Journal of Neuroendocrinology, 2014, 26, 58–67 © 2014 British Society for Neuroendocrinology
ORIGINAL ARTICLE
High-Fat Diet Induces Leptin Resistance in Leptin-Deficient Mice C. E. Koch*, C. Lowe*, D. Pretz*, J. Steger*, L. M. Williams† and A. Tups* *Department of Animal Physiology, Faculty of Biology, Philipps University Marburg, Marburg, Germany. †Metabolic Health Group, Rowett Institute of Nutrition and Health, University of Aberdeen, Aberdeen, UK.
Journal of Neuroendocrinology
Correspondence to: A. Tups, Karl-von-Frisch-Straße 8, 35043 Marburg, Germany (e-mail: [email protected]).
The occurrence of type II diabetes is highly correlated with obesity, although the mechanisms linking the two conditions are incompletely understood. Leptin is a potent insulin sensitiser and, in leptin-deficient, insulin insensitive, Lepob/ob mice, leptin improves glucose tolerance, indicating that leptin resistance may link obesity to insulin insensitivity. Leptin resistance occurs in response to a high-fat diet (HFD) and both hyperleptinaemia and inflammation have been proposed as causative mechanisms. Scrutinising the role of hyperleptinaemia in this process, central hyperleptinaemia in Lepob/ob mice was induced by chronic i.c.v. infusion of leptin (4.2 lg/day) over 10 days. This treatment led to a dramatic decline in body weight and food intake, as well as an improvement in glucose tolerance. Transfer to HFD for 4 days markedly arrested the beneficial effects of leptin on these parameters. Because Lepob/ob mice are exquisitely sensitive to leptin, the possibility that leptin could reverse HFD-induced glucose intolerance in these animals was investigated. HFD led to increased body weight and glucose intolerance compared to a lowfat diet (LFD). Older and heavier Lepob/ob mice were used as body weight-matched controls. Mice in each group received either i.p. leptin (1.25 mg/kg) or vehicle, and glucose tolerance, food intake and the number of phosphorylated signal transducer and activator of transcription (pSTAT)3 immunoreactive cells in the arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) were analysed. Leptin improved glucose tolerance (P = 0. 019) and reduced food intake in Lepob/ob mice on LFD (P ≤ 0.001) but was ineffective in mice on HFD. Furthermore, when leptin was administered centrally, the glucose tolerance of Lepob/ob mice on HFD was significantly impaired (P = 0.007). Although leptin induced the number of pSTAT3 immunoreactive cells in the ARC and VMH of Lepob/ob mice on LFD, HFD was associated with elevated pSTAT3 immunoreactivity in vehicle-treated Lepob/ob mice that was unaffected by leptin treatment, suggesting central leptin resistance. Negating central inflammation by co-administering a c-Jun n-terminal kinase (JNK) inhibitor reinstated the glucose-lowering effects of leptin. These findings demonstrate that Lepob/ob mice develop leptin resistance on a HFD independent of hyperleptinaemia and also indicate that the JNK inflammatory pathway plays a key role in the induction of dietinduced glucose intolerance. Key words: energy and glucose homeostasis, hypothalamus, inflammation, JNK pathway
Obesity and associated disorders, particularly diabetes mellitus type II, have become a leading medical challenge to modern societies. Leptin was discovered in 1994 (1) and was initially characterised to reduce food intake and increase energy expenditure (2–5). In obesity, circulating levels of leptin, secreted primarily by adipocytes, increase, although its effect of reducing food intake and increasing energy expenditure diminishes. Thus, the establishment of leptin resistance is regarded as a leading cause in the onset of obesity.
doi: 10.1111/jne.12131
Despite the well characterised anorexigenic and catabolic properties of leptin, centrally-acting leptin is also essential for the maintenance of glucose homeostasis and is a potent insulin sensitiser (4,6–11). Because obesity gives rise to insulin insensitivity, the first stage in the development of diabetes mellitus type II, leptin resistance has been proposed as a major mechanism linking these two conditions. It has been shown that low doses of leptin that are ineffective in regulating food intake and body weight reduce
HFD induces leptin resistance in Lepob/ob mice
plasma glucose and insulin concentrations (12). However, the molecular mechanism underlying glucose-lowering properties and the insulin-sensitising effects of leptin are not well characterised. An imbalance of increased energy intake and decreased energy expenditure leads to obesity. Amongst other reasons, this imbalance can be caused by the excessive consumption of a diet high in saturated fatty acids (13) or by natural loss of function mutations in the leptin gene or related components of the neuroendocrine network of energy balance (1,5). The loss of leptin function and its consequences for the maintenance of energy balance have been well characterised using the Lepob/ob mouse, which serves as a naturally occurring animal model for this metabolic condition. We recently demonstrated that acute i.p. or i.c.v. injections of leptin improve glucose homeostasis in these mice (14), demonstrating that the presence of leptin is necessary for normal insulin sensitivity and glucose homeostasis. However, in humans, the incidence of naturally occurring loss of leptin function by genetic reasons as a cause for obesity is very low. By contrast, diet-induced obesity (DIO), which is the leading cause for obesity in humans, is characterised by hyperleptinaemia (15). To determine the primary mechanism underlying development of leptin resistance, Lepob/ob mice were administered leptin, via i.c.v. infusion, to induce central hyperleptinaemia and then fed a high-fat diet (HFD) to investigate the role of hyperleptinaemia versus diet in the development of this process. Because HFD appeared to be essential for inducing leptin resistance, we established that acute central leptin resistance occured in DIO wild-type mice particularly considering the glucoselowering properties of the hormone. Furthermore, we characterised whether Lepob/ob mice that are highly sensitive to exogenously applied leptin maintain this sensitivity during high-fat feeding in terms of the acute and chronic metabolic properties of the hormone. We additionally assessed whether these mice retain central leptin sensitivity with respect to the ability to activate the janus kinase 2/signal transducer and activator of transcription 3 (JAK2/ STAT3) pathway because it is well established that leptin regulates energy homeostasis via signalling through this pathway (16,17). Lepob/ob mice on HFD exhibited a morbidly obese phenotype. Recent intriguing evidence suggests that obesity is characterised by central inflammation, particularly increased activation of the c-Jun n-terminal kinase (JNK)-pathway (18,19). By pharmacological reversal of central JNK activity, we investigated whether inflammation via JNK leads to central leptin resistance in terms of its glucose-lowering potential.
Materials and methods Animals All procedures involving animals were licensed under national animal ethics legislation and received approval by the federal public authority for animal ethics. Male C57BL/6JRj wild-type (Lep+/+) and Lepob/ob mice were purchased from Janvier (Le Genest-Saint-Isle, France). They were housed individually under a 12/12 h light/dark cycle with an ambient temperature of 24 °C. All mice had ad lib. access to food and water except the period of food withdrawal (over night) before all of the experiments. Mice were fed either a low-fat diet (LFD; D12450B Research Diets, Brunswick, NJ, USA) with 10% Journal of Neuroendocrinology, 2014, 26, 58–67
59
fat by energy (kcal) or HFD (D12492 Research Diets) with 60% fat by energy (kcal).
Effect of centrally-induced hyperleptinaemia on body weight and glucose tolerance To determine whether hyperleptinaemia or high-fat feeding promotes leptin resistance, in Lepob/ob mice, central hyperleptinaemia was induced by chronic i.c.v leptin infusion before mice underwent dietary interventions. After a 6day observation period, a stainless steel cannula (0.9 mm lateral and 0.1 mm posterior to bregma, 2.5 mm ventral to the surface) was stereotaxically implanted into the lateral cerebral ventricle of 7-week-old Lepob/ob mice as described previously (14). The cannula was connected by silicone tubing to a miniosmotic pump that was implanted subcutaneously in the scapular region (Pump Model 1002; Alzet Inc., Cupertino, CA, USA). Mice were divided into two groups, one group was centrally infused with leptin [4.2 lg in artificial cerebrospinal fluid (aCSF)/day, n = 13], whereas the other group received vehicle (aCSF, n = 6). During this period, mice remained on a LFD. Six days after surgery when the body weight of the leptin-treated group had declined dramatically, this group was subdivided into two groups. One of these groups (n = 6) remained on LFD, whereas the other group and the vehicle-treated mice (n = 6) were transferred to a HFD (n = 7). Four days after the change in diet, an i.p. glucose tolerance test (ipGTT) (0.75 g glucose/kg body weight to reflect the increased glucose insensitivity and increased adiposity of the Lepob/ob mice) was performed. Body weight and food intake were recorded daily throughout the experiment.
Acute effect of peripherally-administered leptin on glucose tolerance in wild-type (Lep+/+) mice on LFD or HFD Male Lep+/+ mice (n = 13–14 per group) were fed either LFD or HFD. After 10 days on a diet, an ipGTT (1 g of glucose/kg body weight) was performed. Sixty minutes before the ipGTT, mice on each diet received either an i.p. injection of leptin (1.25 mg/kg body weight) or vehicle (PBS). Blood was taken from the V. facials and glucose concentration was measured using a glucometer (Accu-Check Performa; Roche, Basel, Switzerland) (14).
Acute action of peripherally-administered leptin on food intake and glucose tolerance of Lepob/ob mice on LFD or HFD Because Lepob/ob mice on HFD gain body weight more rapidly than LFD fed mice and the marked difference in body weight of age matched controls on LFD may result in changes in leptin sensitivity, weight-matched Lepob/ob mice were used for the control group on LFD. These mice were approximately 2 weeks older (8–9 weeks at the ipGTT) than the HFD mice (6– 7 weeks at the ipGTT). Mice were fed the HFD until the body weights matched the older LFD group, which took 9 days (n = 8–12 per group). Leptin (1.25 mg/kg body weight) or vehicle (PBS) was injected i.p. 60 min before an ipGTT (0.75 g glucose/kg body weight) was performed as described above. Cumulative food intake was measured 24 h after injection in a separate cohort of mice (n = 7–8 per group).
Acute action of centrally-administered leptin on glucose tolerance in wild-type and Lepob/ob mice on HFD To establish whether the acute effect of leptin on glucose tolerance in DIO mice differs depending on the route of application, we injected leptin i.c.v. in wild-type and Lepob/ob mice on a HFD. A stainless steel guide cannula (0.9 mm lateral and 0.1 mm posterior to bregma, 2.2 mm ventral to the © 2014 British Society for Neuroendocrinology
60
C. E. Koch et al.
surface) was stereotaxically implanted into the lateral cerebral ventricle (14). After a recovery time of 7 days, an ipGTT was performed in mice fasted over night. Sixty minutes before the ipGTT, both wild-type and Lepob/ob mice on HFD received either an i.c.v. leptin injection (4 lg; Lep+/+, n = 6; Lepob/ob, n = 8) or an i.c.v. vehicle injection (aCSF; Lep+/+, n = 6; Lepob/ob, n = 20). For the ipGTT, 1 g of glucose/kg body weight was used in wild-type mice and 0.75 g glucose/kg body weight for Lepob/ob as described above.
Pharmacological inhibition of central JNK and implication for glucose tolerance in HF fed Lepob/ob mice To determine whether hypothalamic inflammation via the JNK-pathway contributes to leptin resistance and glucose intolerance in Lepob/ob mice on HFD, the JNK-pathway was pharmacologically inhibited before leptin challenge. Lepob/ob mice were injected i.c.v. with the JNK-Inhibitor SP600125 [5 nmol in 5% dimethylsulphoxide (DMSO)/aCSF] or vehicle (5% DMSO/aCSF) followed by a second i.c.v. injection of either leptin (4 lg leptin in aCSF) or vehicle (aCSF) 30 min later (n = 4–8 per group). Sixty minutes after the second i.c.v. injection, an ipGTT (0.75 g glucose/kg body weight) was performed as described above.
