Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Pronounced changes occur in the mammalian circulation in hibernation. These include a slowing of heart rate to as low as 3 beats per minute, a decrease in blood pressure to a mean of about 50 mmHg, a decrease in respiration and metabolic rate, a drop in body temperature to approximately that of the environment and an increase in peripheral vascular resistance (Chatfield and Lyman, 1950; Lyman, 1965; Willis, 1979; Nedergaard and Cannon, 1990). However, relatively little is known about the changes that may occur in perivascular nerves, smooth muscle and endothelium and their specific roles in the increase in peripheral vascular resistance known to occur in the onset and maintenance of hibernation. Similarly, little is known of whether there are changes in vascular function associated with the rapid reversal of these circulatory changes that occur in arousal from hibernation.

In a recent study we showed that after 4 wk of hibernation there is an increase in sympathetic neurotransmission in mesenteric arteries of the golden hamster that appears to involve postjunctional changes since the sensitivity of vasoconstrictor responses to exogenous NE is increased (Ralevic et al., 1997). Increases in sensitivity to NE have also been described in renal arteries, but interestingly not in aorta or femoral arteries, of the hibernating woodchuck (Miller et al., 1986), nor in portal veins of the hibernating hedgehog (Eliassen and Helle, 1975). By contrast, endothelium-dependent vasodilatation is unchanged in hibernation, at least in golden hamster mesenteric arteries (Ralevic et al., 1997).

Our primary aim was to investigate the effects of 8 wk of hibernation on vascular responsiveness of the perfused mesenteric arterial bed of the golden hamster. In the wild, hamsters are known to hibernate in bouts of 3 to 5 days which was also the case in our study. As far as the season of hibernation is concerned, this typically occurs over the winter months, although the actual length varies considerably depending on environmental conditions (Lyman, 1965). The period of hibernation investigated in the present study (8 wk) is twice as long as that previously shown to be associated with an increase in mesenteric arterial sympathetic neurotransmission (Ralevic et al., 1997). It should be noted that hibernation of all mammals including the golden hamster is intermittent, being broken at regular intervals by brief periods of wakefulness, possibly for elimination of waste products (Lyman, 1965; Nedergaard and Cannon, 1990).

During arousal pronounced changes occur in the circulation of hibernating animals. There is an increase in body temperature from 5-10°C to 37°C within the space of a few hours, which requires an intense thermogenic effort (Lyman, 1948, 1982). Arousal starts with an increase in heart rate, respiration rate and oxygen consumption, followed by a rise in body temperature using endogenously generated heat (Lyman and O'Brien, 1988). The increase in body temperature is gradual, with the anterior of the animal warming more quickly than the posterior, due predominantly to differential regional blood flow (Lyman, 1965). Vasoconstriction of cutaneous vessels and differential vasoconstriction between fore and hind parts of the body impede circulation to the posterior and shunt circulation to the anterior, shortening the time involved in the rewarming process. Once the anterior part of the body is well heated, vasodilatation occurs and the posterior warms rapidly (Chatfield and Lyman, 1950; Lyman, 1965). The second aim of this study was to investigate the effects of arousal from hibernation on the function of golden hamster mesenteric arteries.

Acute and chronic decreases in temperature can alter the release of sympathetic neurotransmitters and vascular responses mediated by the smooth muscle. In nonhibernators both augmented and attenuated postjunctional responsiveness have been described with cooling in vitro (Flavahan et al., 1985; Ito and Chiba, 1985; Flavahan and Vanhoutte, 1986; Harker and Webb, 1987; Miller et al., 1986; Yamamoto et al., 1992). Tissue-specific augmented sensitivity of vascular smooth muscle to NE with cooling of hibernated compared with nonhibernated animals has been described (Miller et al., 1986; Harker and Webb, 1987). A change in calcium channels and the regulation of calcium influx with cold-tolerance in hibernating species has also been reported (Hall et al., 1987; Wolowyk et al., 1990). In view of this, our investigation included an assessment of mesenteric arterial vascular function in a cold-exposed control group comprising golden hamsters which had been exposed to the same conditions as the hibernators but which did not themselves hibernate.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Adult male golden hamsters (Mesocricetus auratus) (6-8 mo old) were used in the study. The animals were divided into four groups: one group was left to hibernate for 8 wk (group 1) and were killed while hibernating; as is characteristic, hibernation over this period was intermittent, occurring in bouts of 3 to 5 days. The second group were hamsters taken from group 1 and aroused by being removed from cold conditions, placed at room temperature (22°C) and used at 2 hr after this transfer. A third group, "cold-exposed controls," were exposed to the same levels of temperature and photoperiod as those hamsters that hibernated, but did not themselves hibernate. The fourth group, "age-matched controls," consisted of hamsters of the same age as those that had undergone hibernation but that had not been exposed to any temperature reduction or alteration in photoperiod.

