Siberian hamsters (Phodopus sungorus) undergo bouts of daily torpor during which body temperature decreases by as much as 20°C and provides a significant savings in energy expenditure. Natural torpor in this species is normally triggered by winterlike photoperiods and low ambient temperatures. Intracerebroventricular injection of neuropeptide Y (NPY) reliably induces torporlike hypothermia that resembles natural torpor. NPY-induced torporlike hypothermia is also produced by intracerebroventricular injections of an NPY Y1 receptor agonist but not by injections of an NPY Y5 receptor agonist. In this research, groups of cold-acclimated Siberian hamsters were either coinjected with a Y1 receptor antagonist (1229U91) and NPY or were coinjected with a Y5 receptor antagonist (CGP71683) and NPY in counterbalanced designs. Paired vehicle + NPY induced torporlike hypothermia in 92% of the hamsters, whereas coinjection of Y1 antagonist + NPY induced torporlike hypothermia in 4% of the hamsters. In contrast, paired injections of vehicle + NPY and Y5 antagonist + NPY induced torporlike hypothermia in 100% and 91% of the hamsters, respectively. Although Y5 antagonist treatment alone had no effect on body temperature, Y1 antagonist injections produced hyperthermia compared with controls. Both Y1 antagonist and Y5 antagonist injections significantly reduced food ingestion 24 h after treatment. We conclude that activation of NPY 1 receptors is both sufficient and necessary for NPY-induced torporlike hypothermia.
siberian hamsters employ several physiological and behavioral tactics to survive winter on the Mongolian and Siberian steppes (59, 60). Among them, presumably, Siberian hamsters periodically undergo daily torpor, a form of reversible hypothermia in which body temperature (Tb) typically decreases by 15–20°C for 5–8 h (24, 33). Daily torpor onset is cued by winterlike photoperiods with a short photophase (SP) and a low ambient temperature (Ta) (2, 33, 50). The onset of photoperiod-dependent torpor is significantly delayed, not occurring until photoperiod-dependent decreases in body and fat mass have already reached their seasonal nadir (3, 4, 40). Ablation of various neural mechanisms underlying the circadian and seasonal expression of torpor (e.g., the pineal gland or the suprachiasmatic or paraventricular nuclei of the hypothalamus) can disrupt its timing but cannot eliminate its occurrence (47, 48, 49, 56).
The neuroendocrine substrates mediating torpor initiation, expression, or rewarming remain for the most part only rudimentarily understood. Minimal fat reserves correlate with the seasonal nadir in leptin concentration (2.1 ng/ml) in Siberian hamsters (respectively, Refs. 40 and 34). In addition, high exogenous leptin concentrations inhibit torpor (17, 19, 21) and endogenous serum leptin concentrations of Siberian hamsters entering torpor were never >2.6 ng/ml (17). Reduced chronic concentrations of leptin coincident with minimal fat reserves are an apparent permissive factor for photoperiod-dependent torpor onset (17, 46). Postnatal monosodium glutamate (MSG) treatments produce a characteristic pattern of neural degeneration in the hypothalamic arcuate nucleus (ARC), primarily targeting neurons containing either neuropeptide Y (NPY)/agouti-related peptide (AgRP) or proopiomelanocortin/cocaine and amphetamine-related transcript (8, 12) and colocalizing ARC leptin receptors (1). MSG treatments may thus ablate the neuroendocrine transducer translating reduced leptin feedback into a permissive signal, allowing torpor onset (cf. Ref. 46). MSG-induced ablation of ARC eliminates photoperiod-dependent torpor in Siberian hamsters (46), food deprivation-induced torpor and its enhancement by ghrelin in laboratory mice (22), and torporlike decreases in circadian Tb in suckling rats (52). It is unclear whether destruction of ARC NPY-ergic mechanisms specifically contribute to the effect of MSG treatments on torpor (see below and Ref. 22).