Leptin-dependent STAT3 phosphorylation in the arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) of Lepob/ob mice on HFD We investigated whether the ability of leptin to centrally activate pSTAT3 in the ARC and VMH of Lepob/ob mice is altered by HFD. Lepob/ob mice (8 weeks of age) were fed either the LFD or the HFD. After 9 days on a diet, mice were injected i.c.v. with either vehicle 5% DMSO aCSF) or leptin (4 lg in 5% DMSO in aCSF) (n = 5–7 per group). One hour later, mice were anaesthetised (Narcoren; Merial GmbH, Hallbergmoos, Germany) and trans-cardially perfused with 0.9% saline containing heparin followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Perfused brains were removed and stored in the same solution over night. Thereafter, brains were dehydrated (in 30% sucrose/0.1 M phosphate buffer) until they had sunk, frozen in cooled isopentane, and sectioned coronally at 35 lm. Sections were stored in cryoprotectant at 20 °C. The immunohistochemical analysis of pSTAT3 was performed in accordance with a previously described protocol (20). Immunoreactive cells within one of the bilateral halves of the ARC and VMH were counted by two investigators who were blinded to the treatment, and multiplied by two to estimate the total number of immunoreactive cells per section.
Statistical analysis Data were analysed by one- or two-way ANOVA followed by a Holm–Sidak comparison test, as appropriate, using SIGMASTAT (Systat Software, San Jose, CA, USA). Where data failed equal variance or normality tests, they were analysed by one-way ANOVA on ranks followed by Dunn’s multiple-comparison test. Results are presented as the mean SEM. P ≤ 0.05 was considered statistically significant.
Results Effect of centrally-induced hyperleptinaemia on body weight and glucose tolerance On LFD, centrally-induced hyperleptinaemia (by i.c.v. infusion) led to a dramatic decline in body weight (27.7 0.5 g to 20.5 0.7 g. compared to vehicle-treated Lepob/ob mice, which gradually © 2014 British Society for Neuroendocrinology
increased their body weight from 27.7 0.6 g to 30.3 1.6 g (P ≤ 0.001) within the 6-day infusion regimen. After the switch from LFD to HFD, the reduction in body weight was completely arrested in mice infused with leptin. They exhibited a stable weight on HFD for the next 4 days on a diet (20.4 1.3 after 4 days), whereas leptin-infused mice on LFD continued to lose body weight down to 15.0 0.3 g (P ≤ 0.001). This profound reduction in body weight in the latter group necessitated termination of the experiment (Fig. 1A). Daily food intake of vehicle-injected mice on LFD averages at 4.1 0.4 g from days 9–13, and is significantly elevated compared to leptin-treated mice that consume 0.3 0.1 g per day on average (Fig. 1B; P ≤ 0.001). Average daily food intake was significantly increased when i.c.v. leptin-treated mice were transferred to HFD (2.1 0.1 g) relative to the group that remained on LFD (0.4 0.1 g from days 14–17; P ≤ 0.001) but was still significantly different from vehicle-injected mice on HFD that consumed 4.0 0.2 g (P ≤ 0.001). There was no significant difference in average daily food intake between vehicle-injected mice on LFD versus HFD (4.1. 0.4 g versus 4.0 0.2 g). Before termination of the experiment, after 4 days of HF feeding, glucose tolerance was tested. Vehicle-treated mice on HFD revealed profound glucose intolerance compared to leptin-treated mice on LFD. There was a significant difference in area under the curve (AUC) between these groups (Fig. 1C; P ≤ 0.001). In i.c.v. leptininfused mice fed a HFD, glucose tolerance was significantly reduced compared to the i.c.v. leptin-infused mice on LFD (P ≤ 0.001 in AUC), although it was still significantly improved compared to vehicle-infused mice on HFD (P = 0.003 in AUC).
Acute effect of peripherally-administered leptin on glucose tolerance in wild-type mice on LFD or HFD In normoglycaemic wild-type mice (Lep+/+ mice), we tested whether leptin alters glucose homeostasis. Ten days of HFD in wild-type mice was associated with a significant increase in body weight (mice on HFD after 10 days: 20.6 0.2 g versus mice on LFD: 18.6 0.3, P ≤ 0.001) and impaired glucose tolerance (P ≤ 0.001) compared to mice on LFD. As expected, i.p. leptin injection was not beneficial for further lowering normal glucose levels (Fig. 2; n = 12–13 per group). During states of leptin resistance, as described in wild-type mice on a HFD (21), acute i.p. leptin failed to improve impaired glucose homeostasis (Fig. 2; n = 13 per group). There was no significant difference in glucose concentration at any of the time points tested or in the AUC.
Acute action of peripherally-administered leptin on food intake and glucose tolerance of Lepob/ob mice on LFD or HFD Consistent with previous reports, acute i.p. leptin reduced food intake and improved glucose tolerance in Lepob/ob mice fed a LFD (14,22). Cumulative 24-h mean food intake was 5.1 0.18 g in the vehicle-treated group versus 4.1 0.21 g in the leptin-treated group (Fig. 3B; P = 0.003). Leptin injection 60 min before the ipGTT Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
(A) 40
Vehicle Leptin
61
***
Body weight (g)
35
**
*
30
25 Osmotic pump implantation
** **
*
20
15
0
(B )
High-fat diet
Low-fat diet
10 1
2
3
4
5
6
7
6 *** ** *** ***
5
** ***
8
9 10 11 12 13 14 15 16 17 18 Days
** **
** *
*** **
**
*** ***
4
Food intake (g)
Vehicle LFD Vehicle HFD Leptin LFD Leptin HFD
*** *
*
3
*
*** **
2
1
0
9
10
11
12
13 Days
15
16
20 30 AUC × 103
40
14
17
500
Vehicle HFD Leptin HFD Leptin LFD
400
Vehicle
**
Leptin
200
***
300
***
Glucose concentration (mg/dl)
(C)
Lep
100 0 0
30
60 Time (min)
90
120 0
10
50
Fig. 1. Effect of centrally-induced hyperleptinaemia on body weight and glucose tolerance. (A) Lepob/ob mice that underwent osmotic vehicle pump implantation (aCSF, n = 6) revealed a gradual increase in body weight on a low-fat diet (LFD) that was exacerbated by a high-fat diet (HFD) for 4 days. Relative to this group i.c.v. infused leptin (4.2 lg/day) led to a profound reduction in body weight within 10 days on LFD (P ≤ 0.001, n = 6). Intracerebroventricularly infused mice that were transferred to HFD for 4 days after 6 days on LFD stopped the leptin-induced body weight decline immediately after the change in diet relative to leptin-treated mice that continued to receive LFD (P ≤ 0.001, n = 7). (B) Cumulative 24-h food intake from days 9–17 of mice shown in (A). Change of diet on day 13 from LFD to HFD led to an increase in food intake in i.c.v. leptin-infused mice. (C) Glucose tolerance of mice shown in (A) at day 17. Vehicle infused Lepob/ob mice on HFD had a significantly impaired glucose tolerance relative to i.c.v. leptin infused Lepob/ob mice on LFD (P ≤ 0.001) and HFD (P = 0.003). Glucose tolerance of i.c.v. leptin infused Lepob/ob mice on HFD was significantly reduced compared to i.c.v. leptin infused Lepob/ob mice that remained on LFD (P ≤ 0.001). Data are shown as the mean SEM; *P ≤ 0.05; **P ≤ 0.01; *** P ≤ 0.001. AUC, area under the curve. Journal of Neuroendocrinology, 2014, 26, 58–67
© 2014 British Society for Neuroendocrinology
62
C. E. Koch et al.
250
i.p. Vehicle
200
i.p. Leptin 1.25 mg/kg
***
Glucose concentration (mg/dl)
300
150 100
i.p. Vehicle
50
i.p. Leptin 1.25 mg/kg
0 0
30
60 Time (min)
90
120 0
5
10
15
20
25
30
AUC × 103
Fig. 2. Acute effect of peripherally-administered leptin on glucose tolerance in wild-type (Lep+/+) mice on a low-fat diet (LFD) or a high-fat diet (HFD). High-fat feeding was associated with impaired glucose tolerance compared to mice on LFD (P ≤ 0.001). Relative to controls (n = 13–14 per group), peripherally-administered leptin (1.25 mg/kg in phosphate-buffered saline; n = 13 per group) 60 min before an i.p. glucose tolerance test had no effect on glucose tolerance of mice on LFD( ) or HFD( ). LFD, Low-fat diet; HFD, High-fat diet. The area under the curve (AUC) of leptin-treated mice was comparable to controls. Data are shown as the mean SEM. **P ≤ 0.001.
was sufficient to improve glucose homeostasis significantly (Fig. 3C; P = 0.02). When Lepob/ob mice were fed HFD, the body weight gain was accelerated compared to LFD mice; however, body weight between the two groups was matched with Lepob/ob mice on HFD being 2 weeks younger than LFD mice (LFD: 38.3 g 0.4 g; HFD: 38.8 g 0.6 g; Fig. 3A). Despite having similar body weights, acute i.p. leptin injection in Lepob/ob mice on HFD failed to reduce food intake (24-h mean food intake vehicle: 5.3 0.7 g; 24-h mean food intake leptin 5.8 0.36 g; Fig. 3B, right) or to improve glucose homeostasis (Fig. 3D) compared to Lepob/ob mice on a LFD (Fig. 3C).
Acute action of centrally-administered leptin on glucose tolerance in wild-type and Lepob/ob mice on HFD In this experiment, we investigated whether sensitivity to i.c.v. injected leptin is maintained in wild-type or Lepob/ob mice on HFD. Intracerebroventricular leptin injection 60 min before the ipGTT did not improve glucose homeostasis in wild-type mice on HFD (Fig. 4A). Surprisingly, in Lepob/ob mice on HFD, this treatment further deteriorated the impaired glucose tolerance in these mice. The AUC of Lepob/ob mice on HFD was significantly increased after i.c.v. leptin treatment compared to vehicle-injected mice (Fig. 4B; P = 0.007).
Pharmacological inhibition of central JNK and implication for glucose tolerance in HF fed Lepob/ob mice To investigate whether hypothalamic inflammation via JNK conveys the observed negative action of leptin on glucose homeostasis in Lepob/ob mice on HFD, we pharmacologically inhibited JNK in the central nervous system. Central inhibition of JNK by i.c.v. injection of 5 nmol of SP600125 (JNK-inhibitor) alone was not sufficient to improve glucose homeostasis in these mice (Fig. 4C). Intriguingly, however, i.c.v. injection of SP600125 (5 nmol) before i.c.v. leptin © 2014 British Society for Neuroendocrinology
(4 lg) potently reversed the deleterious effect of leptin on glucose homeostasis in Lepob/ob mice on HFD. The AUC of Lepob/ob mice on HFD treated i.c.v. with SP600125 and leptin was significantly decreased compared to i.c.v. vehicle or JNK-inhibitor treated mice (JNK-inhibitor/leptin versus JNK-inhibitor/vehicle P = 0.017; JNK-inhibitor/leptin versus vehicle/vehicle P = 0.014).
