart and kidney (data not shown). We also did not observe differences between the ob/ob and the ob/ob-B₁^−/− mice on daily food ingestion (Fig. 5B) and on some parameters of adiposity in these animals, including epididymal fat content (Fig. 5C) and adipocyte volume (Fig. 5D and E) and number (Fig. 5F). These data suggest that leanness associated with kinin B₁ receptor deficiency may be dependent on the leptin signaling.

Next, we performed daily injections of exogenous leptin to ob/ob-B₁^−/− and ob/ob mice and measured food intake, body weight, and feed efficiency. Interestingly, we found that the first dose of leptin was more efficient to inhibit food intake (Fig. 6A) and to promote weight loss (Fig. 6B) in ob/ob-B₁^−/− mice in comparison with ob/ob. In addition, ob/ob-B₁^−/− presented a significant reduction of feed efficiency on the first day of leptin administration (Fig. 6C), indicating that weight loss in this condition is not exclusively due to inhibition of food intake. To corroborate these data, we quantified leptin-stimulated STAT3 phosphorylation in the hypothalamus of ob/ob-B₁^−/− and ob/ob mice. As expected, ob/ob-B₁^−/− showed increased levels of phosphorylated STAT3 after leptin induction (Fig. 6D). These results bring forth the hypothesis that the kinin B₁ receptor may play a role in energy homeostasis through the modulation of leptin responsiveness.

To validate our findings using knockout models and to confirm the pathophysiological function of the kinin B₁ receptor in obesity, we pharmacologically blocked this receptor in mice for 28 days by chronic administration of the nonpeptide orally active B₁ receptor antagonist SSR240612 (3 mg · kg⁻¹ body wt · day⁻¹) and measured body weight gain and food intake weekly during this period. In the 1st week, mice were fed with a standard diet and showed no significant differences in body weight (Fig. 7A) and energy intake (Fig. 7B) between the groups receiving SSR240612 or the vehicle. However, after being submitted to a high-fat diet, mice receiving SSR240612, in comparison with the control mice, showed a clear inhibition in body weight gain (Fig. 7A) and energy intake (Fig. 7B). In addition, feed efficiency under high-fat diet was significantly decreased in mice treated with SSR240612 (Fig. 7C). Thus, these data show ample evidence to support the hypothesis that kinin B₁ receptor blockade protects mice from weight gain.

Although significant progress has been made in the identification of genes contributing to metabolic diseases, many pieces of the puzzle are still missing. Our data indicate that one member of the kallikrein-kinin system, the kinin B₁ receptor, may be an important part in this scenario because it seems to play a significant role in the regulation of food intake and energy expenditure exerted by leptin. Before this study, however, few relevant physiological roles had been attributed exclusively to the activation of the kinin B₁ receptor. Moreover, in most physiological processes in which the expression of B₁ receptors was observed, the kinin B₁ receptor was considered a secondary player, serving as an alternative in case of B₂ receptor deficiency (14,27). However, recent results from our group showed that kinin B₁ receptor activation modulates leptin homeostasis (24). In addition, here we present sufficient evidence to support the idea that kinin B₁ receptors are in fact participating in essential metabolic processes.

The kinin B₁ receptor, like many inflammatory molecules, including tumor necrosis factor-α (TNF-α) (28) and LPS (29), may act through redundant pathways to display both its inflammatory and metabolic actions. Similarly, leptin has inflammatory actions and strongly controls energy homeostasis (6), mainly by regulating satiety and potentiating many components of the total energy expenditure, including the energy expended with physical activity (8) and with fat oxidation (7,30). According to our data, leptin may be a key mediator of the metabolic actions of the kinin B₁ receptor. Firstly, despite the gain of fat tissue after the high-fat regimen, B₁^−/− mice are completely protected against high-fat diet–induced hyperleptinemia, suggesting that leptin is not correlated with adiposity levels in these mice. Furthermore, ob/ob-B₁^−/− and ob/ob mice are similarly obese, indicating that leptin may be necessary for the kinin B₁ receptor to affect food intake and energy expenditure. Finally, B₁ receptor deficiency increases leptin responsiveness and exacerbates the effects of exogenous leptin on weight loss in ob/ob mice.

Although the precise mechanism underlying the interaction between the kinin B₁ receptor and the leptin signaling pathway is still unclear, accumulating evidence showing the presence of this receptor and other components of the kallikrein-kinin system in the hypothalamus of rodents and humans (23,31,32) lead us to the hypothesis that the modulation of leptin sensitivity by the B₁ receptor may involve a central phenomenon. In agreement, we observed more pronounced effects of exogenous leptin administration on food intake inhibition in B₁^−/− mice. These experiments, based on previously described protocols (8,33,34), reflect observations obtained from daily injections of high doses of leptin and may therefore represent supra-physiological effects. Nevertheless, increased central leptin sensitivity in B₁^−/− mice is corroborated by increased leptin-induced STAT3 phosphorylation, upregulation of Ob-Rb mRNA, and reduced levels of SOCS3 under basal conditions in the hypothalamus of these animals.