A primary function of arcuate nucleus leptin/NPY-ergic mechanisms may be to conserve white adipose tissue reserves by reducing energy expenditure during energetic challenges (16, 57). During such challenges, reduced leptin concentrations disinhibit ARC NPY neurons, increasing turnover and release of NPY (1, 55). NPY may be a powerful orexigenic neurotransmitter, but it also has the capacity to decrease metabolism (e.g., 35). For example, intracerebroventricular injections of NPY in rats decrease thermogenesis by completely suppressing sympathetic neural input to brown adipose tissue (BAT) (14), decreasing BAT thermogenic activity (GDP binding) (5, 58), reducing whole body metabolic rate (58), and, consequently, lowering Tb (29). NPY injected intracerebroventricularly reliably decreases Tb in warm- and cold-acclimated homeothermic laboratory rats by 1–3°C (e.g., 7, 28, 53, 54). Cold-acclimated heterothermic Siberian hamsters, on the other hand, decrease Tb by as much as 20°C for as long as 18 h after comparable intracerebroventricular injections of NPY (42, 45). Temperature records during NPY-induced torporlike hypothermia resemble those during natural torpor (42). The magnitude of the decrease in Tb after intracerebroventricular injection of NPY may reflect the natural range of Tbs characteristic of the species (cf. Ref. 42).
The neurotransmitter (or neuromodulator), NPY, is a 36-amino acid peptide for which six possible G-protein receptor subtypes have been identified: Y1, Y2, Y3, Y4, Y5, and y6 (e.g., 1, 27). Studies in laboratory mice and rats using Y1 and Y5 receptor agonists, antagonists, and/or genetic knockout animals suggest that both receptor subtypes contribute to NPY-induced food intake (27, 41, 43). In Siberian hamsters, NPY injected intracerebroventricularly reliably increases short-term (1–4 h) food intake (6, 45), but Y1 and Y5 receptor agonists can at best be described as inconsistently affecting short-term food intake (11, 45). Siberian hamsters, on the other hand, markedly increased food hoarding after either an intracerebroventricular injection of NPY or of a Y1 receptor agonist; central Y5 agonist treatment did not affect hoarding (11).
Intracerebroventricularly injected NPY that produces mild hypothermia in the homeothermic rat induces marked torporlike hypothermia in heterothermic Siberian hamsters (42). Because both intracerebroventricularly injected NPY and an NPY Y5 agonist comparably reduced interscapular BAT temperature (TIBAT) in warm-acclimated rats, whereas neither a Y1, Y2, nor a Y4 agonist effected TIBAT (26), we predicted that NPY-induced torporlike hypothermia would most likely be mediated by activation of the NPY Y5 receptor site. To the contrary, NPY Y1 agonist ([d-Arg25]-NPY) treatment, but not Y5 agonist ([d-Trp34]-NPY) treatment, induced torporlike hypothermia (45). NPY Y1 receptor agonist and NPY intracerebroventricular injections produced torporlike hypothermia in a comparable proportion of treated hamsters, and Y1 agonist produced torporlike hypothermia that closely resembled that induced by NPY. Intracerebroventricular injection of the Y5 agonist did produce hypothermia, but it infrequently reached criterion for torporlike hypothermia and failed to resemble either NPY-induced torporlike hypothermia or natural torpor (45).
The purpose of this research was to determine whether activation of the Y1 receptor subtype solely underlies the capacity of NPY to trigger torporlike hypothermia. If NPY acting at the Y1 receptor site is necessary, then coinjection of a Y1 receptor antagonist with NPY should prevent the initiation of torporlike hypothermia. In addition, if activation of the Y5 receptor site by NPY is truly unnecessary for torporlike hypothermia initiation, then injection of a Y5 receptor antagonist immediately prior to NPY treatment should not prevent induction of torporlike hypothermia. To test this, one group of cold-acclimated Siberian hamsters was tested with the Y1 receptor antagonist, 1229U91 (GR231118), and another group of hamsters was tested with the Y5 receptor antagonist, CGP71683A. Each group of hamsters received four different paired treatments over the course of 4 wk in a counterbalanced design. Body temperature data were analyzed for minimum Tb and duration of torporlike hypothermia, if present, and 24-h food intake was calculated.