Induction of hibernation. Animals were placed in a ventilated, refrigerated incubator (model PL3, Leec, Nottingham, UK) with the temperature set at 20°C and the light/dark period set at 8 hr of light per day. The temperature and photoperiod were then gradually reduced by 5°C approximately every 5 days to simulate winter conditions. The light period was reduced by 30 min/day and the temperature was also gradually decreased to reach final conditions of 2 hr of light per day and 9°C ambient temperature. The animals were placed in separate cages with adequate nesting material and transferred to a cold room that was set at 5°C and had a photoperiod of 2 hr. The animals were monitored at regular intervals to check for hibernation. Hamsters were given access to food and water ad libitum.

Arousal from hibernation. Animals were aroused from hibernation by removal from the cold room to room temperature (22°C). This prompted an awakening from hibernation.

Isolated mesenteric arterial bed preparation. Hamsters were killed by asphyxiation with CO₂. Mesenteric beds were isolated and set up for perfusion as described previously (Hill et al., 1996; Ralevic et al., 1997). In brief, the abdomen was opened and the superior mesenteric artery exposed and cannulated with a hypodermic needle. The superior mesenteric vein was cut, blood flushed out through the preparation with approximately 0.5 ml of Krebs solution, the gut dissected away from the mesenteric vasculature and the preparation mounted on a stainless steel grid (7 × 5 cm) in a humid chamber (custom made at University College London). The preparation was perfused at a constant flow rate of 3 ml/min using a peristaltic pump (model 7554-30, Cole-Parmer Instrument Co., Chicago, IL). The perfusate was Krebs solution of the following composition (mM): NaCl 133, KCl 4.7, NaH₂PO₄ 1.35, NaHCO₃ 16.3, MgSO₄ 0.61, CaCl₂ 2.52 and glucose 7.8, gassed with 95% O₂-5% CO₂ and maintained at 37°C. Responses were measured as changes in perfusion pressure (mmHg) with a pressure transducer (model P23XL, Viggo-Spectramed, Oxnard, CA) on a side arm of the perfusion cannula, and recorded on a polygraph (model 7D, Grass Instrument Co., Quincy, MA). Preparations were allowed to equilibrate for 30 min before experimentation. All doses of drugs were applied as 50 µl bolus injections into a neoprene rubber injection port proximal to the preparation. Antagonists were added to the perfusate reservoir.

Vasoconstriction. Electrical field stimulation (EFS; 2-64 Hz, 90 V, 1 ms for 30 s) was applied in an order of increasing frequency to produce frequency-response curves. EFS was followed by the addition of increasing doses of NE (0.05-500 nmol), doses of the pyrimidine UTP (5-5000 nmol) (control and hibernated groups only) and then of the purine ,-meATP (0.5-50 nmol). NE and ,-meATP were used because they mimic the effects of sympathetic transmitters, and UTP (which is released from platelets) allowed further assessment of vascular responsiveness. Because of receptor desensitization a dose-interval for ,-meATP of 10 min was used. Application of KCl (0.15 mmol) was followed by application of doses of histamine (5-5000 nmol).

Vasodilatation. Endothelium-dependent vasodilatation to ACh and UTP was examined in age-matched, cold-exposed and hibernated groups. Methoxamine (3-100 µM) was added to the perfusate to raise the tone of the preparations by approximately 50 mmHg above baseline. Once a stable tone had been achieved, vasodilator dose-response curves to ACh (0.005-50 nmol) and UTP (0.05-500 nmol) were carried out.

Drugs used. ,-methylene ATP (lithium salt), ACh (chloride), histamine (dihydrochloride), methoxamine (hydrochloride), norepinephrine bitartrate and UTP (sodium salt) were from Sigma Chemical Co. (St. Louis, MO).

