ob mice, via indirect effects on the suprachiasmatic nucleus.

NEUROENDOCRINOLOGY

Leptin Normalizes Photic Synchronization in Male ob/ob Mice, via Indirect Effects on the Suprachiasmatic Nucleus Edith Grosbellet, Sylviane Gourmelen, Paul Pévet, François Criscuolo, and Etienne Challet Regulation of Circadian Clocks Team (E.G., S.G., P.P., E.C.), Institute of Cellular and Integrative Neurosciences, Centre National de la Recherche Scientifique UPR3212, and Evolutionary Ecophysiology Team (E.G., F.C.), Department of Ecology, Physiology, and Ethology, Hubert Curien Pluridisciplinary Institute, Centre National de la Recherche Scientifique Unité Mixte de Recherche 7178, University of Strasbourg, 67000 Strasbourg, France

Mounting evidence indicates a strong link between metabolic diseases and circadian dysfunctions. The metabolic hormone leptin, substantially increased in dietary obesity, displays chronobiotic properties. Here we investigated whether leptin is involved in the alteration of timing associated with obesity, via direct or indirect effects on the suprachiasmatic nucleus (SCN), the site of the master clock. Photic synchronization was studied in obese ob/ob mice (deficient in leptin), either injected or not with high doses of recombinant murine leptin (5 mg/kg). This was performed first at a behavioral level, by shifting the light-dark cycle and inducing phase shifts by 30-minute light pulses and then at molecular levels (c-FOS and P-ERK1/2). Moreover, to characterize the targets mediating the chronomodulatory effects of leptin, we studied the induction of phosphorylated signal transducer and activator of transcription 3 (P-STAT3) in the SCN and in different structures projecting to the SCN, including the medial hypothalamus. Ob/ob mice showed altered photic synchronization, including augmented light-induced phase delays. Acute leptin treatment normalized the photic responses of the SCN at both the behavioral and molecular levels (decrease of light-induced c-FOS). Leptin-induced P-STAT3 was modulated by light in the arcuate nucleus and both the ventromedial and dorsomedial hypothalamic nuclei, whereas its expression was independent of the presence of leptin in the SCN. These results suggest an indirect action of leptin on the SCN, possibly mediated by the medial hypothalamus. Taken together, these results highlight a central role of leptin in the relationship between metabolic disturbances and circadian disruptions. (Endocrinology 156: 1080 –1090, 2015)

he master pacemaker located in the suprachiasmatic nucleus (SCN) of the hypothalamus allows metabolic processes to cycle and to be in phase with daily changes of the external environment. The SCN clock, mainly reset by light, synchronizes a network of secondary clocks located in the brain and most peripheral organs (1). The circadian control of metabolism is reflected by rhythmic daily patterns of secretion of most metabolic hormones (2). Among these signals is leptin, synthesized by adipocytes and acting on the medial hypothalamus to inhibit appetite and stim-

T

ulate energy expenditure (3) and which is secreted during the active period in most rodents (4, 5). In turn, nutritional signals can affect circadian rhythmicity. Feeding time and concomitant changes in metabolites as well as hormones are potent entraining cues of peripheral oscillators (6). Furthermore, hypo- and hypercaloric diets alter the functioning of the master clock (7, 8). This reciprocal relationship is of great importance for human health because metabolic disorders like obesity or diabetes are associated with circadian disturbances (1).

ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2015 by the Endocrine Society Received July 8, 2014. Accepted December 8, 2014. First Published Online December 18, 2014

Abbreviations: ARC, arcuate hypothalamic nucleus; DMH, dorsomedial hypothalamic nucleus; IGL, intergeniculate (thalamic) leaflet; NS, nonsignificant; PB, phosphate buffer; PBS-T20, PBS with 0.3% Tween 20; P-ERK1/2, phosphorylated ERK 1/2; P-STAT3, phosphorylated signal transducer and activator of transcription 3; ROI, region of interest; SCN, suprachiasmatic nucleus; VMH, ventromedial hypothalamic nucleus; ZT, Zeitgeber time.

1080

endo.endojournals.org

Endocrinology, March 2015, 156(3):1080 –1090

doi: 10.1210/en.2014-1570

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 April 2015. at 23:21 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2014-1570

Diet-induced obesity and genetic obesity due to leptin deficiency (eg, ob/ob mice) or mutations in leptin receptor (eg, db/db mice) are classic animal models of metabolic disorders. In addition to a shared dampening of behavioral rhythms in high-fat fed and ob/ob mice (8, 9), these models show differential circadian disturbances, in particular regarding light resetting. High-fat fed mice show a lower responsiveness to the synchronizing effects of ambient light, as evidenced by reduced light-induced phase advances and slower reentrainment to advanced light-dark cycle (8). By contrast, light-induced phase delays are larger in ob/ob mice, whereas light-induced phase advances are comparable with those in wild-type controls (10). Highfat feeding leads to hyperleptinemia, whereas this hormone is not synthesized at all in ob/ob mice, suggesting a possible role of circulating leptin in the differential circadian changes between these models of obesity. Even if the mechanisms linking obesity to circadian disturbances are not fully understood, it is noteworthy that in addition to its well-known effects on energy balance, leptin shows chronobiotic properties. Expression of leptin receptors has been detected in the SCN of rodents and humans (11, 12). When isolated in vitro, the SCN clock can be phase advanced by leptin in a dose-dependent fashion (13). Moreover, leptin modulates firing rates of SCN neurons in hypothalamic slices (14). In vivo injections of leptin modulate the photic synchronization of the master clock in wild-type mice (15) and produce phase advances of the locomotor activity rhythm in Syrian hamsters (16). Furthermore, rhythmic release of leptin has been implicated in the body mass gain induced by desynchronized feeding (17). Because leptin may act as a time giver and shows high levels in dietary obesity, it could be involved in altered timing associated with obesity, via direct effects on the master clock in vivo. To test these assumptions, we investigated photic synchronization in ob/ob mice, first at the behavioral level by shifting the light-dark cycle and inducing phase shifts by light pulses, and then at the molecular level. To study whether treatment with leptin could modulate the responses to light, ob/ob mice were injected with recombinant leptin before light exposure at night. Finally, by investigating intracellular signaling activated by leptin, we determined whether the chronomodulatory effects of leptin are due to a direct action of leptin on the SCN or if they are mediated by brain structures known to integrate metabolic cues and to project to the SCN, such as the medial hypothalamus, including arcuate (ARC), ventromedial (VMH), and dorsomedial (DMH) nuclei (18 –20) and the intergeniculate leaflet (IGL) of the thalamus (21–23).