MATERIALS AND METHODS
This research was carried out in laboratory facilities approved by the Association for the Assessment and Accreditation of Laboratory Animal Care. All procedures were approved by the Animal Care and Use Committee of the University of California at Berkeley and conformed to the standards established by the National Institutes of Health.
Fifty-two adult female Siberian hamsters reared in a photoperiod (14:10-h light-dark cycle) with a long photophase (LP) and Ta = 22°C were housed individually in polypropylene tub cages (12.5 × 8.0 × 5.5 cm) with Care Fresh Bedding (Harlan, San Diego, CA) and provided water and food (Lab Diet, #5015) ad libitum, except where described otherwise.
After 3 wk of cold acclimation to Ta = 10°C, all hamsters were deeply anesthetized with a ketamine-based anesthetic (0.34 ml/kg body mass ip, containing 21 mg ketamine + 2.4 mg xylazine + 0.3 mg acepromazine/ml) to undergo a single surgery, in which an intracerebroventricular guide cannula was implanted intracranially and a telemetric temperature transmitter was implanted intra-abdominally. In an initial procedure, a 26-gauge guide cannula was stereotaxically lowered to just above the third ventricle using the following coordinates: 0.0 mm anterior to bregma, 0.0 mm lateral to the midsagittal line, and 5.5 mm ventral to dura. The cannula was fixed to the skull with cyanoacrylate glue, stainless-steel screws, and dental acrylic. A stainless-steel stylet remained in the cannula, except during injections, to maintain patency. The wound was sutured shut with sterile sutures, if necessary.
With completion of the cannula implantation, the abdomen was shaved, a midline incision made, and a transmitter inserted into the peritoneal cavity. The peritoneum and skin were closed with sterile sutures. At the completion of surgery, the hamsters received an injection of 0.1 ml buprenorphine (0.015 mg/ml) to provide postsurgical analgesia. The hamsters were allowed at least 7 days to recover before the first injection.
Pharmacological treatments and injection procedure.
NPY Y1 subtype receptors were blocked by intracerebroventricular injections of the NPY Y1 antagonist, 1229U91 (GR231118, AnaSpec, San Jose, CA) (31, 32). The dose of Y1 antagonist was 40 μm 1229U91 in 1.0 μl sterile saline, and vehicle injections were 1.0 μl sterile saline. NPY Y5 subtype receptors were blocked by intraperitoneal injections of the NPY Y5 antagonist, CGP71683A (Sigma) (9, 41). The dose of Y5 antagonist was 10 μg CGP71683A/g body mass. The concentration of CGP71683A was 10 μg/2 μl 10% DMSO in sterile saline, and the vehicle injections were 2 μl DMSO/g body mass. The animals also received intracerebroventricular injections of NPY (AnaSpec). The NPY dose was 7.5 μg in 1.0 μl sterile saline.
For the intracerebroventricular injection procedure, the stylet was removed from the guide cannula, and a 33-gauge injection cannula was lowered until it extended 1.0 mm beyond the guide cannula. A Hamilton microsyringe was connected to the injection cannula with polyethylene tubing, and the injectate was slowly injected over 30 s. The injection cannula was then left in place for ∼30–45 s to minimize reflux before it was removed and the stylet returned to the guide cannula.
Body temperature measurements and torpor criterion.