Data analysis. Vasoconstrictor responses were evaluated as the increase in perfusion pressure in mmHg above baseline. Results are presented as means ± S.E., with numbers of observations in parentheses (n). pD₂ values were calculated as the negative log of the dose (in moles) that produced a half-maximal response. This method was also used to calculate the frequency that produced a half-maximal constrictor response to stimulation of sympathetic nerves. Comparisons between the groups were made by ANOVA with P < .05 considered significant. When maximal responses were not achieved, responses at each dose were compared using ANOVA. Post hoc analysis was with Tukey's test and differences were considered significant when P < .05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. There was a significant difference in body weight among the four groups of hamsters. The age-matched controls were heavier than any of the other groups, the cold-exposed controls were heavier than the hibernated and arousal groups and there was no significant difference in mean body weight between the hibernated and arousal groups. Body weights were: age-matched controls, 160.4 ± 6.0 g (n = 13); cold-exposed controls, 132.8 ± 2.9 g (n = 13); 8 wk of hibernation, 103.6 ± 4.9 g (n = 7); arousal from hibernation, 96.6 ± 3.6 g (n = 6).

Unconstricted preparations. Basal perfusion pressure of the isolated perfused mesenteric arterial preparations was not significantly different among the groups. These were: 23.1 ± 3.6 mmHg (n = 13) in age-matched controls; 37.1 ± 5.6 mmHg (n = 13) in cold-exposed controls, 25.0 ± 5.0 mmHg (n = 7) at 8 weeks of hibernation and 36.2 ± 6.8 mmHg (n = 6) in the arousal from hibernation group.

EFS. EFS (2-64 Hz, 90 V, 1 ms, 30 s) elicited frequencydependent vasoconstriction due to the activation of sympathetic nerves (18). At 8 wk of hibernation the sensitivity of responses was augmented resulting in a statistically significant leftward shift in the frequency-response curve (fig. 1; table 1). There was no difference among the groups in the maximum contractile responses to EFS.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Vasoconstrictor responses of golden hamster isolated mesenteric arterial beds to electrical field stimulation (EFS; 2-64 Hz, 90V, 1 ms, 30 s) in hibernation. Responses shown are to mesenteric arterial beds isolated from: age-matched controls (, n = 12); cold-exposed controls (, n = 8); 8 wk hibernators (, n = 7); arousal (, n = 6). The sensitivity of responses in the hibernation groups was significantly greater compared with the age-matched controls.

NE. Exogenous NE elicited dose-dependent vasoconstriction of the mesenteric arterial preparations. There was no significant difference in the sensitivity of responses between the groups. However, the maximal response was significantly greater in the arousal from hibernation group than the hibernation, cold-exposed and age-matched control groups (fig. 2). pD₂ values and maximal heights of responses are given in table 1.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Vasoconstrictor responses of golden hamster isolated mesenteric arterial beds to norepinephrine (NE; 0.05-500 nmol). Responses shown are those of mesenteric arterial beds isolated from: age-matched controls (, n = 12); cold-exposed controls (, n = 8); 8 wk hibernators (, n = 7); arousal from hibernation (, n = 6). Maximum responses of the arousal group were significantly different from the hibernation, cold-exposed and age-matched control groups.

,-meATP. Vasoconstrictor responses in the group aroused from hibernation were significantly augmented at 0.5 nmol ,-meATP, but were not significantly different from other groups at 5 nmol ,-meATP. Responses at 0.5 nmol ,-meATP were: age-matched controls, 9.6 ± 4.3 mmHg (n = 13); cold-exposed controls, 18.0 ± 2.6 mmHg (n = 8); hibernated, 8.8 ± 2.2 mmHg (n = 7); arousal, 40.2 ± 9.5 mmHg (n = 6). Responses at 5 nmol ,-meATP were: age-matched controls, 46.9 ± 4.2 mmHg (n = 13); cold-exposed controls, 54.5 ± 9.5 mmHg (n = 8); hibernation, 51.6 ± 11.5 mmHg (n = 7); arousal, 79.3 ± 9.8 mmHg (n = 6).

UTP. UTP (5-5000 nmol) elicited dose-dependent contractions in unconstricted preparations from age-matched controls, cold-exposed controls and hibernated hamsters. There was no significant difference in responses to UTP between the 8 wk hibernated and the two control groups (fig. 3).