1081

Materials and Methods Ethics statement All experiments were performed in accordance with the US National Institutes of Health Guide for the Care and Use of Laboratory Animals (1996) and the French National Law (implementing the European Union Directive 2010/63/EU) and approved by the Regional Ethical Committee of Strasbourg for Animal Experimentation (AL/01/02/03/10).

Animals, housing, and diet Six-week-old male C57BL6JRj ob/ob mice and their heterozygous control ob/⫹ mice were purchased from Janvier Labs breeding center. They were housed in individual cages, kept at 22 ⫾ 1°C with food ad libitum (standard low fat diet, 105; SAFE) and maintained under a constant light-dark cycle (12 h, 12 dark cycle; lights on 7:00 AM, lights off 7:00 PM) unless stated otherwise.

Experimental design In a first series, we investigated the resynchronization after jet lags (see Supplemental Figure 1). Twenty-four mice (n ⫽ 12 per genotype) were transferred to constant darkness for 10 days to determine the endogenous period. Two weeks after the return to the initial light-dark cycle, mice were exposed to a 6-hour phase advance of the light-dark cycle (first jet lag test, advance shift). After two weeks under the new light-dark cycle (12 h, 12 dark cycle; lights on 1:00 AM, lights off 1:00 pm), mice were exposed to a 6-hour phase delay of the light-dark cycle (second jet lag test, delay shift). Mice were thus exposed again to the initial lightdark cycle (12 h, 12 dark cycle; lights on 7:00 AM, lights off 7:00 PM) for 2 weeks. In a second series, we studied the behavioral responses after a light pulse (see Supplemental Figure 1). Twenty-four mice (n ⫽ 12 per genotype) were transferred under dark conditions. On the first night, animals were exposed to a 30-minute fluorescent white light pulse (200 lux at the level of the animals) at projected Zeitgeber time (ZT) 13 (projected ZT12 corresponding to the time of light offset the day before). Thirty minutes before the light pulse, all mice received an ip injection of vehicle (0.01 M PBS ⫹ 0.1% BSA) or recombinant murine leptin (5 mg/kg; 1 mg/mL; Peprotech; catalog number 450 –31; n ⫽ 6 per genotype per group). The pharmacological dose of leptin and time of injection were chosen based on previous data (15). In a pilot study, we determined that plasma leptin levels ranged between 100 and 500 ng/mL, 3 hours after ip administration of this dose of recombinant leptin in control mice. Mice were then kept for 10 days to assess light-induced phase delays and putative modulation by leptin treatment. In a third series, to investigate the molecular responses after a light pulse (see Supplemental Figure 1), 24 mice (n ⫽ 12 per genotype) were transferred to dark conditions, exposed to a 30minute light pulse at projected ZT13 and injected either with vehicle or leptin as described above. Mice were euthanized under a dim red light (TL-D 18W Red SLV; Philips; ⬍ 3 lux at the level of animals) 1 hour after the beginning of the light pulse. Dark control animals (24 mice, n ⫽ 12 per genotype) also received the injection of vehicle or leptin at projected ZT13 and were killed at the same time as light-exposed mice. One ob/ob mouse in the group of dark control receiving vehicle injection died at the be-


1082

Grosbellet et al

Photic Synchronization in ob/ob Mice

ginning of the experiment, for unknown reasons (n ⫽ 5 in this group).

Behavioral recordings and analysis of activity data General locomotor activity was detected by infrared sensor and recorded every 5 minutes by a personal computer-based acquisition system (Circadian Activity Motor System; INSERM). Activity data were analyzed with ClockLab software (Actimetrics). In constant darkness, calculation of the endogenous period with ␹2 periodogram was performed with ClockLab. The rate of reentrainment to a new light-dark cycle was defined as the number of days necessary for the animal to present activity onsets/ offsets at fixed phases relative to lights offset/onset, respectively, and a duration of activity period similar to baseline values. In light pulse experiments, activity onsets were fitted by linear regressions, projecting the phase of the postpulse free-run back to the mean phase under light-entrained conditions (ie, eight cycles before and eight cycles after light exposure). The phase shift was calculated as the difference between the two fitted lines on the first day of darkness (ie, the day of treatment).

Immunohistochemistry Animals were deeply anesthetized with sodium pentobarbital (ip, 150 mg/kg) and intracardially perfused with saline (0.9% NaCl) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB). Each brain was rapidly removed and postfixed in the same fixative for 24 hours and then cryoprotected in 30% sucrose for 48 –72 hours. Brains were quickly frozen in isopentane at ⫺40°C and stored at ⫺80°C. The coronal cryosections of 30 ␮m through the SCN and the medial hypothalamus (ARC, VMH, and DMH) were prepared by a cryostat at ⫺20°C. For c-FOS and phosphorylated ERK 1/2 (P-ERK1/2) immunostaining, free-floating sections were washed in 0.01 M PBS and incubated in a solution of 3% of H2O2 (Sigma) in PBS for 30 minutes. Brain sections were then rinsed in PBS for 3 ⫻ 10 minutes and incubated for 2 hours in a blocking solution in 10% normal goat serum in PBS with 0.3% Tween 20 (PBS-T20), followed by an incubation with a rabbit polyclonal anti-c-FOS antibody (1:10 000, SC-52; Santa Cruz Biotechnology) or a rabbit monoclonal anti-P-ERK1/2 (1:40 000; number 4370; Cell Signaling Technologies) for 18 hours at 4°C. Sections were rinsed 3 ⫻ 10 minutes in PBS-T20 and incubated for 2 hours at 4°C with a biotinylated goat antirabbit IgG (Vectastain Standard Elite ABC Kit PK6101; Vector Laboratories, Inc) diluted 1:500 in PBS-T20. Sections were rinsed in PBST20 and incubated for 1 hour at room temperature with an avidin-biotin-peroxidase complex (Vectastain Standard Elite ABC kit; Vector Laboratories). Sections were rinsed 3 ⫻ 10 minutes in PBS and incubated with 0.05% 3,3⬘-diaminobenzidine (Sigma) with 0.015% H2O2 in tap water. Then sections were rinsed with PBS, wet mounted onto gelatin-coated slides, dehydrated through a series of alcohols, soaked in toluene, and coverslipped with Eukitt (Chem Lab). For phosphorylated signal transducer and activator of transcription 3 (P-STAT3) immunostaining, sections were pretreated with 1% NaOH and 1% H2O2 in H2O for 20 minutes, 0.3% glycine in PB for 10 minutes and 0.03% sodium dodecyl sulfate in PB for 10 minutes. Subsequently, sections were blocked for 1 hour with 3% normal goat serum in PBS ⫹ 0.25% Triton X-100.