Leptin-dependent STAT3 phosphorylation in the ARC and VMH of Lepob/ob mice on LFD and HFD We next investigated whether high-fat feeding in Lepob/ob mice alters central leptin sensitivity at the molecular level. In highly-sensitive Lepob/ob mice on LFD, an i.c.v. leptin injection was associated with a profound increase in the number of pSTAT3(Tyr705) positive cells in the ARC (by 4.4-fold; Fig. 5A,B) and VMH (by 6.5-fold; Fig. 5C,D) compared to vehicle (vehicle versus leptin P ≤ 0.001). In both brain regions, vehicle-treated Lepob/ob mice on HFD revealed a level of pSTAT3 immunoreactivity that was comparable to the leptin-treated LFD-group. There was no significant difference between the two groups (in the ARC LFD leptin versus HFD vehicle was P = 0.057) but a significant difference to the LFD vehicle-treated mice (ARC: LFD vehicle versus HFD vehicle P = 0.008; VMH: LFD vehicle versus HFD vehicle P = 0.004) in both regions. In Lepob/ob mice on HFD, i.c.v. leptin did not increase the number of pSTAT3 (Tyr705) immunoreactive cells in the ARC and VMH relative to the vehicle group.
Discussion It is well established that the regulation of glucose homeostasis is one of the essential pleiotropic actions of leptin. The glucose-lowering properties of the hormone appear to largely originate from the hypothalamic sensitisation of the insulin signalling pathway (14). Nonetheless, whether or not leptin exhibits a similar glucose-lowering Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
Low-fat diet
Body weight (g)
40 35 30 25 0 0
2
4 6 Days
8
10
Glucose concentration (mg/dl)
(C)
6 5 ** 4 3 2 1 Vehicle Leptin
0
Vehicle Leptin
Low-fat diet 500 400 i.p. Vehicle
300 200
i.p. Leptin 1.25 mg/kg
*
100 0
0
30
60 90 120 Time (min)
150
(D)
180
0
20
40 AUC × 103
60
80
60
80
High-fat diet
600 Glucose concentration (mg/dl)
High-fat diet
(B ) Cumulative food intake of 24 h (g)
(A)
63
500 400
i.p. Leptin 1.25 mg/kg
300 i.p. Vehicle
200 100 0
0
30
60
90 120 Time (min)
150
180
0
20
40 AUC × 103
Fig. 3. Acute action of peripherally-administered leptin on food intake and glucose tolerance of Lepob/ob mice on a low-fat diet (LFD) or a high-fat diet (HFD). (A) During high-fat feeding, Lepob/ob mice revealed a rapid increase in body weight gain relative to Lepob/ob mice on LFD. Mice on HFD were 2 weeks younger than mice on LFD so that body weight was identical in both groups after 9 days of HF feeding. At this time point, peripherally-administered leptin (1.25 mg/ kg) 60 min before an i.p. glucose tolerance test had no effect on (B) 24-h food intake or (C) glucose tolerance in Lepob/ob mice on LFD compared to (D) Lepob/ob mice on HFD (n = 7–8 per group). AUC, area under the curve. In the latter group, leptin reduced (B) food intake over 24 h (n = 7–8 per group; P = 0.003) and (C) improved glucose tolerance 60 min after injection (n = 11–12 per group; P = 0.02). Data are shown as the mean SEM; *P ≤ 0.05; **P ≤ 0.001. Δ, LFD; ○, HFD.
capacity in DIO, the major cause of obesity in humans, remains unclear. In the present study, we investigated the development of leptin insensitivity after DIO in both the presence and absence of leptin in wild-type and Lepob/ob mice. The role of the JNK inflammatory pathway in the development of leptin resistance in DIO was also investigated. It is well known that leptin fails to reduce food intake and increase energy expenditure during DIO in humans and mice (23). Journal of Neuroendocrinology, 2014, 26, 58–67
In mice fed a HFD, leptin signalling via the JAK2/STAT3-pathway has been shown to be impaired in the ARC, which is the neuronal centre for the regulation of energy metabolism (21,24), although failure of leptin to regulate the long form of the leptin receptor (L-Rb) in the ARC can occur before failure of the leptin-induced JAK2/STAT3-pathway (25). In the present study, we first addressed the intriguing question of whether hyperleptinaemia per se or DIO induces leptin resistance © 2014 British Society for Neuroendocrinology
64
C. E. Koch et al.
Glucose concentration (mg/dl)
(A)
Lep+/+ High-fat diet
400 300
i.c.v. Leptin 4 µg
200 i.c.v. Vehicle 100 0
0
30
10
0
20 30 AUC × 103
40
50
Lepob/ob High-fat diet
(B) 600 500 400
i.c.v. Leptin 4 µg 300
**
Glucose concentration (mg/dl)
60 90 120 150 180 Time (min)
i.c.v. Vehicle 200 100 0
0
30
60 90 120 150 180 0 Time (min)
20
60 40 AUC × 103
80
100
Lepob/ob High-fat diet
500 400
5 nmol JNK-Inhib.
300
Vehicle
200
5 nmol JNK-Inhib. + 4 µg Leptin
* *
Glucose concentration (mg/dl)
(C) 600
100 0
0
30
60 90 120 150 180 0 Time (min)
20
40 60 AUC × 103
80
100
Fig. 4. Central implication of leptin and hypothalamic inflammation on glucose homeostasis. (A) Centrally-administered leptin (4 lg) 60 min before an i.p. glucose tolerance test (ipGTT) did not improve glucose tolerance of Lep+/+ mice on a high-fat diet (HFD). The area under the curve (AUC) of leptin-treated mice was comparable with vehicle-treated controls (n = 6 per group). (B) In Lepob/ob mice on HFD, centrally-injected leptin (4 lg; n = 8) 60 min before an ipGTT impaired glucose tolerance. The calculated AUC of leptin treated mice was significantly increased compared to vehicle-treated controls (P = 0.007). (C) Intracerebroventricular injection of SP600125 (5 nmol; n = 8), a pharmacological inhibitor of c-Jun n-terminal kinase (JNK), followed by an i.c.v. vehicle injection 30 min later, and an ipGTT 60 min after the second injection, did not alter glucose tolerance of Lepob/ob mice on HFD. Consecutive i.c.v. injections of the JNKinhibitor and i.c.v. leptin (4 lg; n = 4) 30 min apart followed by an ipGTT 60 min later improved glucose tolerance significantly compared to vehicle-treated controls (n = 6; P = 0.014) and SP600125-treated mice (P = 0.017). Data are shown as the mean SEM; *P ≤ 0.05; **P ≤ 0.01.
utilising an i.c.v leptin infusion regimen in Lepob/ob mice. In these mice that are exquisitely leptin sensitive, central hyperleptinaemia, induced by chronic i.c.v. leptin infusion (4.2 lg/day) for 10 days, led to a dramatic decline in body weight and food intake, as well as an improvement in glucose tolerance, ruling out the possibility of hyperleptinaemia being the cause of leptin resistance. Furthermore, high-fat feeding attenuated the leptin-induced decline in the body weight trajectory, cumulative food intake and impaired glucose tolerance, suggesting that the consumption of a HFD is the leading © 2014 British Society for Neuroendocrinology
cause of leptin resistance. Because the diet-induced loss of the glucose-lowering properties of leptin might be specific to the ob/ob genotype, we investigated the acute effect of i.p. leptin on glucose tolerance in wild-type mice fed a HFD. In line with these findings also in wild-type mice on HFD, acute i.p. leptin application failed to improve glucose tolerance. Consistent with the hypothesis that HFD induces leptin resistance, acute leptin was unable to reduce 24-h food intake or glucose tolerance in leptin-deficient, Lepob/ob mice on a HFD in contrast to weight-matched Lepob/ob control mice fed a Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
(B)
(A)
Lepob/ob
LFD-Leptin
ARC
100 µm HFD-Vehicle
HFD-Leptin
pSTAT3 (Tyr705) immunoreactive cells in ARC
250 LFD-Vehicle
*** ** P = 0.0569
200 *** 150
100
50
Vehicle 0
HFD-Leptin
Vehicle
Leptin HFD
(D) 225 pSTAT3 (Y705) counted cells in the VMH
HFD-Vehicle
LFD-Leptin
Leptin LFD
(C) LFD-Vehicle
65
200 175 **
150
***
125 100 75 50 25 0
Vehicle
Leptin LFD
Vehicle
Leptin
HFD
Fig. 5. Leptin-dependent signal transducer and activator of transcription (STAT)3 phosphorylation in the arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) of Lepob/ob mice on a low-fat diet (LFD) and high-fat diet (HFD). Centrally-administered leptin (4 lg) led to a significant increase in phosphorylated (p)STAT3(Tyr705) immunoreactive cells in the ARC and VMH of Lepob/ob mice fed a LFD compared to vehicle (LFD: vehicle versus leptin P ≤ 0.001). In Lepob/ob mice fed a HFD, vehicle treatment was associated with an increased number of pSTAT3(Tyr705) immunoreactive cells relative to the LFD vehicle group (ARC: LFD vehicle versus HFD vehicle P = 0.008; VMH: LFD vehicle versus HFD vehicle P = 0.004) and central leptin treatment had no effect in these mice (n = 5–7 per group). (A) Shown are representative images of pSTAT3(Tyr705) immunoreactivity in the ARC of each treatment group (B) and the respective quantification of pSTAT3(Tyr705) positive cells. (C) Shown are representative images of pSTAT3(Tyr705) immunoreactivity in the VMH of each treatment group (D) and the respective quantification of pSTAT3(Tyr705) positive cells. Data are shown as the mean SEM; ** P ≤ 0.01; ***P ≤ 0.001.
LFD. The development of leptin insensitivity in this instance was independent of adiposity because Lepob/ob mice on the HFD and LFD were weight matched. These data collectively demonstrate that a HFD can lead to leptin insensitivity in the absence of endogenous leptin and during a profound i.c.v. leptin-induced decline in body weight, ruling out a role for hyperleptinaemia in this process. Our data are consistent with the findings of White et al. (26) who reported that the anorexigenic effect of i.p. leptin was abolished in Lepob/ob mice fed the HFD used in the present study for 14 days. By contrast, however, a study performed by Knight et al. (27) in which Lepob/ob mice were fed a high-fat diet for a very long Journal of Neuroendocrinology, 2014, 26, 58–67
period (20 weeks), and leptin was replaced at low doses during this time, suggested that hyperleptinaemia is required for the development of leptin resistance. However, this discrepancy might be explained by the very different experimental approaches, as well as by the time period, 9 days versus 20 weeks, on the diet. A complex time course in hypothalamic inflammation has been suggested, and this might obscure the analysis of hyperleptinaemia as a contributing factor to the development of leptin resistance (27). It has been proposed that HF feeding leads to impaired entry of leptin into the brain trough the blood–brain barrier (28). This may be a result of the increase in circulating triglycerides typically observed © 2014 British Society for Neuroendocrinology