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Vasoconstrictor responses of unconstricted golden hamster mesenteric arterial beds to uridine 5'-triphosphate (UTP; 5-5000 nmol). Responses shown are those of mesenteric arterial beds isolated from: age-matched controls (, n = 12); cold-exposed controls (, n = 8); 8 wk hibernators (, n = 7).

Histamine. There was no significant difference in vasoconstrictor responsiveness to histamine (5 µmol) among the groups. Responses were: age-matched controls, 14.9 ± 2.6 mmHg (n = 10); cold-exposed controls, 30.9 ± 5.5 mmHg (n = 8); hibernation, 27.3 ± 6.3 mmHg (n = 6); arousal, 16.0 ± 3.7 mmHg (n = 6).

KCl. Mean contractions elicited by KCl (0.15 mmol) were similar among the four groups: age-matched, 36.9 ± 3.0 mmHg (n = 13); cold-exposed control, 38.0 ± 5.9 mmHg (n = 8); hibernated, 35.1 ± 5.1 mmHg (n = 8); arousal, 41.3 ± 6.2 (n = 6).

Preconstricted preparations. There was no significant difference in the increase in perfusion pressure above baseline that was achieved by continuous perfusion with methoxamine (3-100 µM) in the three groups of preparations. The increase in tone above basal was: 49.4 ± 3.5 mmHg (n = 11) in age-matched controls, 48.1 ± 3.0 mmHg (n = 13) in cold-exposed controls and 55.0 ± 0.9 mmHg (n = 5) at 8 wk after hibernation. There was no significant difference among the groups with respect to the mean concentrations of methoxamine used to increase the tone of the preparations: 31.7 ± 9.7 µM (n = 11) in age-matched controls; 19.0 ± 7.0 µM (n = 13) in cold-exposed controls and 4.5 ± 0.5 µM (n = 5) at 8 wk of hibernation.

ACh and UTP. ACh and UTP, endothelium-dependent vasodilators, elicited dose-dependent vasodilatation of the hamster mesenteric arterial beds that was similar in all the groups (fig. 4). Mean pD₂ values and maximal relaxation to ACh and UTP are given in table 1.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Vasodilator responses of methoxamine-preconstricted golden hamster mesenteric arterial beds to uridine 5'-triphosphate (UTP; 0.05-500 nmol). Responses shown are those of mesenteric arterial beds isolated from: age-matched controls (, n = 11); cold-exposed controls (, n = 13); 8 wk hibernators (, n = 5). There was no significant difference in responses between the groups.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Our results show that there is an increase in sensitivity of sympathetic neurotransmission in the golden hamster mesenteric arterial bed at 8 wk of hibernation, and that by 2 hr of arousal this has reversed to the normal state.

The increase in sympathetic neurotransmission is quantitatively similar to that reported by us previously for golden hamster mesenteric arteries hibernated for 4 wk (Ralevic et al., 1997), suggesting that augmentation of sympathetic neurotransmission has reached a maximum by 4 wk and that this does not change with a longer period of hibernation. This is perhaps not surprising given that hibernation of all mammals is intermittent, being interspersed by regular bouts of wakefulness (Lyman, 1965; Nedergaard and Cannon, 1990). The ready reversibility of the changes in the mesenteric vasculature at only 2 hr after arousal from hibernation is a novel finding that is consistent with the concept that there is a defined increase in sympathetic neurotransmission, which, although it may vary within bouts of hibernation, does not change with the overall duration of hibernation. These findings illustrate the remarkable plasticity of the sympathetic nervous system.

Previous reports (Eliassen and Helle, 1975; Miller et al., 1986) have shown an increased sensitivity to NE in vessels from hibernating woodchuck and hedgehog. Our earlier study (Ralevic et al., 1997) showed a small, but statistically significant increase in sensitivity to NE in the mesenteric arterial bed of the hamster, after 4 wk of hibernation. In contrast, in our study, despite a clear increase in sensitivity to sympathetic nerve stimulation, responses to exogenously applied NE at 8 wk of hibernation were not significantly different from those of either of the control groups. It is not clear whether different mechanisms are involved in the increased sensitivity to NE between species and under different experimental conditions. However, it is important to note that the results of all of these studies are consistent with an increased vascular response to sympathetic nerve activation in hibernation.