They were then incubated in rabbit monoclonal P-STAT3 antibody (1:2000, number 9145; Cell Signaling Technologies) for 18 hours at 4°C. On the next day, the procedure was performed as described above.

Quantification of immunostaining Images were taken on a Leica DMRB microscope (Leica Microsystems) connected to an Olympus DP50 digital camera (Olympus France). For staining intensity, all lighting parameters on the microscope and the camera software (Viewfinder Lite; Olympus) were standardized to ensure stable lighting throughout the image capture procedure. Using the atlas of Paxinos and Franklin (24), one brain section per structure per animal was carefully selected to compare the same level of structures between animals, corresponding to bregma levels ⫺0.46 mm for SCN (Figure 35 of the atlas), ⫺1.58 mm for ARC and VMH (Figure 44), ⫺1.82 mm for DMH (Figure 46) and ⫺2.30 mm for IGL (Figure 50). Boundaries of structures were chosen by comparison with sections stained with Cresyl Violet and by using anatomical landmarks (eg, third ventricle, fornix, and mammillothalamic tract for DMH). Quantifications were performed with ImageJ software (W. S. Rasband, US National Institutes of Health, Bethesda, Maryland) by delimiting so-called regions of interest (ROIs) covering each side of the nuclei. For each animal, three pictures were taken: one of the target structure, one part of the section without specific staining (internal capsule), and one of the slide without section. The latter was subtracted from the two other images to compensate for variations in the illumination of the image field. Then the mean intensity of gray pixel value was calculated and averaged for both sides of target structures and for unstained tissue, which corresponds to tissue background, varying between animals. By subtracting the mean intensity of background tissue to the mean intensity of selected structure, we obtained the final value of intensity, proportional to the quantity of targeted protein.

Statistical analysis Data are presented as mean ⫾ SEM. Normality was assessed with a Shapiro-Wilk test. Unpaired Student’s t tests or MannWhitney U tests were used to compare two groups. A two-way ANOVA were performed to assess the effects of genotype and treatment (leptin or vehicle) conditions and the interaction between these factors. A three-way ANOVA was used to test the aforementioned factors plus the effect of light (Statistica 10.0; StatSoft). Values of P ⬎ .05 were considered nonsignificant (NS).

Results Faster resynchronization in ob/ob mice after a 6-hour phase delay Ob/ob mice slowed their resynchronization after phase advance by 23% (Figure 1, A and B, left panel: 7.3 ⫾ 0.5 for ob/ob vs 5.6 ⫾ 0.3 d for ob/⫹; t test: t22 ⫽ 2.8, Pone-tailed ⫽ .005) but were resynchronized faster after a phase delay compared with the ob/⫹ control mice (Figure 1B: 4.3 ⫾ 0.4 vs 5.9 ⫾ 0.5 d, respectively; U(Mann-Whitney) ⫽ 27, P ⫽ .009). These changes cannot be explained by a mod-