66
C. E. Koch et al.
during HFD (29). In the present study, Lepob/ob mice on HFD were resistant to peripheral leptin but sensitive to an acute i.c.v. leptin injection in terms of the hormone’s glycaemic effect. Surprisingly, however, in contrast to chronic i.c.v. administration, acute i.c.v. leptin injection increased glucose intolerance. This difference might be explained by a much higher peak concentration that is achieved by acute injection of a 4 lg dose compared to infusion of 4.2 lg over 24 h. This high peak dose might induce pro-inflammatory properties of the hormone. Leptin is a cytokine and, similar to other cytokines, such as interleukin-6 and tumour necrosis factor (TNF)a, can induce a pro-inflammatory response via its activation of the JAK2/STAT3pathway (30,31). Synergistic actions of leptin and these other cytokines have been reported. In a study by Rizk et al. (32), i.v. injected TNFa was much more potent in activating pSTAT3 in hypothalamic lysates than i.v. injected leptin. A dose of 3 mg/kg of leptin compared to 12 lg/kg TNFa led to an activation of pSTAT3 in the hypothalamus that was of a similar magnitude (32). Systemic co-administration of both cytokines was associated with a marked synergistic activation of pSTAT3 in hypothalamic lysates. Intriguingly, we detected a marked increase in the number of pSTAT3 immunoreactive cells in the ARC and VMH of vehicle-treated Lepob/ob mice on HFD compared to LFD that was comparable to leptin-induced pSTAT3 in the latter group. To our knowledge, this diet-induced induction of pSTAT3 has not been previously reported in wild-type mice. Munzberg et al. (21) have shown that wild-type mice on HFD for 16 weeks did not reveal an increase of pSTAT3 immunoreactive cells in the ARC. To determine whether this effect is specific to the ob/ob genotype, a detailed time-course study of pSTAT3 immunoreactivity at different periods of HFD is required in future studies. It is plausible that, despite the lack of leptin in Lepob/ob mice fed a HFD, other circulating cytokines, whose concentration in the blood is elevated during HFD, may cause pSTAT3 activation in the ARC and VMH. The concentration and composition of these cytokines might be different in Lepob/ob mice on HFD relative to wild-type mice on this diet. Additional studies are required to address this question. Interestingly, during DIO, leptin failed to induce pSTAT3 in both hypothalamic regions of Lepob/ob mice, which suggests that pSTAT3 in the control group is already maximally activated. It is an unexpected finding that Lepob/ob mice on HFD are still obese and glucose intolerant despite elevated pSTAT3 in the ARC. Consistently, however, it has been reported that genetically enhanced STAT3 activation in pro-opiomelanocortin-expressing neurones in the hypothalamus provokes negative-feedback inhibition of leptin signalling in obesity (33). Interestingly, mice with a mutated STAT3 binding site in the leptin receptor (homozygous for L-RbSer1138 expression) have impaired energy homeostasis similar to Lepdb/db mice with a total loss of function of L-Rb; however, in contrast to Lepdb/db mice, they retain normal blood glucose when pair-fed with wild-type mice (34). Together with accumulating evidence that leptin sensitises central insulin action via the phosphatidylinositol-3-kinase/AKT-pathway (14), this suggests that the glucose-lowering properties of leptin are mediated via JAK2/STAT3 independent signalling pathways. In addition, our finding that i.c.v. leptin impairs glucose tolerance in Lepob/ ob mice on HFD despite the lack of pSTAT3 activation supports STAT3 independent regulation of glucose homeostasis by leptin. © 2014 British Society for Neuroendocrinology
A large body of evidence suggests that hypothalamic inflammation during HFD is associated with (and may be causative in) the development of obesity (35). The JNK inflammatory pathway appears to play a crucial role in this process and the development of obesity and insulin insensitivity because JNK is elevated in the ARC of mice fed a HFD and in Lepob/ob mice (36). In the present study, we tested the hypothesis that acute pharmacological inhibition of central JNK reverses the deleterious effect of leptin on glucose homeostasis in Lepob/ob mice on a HFD. Although i.c.v. injection of the JNK-inhibitor 30 min before i.c.v. leptin injection restored the glucose-lowering properties of leptin, it failed to alter glucose tolerance when given alone. This suggests that the degree of hypothalamic inflammation via JNK activity that was reversed by i.c.v. injection of 5 nmol SP600125 was not sufficient to solely improve glucose homeostasis. These findings imply that elevated hypothalamic inflammation via JNK mediates pro-diabetic properties of central leptin via the additive effects of the exogenous application of this pro-inflammatory cytokine at a metabolic state at which other cytokines are already elevated. The deterioration of glucose homeostasis by leptin that was observed in DIO Lepob/ob mice may be mediated via as yet unknown leptin signalling pathways that become activated at this particular metabolic state of extreme obesity. However, the involvement of other pathways (e.g. phosphoinositide 3-kinase/AKT signalling or WNT signalling) that have been implicated with central regulation of glucose homeostasis remains to be investigated further. It is also plausible that leptin indirectly worsens glucose homeostasis in these mice. A series of experiments is urgently required to delineate the precise molecular mechanisms with respect to how central leptin regulates glucose metabolism. Received 18 July 2013, revised 27 November 2013, accepted 23 December 2013
References 1 Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994; 372: 425–432. 2 Campfield LA, Smith FJ, Burn P. The OB protein (leptin) pathway – a link between adipose tissue mass and central neural networks. Horm Metab Res 1996; 28: 619–632. 3 Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998; 395: 763–770. 4 Farooqi IS, Jebb SA, Langmack G, Lawrence E, Cheetham CH, Prentice AM, Hughes IA, McCamish MA, O’Rahilly S. Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med 1999; 341: 879–884. 5 Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst JA, Cheetham CH, Earley AR, Barnett AH, Prins JB, O’Rahilly S. Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997; 387: 903–908. 6 Kamohara S, Burcelin R, Halaas JL, Friedman JM, Charron MJ. Acute stimulation of glucose metabolism in mice by leptin treatment. Nature 1997; 389: 374–377.
Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
7 Muzzin P, Eisensmith RC, Copeland KC, Woo SL. Correction of obesity and diabetes in genetically obese mice by leptin gene therapy. Proc Natl Acad Sci USA 1996; 93: 14804–14808. 8 Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F. Effects of the obese gene product on body weight regulation in ob/ob mice. Science 1995; 269: 540–543. 9 Schwartz MW, Baskin DG, Bukowski TR, Kuijper JL, Foster D, Lasser G, Prunkard DE, Porte D Jr, Woods SC, Seeley RJ, Weigle DS. Specificity of leptin action on elevated blood glucose levels and hypothalamic neuropeptide Y gene expression in ob/ob mice. Diabetes 1996; 45: 531– 535. 10 Sivitz WI, Walsh SA, Morgan DA, Thomas MJ, Haynes WG. Effects of leptin on insulin sensitivity in normal rats. Endocrinology 1997; 138: 3395–3401. 11 Yu X, Park BH, Wang MY, Wang ZV, Unger RH. Making insulin-deficient type 1 diabetic rodents thrive without insulin. Proc Natl Acad Sci USA 2008; 105: 14070–14075. 12 Hedbacker K, Birsoy K, Wysocki RW, Asilmaz E, Ahima RS, Farooqi IS, Friedman JM. Antidiabetic effects of IGFBP2, a leptin-regulated gene. Cell Metab 2010; 11: 11–22. 13 Van HM, Compton DS, France CF, Tedesco RP, Fawzi AB, Graziano MP, Sybertz EJ, Strader CD, Davis HR Jr. Diet-induced obese mice develop peripheral, but not central, resistance to leptin. J Clin Invest 1997; 99: 385–390. 14 Koch C, Augustine RA, Steger J, Ganjam GK, Benzler J, Pracht C, Lowe C, Schwartz MW, Shepherd PR, Anderson GM, Grattan DR, Tups A. Leptin rapidly improves glucose homeostasis in obese mice by increasing hypothalamic insulin sensitivity. J Neurosci 2010; 30: 16180–16187. 15 Prolo P, Wong ML, Licinio J. Leptin. Int J Biochem Cell Biol 1998; 30: 1285–1290. 16 Hubschle T, Thom E, Watson A, Roth J, Klaus S, Meyerhof W. Leptininduced nuclear translocation of STAT3 immunoreactivity in hypothalamic nuclei involved in body weight regulation. J Neurosci 2001; 21: 2413–2424. 17 Bjorbaek C, Uotani S, da SB, Flier JS. Divergent signaling capacities of the long and short isoforms of the leptin receptor. J Biol Chem 1997; 272: 32686–32695. 18 Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS. A central role for JNK in obesity and insulin resistance. Nature 2002; 420: 333–336. 19 Tuncman G, Hirosumi J, Solinas G, Chang L, Karin M, Hotamisligil GS. Functional in vivo interactions between JNK1 and JNK2 isoforms in obesity and insulin resistance. Proc Natl Acad Sci USA 2006; 103: 10741–10746. 20 Munzberg H, Huo L, Nillni EA, Hollenberg AN, Bjorbaek C. Role of signal transducer and activator of transcription 3 in regulation of hypothalamic proopiomelanocortin gene expression by leptin. Endocrinology 2003; 144: 2121–2131. 21 Munzberg H, Flier JS, Bjorbaek C. Region-specific leptin resistance within the hypothalamus of diet-induced obese mice. Endocrinology 2004; 145: 4880–4889.
Journal of Neuroendocrinology, 2014, 26, 58–67
67
22 Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, Friedman JM. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 1995; 269: 543–546. 23 Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995; 269: 546–549. 24 El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, Flier JS. Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obesity. J Clin Invest 2000; 105: 1827–1832. 25 Mitchell SE, Nogueiras R, Morris A, Tovar S, Grant C, Cruickshank M, Rayner DV, Dieguez C, Williams LM. Leptin receptor gene expression and number in the brain are regulated by leptin level and nutritional status. J Physiol 2009; 587: 3573–3585. 26 White CL, Whittington A, Barnes MJ, Wang Z, Bray GA, Morrison CD. HF diets increase hypothalamic PTP1B and induce leptin resistance through both leptin-dependent and -independent mechanisms. Am J Physiol Endocrinol Metab 2009; 296: E291–E299. 27 Knight ZA, Hannan KS, Greenberg ML, Friedman JM. Hyperleptinemia is required for the development of leptin resistance. PLoS ONE 2010; 5: e11376. 28 Banks WA. Is obesity a disease of the blood–brain barrier? Physiological, pathological, and evolutionary considerations. Curr Pharm Des 2003; 9: 801–809. 29 Banks WA, Coon AB, Robinson SM, Moinuddin A, Shultz JM, Nakaoke R, Morley JE. Triglycerides induce leptin resistance at the blood–brain barrier. Diabetes 2004; 53: 1253–1260. 30 Romanatto T, Cesquini M, Amaral ME, Roman EA, Moraes JC, Torsoni MA, Cruz-Neto AP, Velloso LA. TNF-alpha acts in the hypothalamus inhibiting food intake and increasing the respiratory quotient – effects on leptin and insulin signaling pathways. Peptides 2007; 28: 1050–1058. 31 Piekorz R, Schlierf B, Burger R, Hocke GM. Reconstitution of IL6-inducible differentiation of a myeloid leukemia cell line by activated Stat factors. Biochem Biophys Res Commun 1998; 250: 436–443. 32 Rizk NM, Stammsen D, Preibisch G, Eckel J. Leptin and tumor necrosis factor-alpha induce the tyrosine phosphorylation of signal transducer and activator of transcription proteins in the hypothalamus of normal rats in vivo. Endocrinology 2001; 142: 3027–3032. 33 Ernst MB, Wunderlich CM, Hess S, Paehler M, Mesaros A, Koralov SB, Kleinridders A, Husch A, Munzberg H, Hampel B, Alber J, Kloppenburg P, Bruning JC, Wunderlich FT. Enhanced Stat3 activation in POMC neurons provokes negative feedback inhibition of leptin and insulin signaling in obesity. J Neurosci 2009; 29: 11582–11593. 34 Myers MG Jr. Lecture Outstanding Scientific Achievement Award deconstructing leptin: from signals to circuits. Diabetes 2010; 59: 2708–2714. 35 Thaler JP, Choi SJ, Schwartz MW, Wisse BE. Hypothalamic inflammation and energy homeostasis: resolving the paradox. Front Neuroendocrinol 2010; 31: 79–84. 36 Benzler GK, Ganjam J, Legler K, Steger J, Stohrz S, Kruger M, Tups A. Acute inhibition of central c-Jun N-terminal kinase restores hypothalamic insulin signalling and alleviates glucose intolerance in diabetic mice. J Neuroendocrinol 2013; 25: 446–454.