SOCS3 is a member of a family of cytokine-inducible suppressors of cytokine signaling that particularly inhibits leptin signal transduction (13). In mouse embryonic fibroblasts, TNF-α–dependent activation of the nuclear factor-κB (NF-κB) is able to induce SOCS3 expression and, consequently, elicit negative effects on STAT signaling (35). As an inflammatory mediator, the kinin B₁ receptor can also activate NF-κB (36). Therefore, a possible mechanism to explain the downregulation of SOCS3 in B₁^−/− mice would be a reduction of B₁ receptor–induced NF-κB activity in Ob-Rb neurons. Moreover, B₁ receptor deficiency could lead to an understimulation of the hypothalamic cytokine signaling due to a decrease in cytokine production. Studies have shown that kinins can induce TNF-α and interleukin (IL)-1 release from macrophages (37) and IL-6 expression by astrocytes (38).

Obesity has also been linked to increased inflammation and macrophage infiltration in the adipose tissue (39). It is postulated that cytokines produced by the adipose tissue, which include leptin, TNF-α, and some ILs, are pivotal factors in the control of metabolism. In previous studies, B₁ receptor deficiency has been associated with hypoalgesia and reduced inflammatory response (18,19). It was demonstrated that the hypotensive effect of LPS was blunted in B₁^−/− mice and the infiltration of polymorphonuclear leukocytes into tissues. The contributions of the kinin B₁ receptor to the inflammatory status of the adipose tissue are, however, not clear. Thus, future experiments may be proposed to understand how the absence of B₁ receptors in the adipose tissue could modulate inflammatory mediators to participate in the phenotypes observed in B₁^−/− mice.

On the other hand, the regulation of the energy balance by kinin B₁ receptor may not be related exclusively to activation by kinins. Recently, Lai et al. (40) presented strong evidence that kinins are not the only agonists that bind to kinin receptors. They showed that the endogenous opioid peptide dynorphin A promotes neuropathic pain through the activation of both kinin B₁ and B₂ receptors. In the early 80s, 3 years after dynorphins were identified as opioid regulators of pain and stress responses, these peptides were described as potent inducers of food intake (41,42). However, in 2 decades, modest advances have been made to provide a solid understanding of the mechanisms mediating the regulation of energy balance by dynorphin. Nevertheless, evidence showing that dynorphin neurons in the arcuate nucleus are leptin responsive (43) and that the effect of κ-opioid agonists on food intake cannot be reproduced in ob/ob mice (44) indicates that leptin may be required for this event to happen. Thus, we speculate that dynorphin may act through the kinin B₁ receptor to control food intake via modulation of the leptin signaling.

In general, protection against obesity is a common feature of mice with leptin hypersensitivity. SOCS3 gene disruption in the brain, for instance, is able to increase leptin sensitivity and promote resistance to high-fat diet–induced weight gain and hyperleptinemia (33). Moreover, the neuron-specific tyrosine phosphatase protein-tyrosine phosphatase 1B (PTP1B) knockout mice are over-responsive to the effects of exogenous leptin and show reduced adiposity and elevated energy expenditure with activity (34). Thus, targeting SOCS3 and targeting PTP1B are currently considered attractive strategies to fight obesity and diabetes.

Similarly, CB₁ cannabinoid receptor knockout mice (45,46) are more sensitive to intracerebroventricular injections of leptin, have decreased body weight, and are slightly hypophagic. In addition, they are resistant to high-fat diet–induced weight gain due to an increase in energy expenditure. Based on this animal model, researchers evidenced a strong anti-obesity effect of the CB₁ receptor antagonist SR141716A (46), and currently, under the name of rimonabant, this drug has been considered the most efficient clinical therapy to treat human obesity (47). Therefore, based on the presented data, here we suggest the orally active kinin B₁ receptor antagonist SSR240612 as a potential tool to prevent obesity. Although several experiments still must be addressed to verify the clinical relevance of this drug, we believe that our results and the grown body of evidence showing limited expression of B₁ receptors under physiological conditions (14) may render the kinin B₁ receptor an interesting potential target to treat obesity with low risks of severe side effects.

In summary, we have presented in this study consistent data indicating that functional deficiency of the kinin B₁ receptor can protect mice from obesity through a mechanism that may involve leptin signaling.

This work was supported by grants from the Fundação de Amparo à Pesquisa do Estado de São Paulo, Conselho Nacional de Desenvolvimento Científico e Tecnológico, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, and the Institut National de la Santé et de la Recherche Médicale.

We thank Élton D. da Silva, Juliana G. Pessoa, Martin Würtele, and Vicência M.T. Sales for technical or intellectual assistance.