doi: 10.1210/en.2014-1570


1083

6.62, P ⫽ .01 followed by Tukey, P ⫽ .008] and decreased by 30% the levels of c-FOS in ob/ob mice receiving light (Figure 3, A and C: treatment effect F1,38 ⫽ 6.9, P ⫽ .01 followed by Tukey, P ⫽ .04). Thus, the reduction of light-induced phase delay due to leptin in ob/ob mice correlates with a decrease of c-FOS level in the master clock. These modulatory effects of leptin occurred predominantly in the ventral SCN (Figure 3C and Supplemental Figure 2). Differential effects of leptin on the medial hypothalamus in ob/ⴙ and ob/ob mice Contrary to P-ERK1/2 (Figure 4), the level of c-FOS in the medial hypothalamus was not affected by light or leptin (data not shown). As in the SCN, leptin almost doubled the levFigure 1. A, Representative double-plotted actograms of general activity of ob/⫹ (left panel) els of P-ERK1/2 in the ARC and and ob/ob (right panel) mice successively challenged by 6-hour advance and 6-hour delay of the VMH of ob/⫹ mice held in darkness light-dark cycle. Gray areas represent dark periods. In the first jet-lag test (phase advance), the arrow indicates the activity onset on the first day of resynchronization to the new light offset. By (Figure 4, Tukey P ⫽ .002 and P ⫽ contrast, in the second jet-lag test (phase delay), the arrow indicates the activity offset on the .01 in ARC and VMH, respectively; first day of resynchronization to the new light onset. B, Rates of reentrainment in ob/⫹ and NS increase in DMH). In the VMH, ob/ob mice during the first jet-lag test with 6-hour phase advance (left panel) and the second the induction of P-ERK1/2 occurred jet-lag test with 6-hour phase delay (right panel). Groups with different letters are significantly different (P ⬍ .05). mainly in the lateral part (Figure 4C). A light effect (F1,37 ⫽ 13.3, P ⬍ .001) was observed only in ARC of ob/⫹ ification of the endogenous period in ob/ob mice (23.97 ⫾ 0.08 h in ob/⫹ mice vs 23.97 ⫾ 0.07 h in ob/ob mice, NS). mice. In dark conditions, a higher basal activation of P-ERK1/2 was observed in ARC of ob/ob compared with ob/⫹ mice (P ⫽ .003 in ARC, a trend in VMH and DMH, Higher light-induced phase delay in ob/ob mice, NS). Of note, in dark conditions leptin tended to normalnormalized by acute leptin injection In mice injected with a vehicle solution, the phase delay ize P-ERK1/2 levels in ob/ob mice, especially in VMH and induced by light was higher in ob/ob mice [Figure 2, A and DMH, although the decrease did not reach the threshold B: (genotype ⫻ treatment) interaction, F3,22 ⫽ 6.1, P ⫽ .02 of significance (NS). followed by Tukey’s test, P ⫽ .01]. Leptin injection did not change the photic response in ob/⫹ mice. By contrast, Indirect action of leptin on the SCN In dark conditions, acute leptin treatment did not injection of leptin decreased the phase delay in ob/ob mice change the level of P-STAT3 in the SCN of both genotypes by 25% (Tukey, P ⫽ .01), thus normalizing the delay to (Figure 5A). The combined effect of light and leptin led to the same duration as that observed in the ob/⫹ mice (NS). an increase of P-STAT3, [(light ⫻ treatment) interaction, F1,36 ⫽ 5.9, P ⫽ .02]. The level of P-STAT3 in the SCN Leptin injection decreased c-FOS in SCN of lightbeing similar between ob/⫹ and ob/ob mice injected with exposed ob/ob mice Light pulse treatment increased the levels of c-FOS and vehicle, these results strongly suggest the expression of P-ERK1/2 independently of genotypes and leptin treat- P-STAT3 in the SCN is independent of leptin. By contrast, the quantity of P-STAT3 was very low in ment (Figure 3: light effect, F1,38 ⫽ 51.2 and F1,36 ⫽ 47.0, respectively, P ⬍ .001). Leptin doubled the levels of the medial hypothalamus of ob/ob mice (Figure 5, B–F, P-ERK1/2 in ob/⫹ mice held in dark conditions [Figure 3, P ⬍ .05 in the ARC and VMH; a trend observed in DMH, B and D: (genotype/light/treatment) interaction, F1,36 ⫽ NS) and reached the same level as in ob/⫹ mice after leptin


1084

Grosbellet et al



duced expression of c-FOS by 50% in ob/ob mice (Figure 6B, P ⫽ .02).

Discussion Ob/ob mice showed faster resynchronization to delayed light-dark cycles and higher phase delays induced by light pulses. High doses of exogenous leptin normalized the photic responses of the master clock in ob/ob mice at both the behavioral (decrease of phase delay) and molecular levels (decrease of c-FOS). P-STAT3 immunostaining revealed that peripheral leptin acts on the ARC, VMH, and DMH, but not on the SCN. All together, these results demonstrate an indirect chronomodulatory effect of leptin on the master clock, possibly mediated by the medial hypothalamus. Alterations of photic responses to light in ob/ob mice Compared with ob/⫹ animals, activity rhythms in ob/ob mice resynchronize more slowly after a 6-hour advance of the light-dark cycle but faster after a 6-hour delay (Figure 1). This differential response may be Figure 2. A, Representative double-plotted actograms of general activity illustrating phase delay based on distinct molecular signalinduced by a 30-minute light pulse at projected ZT13 in ob/⫹ and ob/ob mice injected either ing pathways involved in phase adwith vehicle or leptin solution. Arrows indicate the first day of dark conditions, during which the vances and delays (25). High-fat-fed light pulse was applied. B, Mean phase delays in response to light in ob/⫹ and ob/ob mice injected either with vehicle or leptin solution. Groups with different letters are significantly mice, exhibiting high levels of circudifferent (P ⬍ .05). lating leptin, and ob/ob mice show the same slower rate of resynchronization after a 6-hour phase advance injection in dark conditions (NS). Moreover, in VMH and of the light-dark cycle (this study and reference 8). This DMH, the leptin-induced increase of P-STAT3 was posuggests that circulating leptin does not affect the intratentiated by light in a similar way between ob/⫹ and ob/ob mice [(light ⫻ treatment) interaction: F1,37 ⫽ 26, P ⬍ .001 cellular cascades of light-induced phase advances. By conand F1,35 ⫽ 6.9, P ⫽ .01 in the VMH and DMH, respec- trast, ob/ob mice resynchronize faster than lean mice after tively], whereas the P-STAT3 levels were not affected by a 6-hour phase delay of the light-dark cycle, whereas the leptin in the ARC of ob/⫹ mice, regardless of the light resetting of high-fat-fed mice challenged by the same delay is unaffected (8). This strongly suggests that leptin plays a conditions (NS). In the IGL of both genotypes, the levels of P-ERK1/2 critical role in photic phase delays, leading us to focus on were not affected by light or treatment conditions (Figure the involvement of leptin in these shifts. To dissociate al6A), despite an upward trend in ob/⫹ mice after leptin tered photic synchronization from alterations of masking injection in dark conditions (NS). Light increased levels of responses to light (ie, direct inhibitory effects of bright c-FOS in both genotypes (Figure 6B, light effect: F1,38 ⫽ light on motor activity in mice), we investigated further the 5.6, P ⫽ .02). Remarkably, leptin reduced the light-in- phase delays induced by a light pulse at early night.