© 2014 British Society for Neuroendocrinology
ORIGINAL ARTICLE
High-Fat Diet Induces Leptin Resistance in Leptin-Deficient Mice C. E. Koch*, C. Lowe*, D. Pretz*, J. Steger*, L. M. Williams† and A. Tups* *Department of Animal Physiology, Faculty of Biology, Philipps University Marburg, Marburg, Germany. †Metabolic Health Group, Rowett Institute of Nutrition and Health, University of Aberdeen, Aberdeen, UK.
Journal of Neuroendocrinology
Correspondence to: A. Tups, Karl-von-Frisch-Straße 8, 35043 Marburg, Germany (e-mail: [email protected]).
The occurrence of type II diabetes is highly correlated with obesity, although the mechanisms linking the two conditions are incompletely understood. Leptin is a potent insulin sensitiser and, in leptin-deficient, insulin insensitive, Lepob/ob mice, leptin improves glucose tolerance, indicating that leptin resistance may link obesity to insulin insensitivity. Leptin resistance occurs in response to a high-fat diet (HFD) and both hyperleptinaemia and inflammation have been proposed as causative mechanisms. Scrutinising the role of hyperleptinaemia in this process, central hyperleptinaemia in Lepob/ob mice was induced by chronic i.c.v. infusion of leptin (4.2 lg/day) over 10 days. This treatment led to a dramatic decline in body weight and food intake, as well as an improvement in glucose tolerance. Transfer to HFD for 4 days markedly arrested the beneficial effects of leptin on these parameters. Because Lepob/ob mice are exquisitely sensitive to leptin, the possibility that leptin could reverse HFD-induced glucose intolerance in these animals was investigated. HFD led to increased body weight and glucose intolerance compared to a lowfat diet (LFD). Older and heavier Lepob/ob mice were used as body weight-matched controls. Mice in each group received either i.p. leptin (1.25 mg/kg) or vehicle, and glucose tolerance, food intake and the number of phosphorylated signal transducer and activator of transcription (pSTAT)3 immunoreactive cells in the arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) were analysed. Leptin improved glucose tolerance (P = 0. 019) and reduced food intake in Lepob/ob mice on LFD (P ≤ 0.001) but was ineffective in mice on HFD. Furthermore, when leptin was administered centrally, the glucose tolerance of Lepob/ob mice on HFD was significantly impaired (P = 0.007). Although leptin induced the number of pSTAT3 immunoreactive cells in the ARC and VMH of Lepob/ob mice on LFD, HFD was associated with elevated pSTAT3 immunoreactivity in vehicle-treated Lepob/ob mice that was unaffected by leptin treatment, suggesting central leptin resistance. Negating central inflammation by co-administering a c-Jun n-terminal kinase (JNK) inhibitor reinstated the glucose-lowering effects of leptin. These findings demonstrate that Lepob/ob mice develop leptin resistance on a HFD independent of hyperleptinaemia and also indicate that the JNK inflammatory pathway plays a key role in the induction of dietinduced glucose intolerance. Key words: energy and glucose homeostasis, hypothalamus, inflammation, JNK pathway
Obesity and associated disorders, particularly diabetes mellitus type II, have become a leading medical challenge to modern societies. Leptin was discovered in 1994 (1) and was initially characterised to reduce food intake and increase energy expenditure (2–5). In obesity, circulating levels of leptin, secreted primarily by adipocytes, increase, although its effect of reducing food intake and increasing energy expenditure diminishes. Thus, the establishment of leptin resistance is regarded as a leading cause in the onset of obesity.
doi: 10.1111/jne.12131
Despite the well characterised anorexigenic and catabolic properties of leptin, centrally-acting leptin is also essential for the maintenance of glucose homeostasis and is a potent insulin sensitiser (4,6–11). Because obesity gives rise to insulin insensitivity, the first stage in the development of diabetes mellitus type II, leptin resistance has been proposed as a major mechanism linking these two conditions. It has been shown that low doses of leptin that are ineffective in regulating food intake and body weight reduce
HFD induces leptin resistance in Lepob/ob mice
plasma glucose and insulin concentrations (12). However, the molecular mechanism underlying glucose-lowering properties and the insulin-sensitising effects of leptin are not well characterised. An imbalance of increased energy intake and decreased energy expenditure leads to obesity. Amongst other reasons, this imbalance can be caused by the excessive consumption of a diet high in saturated fatty acids (13) or by natural loss of function mutations in the leptin gene or related components of the neuroendocrine network of energy balance (1,5). The loss of leptin function and its consequences for the maintenance of energy balance have been well characterised using the Lepob/ob mouse, which serves as a naturally occurring animal model for this metabolic condition. We recently demonstrated that acute i.p. or i.c.v. injections of leptin improve glucose homeostasis in these mice (14), demonstrating that the presence of leptin is necessary for normal insulin sensitivity and glucose homeostasis. However, in humans, the incidence of naturally occurring loss of leptin function by genetic reasons as a cause for obesity is very low. By contrast, diet-induced obesity (DIO), which is the leading cause for obesity in humans, is characterised by hyperleptinaemia (15). To determine the primary mechanism underlying development of leptin resistance, Lepob/ob mice were administered leptin, via i.c.v. infusion, to induce central hyperleptinaemia and then fed a high-fat diet (HFD) to investigate the role of hyperleptinaemia versus diet in the development of this process. Because HFD appeared to be essential for inducing leptin resistance, we established that acute central leptin resistance occured in DIO wild-type mice particularly considering the glucoselowering properties of the hormone. Furthermore, we characterised whether Lepob/ob mice that are highly sensitive to exogenously applied leptin maintain this sensitivity during high-fat feeding in terms of the acute and chronic metabolic properties of the hormone. We additionally assessed whether these mice retain central leptin sensitivity with respect to the ability to activate the janus kinase 2/signal transducer and activator of transcription 3 (JAK2/ STAT3) pathway because it is well established that leptin regulates energy homeostasis via signalling through this pathway (16,17). Lepob/ob mice on HFD exhibited a morbidly obese phenotype. Recent intriguing evidence suggests that obesity is characterised by central inflammation, particularly increased activation of the c-Jun n-terminal kinase (JNK)-pathway (18,19). By pharmacological reversal of central JNK activity, we investigated whether inflammation via JNK leads to central leptin resistance in terms of its glucose-lowering potential.
Materials and methods Animals All procedures involving animals were licensed under national animal ethics legislation and received approval by the federal public authority for animal ethics. Male C57BL/6JRj wild-type (Lep+/+) and Lepob/ob mice were purchased from Janvier (Le Genest-Saint-Isle, France). They were housed individually under a 12/12 h light/dark cycle with an ambient temperature of 24 °C. All mice had ad lib. access to food and water except the period of food withdrawal (over night) before all of the experiments. Mice were fed either a low-fat diet (LFD; D12450B Research Diets, Brunswick, NJ, USA) with 10% Journal of Neuroendocrinology, 2014, 26, 58–67
59
fat by energy (kcal) or HFD (D12492 Research Diets) with 60% fat by energy (kcal).
Effect of centrally-induced hyperleptinaemia on body weight and glucose tolerance To determine whether hyperleptinaemia or high-fat feeding promotes leptin resistance, in Lepob/ob mice, central hyperleptinaemia was induced by chronic i.c.v leptin infusion before mice underwent dietary interventions. After a 6day observation period, a stainless steel cannula (0.9 mm lateral and 0.1 mm posterior to bregma, 2.5 mm ventral to the surface) was stereotaxically implanted into the lateral cerebral ventricle of 7-week-old Lepob/ob mice as described previously (14). The cannula was connected by silicone tubing to a miniosmotic pump that was implanted subcutaneously in the scapular region (Pump Model 1002; Alzet Inc., Cupertino, CA, USA). Mice were divided into two groups, one group was centrally infused with leptin [4.2 lg in artificial cerebrospinal fluid (aCSF)/day, n = 13], whereas the other group received vehicle (aCSF, n = 6). During this period, mice remained on a LFD. Six days after surgery when the body weight of the leptin-treated group had declined dramatically, this group was subdivided into two groups. One of these groups (n = 6) remained on LFD, whereas the other group and the vehicle-treated mice (n = 6) were transferred to a HFD (n = 7). Four days after the change in diet, an i.p. glucose tolerance test (ipGTT) (0.75 g glucose/kg body weight to reflect the increased glucose insensitivity and increased adiposity of the Lepob/ob mice) was performed. Body weight and food intake were recorded daily throughout the experiment.
Acute effect of peripherally-administered leptin on glucose tolerance in wild-type (Lep+/+) mice on LFD or HFD Male Lep+/+ mice (n = 13–14 per group) were fed either LFD or HFD. After 10 days on a diet, an ipGTT (1 g of glucose/kg body weight) was performed. Sixty minutes before the ipGTT, mice on each diet received either an i.p. injection of leptin (1.25 mg/kg body weight) or vehicle (PBS). Blood was taken from the V. facials and glucose concentration was measured using a glucometer (Accu-Check Performa; Roche, Basel, Switzerland) (14).
Acute action of peripherally-administered leptin on food intake and glucose tolerance of Lepob/ob mice on LFD or HFD Because Lepob/ob mice on HFD gain body weight more rapidly than LFD fed mice and the marked difference in body weight of age matched controls on LFD may result in changes in leptin sensitivity, weight-matched Lepob/ob mice were used for the control group on LFD. These mice were approximately 2 weeks older (8–9 weeks at the ipGTT) than the HFD mice (6– 7 weeks at the ipGTT). Mice were fed the HFD until the body weights matched the older LFD group, which took 9 days (n = 8–12 per group). Leptin (1.25 mg/kg body weight) or vehicle (PBS) was injected i.p. 60 min before an ipGTT (0.75 g glucose/kg body weight) was performed as described above. Cumulative food intake was measured 24 h after injection in a separate cohort of mice (n = 7–8 per group).
Acute action of centrally-administered leptin on glucose tolerance in wild-type and Lepob/ob mice on HFD To establish whether the acute effect of leptin on glucose tolerance in DIO mice differs depending on the route of application, we injected leptin i.c.v. in wild-type and Lepob/ob mice on a HFD. A stainless steel guide cannula (0.9 mm lateral and 0.1 mm posterior to bregma, 2.2 mm ventral to the © 2014 British Society for Neuroendocrinology
60
C. E. Koch et al.
surface) was stereotaxically implanted into the lateral cerebral ventricle (14). After a recovery time of 7 days, an ipGTT was performed in mice fasted over night. Sixty minutes before the ipGTT, both wild-type and Lepob/ob mice on HFD received either an i.c.v. leptin injection (4 lg; Lep+/+, n = 6; Lepob/ob, n = 8) or an i.c.v. vehicle injection (aCSF; Lep+/+, n = 6; Lepob/ob, n = 20). For the ipGTT, 1 g of glucose/kg body weight was used in wild-type mice and 0.75 g glucose/kg body weight for Lepob/ob as described above.
Pharmacological inhibition of central JNK and implication for glucose tolerance in HF fed Lepob/ob mice To determine whether hypothalamic inflammation via the JNK-pathway contributes to leptin resistance and glucose intolerance in Lepob/ob mice on HFD, the JNK-pathway was pharmacologically inhibited before leptin challenge. Lepob/ob mice were injected i.c.v. with the JNK-Inhibitor SP600125 [5 nmol in 5% dimethylsulphoxide (DMSO)/aCSF] or vehicle (5% DMSO/aCSF) followed by a second i.c.v. injection of either leptin (4 lg leptin in aCSF) or vehicle (aCSF) 30 min later (n = 4–8 per group). Sixty minutes after the second i.c.v. injection, an ipGTT (0.75 g glucose/kg body weight) was performed as described above.