Cages with Siberian hamsters bearing telemetric transmitters (Model VM-FH-LT, Minimitter, Sunriver, OR) for remote recording of Tb were placed on individual receiver boards. The boards captured signals from the transmitters, and the signals were then averaged every 10 min and stored on a computer (DataQuest, St. Paul, MN). The Tb data were analyzed for minimum Tb (Tb min), which was determined as the lowest Tb during the 2 h after treatment or if torporlike hypothermia occurred, the absolute minimum Tb during the bout of hypothermia. Criterion for torporlike hypothermia was the same as we and others have used for natural torpor, i.e., Tb min < 32.0°C for a minimum of 30 consecutive min (46, 50). Duration of torporlike hypothermia was measured as the total time from the first Tb < 32.0°C to the first Tb > 32.0°C. Tb data were also analyzed for the proportion of animals entering torporlike hypothermia subsequent to treatments.
The adult Siberian hamsters were moved into an environmental chamber with the same LP in which they had been reared but Ta was lowered to 10°C. The hamsters were allowed to acclimate to the new temperature for 3 wk. The hamsters were weighed and divided into 2 weight-balanced groups: 1 for testing the NPY Y1 receptor antagonist (1229U91; n = 28) and the other for testing the NPY Y5 receptor antagonist (CGP71683A; n = 24). All injections were made within the first 1–3 h of the light (rest) phase when natural torpor normally occurs.
NPY Y1 receptor antagonist treatments.
All food was removed from the animals' cages ∼1 h after light onset, and injections were begun at this time. In the first week, one-quarter of the hamsters were injected intracerebroventricularly with 1229U91 and then were immediately injected intracerebroventricularly with NPY (Y1A + NPY group), the next one-quarter was injected intracerebroventricularly with saline then immediately injected intracerebroventricularly with NPY (Sal + NPY group), the third one-quarter was injected intracerebroventricularly with 1229U91 and then saline (Y1A + Sal group), and the final one-quarter was injected intracerebroventricularly with saline and then given another intracerebroventricular saline injection (Sal + Sal group) (see Table 1). In the successive 3 wk, the treatments were repeated in a counterbalanced design, such that each hamster received each treatment pairing (Table 1). Upon completion of a pair of injections, the hamster was immediately returned to its cage and Tb measurements continued. A preweighed amount of food was provided to each hamster's cage at the end of the light phase, ∼8 h later, after the likely termination of any NPY-induced torporlike hypothermia. The food was reweighed 24 h from the time of treatment, and intake was calculated.
NPY Y5 receptor antagonist treatments.
This same design was repeated with the hamsters receiving Y5 antagonist treatments (see Table 1). Approximately 1 h after light onset, food was removed from the animals' cages and treatments were begun. In the first week, one-quarter of the hamsters were injected intraperitoneally with CGP71683A, then were injected intracerebroventricularly with NPY (Y5A + NPY group), the second quarter was injected intraperitoneally with DMSO vehicle and then intracerebroventricularly with NPY (DMSO + NPY group), another one-quarter received CGP71683A injected intraperitoneally and saline injected intracerebroventricularly (Y5A + Sal group), and the final one-quarter was injected intraperitoneally with DMSO vehicle and intracerebroventricularly with saline (DMSO + Sal group). All animals received all treatments in a counterbalanced design over 4 wk (Table 1). The animals were immediately returned to their cages after treatment, and Tb and 24-h food intake measurements were performed as above.
The hamsters were administered a lethal injection of pentobarbital sodium (euthanasia solution) with the completion of testing. While the animals were deeply anesthetized, India ink was injected through the injection cannula, and the hamsters were transcardially perfused with 0.1 M PBS followed by buffered formalin. The brain was removed and stored in buffered formalin until frozen coronal sections (40 μm) were cut, mounted on prepared slides, stained with cresyl violet, and coverslipped. Locus of cannula termination and presence or absence of India ink in third ventricle were verified by two observers.