doi: 10.1210/en.2014-1570


1085

photic resetting at night, phase advances should be smaller or even absent in ob/ob mice, whereas phase delays should be larger due to the absence of leptinergic phase advance. This hypothesis fits with our behavioral results, especially because the injection of leptin reduces photic phase delays in ob/ob mice. However, the modulatory effect of leptin on behavioral phase shifts correlates with a decrease in light-induced c-FOS expression in the SCN. This response suggests a modulation of molecular photic changes within SCN cells (eg, in the ventrolateral, retinorecipient region of the SCN). Leptin acts indirectly on the SCN P-STAT3 levels in the SCN do not differ between ob/⫹ and ob/ob mice and are not modulated by exogenous leptin (Figure 5A). This is consistent with a previous study of Caron et al (27), which did not detect the mRNA of long-form leptin receptor in the SCN. Basal levels of P-STAT3 measured in the SCN could be due to other signaling pathways involving P-STAT3, such as cytokine and growth Figure 3. A and B, c-FOS and P-ERK1/2 mean responses in the SCNs of ob/⫹ and ob/ob mice factor signaling (28 –30). This obserafter injection of vehicle or leptin followed (light pulse) or not (dark control) by a 30-minute light vation implies an indirect action of pulse at projected ZT13. Groups with different letters are significantly different (P ⬍ .05). C and D, Representative photomicrographs of c-FOS (C) and P-ERK1/2 (D) staining in the SCN of ob/⫹ leptin on the SCN. The leptin-inand ob/ob mice. Scale bar, 100 ␮m. duced increase of P-ERK1/2 in SCN of ob/⫹ mice (Figure 3C) would thus be due to an indirect action of leptin, Light-induced phase delays are higher in ob/ob mice (Figure 2), thus providing a likely explanation for the given that ERK is a component of the MAPK family, posmaller rate of reentrainment to a 6-hour delay in light- tently activated by excitatory synaptic activity (31). We thus looked for structures targeted by leptin and dark cycle (ie, higher phase delays every day allow them to resynchronize faster to the new light-dark cycle). This con- projecting to the SCN that could convey such information. firms a previous study (10), showing stronger phase delays Because the IGL can convey metabolic cues to the SCN via in ob/ob mice exposed to a light pulse at ZT15. c-FOS neuropeptide Y projections (21–23), they formed a polevels were not increased in SCN of ob/ob mice (Figure 3), tential candidate. Leptin injection did not lead to a sizable ruling out higher sensitivity to light of SCN cells or im- induction of P-STAT3 in the IGL (data not shown), paired photo-detection or transduction in the retina. Be- whereas they tended to display increased P-ERK1/2 in cause the output of the circadian system is weaker in ob/ob ob/⫹ mice kept in dark conditions (Figure 6). These results mice (9, 10), light may have a stronger resetting effect on strongly suggest an indirect effect of leptin on the IGL, in the poorly coupled clock (26), explaining higher light- accordance with the absence of leptin receptor mRNA in induced phase delays. Moreover, leptin induces phase ad- that region (32). Moreover, because leptin reduces c-FOS vances of the SCN in vitro during most of the circadian in the SCN of ob/ob mice receiving light, an involvement cycle (13). Considering additive shifting effects during of IGL would suggest a higher inhibition of SCN by acti-


1086

Grosbellet et al



Because the binding of leptin on its receptor leads to the phosphorylation of ERK1/2 (35), it is not surprising that leptin increased the levels of P-ERK1/2 in ob/⫹ mice held in dark conditions (Figure 4). By contrast, basal P-ERK1/2 levels at night are already high in the ARC, VMH, and DMH of ob/ob mice and tend to decrease after leptin injection. A hyperphosphorylation of ERK in the ARC of ob/ob mice has already been reported (36) and could be due to high levels of glucose, as shown in the sensory neurons of diabetic rats (37). Acute leptin can thus normalize the phosphorylation state of ERK1/2, either directly or via possible metabolic changes. Basal levels of P-STAT3 in dark conditions are higher in medial hypothalamus of ob/⫹ mice than in Figure 4. A, B, and D, P-ERK1/2 mean responses in the ARC (panel A), VMH (panel B), and ob/ob mice and are not affected by DMH (panel D) nuclei of ob/⫹ and ob/ob mice after injection of vehicle or leptin followed (light leptin injection. Plasma leptin in nocpulse) or not (dark control) by a 30-minute light pulse at projected ZT13. Groups with different letters are significantly different (P ⬍ .05). C, Representative photomicrographs of P-ERK1/2 turnal rodents displays a clear daily staining in ARC and VMH nuclei of ob/⫹ and ob/ob mice injected with vehicle or leptin in dark rhythm with a peak before midnight conditions. The ellipse and triangle illustrate the location of VMH and ARC, respectively. The (4, 5, 38). Because our mice were outlines are not meant to denote the exact ROIs used for quantification. For the ROI analysis performed, please see Materials and Methods. Scale bar, 100 ␮m. f, fornix; 3V, third ventricle. sampled in the early night, this corresponds to the acrophase of their leptin rhythm, reaching values apvated IGL. Such a prediction does not fit with the decreased c-FOS levels observed in IGL of ob/ob mice in proximately 8 ng/mL in ob/⫹ mice (data not shown). The these conditions (Figure 6). Hence, our results suggest that difference of P-STAT3 levels in the medial hypothalamus the IGL are not directly involved in transmitting leptiner- between ob/⫹ and ob/ob mice might then be explained by gic signals to the SCN. We also excluded the lateral hy- an activating effect of endogenous leptin in ob/⫹ mice. pothalamic area as a relay of leptin effect to the master This could thus prevent a further induction of P-STAT3 clock (33) because no modulation of P-STAT3, c-FOS, by exogenous leptin in ob/⫹ mice. In sharp contrast, and P-ERK1/2 by light and/or leptin could be observed in treatment with leptin in ob/ob mice strongly increases P-STAT3 in the ARC, VMH, and DMH, confirming the this structure (data not shown). functionality of the hypothalamic leptinergic pathways in Modulation of P-ERK1/2 and P-STAT3 in medial these mice as previously described (eg, reference 39). In hypothalamus by light and/or leptin ob/⫹ mice, light potentiates P-STAT3 induction in the Contrary to P-ERK1/2 and P-STAT3, c-FOS levels were VMH and DMH but not in the ARC. This difference (Fignot affected by light and/or leptin in medial hypothalamus ure 5) could be due to a time-course effect of leptin. The (data not shown). Endogenous expression of c-FOS ex- ARC being targeted by exogenous leptin more rapidly hibits a daily rhythm in this region, reaching a maximum than other hypothalamic sites, leptinergic signals would during the night (34). Expression of c-FOS is induced in be terminated in the ARC of ob/⫹ mice at the time the the medial hypothalamus by leptin injected during day- animals were killed, ie, 90 minutes after the leptin injection time, as shown in numerous studies (eg, reference 18). The (40, 41). By contrast, the light potentiation of P-STAT3 is high basal values of c-FOS in the medial hypothalamus at observed in the ARC of ob/ob mice. This difference benight, in response to endogenous factors associated with tween the two genotypes could come from neuroanatomifeeding, could prevent further activation by exogenous cal alterations observed in the ob/ob mice, in particular in leptin (ceiling effect). the ARC (42). The use of an inducible genetic mouse