Leptin-dependent STAT3 phosphorylation in the arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) of Lepob/ob mice on HFD We investigated whether the ability of leptin to centrally activate pSTAT3 in the ARC and VMH of Lepob/ob mice is altered by HFD. Lepob/ob mice (8 weeks of age) were fed either the LFD or the HFD. After 9 days on a diet, mice were injected i.c.v. with either vehicle 5% DMSO aCSF) or leptin (4 lg in 5% DMSO in aCSF) (n = 5–7 per group). One hour later, mice were anaesthetised (Narcoren; Merial GmbH, Hallbergmoos, Germany) and trans-cardially perfused with 0.9% saline containing heparin followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Perfused brains were removed and stored in the same solution over night. Thereafter, brains were dehydrated (in 30% sucrose/0.1 M phosphate buffer) until they had sunk, frozen in cooled isopentane, and sectioned coronally at 35 lm. Sections were stored in cryoprotectant at 20 °C. The immunohistochemical analysis of pSTAT3 was performed in accordance with a previously described protocol (20). Immunoreactive cells within one of the bilateral halves of the ARC and VMH were counted by two investigators who were blinded to the treatment, and multiplied by two to estimate the total number of immunoreactive cells per section.
Statistical analysis Data were analysed by one- or two-way ANOVA followed by a Holm–Sidak comparison test, as appropriate, using SIGMASTAT (Systat Software, San Jose, CA, USA). Where data failed equal variance or normality tests, they were analysed by one-way ANOVA on ranks followed by Dunn’s multiple-comparison test. Results are presented as the mean SEM. P ≤ 0.05 was considered statistically significant.
Results Effect of centrally-induced hyperleptinaemia on body weight and glucose tolerance On LFD, centrally-induced hyperleptinaemia (by i.c.v. infusion) led to a dramatic decline in body weight (27.7 0.5 g to 20.5 0.7 g. compared to vehicle-treated Lepob/ob mice, which gradually © 2014 British Society for Neuroendocrinology
increased their body weight from 27.7 0.6 g to 30.3 1.6 g (P ≤ 0.001) within the 6-day infusion regimen. After the switch from LFD to HFD, the reduction in body weight was completely arrested in mice infused with leptin. They exhibited a stable weight on HFD for the next 4 days on a diet (20.4 1.3 after 4 days), whereas leptin-infused mice on LFD continued to lose body weight down to 15.0 0.3 g (P ≤ 0.001). This profound reduction in body weight in the latter group necessitated termination of the experiment (Fig. 1A). Daily food intake of vehicle-injected mice on LFD averages at 4.1 0.4 g from days 9–13, and is significantly elevated compared to leptin-treated mice that consume 0.3 0.1 g per day on average (Fig. 1B; P ≤ 0.001). Average daily food intake was significantly increased when i.c.v. leptin-treated mice were transferred to HFD (2.1 0.1 g) relative to the group that remained on LFD (0.4 0.1 g from days 14–17; P ≤ 0.001) but was still significantly different from vehicle-injected mice on HFD that consumed 4.0 0.2 g (P ≤ 0.001). There was no significant difference in average daily food intake between vehicle-injected mice on LFD versus HFD (4.1. 0.4 g versus 4.0 0.2 g). Before termination of the experiment, after 4 days of HF feeding, glucose tolerance was tested. Vehicle-treated mice on HFD revealed profound glucose intolerance compared to leptin-treated mice on LFD. There was a significant difference in area under the curve (AUC) between these groups (Fig. 1C; P ≤ 0.001). In i.c.v. leptininfused mice fed a HFD, glucose tolerance was significantly reduced compared to the i.c.v. leptin-infused mice on LFD (P ≤ 0.001 in AUC), although it was still significantly improved compared to vehicle-infused mice on HFD (P = 0.003 in AUC).
Acute effect of peripherally-administered leptin on glucose tolerance in wild-type mice on LFD or HFD In normoglycaemic wild-type mice (Lep+/+ mice), we tested whether leptin alters glucose homeostasis. Ten days of HFD in wild-type mice was associated with a significant increase in body weight (mice on HFD after 10 days: 20.6 0.2 g versus mice on LFD: 18.6 0.3, P ≤ 0.001) and impaired glucose tolerance (P ≤ 0.001) compared to mice on LFD. As expected, i.p. leptin injection was not beneficial for further lowering normal glucose levels (Fig. 2; n = 12–13 per group). During states of leptin resistance, as described in wild-type mice on a HFD (21), acute i.p. leptin failed to improve impaired glucose homeostasis (Fig. 2; n = 13 per group). There was no significant difference in glucose concentration at any of the time points tested or in the AUC.
Acute action of peripherally-administered leptin on food intake and glucose tolerance of Lepob/ob mice on LFD or HFD Consistent with previous reports, acute i.p. leptin reduced food intake and improved glucose tolerance in Lepob/ob mice fed a LFD (14,22). Cumulative 24-h mean food intake was 5.1 0.18 g in the vehicle-treated group versus 4.1 0.21 g in the leptin-treated group (Fig. 3B; P = 0.003). Leptin injection 60 min before the ipGTT Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
(A) 40
Vehicle Leptin
61
***
Body weight (g)
35
**
*
30
25 Osmotic pump implantation
** **
*
20
15
0
(B )
High-fat diet
Low-fat diet
10 1
2
3
4
5
6
7
6 *** ** *** ***
5
** ***
8
9 10 11 12 13 14 15 16 17 18 Days
** **
** *
*** **
**
*** ***
4
Food intake (g)
Vehicle LFD Vehicle HFD Leptin LFD Leptin HFD
*** *
*
3
*
*** **
2
1
0
9
10
11
12
13 Days
15
16
20 30 AUC × 103
40
14
17
500
Vehicle HFD Leptin HFD Leptin LFD
400
Vehicle
**
Leptin
200
***
300
***
Glucose concentration (mg/dl)
(C)
Lep
100 0 0
30
60 Time (min)
90
120 0
10
50
Fig. 1. Effect of centrally-induced hyperleptinaemia on body weight and glucose tolerance. (A) Lepob/ob mice that underwent osmotic vehicle pump implantation (aCSF, n = 6) revealed a gradual increase in body weight on a low-fat diet (LFD) that was exacerbated by a high-fat diet (HFD) for 4 days. Relative to this group i.c.v. infused leptin (4.2 lg/day) led to a profound reduction in body weight within 10 days on LFD (P ≤ 0.001, n = 6). Intracerebroventricularly infused mice that were transferred to HFD for 4 days after 6 days on LFD stopped the leptin-induced body weight decline immediately after the change in diet relative to leptin-treated mice that continued to receive LFD (P ≤ 0.001, n = 7). (B) Cumulative 24-h food intake from days 9–17 of mice shown in (A). Change of diet on day 13 from LFD to HFD led to an increase in food intake in i.c.v. leptin-infused mice. (C) Glucose tolerance of mice shown in (A) at day 17. Vehicle infused Lepob/ob mice on HFD had a significantly impaired glucose tolerance relative to i.c.v. leptin infused Lepob/ob mice on LFD (P ≤ 0.001) and HFD (P = 0.003). Glucose tolerance of i.c.v. leptin infused Lepob/ob mice on HFD was significantly reduced compared to i.c.v. leptin infused Lepob/ob mice that remained on LFD (P ≤ 0.001). Data are shown as the mean SEM; *P ≤ 0.05; **P ≤ 0.01; *** P ≤ 0.001. AUC, area under the curve. Journal of Neuroendocrinology, 2014, 26, 58–67
© 2014 British Society for Neuroendocrinology
62
C. E. Koch et al.
250
i.p. Vehicle
200
i.p. Leptin 1.25 mg/kg
***
Glucose concentration (mg/dl)
300
150 100
i.p. Vehicle
50
i.p. Leptin 1.25 mg/kg
0 0
30
60 Time (min)
90
120 0
5
10
15
20
25
30
AUC × 103
Fig. 2. Acute effect of peripherally-administered leptin on glucose tolerance in wild-type (Lep+/+) mice on a low-fat diet (LFD) or a high-fat diet (HFD). High-fat feeding was associated with impaired glucose tolerance compared to mice on LFD (P ≤ 0.001). Relative to controls (n = 13–14 per group), peripherally-administered leptin (1.25 mg/kg in phosphate-buffered saline; n = 13 per group) 60 min before an i.p. glucose tolerance test had no effect on glucose tolerance of mice on LFD( ) or HFD( ). LFD, Low-fat diet; HFD, High-fat diet. The area under the curve (AUC) of leptin-treated mice was comparable to controls. Data are shown as the mean SEM. **P ≤ 0.001.
was sufficient to improve glucose homeostasis significantly (Fig. 3C; P = 0.02). When Lepob/ob mice were fed HFD, the body weight gain was accelerated compared to LFD mice; however, body weight between the two groups was matched with Lepob/ob mice on HFD being 2 weeks younger than LFD mice (LFD: 38.3 g 0.4 g; HFD: 38.8 g 0.6 g; Fig. 3A). Despite having similar body weights, acute i.p. leptin injection in Lepob/ob mice on HFD failed to reduce food intake (24-h mean food intake vehicle: 5.3 0.7 g; 24-h mean food intake leptin 5.8 0.36 g; Fig. 3B, right) or to improve glucose homeostasis (Fig. 3D) compared to Lepob/ob mice on a LFD (Fig. 3C).
Acute action of centrally-administered leptin on glucose tolerance in wild-type and Lepob/ob mice on HFD In this experiment, we investigated whether sensitivity to i.c.v. injected leptin is maintained in wild-type or Lepob/ob mice on HFD. Intracerebroventricular leptin injection 60 min before the ipGTT did not improve glucose homeostasis in wild-type mice on HFD (Fig. 4A). Surprisingly, in Lepob/ob mice on HFD, this treatment further deteriorated the impaired glucose tolerance in these mice. The AUC of Lepob/ob mice on HFD was significantly increased after i.c.v. leptin treatment compared to vehicle-injected mice (Fig. 4B; P = 0.007).
Pharmacological inhibition of central JNK and implication for glucose tolerance in HF fed Lepob/ob mice To investigate whether hypothalamic inflammation via JNK conveys the observed negative action of leptin on glucose homeostasis in Lepob/ob mice on HFD, we pharmacologically inhibited JNK in the central nervous system. Central inhibition of JNK by i.c.v. injection of 5 nmol of SP600125 (JNK-inhibitor) alone was not sufficient to improve glucose homeostasis in these mice (Fig. 4C). Intriguingly, however, i.c.v. injection of SP600125 (5 nmol) before i.c.v. leptin © 2014 British Society for Neuroendocrinology
(4 lg) potently reversed the deleterious effect of leptin on glucose homeostasis in Lepob/ob mice on HFD. The AUC of Lepob/ob mice on HFD treated i.c.v. with SP600125 and leptin was significantly decreased compared to i.c.v. vehicle or JNK-inhibitor treated mice (JNK-inhibitor/leptin versus JNK-inhibitor/vehicle P = 0.017; JNK-inhibitor/leptin versus vehicle/vehicle P = 0.014).
Leptin-dependent STAT3 phosphorylation in the ARC and VMH of Lepob/ob mice on LFD and HFD We next investigated whether high-fat feeding in Lepob/ob mice alters central leptin sensitivity at the molecular level. In highly-sensitive Lepob/ob mice on LFD, an i.c.v. leptin injection was associated with a profound increase in the number of pSTAT3(Tyr705) positive cells in the ARC (by 4.4-fold; Fig. 5A,B) and VMH (by 6.5-fold; Fig. 5C,D) compared to vehicle (vehicle versus leptin P ≤ 0.001). In both brain regions, vehicle-treated Lepob/ob mice on HFD revealed a level of pSTAT3 immunoreactivity that was comparable to the leptin-treated LFD-group. There was no significant difference between the two groups (in the ARC LFD leptin versus HFD vehicle was P = 0.057) but a significant difference to the LFD vehicle-treated mice (ARC: LFD vehicle versus HFD vehicle P = 0.008; VMH: LFD vehicle versus HFD vehicle P = 0.004) in both regions. In Lepob/ob mice on HFD, i.c.v. leptin did not increase the number of pSTAT3 (Tyr705) immunoreactive cells in the ARC and VMH relative to the vehicle group.