The proportion of hamsters undergoing torporlike hypothermia after Y1 or Y5 antagonist treatments was analyzed by χ2-test Minimum Tb and 24-h food intake were each analyzed by repeated-measures ANOVA with post hoc comparisons between means (Student-Newman-Keuls). Minimum Tb during torporlike hypothermia and duration of torporlike hypothermia were compared between the DMSO + NPY and Y5A + NPY groups using t-tests. Maximum Tb after vehicle + vehicle vs. Y1A + Sal vs. Y5A + Sal treatments was analyzed using one-way ANOVA with post hoc comparisons. The relation between food intake as a function of Tb min during torporlike hypothermia was determined by linear regression. Statistics were performed using the SigmaStat 3.0 program (SPSS, Chicago, IL). All differences were considered statistically significant if P < 0.05. Precise P values are given in accordance with American Physiological Society's Guidelines for Authors.
Twenty-seven of the hamsters in the Y1 receptor antagonist group and 24 of the hamsters in the Y5 receptor antagonist group had cannulas verified as terminating in the third ventricle. Hamsters in which the cannulas did not successfully terminate in the third ventricle were not considered in the analyses.
NPY Y1 receptor antagonist treatments.
The various treatment pairings significantly affected the proportion of Siberian hamsters entering torporlike hypothermia (χ2 = 90.4, P < 0.1×10−15; Fig. 1). Coinjection of Sal + Sal or Y1A + Sal induced torporlike hypothermia in 0% of the hamsters. Injection of Sal + NPY initiated torporlike hypothermic bouts in 92.3% of the hamsters (Fig. 2). Coinjection of Y1A + NPY, on the other hand, induced torporlike hypothermia in 1 of 28 hamsters (3.8%; Fig. 2). There was also a significant effect of treatment on Tb min (F = 75.5, P < 0.1 × 10−15; Table 2). The Tb min of hamsters treated with Sal + NPY was significantly different from each of the other treatments (P = 0.0001, for each), whereas the Tb min of none of the other three treatments (Sal + Sal, Y1A + Sal and Y1A + NPY) differed from each other (P = 0.21, P = 0.36, and P = 0.08, respectively; Table 2).
NPY receptor antagonist treatments significantly affected maximum Tb during the first hour after treatment (ANOVA on ranks, P = 0.3 × 10−12; Table 3). The maximum Tb of the hamsters receiving the Y1A + Sal treatment was significantly greater than that of the vehicle + vehicle controls (P = 0.0002; Table 3).
Unplanned observations of the Siberian hamsters coinjected with Y1A + NPY revealed very unusual patterns of behavior, which were characterized by rapid breathing and shivering and limb twitching. They either lied on their back with their dorsum exposed or underwent “shaky and wobbly” walking. Although hamsters treated with Y1A + Sal also exhibited rapid breathing and some shivering, they typically assumed the curled-up posture characteristic of torpor even though they were not hypothermic.
Twenty-four hour food intake was significantly affected by NPY Y1 receptor antagonist treatments (F = 47.2, P < 0.1 × 10−15; Fig. 3). The Y1 antagonist significantly reduced 24-h intake. The Y1A + Sal and the Y1A + NPY treatment pairing groups both reduced consumption compared with the Sal + Sal or Sal + NPY treatments (P = 0.0002 for all; Fig. 3). Food intake of the Y1A + Sal group did not differ from that of Y1A + NPY-treated hamsters (P = 0.50).
Twenty-four-hour food ingestion of hamsters coinjected with Sal + NPY did not differ from that of Sal + Sal injected controls (P = 0.47, Fig. 3). Food intake of hamsters after undergoing torporlike hypothermia due to Sal + NPY treatment also did not vary as a function of minimum Tb (F = 2.2, P = 0.16; not shown).
NPY Y5 receptor antagonist treatments.