doi: 10.1210/en.2014-1570


1087

ARC, VMH, and DMH are sensitive to light information (20, 43), the higher induction of P-STAT3 could reflect crosstalk between leptin and a pathway indirectly conveying light information, the latter possibly down-regulating inhibitors of STAT3 phosphorylation such as protein tyrosine phosphatase 1B (44). Medial hypothalamus as a likely candidate to mediate chronomodulatory actions of leptin The strong alterations of photic resetting of the master clock in ob/ob mice are normalized by an acute injection of leptin, demonstrating that leptin can modulate the sensitivity to light of the master clock. Among the different structures known to convey metabolic information to the SCN (eg, IGL, lateral hypothalamic area, medial hypothalamus), the leptinergic pathway (assessed by P-STAT3) and cellular activation (assessed by P-ERK1/2) are modulated by light and/or leptin only in the medial hypothalamus. The medial hypothalamus has already been demonstrated to integrate metabolic cues that in turn affect the master clock and its resetting properties by light. This includes ghrelin signals, glucoprivic factors, and hypocaloric feeding (19, 45, 46). All together, these results suggest that medial hypothalamus is a good candidate to relay leptinergic Figure 5. A–D, P-STAT3 mean responses in the SCN (panel A), ARC (panel B), VMH (panel C), information to the SCN. In view of and DMH (panel D) nuclei of ob/⫹ and ob/ob mice after injection of vehicle or leptin followed reduced c-FOS induction in SCN of (light pulse) or not (dark control) by a 30-minute light pulse at projected ZT13. Groups with different letters are significantly different (P ⬍ .05). In panel A, the star represents a significant ob/ob mice after leptin injection difference (P ⬍ .05) for the overall effect of leptin between dark and light conditions. E and F, with light exposure, and the resultRepresentative photomicrographs of P-STAT3 staining in the ARC, VMH (E), and DMH (F) nuclei ing decrease in photic phase delays, of ob/⫹ and ob/ob mice injected with vehicle or leptin and exposed (light pulse) or not (dark control) to a light pulse at projected ZT13. E, The ellipse and triangle illustrate the location of it is plausible to explain this negaVMH and ARC, respectively. F, The round-cornered rectangle represents the location of DMH. tive modulation either by inhibiThe outlines are not meant to denote exact ROIs used for quantification. For the ROI analysis tory afferents to excitatory (eg, gluperformed, please see Materials and Methods. Scale bar, 100 ␮m. f, fornix; 3V, third ventricle. tamatergic) SCN neurons or by the stimulation of ␥-aminobutyric acid model in future studies could circumvent the issue of neucontaining neurons in the SCN. Our results identify three romorphological differences. How light potentiates leptin-induced P-STAT3 in the candidate structures for transmitting leptinergic cues to medial hypothalamus remains unclear. Because the the SCN, namely the ARC, VMH, and DMH. The SCN


1088

Grosbellet et al



toxin) or local intracerebral applications of leptin in targeted hypothalamic areas would be helpful to discriminate the relative importance of these nuclei in modulation of photic resetting of the master clock in mice. Moreover, because these nuclei are tightly interconnected, we can only hypothesize that leptinergic stimulation of the medial hypothalamus conveys a global inhibitory input to the SCN (independent of the stimulation of IGL), which ultimately decreases photic phase delays of the master clock, in association with decreased light-induced c-FOS levels (Figure 7). In conclusion, our study shows that pharmacological doses of leptin indirectly modulate circadian synchronization to the light of the master clock, possibly via integraFigure 6. A and B, P-ERK1/2 and c-FOS mean responses in the IGL of ob/⫹ and ob/ob mice after tion by the medial hypothalamus. injection of vehicle or leptin followed (light pulse) or not (dark control) by a 30-minute light pulse at Therefore, pathophysiological sitprojected ZT13. Groups with different letters are significantly different (P ⬍ .05). C, Representative photomicrographs of c-FOS staining in IGL of ob/⫹ and ob/ob mice injected with vehicle or leptin in uations in humans that reduce the dark conditions. The rectangle represents the location of IGL. The outlines are not meant to denote daily amplitude of circulating lepexact ROIs used for quantification. For the ROI analysis performed, please see Materials and Methods. tin, such as sleep deprivation or Scale bar, 100 ␮m. shift work (49, 50), as well as human receives both inhibitory and excitatory inputs from these metabolic disorders associated with central leptin resisnuclei, as shown in rats (19, 20, 47, 48). Nonetheless, tance, are likely prone to generate circadian disturbances further studies using neuronal inhibition (eg, tetrodo- that are by themselves deleterious for metabolic health.

Figure 7. Schema for the transmission of chronomodulatory effects of leptin by the medial hypothalamus in ob/ob mice. The medial hypothalamus, whose nuclei (ARC, VMH, and DMH) are interconnected, integrates photic and leptinergic signals. Global inhibition of the SCN from these nuclei could explain the decreased light-induced phase delay of the master clock, associated with decreased c-FOS. RHT, retinohypothalamic tract.


doi: 10.1210/en.2014-1570


Acknowledgments We thank S. Dumont and Dr D. Sage-Ciocca for their expert assistance with immunohistochemistry and actimetry, respectively. We also thank Dr J. Mendoza and Dr D. Hicks for their helpful comments on the manuscript. Address all correspondence and requests for reprints to: Etienne Challet, PhD, Team Leader, Regulation of Circadian Clocks, University of Strasbourg, 5 Rue Blaise Pascal, Strasbourg 67000, France. E-mail: [email protected]. This work was supported by the Centre National de Recherche Scientifique (to P.P., F.C., and E.C.); University of Strasbourg (to P.P., F.C., and E.C.); Agence Nationale pour la Recherche, “Jeunes Chercheurs/Jeunes Chercheuses,” Grant ANR-07-JCJC-0111 (to E.C.); ProjEx H2E, “Contributions of Exotic Animal Models in the Discovery of New Therapeutic Approaches in Human Pathophysiology,” University of Strasbourg (to F.C. and E.C.); and a doctoral fellowship from the French Ministry of Higher Education and Research (to E.G.). Disclosure Summary: The authors have nothing to disclose.