Discussion It is well established that the regulation of glucose homeostasis is one of the essential pleiotropic actions of leptin. The glucose-lowering properties of the hormone appear to largely originate from the hypothalamic sensitisation of the insulin signalling pathway (14). Nonetheless, whether or not leptin exhibits a similar glucose-lowering Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
Low-fat diet
Body weight (g)
40 35 30 25 0 0
2
4 6 Days
8
10
Glucose concentration (mg/dl)
(C)
6 5 ** 4 3 2 1 Vehicle Leptin
0
Vehicle Leptin
Low-fat diet 500 400 i.p. Vehicle
300 200
i.p. Leptin 1.25 mg/kg
*
100 0
0
30
60 90 120 Time (min)
150
(D)
180
0
20
40 AUC × 103
60
80
60
80
High-fat diet
600 Glucose concentration (mg/dl)
High-fat diet
(B ) Cumulative food intake of 24 h (g)
(A)
63
500 400
i.p. Leptin 1.25 mg/kg
300 i.p. Vehicle
200 100 0
0
30
60
90 120 Time (min)
150
180
0
20
40 AUC × 103
Fig. 3. Acute action of peripherally-administered leptin on food intake and glucose tolerance of Lepob/ob mice on a low-fat diet (LFD) or a high-fat diet (HFD). (A) During high-fat feeding, Lepob/ob mice revealed a rapid increase in body weight gain relative to Lepob/ob mice on LFD. Mice on HFD were 2 weeks younger than mice on LFD so that body weight was identical in both groups after 9 days of HF feeding. At this time point, peripherally-administered leptin (1.25 mg/ kg) 60 min before an i.p. glucose tolerance test had no effect on (B) 24-h food intake or (C) glucose tolerance in Lepob/ob mice on LFD compared to (D) Lepob/ob mice on HFD (n = 7–8 per group). AUC, area under the curve. In the latter group, leptin reduced (B) food intake over 24 h (n = 7–8 per group; P = 0.003) and (C) improved glucose tolerance 60 min after injection (n = 11–12 per group; P = 0.02). Data are shown as the mean SEM; *P ≤ 0.05; **P ≤ 0.001. Δ, LFD; ○, HFD.
capacity in DIO, the major cause of obesity in humans, remains unclear. In the present study, we investigated the development of leptin insensitivity after DIO in both the presence and absence of leptin in wild-type and Lepob/ob mice. The role of the JNK inflammatory pathway in the development of leptin resistance in DIO was also investigated. It is well known that leptin fails to reduce food intake and increase energy expenditure during DIO in humans and mice (23). Journal of Neuroendocrinology, 2014, 26, 58–67
In mice fed a HFD, leptin signalling via the JAK2/STAT3-pathway has been shown to be impaired in the ARC, which is the neuronal centre for the regulation of energy metabolism (21,24), although failure of leptin to regulate the long form of the leptin receptor (L-Rb) in the ARC can occur before failure of the leptin-induced JAK2/STAT3-pathway (25). In the present study, we first addressed the intriguing question of whether hyperleptinaemia per se or DIO induces leptin resistance © 2014 British Society for Neuroendocrinology
64
C. E. Koch et al.
Glucose concentration (mg/dl)
(A)
Lep+/+ High-fat diet
400 300
i.c.v. Leptin 4 µg
200 i.c.v. Vehicle 100 0
0
30
10
0
20 30 AUC × 103
40
50
Lepob/ob High-fat diet
(B) 600 500 400
i.c.v. Leptin 4 µg 300
**
Glucose concentration (mg/dl)
60 90 120 150 180 Time (min)
i.c.v. Vehicle 200 100 0
0
30
60 90 120 150 180 0 Time (min)
20
60 40 AUC × 103
80
100
Lepob/ob High-fat diet
500 400
5 nmol JNK-Inhib.
300
Vehicle
200
5 nmol JNK-Inhib. + 4 µg Leptin
* *
Glucose concentration (mg/dl)
(C) 600
100 0
0
30
60 90 120 150 180 0 Time (min)
20
40 60 AUC × 103
80
100
Fig. 4. Central implication of leptin and hypothalamic inflammation on glucose homeostasis. (A) Centrally-administered leptin (4 lg) 60 min before an i.p. glucose tolerance test (ipGTT) did not improve glucose tolerance of Lep+/+ mice on a high-fat diet (HFD). The area under the curve (AUC) of leptin-treated mice was comparable with vehicle-treated controls (n = 6 per group). (B) In Lepob/ob mice on HFD, centrally-injected leptin (4 lg; n = 8) 60 min before an ipGTT impaired glucose tolerance. The calculated AUC of leptin treated mice was significantly increased compared to vehicle-treated controls (P = 0.007). (C) Intracerebroventricular injection of SP600125 (5 nmol; n = 8), a pharmacological inhibitor of c-Jun n-terminal kinase (JNK), followed by an i.c.v. vehicle injection 30 min later, and an ipGTT 60 min after the second injection, did not alter glucose tolerance of Lepob/ob mice on HFD. Consecutive i.c.v. injections of the JNKinhibitor and i.c.v. leptin (4 lg; n = 4) 30 min apart followed by an ipGTT 60 min later improved glucose tolerance significantly compared to vehicle-treated controls (n = 6; P = 0.014) and SP600125-treated mice (P = 0.017). Data are shown as the mean SEM; *P ≤ 0.05; **P ≤ 0.01.
utilising an i.c.v leptin infusion regimen in Lepob/ob mice. In these mice that are exquisitely leptin sensitive, central hyperleptinaemia, induced by chronic i.c.v. leptin infusion (4.2 lg/day) for 10 days, led to a dramatic decline in body weight and food intake, as well as an improvement in glucose tolerance, ruling out the possibility of hyperleptinaemia being the cause of leptin resistance. Furthermore, high-fat feeding attenuated the leptin-induced decline in the body weight trajectory, cumulative food intake and impaired glucose tolerance, suggesting that the consumption of a HFD is the leading © 2014 British Society for Neuroendocrinology
cause of leptin resistance. Because the diet-induced loss of the glucose-lowering properties of leptin might be specific to the ob/ob genotype, we investigated the acute effect of i.p. leptin on glucose tolerance in wild-type mice fed a HFD. In line with these findings also in wild-type mice on HFD, acute i.p. leptin application failed to improve glucose tolerance. Consistent with the hypothesis that HFD induces leptin resistance, acute leptin was unable to reduce 24-h food intake or glucose tolerance in leptin-deficient, Lepob/ob mice on a HFD in contrast to weight-matched Lepob/ob control mice fed a Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
(B)
(A)
Lepob/ob
LFD-Leptin
ARC
100 µm HFD-Vehicle
HFD-Leptin
pSTAT3 (Tyr705) immunoreactive cells in ARC
250 LFD-Vehicle
*** ** P = 0.0569
200 *** 150
100
50
Vehicle 0
HFD-Leptin
Vehicle
Leptin HFD
(D) 225 pSTAT3 (Y705) counted cells in the VMH
HFD-Vehicle
LFD-Leptin
Leptin LFD
(C) LFD-Vehicle
65
200 175 **
150
***
125 100 75 50 25 0
Vehicle
Leptin LFD
Vehicle
Leptin
HFD
Fig. 5. Leptin-dependent signal transducer and activator of transcription (STAT)3 phosphorylation in the arcuate nucleus (ARC) and ventromedial hypothalamus (VMH) of Lepob/ob mice on a low-fat diet (LFD) and high-fat diet (HFD). Centrally-administered leptin (4 lg) led to a significant increase in phosphorylated (p)STAT3(Tyr705) immunoreactive cells in the ARC and VMH of Lepob/ob mice fed a LFD compared to vehicle (LFD: vehicle versus leptin P ≤ 0.001). In Lepob/ob mice fed a HFD, vehicle treatment was associated with an increased number of pSTAT3(Tyr705) immunoreactive cells relative to the LFD vehicle group (ARC: LFD vehicle versus HFD vehicle P = 0.008; VMH: LFD vehicle versus HFD vehicle P = 0.004) and central leptin treatment had no effect in these mice (n = 5–7 per group). (A) Shown are representative images of pSTAT3(Tyr705) immunoreactivity in the ARC of each treatment group (B) and the respective quantification of pSTAT3(Tyr705) positive cells. (C) Shown are representative images of pSTAT3(Tyr705) immunoreactivity in the VMH of each treatment group (D) and the respective quantification of pSTAT3(Tyr705) positive cells. Data are shown as the mean SEM; ** P ≤ 0.01; ***P ≤ 0.001.
LFD. The development of leptin insensitivity in this instance was independent of adiposity because Lepob/ob mice on the HFD and LFD were weight matched. These data collectively demonstrate that a HFD can lead to leptin insensitivity in the absence of endogenous leptin and during a profound i.c.v. leptin-induced decline in body weight, ruling out a role for hyperleptinaemia in this process. Our data are consistent with the findings of White et al. (26) who reported that the anorexigenic effect of i.p. leptin was abolished in Lepob/ob mice fed the HFD used in the present study for 14 days. By contrast, however, a study performed by Knight et al. (27) in which Lepob/ob mice were fed a high-fat diet for a very long Journal of Neuroendocrinology, 2014, 26, 58–67
period (20 weeks), and leptin was replaced at low doses during this time, suggested that hyperleptinaemia is required for the development of leptin resistance. However, this discrepancy might be explained by the very different experimental approaches, as well as by the time period, 9 days versus 20 weeks, on the diet. A complex time course in hypothalamic inflammation has been suggested, and this might obscure the analysis of hyperleptinaemia as a contributing factor to the development of leptin resistance (27). It has been proposed that HF feeding leads to impaired entry of leptin into the brain trough the blood–brain barrier (28). This may be a result of the increase in circulating triglycerides typically observed © 2014 British Society for Neuroendocrinology