The proportions of hamsters entering torporlike hypothermia were affected by the different treatment pairings (χ2 = 87.7, P < 0.1 × 10−15; Fig. 1). Neither coinjection of DMSO + Sal nor Y5A + Sal produced torporlike hypothermia (0%, for both). 100% of the hamsters treated with DMSO + NPY entered torporlike hypothermia and 91.3% of the hamsters receiving Y5A + NPY underwent torporlike hypothermic bouts (Fig. 4). Tb min was significantly affected in the four treatment groups (F = 220.8, P < 0.1 × 10−15; Table 4). With the exception of the DMSO + Sal vs. Y5A + Sal group comparison (P = 0.24), all pairwise comparisons were significantly different, including DMSO + NPY vs. Y5A + NPY (P = 0.0003, for all; Table 4). Similarly, when comparing only those animals entering torporlike hypothermia in the latter two pairings, there was a significant difference in Tb min during torporlike hypothermia (t-test = 2.7, P = 0.011; Table 4) and in torporike hypothermia duration (t-test = 3.4, P = 0.002; Table 4).
Maximum Tb during the 1 h following Y5 antagonist treatments (Y5A + Sal) did not differ from that of vehicle + vehicle controls (P = 0.27, Table 3) but was significantly less than the maximum Tb of Y1 antagonist treated hamsters (P = 0.0002, Table 3).
NPY Y5 antagonist treatments significantly affected 24-h food intake (F = 12.5, P = 0.000003; Fig. 3). The Y5A + Sal treatment pairing resulted in lower 24-h food intakes than the other three treatment pairings (DMSO + Sal, DMSO + NPY, and Y5A + NPY) (P = 0.0002, P = 0.0002, and P = 0.011, respectively; Fig. 3). The Y5A + NPY pairing consumed less food than the DMSO + Sal group (P = 0.010).
DMSO + NPY treatment group vs. DMSO + Sal group did not differentially affect 24-h food ingestion (P = 0.20, Fig. 3). After the DMSO + NPY treatment, 24-h food intake also was not a function of minimum Tb (F = 0.7, P = 0.42; not shown). Food intake was a significant function of Tb min after the Y5A + NPY paired treatment (F = 5.2, P = 0.036; not shown) but was negatively correlated to Tb min (r = −0.48).
Pretreatment of cold-acclimated Siberian hamsters with an NPY Y1 receptor antagonist completely prevented NPY from inducing the torporlike hypothermia characteristic of NPY treatment alone. Previously, intracerebroventricular injections of a Y1 receptor agonist mimicked NPY injections, initiating comparable bouts of torporlike hypothermia (45). Activation of the NPY Y1 receptor subtype is thus both sufficient and necessary for the ability of NPY to initiate torporlike hypothermia in Siberian hamsters. Conversely, cotreatment of hamsters with a Y5 receptor antagonist and NPY completely failed to prevent the onset of NPY-induced torporlike hypothermia. In addition, intracerebroventricular injections of a Y5 agonist previously failed to produce torporlike hypothermia comparable to that after NPY treatments (45). NPY activation of the Y5 receptor site, unlike the Y1 receptor, appears neither sufficient nor necessary for the initiation of torporlike hypothermia. Intracerebroventricular injected NPY appears to have the capacity to induce a pattern of hypothermia comparable to that during natural torpor (see below), and this ability requires activation of the NPY Y1 receptor.