References 1. Froy O. Metabolism and circadian rhythms—implications for obesity. Endocr Rev. 2010;31(1):1–24. 2. Tsang AH, Barclay JL, Oster H. Interactions between endocrine and circadian systems. J Mol Endocrinol. 2014;52(1):R1–R16. 3. Ahima RS, Lazar MA. Adipokines and the peripheral and neural control of energy balance. Mol Endocrinol. 2008;22(5):1023–1031. 4. Kalsbeek A, Fliers E, Romijn JA, et al. The suprachiasmatic nucleus generates the diurnal changes in plasma leptin levels. Endocrinology. 2001;142(6):2677–2685. 5. Cuesta M, Clesse D, Pévet P, Challet E. From daily behavior to hormonal and neurotransmitters rhythms: comparison between diurnal and nocturnal rat species. Horm Behav. 2009;55(2):338 –347. 6. Schibler U, Ripperger J, Brown SA. Peripheral circadian oscillators in mammals: time and food. J Biol Rhythms. 2003;18(3):250 –260. 7. Challet E, Pévet P, Vivien-Roels B, Malan A. Phase-advanced daily rhythms of melatonin, body temperature, and locomotor activity in food-restricted rats fed during daytime. J Biol Rhythms. 1997;12(1): 65–79. 8. Mendoza J, Pévet P, Challet E. High-fat feeding alters the clock synchronization to light. J Physiol. 2008;586(Pt 24):5901–5910. 9. Ando H, Kumazaki M, Motosugi Y, et al. Impairment of peripheral circadian clocks precedes metabolic abnormalities in ob/ob mice. Endocrinology. 2011;152(4):1347–1354. 10. Sans-Fuentes MA, Díez-Noguera A, Cambras T. Light responses of the circadian system in leptin deficient mice. Physiol Behav. 2010; 99(4):487– 494. 11. Couce ME, Burguera B, Parisi JE, Jensen MD, Lloyd R V. Localization of leptin receptor in the human brain. Neuroendocrinology. 1997;66(3):145–150. 12. Guan XM, Hess JF, Yu H, Hey PJ, Van der Ploeg LH. Differential expression of mRNA for leptin receptor isoforms in the rat brain. Mol Cell Endocrinol. 1997;133(1):1–7. 13. Prosser R, Bergeron H. Leptin phase-advances the rat suprachiasmatic circadian clock in vitro. Neurosci Lett. 2003;336(3):139 – 142. 14. Inyushkin AN, Bhumbra GS, Dyball REJ. Leptin modulates spike

15.

16.

17.

18.

19.

20.

21.

22.

23.

24. 25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

1089

coding in the rat suprachiasmatic nucleus. J Neuroendocrinol. 2009; 21(8):705–714. Mendoza J, Lopez-Lopez C, Revel FG, et al. Dimorphic effects of leptin on the circadian and hypocretinergic systems of mice. J Neuroendocrinol. 2011;23(1):28 –38. Gündüz B, Karakas¸ A. The effects of leptin hormone on locomotor activity in Syrian hamsters (Mesocricetus auratus). Turk J Biol. 2011;35(6):727–734. Arble DM, Vitaterna MH, Turek FW. Rhythmic leptin is required for weight gain from circadian desynchronized feeding in the mouse. PLoS One. 2011;6(9):e25079. Elmquist JK, Ahima RS, Maratos-Flier E, Flier JS, Saper CB. Leptin activates neurons in ventrobasal hypothalamus and brainstem. Endocrinology. 1997;138(2):839 – 842. Yi C-X, Van der Vliet J, Dai J, Yin G, Ru L, Buijs RM. Ventromedial arcuate nucleus communicates peripheral metabolic information to the suprachiasmatic nucleus. Endocrinology. 2006;147(1):283– 294. Saeb-Parsy K, Lombardelli S, Khan FZ, McDowall K, Au-Yong IT, Dyball RE. Neural connections of hypothalamic neuroendocrine nuclei in the rat. J Neuroendocrinol. 2000;12(7):635– 648. Saderi N, Cazarez-Márquez F, Buijs FN, et al. The NPY intergeniculate leaflet projections to the suprachiasmatic nucleus transmit metabolic conditions. Neuroscience. 2013;246:291–300. Challet E, Pévet P, Malan A. Intergeniculate leaflet lesion and daily rhythms in food-restricted rats fed during daytime. Neurosci Lett. 1996;216(3):214 –218. Challet E, Jacob N, Vuillez P, Pévet P, Malan A. Fos-like immunoreactivity in the circadian timing system of calorie-restricted rats fed at dawn: daily rhythms and light pulse-induced changes. Brain Res. 1997;770(1–2):228 –236. Paxinos G, Franklin KB. The Mouse Brain in Stereotaxic Coordinates. San Diego: Academic Press. 2004. Gillette MU, Mitchell JW. Signaling in the suprachiasmatic nucleus: selectively responsive and integrative. Cell Tissue Res. 2002;309(1): 99 –107. Achermann P, Kunz H. Modeling circadian rhythm generation in the suprachiasmatic nucleus with locally coupled self-sustained oscillators: phase shifts and phase response curves. J Biol Rhythms. 1999; 14(6):460 – 468. Caron E, Sachot C, Prevot V, Bouret SG. Distribution of leptinsensitive cells in the postnatal and adult mouse brain. J Comp Neurol. 2010;518(4):459 – 476. O’Shea JJ, Gadina M, Schreiber RD. Cytokine signaling in 2002: new surprises in the Jak/Stat pathway. Cell. 2002;109(suppl):S121– S131. Lundkvist GB, Robertson B, Mhlanga JD, Rottenberg ME, Kristensson K. Expression of an oscillating interferon-␥ receptor in the suprachiasmatic nuclei. Neuroreport. 1998;9(6):1059 –1063. Duhart JM, Leone MJ, Paladino N, et al. Suprachiasmatic astrocytes modulate the circadian clock in response to TNF-␣. J Immunol. 2013;191(9):4656 – 4664. Impey S, Obrietan K, Storm DR. Making new connections: role of ERK/MAP kinase signaling in neuronal plasticity. Neuron. 1999; 23(1):11–14. Elmquist JK, Bjørbaek C, Ahima RS, Flier JS, Saper CB. Distributions of leptin receptor mRNA isoforms in the rat brain. J Comp Neurol. 1998;395(4):535–547. McGranaghan PA, Piggins HD. Orexin A-like immunoreactivity in the hypothalamus and thalamus of the Syrian hamster (Mesocricetus auratus) and Siberian hamster (Phodopus sungorus), with special reference to circadian structures. Brain Res. 2001;904(2):234 –244. Lamont EW, Bruton J, Blum ID, Abizaid A. Ghrelin receptor-knockout mice display alterations in circadian rhythms of activity and feeding under constant lighting conditions. Eur J Neurosci. 2014; 39(2):207–217.