66
C. E. Koch et al.
during HFD (29). In the present study, Lepob/ob mice on HFD were resistant to peripheral leptin but sensitive to an acute i.c.v. leptin injection in terms of the hormone’s glycaemic effect. Surprisingly, however, in contrast to chronic i.c.v. administration, acute i.c.v. leptin injection increased glucose intolerance. This difference might be explained by a much higher peak concentration that is achieved by acute injection of a 4 lg dose compared to infusion of 4.2 lg over 24 h. This high peak dose might induce pro-inflammatory properties of the hormone. Leptin is a cytokine and, similar to other cytokines, such as interleukin-6 and tumour necrosis factor (TNF)a, can induce a pro-inflammatory response via its activation of the JAK2/STAT3pathway (30,31). Synergistic actions of leptin and these other cytokines have been reported. In a study by Rizk et al. (32), i.v. injected TNFa was much more potent in activating pSTAT3 in hypothalamic lysates than i.v. injected leptin. A dose of 3 mg/kg of leptin compared to 12 lg/kg TNFa led to an activation of pSTAT3 in the hypothalamus that was of a similar magnitude (32). Systemic co-administration of both cytokines was associated with a marked synergistic activation of pSTAT3 in hypothalamic lysates. Intriguingly, we detected a marked increase in the number of pSTAT3 immunoreactive cells in the ARC and VMH of vehicle-treated Lepob/ob mice on HFD compared to LFD that was comparable to leptin-induced pSTAT3 in the latter group. To our knowledge, this diet-induced induction of pSTAT3 has not been previously reported in wild-type mice. Munzberg et al. (21) have shown that wild-type mice on HFD for 16 weeks did not reveal an increase of pSTAT3 immunoreactive cells in the ARC. To determine whether this effect is specific to the ob/ob genotype, a detailed time-course study of pSTAT3 immunoreactivity at different periods of HFD is required in future studies. It is plausible that, despite the lack of leptin in Lepob/ob mice fed a HFD, other circulating cytokines, whose concentration in the blood is elevated during HFD, may cause pSTAT3 activation in the ARC and VMH. The concentration and composition of these cytokines might be different in Lepob/ob mice on HFD relative to wild-type mice on this diet. Additional studies are required to address this question. Interestingly, during DIO, leptin failed to induce pSTAT3 in both hypothalamic regions of Lepob/ob mice, which suggests that pSTAT3 in the control group is already maximally activated. It is an unexpected finding that Lepob/ob mice on HFD are still obese and glucose intolerant despite elevated pSTAT3 in the ARC. Consistently, however, it has been reported that genetically enhanced STAT3 activation in pro-opiomelanocortin-expressing neurones in the hypothalamus provokes negative-feedback inhibition of leptin signalling in obesity (33). Interestingly, mice with a mutated STAT3 binding site in the leptin receptor (homozygous for L-RbSer1138 expression) have impaired energy homeostasis similar to Lepdb/db mice with a total loss of function of L-Rb; however, in contrast to Lepdb/db mice, they retain normal blood glucose when pair-fed with wild-type mice (34). Together with accumulating evidence that leptin sensitises central insulin action via the phosphatidylinositol-3-kinase/AKT-pathway (14), this suggests that the glucose-lowering properties of leptin are mediated via JAK2/STAT3 independent signalling pathways. In addition, our finding that i.c.v. leptin impairs glucose tolerance in Lepob/ ob mice on HFD despite the lack of pSTAT3 activation supports STAT3 independent regulation of glucose homeostasis by leptin. © 2014 British Society for Neuroendocrinology
A large body of evidence suggests that hypothalamic inflammation during HFD is associated with (and may be causative in) the development of obesity (35). The JNK inflammatory pathway appears to play a crucial role in this process and the development of obesity and insulin insensitivity because JNK is elevated in the ARC of mice fed a HFD and in Lepob/ob mice (36). In the present study, we tested the hypothesis that acute pharmacological inhibition of central JNK reverses the deleterious effect of leptin on glucose homeostasis in Lepob/ob mice on a HFD. Although i.c.v. injection of the JNK-inhibitor 30 min before i.c.v. leptin injection restored the glucose-lowering properties of leptin, it failed to alter glucose tolerance when given alone. This suggests that the degree of hypothalamic inflammation via JNK activity that was reversed by i.c.v. injection of 5 nmol SP600125 was not sufficient to solely improve glucose homeostasis. These findings imply that elevated hypothalamic inflammation via JNK mediates pro-diabetic properties of central leptin via the additive effects of the exogenous application of this pro-inflammatory cytokine at a metabolic state at which other cytokines are already elevated. The deterioration of glucose homeostasis by leptin that was observed in DIO Lepob/ob mice may be mediated via as yet unknown leptin signalling pathways that become activated at this particular metabolic state of extreme obesity. However, the involvement of other pathways (e.g. phosphoinositide 3-kinase/AKT signalling or WNT signalling) that have been implicated with central regulation of glucose homeostasis remains to be investigated further. It is also plausible that leptin indirectly worsens glucose homeostasis in these mice. A series of experiments is urgently required to delineate the precise molecular mechanisms with respect to how central leptin regulates glucose metabolism. Received 18 July 2013, revised 27 November 2013, accepted 23 December 2013
References 1 Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 1994; 372: 425–432. 2 Campfield LA, Smith FJ, Burn P. The OB protein (leptin) pathway – a link between adipose tissue mass and central neural networks. Horm Metab Res 1996; 28: 619–632. 3 Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998; 395: 763–770. 4 Farooqi IS, Jebb SA, Langmack G, Lawrence E, Cheetham CH, Prentice AM, Hughes IA, McCamish MA, O’Rahilly S. Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med 1999; 341: 879–884. 5 Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, Wareham NJ, Sewter CP, Digby JE, Mohammed SN, Hurst JA, Cheetham CH, Earley AR, Barnett AH, Prins JB, O’Rahilly S. Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997; 387: 903–908. 6 Kamohara S, Burcelin R, Halaas JL, Friedman JM, Charron MJ. Acute stimulation of glucose metabolism in mice by leptin treatment. Nature 1997; 389: 374–377.
Journal of Neuroendocrinology, 2014, 26, 58–67
HFD induces leptin resistance in Lepob/ob mice
7 Muzzin P, Eisensmith RC, Copeland KC, Woo SL. Correction of obesity and diabetes in genetically obese mice by leptin gene therapy. Proc Natl Acad Sci USA 1996; 93: 14804–14808. 8 Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F. Effects of the obese gene product on body weight regulation in ob/ob mice. Science 1995; 269: 540–543. 9 Schwartz MW, Baskin DG, Bukowski TR, Kuijper JL, Foster D, Lasser G, Prunkard DE, Porte D Jr, Woods SC, Seeley RJ, Weigle DS. Specificity of leptin action on elevated blood glucose levels and hypothalamic neuropeptide Y gene expression in ob/ob mice. Diabetes 1996; 45: 531– 535. 10 Sivitz WI, Walsh SA, Morgan DA, Thomas MJ, Haynes WG. Effects of leptin on insulin sensitivity in normal rats. Endocrinology 1997; 138: 3395–3401. 11 Yu X, Park BH, Wang MY, Wang ZV, Unger RH. Making insulin-deficient type 1 diabetic rodents thrive without insulin. Proc Natl Acad Sci USA 2008; 105: 14070–14075. 12 Hedbacker K, Birsoy K, Wysocki RW, Asilmaz E, Ahima RS, Farooqi IS, Friedman JM. Antidiabetic effects of IGFBP2, a leptin-regulated gene. Cell Metab 2010; 11: 11–22. 13 Van HM, Compton DS, France CF, Tedesco RP, Fawzi AB, Graziano MP, Sybertz EJ, Strader CD, Davis HR Jr. Diet-induced obese mice develop peripheral, but not central, resistance to leptin. J Clin Invest 1997; 99: 385–390. 14 Koch C, Augustine RA, Steger J, Ganjam GK, Benzler J, Pracht C, Lowe C, Schwartz MW, Shepherd PR, Anderson GM, Grattan DR, Tups A. Leptin rapidly improves glucose homeostasis in obese mice by increasing hypothalamic insulin sensitivity. J Neurosci 2010; 30: 16180–16187. 15 Prolo P, Wong ML, Licinio J. Leptin. Int J Biochem Cell Biol 1998; 30: 1285–1290. 16 Hubschle T, Thom E, Watson A, Roth J, Klaus S, Meyerhof W. Leptininduced nuclear translocation of STAT3 immunoreactivity in hypothalamic nuclei involved in body weight regulation. J Neurosci 2001; 21: 2413–2424. 17 Bjorbaek C, Uotani S, da SB, Flier JS. Divergent signaling capacities of the long and short isoforms of the leptin receptor. J Biol Chem 1997; 272: 32686–32695. 18 Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS. A central role for JNK in obesity and insulin resistance. Nature 2002; 420: 333–336. 19 Tuncman G, Hirosumi J, Solinas G, Chang L, Karin M, Hotamisligil GS. Functional in vivo interactions between JNK1 and JNK2 isoforms in obesity and insulin resistance. Proc Natl Acad Sci USA 2006; 103: 10741–10746. 20 Munzberg H, Huo L, Nillni EA, Hollenberg AN, Bjorbaek C. Role of signal transducer and activator of transcription 3 in regulation of hypothalamic proopiomelanocortin gene expression by leptin. Endocrinology 2003; 144: 2121–2131. 21 Munzberg H, Flier JS, Bjorbaek C. Region-specific leptin resistance within the hypothalamus of diet-induced obese mice. Endocrinology 2004; 145: 4880–4889.
Journal of Neuroendocrinology, 2014, 26, 58–67
67
22 Halaas JL, Gajiwala KS, Maffei M, Cohen SL, Chait BT, Rabinowitz D, Lallone RL, Burley SK, Friedman JM. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 1995; 269: 543–546. 23 Campfield LA, Smith FJ, Guisez Y, Devos R, Burn P. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995; 269: 546–549. 24 El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, Flier JS. Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obesity. J Clin Invest 2000; 105: 1827–1832. 25 Mitchell SE, Nogueiras R, Morris A, Tovar S, Grant C, Cruickshank M, Rayner DV, Dieguez C, Williams LM. Leptin receptor gene expression and number in the brain are regulated by leptin level and nutritional status. J Physiol 2009; 587: 3573–3585. 26 White CL, Whittington A, Barnes MJ, Wang Z, Bray GA, Morrison CD. HF diets increase hypothalamic PTP1B and induce leptin resistance through both leptin-dependent and -independent mechanisms. Am J Physiol Endocrinol Metab 2009; 296: E291–E299. 27 Knight ZA, Hannan KS, Greenberg ML, Friedman JM. Hyperleptinemia is required for the development of leptin resistance. PLoS ONE 2010; 5: e11376. 28 Banks WA. Is obesity a disease of the blood–brain barrier? Physiological, pathological, and evolutionary considerations. Curr Pharm Des 2003; 9: 801–809. 29 Banks WA, Coon AB, Robinson SM, Moinuddin A, Shultz JM, Nakaoke R, Morley JE. Triglycerides induce leptin resistance at the blood–brain barrier. Diabetes 2004; 53: 1253–1260. 30 Romanatto T, Cesquini M, Amaral ME, Roman EA, Moraes JC, Torsoni MA, Cruz-Neto AP, Velloso LA. TNF-alpha acts in the hypothalamus inhibiting food intake and increasing the respiratory quotient – effects on leptin and insulin signaling pathways. Peptides 2007; 28: 1050–1058. 31 Piekorz R, Schlierf B, Burger R, Hocke GM. Reconstitution of IL6-inducible differentiation of a myeloid leukemia cell line by activated Stat factors. Biochem Biophys Res Commun 1998; 250: 436–443. 32 Rizk NM, Stammsen D, Preibisch G, Eckel J. Leptin and tumor necrosis factor-alpha induce the tyrosine phosphorylation of signal transducer and activator of transcription proteins in the hypothalamus of normal rats in vivo. Endocrinology 2001; 142: 3027–3032. 33 Ernst MB, Wunderlich CM, Hess S, Paehler M, Mesaros A, Koralov SB, Kleinridders A, Husch A, Munzberg H, Hampel B, Alber J, Kloppenburg P, Bruning JC, Wunderlich FT. Enhanced Stat3 activation in POMC neurons provokes negative feedback inhibition of leptin and insulin signaling in obesity. J Neurosci 2009; 29: 11582–11593. 34 Myers MG Jr. Lecture Outstanding Scientific Achievement Award deconstructing leptin: from signals to circuits. Diabetes 2010; 59: 2708–2714. 35 Thaler JP, Choi SJ, Schwartz MW, Wisse BE. Hypothalamic inflammation and energy homeostasis: resolving the paradox. Front Neuroendocrinol 2010; 31: 79–84. 36 Benzler GK, Ganjam J, Legler K, Steger J, Stohrz S, Kruger M, Tups A. Acute inhibition of central c-Jun N-terminal kinase restores hypothalamic insulin signalling and alleviates glucose intolerance in diabetic mice. J Neuroendocrinol 2013; 25: 446–454.
© 2014 British Society for Neuroendocrinology