We propose that NPY Y1 receptor sites activated by NPY injected intracerebroventricularly directly or multisynaptically affect thermoregulatory neurons underlying defended Tb. Intracerebroventricular Y1 agonist injections thus induce torporlike hypothermia (45), whereas intracerebroventricular Y1 antagonist injections produce hyperthermia. Activation of the Y1 receptor by NPY produces a prolonged reduction in regulated Tb that appears similar to that recorded during natural torpor. Although Tb typically decreases by 15–20°C for 5–8 h during photoperiod-dependent torpor in Siberian hamsters (24, 33), this is also somewhat misleading. For example, in a recent study (46), only 68% of the hamsters underwent torpor between weeks 12 and 18 of SP/cold exposure. The mean minimum Tb of those hamsters entering torpor was 22°C (15°C < euthermia) and the mean duration of torpor was 5 h; nevertheless, the range of minimum Tbs was 29.8-17.4°C, and the range of torpor durations was 40–710 min. While the mean minimum Tbs and the mean durations during NPY-induced torporlike hypothermia may tend to be less than those seen during natural torpor (e.g., Tables 2 and 4), the range of minimum Tbs and torpor durations typically observed during NPY-induced torporlike hypothermia (42, 45, present study) are comparable (Fig. 5). In addition, NPY-induced torporlike hypothermia is not a passive, obligatory hypothermia, but like natural torpor is sensitive to external disturbances. Handling and even loud noises are sufficient to prematurely trigger rewarming from torporlike hypothermia (J. Dark and K. M. Pelz, unpublished observations). In a manner also similar to natural torpor, NPY-induced torporlike hypothermia has a minimum Tb of ∼15°C which it does not exceed regardless of NPY dose or Ta (42, 45, present study).
Inhibition of Y1 receptor activation by the Y1 antagonist, on the other hand, produces a brief, phasic hyperthermia. Y1 antagonist presumably induces hyperthermia by temporarily removing the tonic influence of baseline endogenous NPY from the same thermoregulatory neurons mediating defended Tb. A temporary elevation in regulated Tb is thus likely achieved, at least partially, by the observed thermogenic shivering. Intracerebroventricular injections of Y1 receptor antisense oligodeoxynucleotides in laboratory rats, which temporarily impair receptor functioning, similarly produce an intense but transient hyperthermia (36).
Prior to the present findings in cold-acclimated Siberian hamsters, the NPY Y5 receptor appeared the more likely candidate to mediate the effect of NPY on Tb. Y1 receptor activation (as well as Y2 and Y4 receptors) failed to affect interscapular BAT temperature in rats (26). The intracerebroventricular injection of a Y5 agonist reduced interscapular BAT temperature and energy expenditure comparably to that observed after NPY treatment (26). In addition, long-term Y5 agonist treatments in mice decreased BAT UCP 1 mRNA content and reduced BAT nonshivering thermogenesis (39), while long-term Y5 antagonist treatments increased BAT UCP 1 mRNA and presumably increased nonshivering thermogenesis (38). Stimulation of Y5 receptors by NPY clearly has the potential to affect thermogenesis and/or energy expenditure in rats and mice. In fact, the limited and inconsistent hypothermia observed previously after intracerebroventricular Y5 receptor agonist injections in cold-acclimated Siberian hamsters may possibly represent a compromised ability to maintain Tb (45).
The regulated change in defended Tb necessary for NPY-induced torporlike hypothermia may require activation of Y1 receptors; nevertheless, it is just as likely that NPY activation of Y5 receptors contribute to the full expression of NPY-induced torporlike hypothermia. We are unaware of any data suggesting Y1 receptor activation capable of affecting BAT nonshivering thermogenesic capacity or activity or BAT temperature (e.g., 26, 43). Y5 receptor activation apparently underlies the inhibition of BAT nonshivering thermogenesis characteristic of intracerebroventricularly injected NPY (5, 14, 26, 58). Thermogenesis in cold-adapted Siberian hamsters is achieved primarily by BAT nonshivering thermogenesis, and natural torpor most likely involves an inhibition of BAT nonshivering thermogenesis (20, 23, 25). Even though coinjection of Y5 antagonist with NPY failed to prevent torporlike hypothermia, torporlike hypothermia was significantly less pronounced. Presumably, NPY continues inducing torporlike hypothermia via Y1 receptor activation, but the presence of Y5 antagonist disrupts the neural pathway mediating the usual inhibition of BAT nonshivering thermogenesis. Torporlike hypothermia may occur despite persistent, uninhibited BAT thermogenesis. There are several ways animals can reduce Tb that do not involve BAT. First, they can increase heat loss via changes in conductance. Second, intracerebroventricular NPY, Y1 agonist, or Y5 agonist injections all produce a state of central hypothyroidism characterized by markedly reduced hypothalamic concentrations of proTRH mRNA and plasma concentrations of TSH, T3, and T4 (e.g., 15). Finally, AgRP can directly inhibit the sympathetic innervation of BAT and reduce BAT temperature (62). Animals kept in ambient temperatures above thermoneutrality, for example, can lose heat, but it requires increased energy expenditure, which may explain the negative correlation between food intake and minimum Tb in hamsters receiving paired Y5 antagonist + NPY injections.