1090

Grosbellet et al


35. Frühbeck G. Intracellular signalling pathways activated by leptin. Biochem J. 2006;393(Pt 1):7–20. 36. Morikawa Y, Ueyama E, Senba E. Fasting-induced activation of mitogen-activated protein kinases (ERK/p38) in the mouse hypothalamus. J Neuroendocrinol. 2004;16(2):105–112. 37. Purves T, Middlemas A, Agthong S, et al. A role for mitogen-activated protein kinases in the etiology of diabetic neuropathy. FASEB J. 2001;15(13):2508 –2514. 38. Ahrén B. Diurnal variation in circulating leptin is dependent on gender, food intake and circulating insulin in mice. Acta Physiol Scand. 2000;169(4):325–331. 39. Pinto S, Roseberry AG, Liu H, et al. Rapid rewiring of arcuate nucleus feeding circuits by leptin. Science. 2004;304(5667):110 –115. 40. Münzberg H. Differential leptin access into the brain—a hierarchical organization of hypothalamic leptin target sites? Physiol Behav. 2008;94(5):664 – 669. 41. Balland E, Dam J, Langlet F, et al. Hypothalamic tanycytes are an ERK-gated conduit for leptin into the brain. Cell Metab. 2014;19(2): 293–301. 42. Bouret SG, Draper SJ, Simerly RB. Trophic action of leptin on hypothalamic neurons that regulate feeding. Science. 2004;304(5667): 108 –110. 43. Canteras NS, Ribeiro-Barbosa ER, Goto M, Cipolla-Neto J, Swanson LW. The retinohypothalamic tract: comparison of axonal projection patterns from four major targets. Brain Res Rev. 2011;65(2): 150 –183.


44. Feldhammer M, Uetani N, Miranda-Saavedra D, Tremblay ML. PTP1B: a simple enzyme for a complex world. Crit Rev Biochem Mol Biol. 2013;48(5):430 – 445. 45. Challet E, Pévet P, Lakhdar-Ghazal N, Malan A. Ventromedial nuclei of the hypothalamus are involved in the phase advance of temperature and activity rhythms in food-restricted rats fed during daytime. Brain Res Bull. 1997;43(2):209 –218. 46. Yi C-X, Challet E, Pévet P, Kalsbeek A, Escobar C, Buijs RM. A circulating ghrelin mimetic attenuates light-induced phase delay of mice and light-induced Fos expression in the suprachiasmatic nucleus of rats. Eur J Neurosci. 2008;27(8):1965–1972. 47. Moga MM, Moore RY. Putative excitatory amino acid projections to the suprachiasmatic nucleus in the rat. Brain Res. 1996;743(1–2): 171–177. 48. Acosta-Galvan G, Yi C-X, Van der Vliet J, et al. Interaction between hypothalamic dorsomedial nucleus and the suprachiasmatic nucleus determines intensity of food anticipatory behavior. Proc Natl Acad Sci USA. 2011;108(14):5813–5818. 49. Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med. 2004;141(11):846 – 850. 50. Scheer FAJL, Hilton MF, Mantzoros CS, Shea SA. Adverse metabolic and cardiovascular consequences of circadian misalignment. Proc Natl Acad Sci USA. 2009;106(11):4453– 4458.


Pancreatic polypeptide controls energy homeostasis via Npy6r signaling in the suprachiasmatic nucleus in mice.

Lithium effects on circadian rhythms in fibroblasts and suprachiasmatic nucleus slices from Cry knockout mice.

ob) mice.

ob mice.

ob) mice.

ob) mice.

ob mice.

The effects of GABA and benzodiazepines on neurones in the suprachiasmatic nucleus (SCN) of Syrian hamsters.

ob mice.

Functional network inference of the suprachiasmatic nucleus.

Synaptology of the rat suprachiasmatic nucleus.

Neurogenesis of the hamster suprachiasmatic nucleus.

ob) mice: are they hyperoxaluric?

Disruption of the Suprachiasmatic Nucleus in Fibroblast Growth Factor Signaling-Deficient Mice.

ob mice.

Effects of histamine on spontaneous electrical activity of neurons in rat suprachiasmatic nucleus.

ob mice.

Circadian and ultradian rhythms of clock gene expression in the suprachiasmatic nucleus of freely moving mice.

ob mice.

Persistent neuronal Ube3a expression in the suprachiasmatic nucleus of Angelman syndrome model mice.

An enlarged suprachiasmatic nucleus in homosexual men.

Differential modulation of clock gene expression in the suprachiasmatic nucleus, liver and heart of aged mice.

Transplanted suprachiasmatic nucleus determines circadian period.

The efferent connections of the suprachiasmatic nucleus of the hypothalamus.