In contrast, the opposite situation may prevail during coinjection of Y1 antagonist with NPY and explain the hamsters' unusual behavior. The presence of the Y1 antagonist prevents a downward shift in Tb setpoint and initiation of torporlike hypothermia; nevertheless, NPY continues activating Y5 receptors and inhibiting BAT nonshivering thermogenesis. Compromised BAT function may have necessitated increased shivering thermogenesis (rapid breathing, shivering, and limb twitching) for the hamsters to maintain euthermia. The hamsters' behavior was interestingly reminiscent of the behavior of ground squirrels rewarming from hibernation, in which shivering thermogenesis plays a major role and BAT nonshivering thermogenesis a lesser role (10, 20).
Ad libitum food intake of cold-acclimated Siberian hamsters apparently requires both NPY Y1 and Y5 mechanisms. Both the Y1 antagonist + saline and Y5 antagonist + saline groups consumed significantly less food than their respective control groups; the importance of NPY pathways during metabolic challenges has been long recognized (30, 61). Food ingestion of the Y1 antagonist + saline and Y1 antagonist + NPY pairings was comparable, whereas the Y5 antagonist + NPY group ingested significantly more food than the Y5 antagonist + saline group. This would seem to confirm in cold-acclimated Siberian hamsters the greater importance of NPY Y1 vs. Y5 receptors for NPY-induced food intake as suggested by experiments using Y1 vs. Y5 receptor KO mice (e.g., 43). Previously, torporlike hypothermia induced by NPY injected intracerebroventricular provided an energy intake savings, reducing 24-h food intake compared with euthermic vehicle-injected controls (42, 45). Twenty-four-hour food ingestion of the vehicle + NPY pairing in neither the Y1 antagonist nor the Y5 antagonist treatment groups differed significantly from the appropriate control group. Although the basis for this apparent absence of a hypothermia-dependent energy savings is not currently evident, the most likely cause may be disturbances arising from the paired injection paradigm.
Activation of the NPY Y1 receptor has been repeatedly associated with adaptive responses to reduced energy intake. For example, refeeding after a 24-h fast is markedly diminished in Y1 receptor KO mice (44) but not in Y5 receptor KO mice (37). A Y1 receptor antagonist (1229U91) injected intracerebroventricularly similarly inhibits feeding after an overnight fast (31), and Y1 receptor antisense oligodeoxynucleotide treatments impair the increased energy intake after overnight fasting (51). Although refeeding after either food restriction or fasting is comparatively limited in Siberian hamsters, food restriction or an intracerebroventricular injection of Y1 agonist significantly increases foraging and hoarding (11). Intracerebroventricular injection of a Y1 antagonist inhibited food intake and foraging induced by fasting but, surprisingly, had a less pronounced effect on hoarding (32). The NPY Y1 receptor also apparently contributes to the delayed puberty due to reduced energy intake (e.g., 18, 43). Because both photoperiod-dependent and food restriction-induced torpor are associated with reduced caloric intake, the dependence of NPY-induced torporlike hypothermia upon NPY Y1 receptor activation suggests a possible function for the Y1 receptor in natural torpor as well.
This research was supported by National Institutes of Health Grant, NS-30816.
We are grateful to David Routman, Chris Tuthill, and Elanor E. Shoomer for their excellent technical assistance. We also thank David Piekarski and Lance Kriegsfeld for their valuable comments on an earlier version of the manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society