The increase in sympathetic neurotransmission at 8 wk of hibernation may involve an increase in the amount of transmitter, an increase involving altered mechanisms of transmitter release or a decrease in mechanisms of transmitter degradation or uptake. The precise mechanism needs further investigation. In rat mesenteric arteries, vasodilator transmitters released from sensory nerves may negatively modulate responses mediated by sympathetic nerves (Li and Duckles, 1992; Ralevic et al., 1995). Thus, responses of these vessels to electrical field stimulation are the resultant of opposing sympathetic constrictor and sensory-motor vasodilator actions. However, unlike mesenteric arteries of the rat, perivascular nerves of golden hamster mesenteric arteries showing positive immunoreactivity for the classical sensory neurotransmitter calcitonin gene-related peptide do not appear to have an afferent vasodilater function (Hill et al., 1996).

During arousal from hibernation, an increase in the maximal response but not the sensitivity of responses to exogenously applied NE and an increase in submaximal responses to ,-meATP (a metabolically stable analogue of the sympathetic cotransmitter ATP), but not in those to KCl, histamine and UTP, implies a specific increase in the number of postjunctional receptors for sympathetic neurotransmitters, or an increase in the efficiency of receptor coupling. It is possible that as part of the dramatic changes that take place during arousal, postjunctional receptors for sympathetic neurotransmitters are up-regulated in consequence of and to compensate for the decrease in sympathetic neurotransmission relative to that in hibernation. It would be of interest to examine mesenteric arterial responses to NE at more than 2 hr after arousal to see if the augmentation of maximal vasoconstriction is indeed due to overcompensation by receptor up-regulation. Whether this is physiologically significant in the process of recovery from hibernation remains to be determined.

Endothelium-dependent vasodilatation to ACh and UTP (Ralevic and Burnstock, 1996) was unchanged at the conclusion of 8 wk of hibernation, as reported previously for golden hamster mesenteric arteries at 4 wk of hibernation (Ralevic et al., 1997). It is interesting to note, however, that a decrease in the percentage of endothelial cells showing positive immunoreactivity for nitric oxide synthase in renal arteries of 8 wk hibernated golden hamster has been reported, and this reverts to normal with arousal (Saitongdee et al., 1998).

The increase in sympathetic neurotransmission of hamster mesenteric arteries in hibernation (Ralevic et al., 1997; our study) allows speculation that this has an important role in the increase in peripheral vascular resistance known to occur during hibernation. How the changes in mesenteric arterial function in hibernation relate to the patterns of feeding and abstinence from feeding in hibernation is unclear. The ready reversibility of the increased sympathetic neurotransmission described in our study is consistent with a requirement for rapid normalization of vascular tone as the animal wakes from hibernation and resumes activity. Regional differences in the hemodynamic changes that take place in the hamster as it starts to warm up after hibernation have been described (Lyman, 1965), and it would be of interest to examine mesenteric arterial function within minutes of the arousal process.

The increase in mesenteric arterial sympathetic neurotransmission is consistent with the increase in peripheral vascular resistance known to occur in hibernation (Lyman, 1965; Willis, 1979; Nedergaard and Cannon, 1990), although the functional relevance of this requires consideration of central control of the autonomic nervous system and regional differences in sympathetic discharge. For instance, although there is a propensity for an increase in peripheral sympathetic neurotransmission, there also may be a withdrawal of sympathetic influence on the heart, which has been suggested to contribute to slowing of the heart rate in hibernation, in addition to an increase in parasympathetic activity (Lyman, 1965).

In our study, long-term exposure to cold had no significant effect on mesenteric arterial responsiveness to vasoconstrictors or vasodilators, consistent with our previous findings (Ralevic et al., 1997). Cold acclimation similarly had no effect on sensitivity to NE in the aorta, renal and femoral arteries in woodchucks (Miller et al., 1986). In contrast, decreased responsiveness to NE was observed in carotid and aortic tissues from cold-acclimated nonhibernators, namely rats and rabbits (Fregly et al., 1977; Flaim and Hsieh, 1978).

We have previously shown that changes in mesenteric arterial responsiveness due to hibernation are detectable at 37°C (Ralevic et al., 1997). As vascular responsiveness may be influenced by temperature, this was an important consideration in the design of our study. In view of the fact that the present investigation is a continuation of our earlier study, it was considered important to maintain consistency with respect to the experimental temperatures. This temperature is also appropriate for the arousal from hibernation group, as the body temperature of animals in this group had returned to similar values as those of the age-matched controls. However, our novel findings concerning the rapid reversibility of changes in hibernation raises the question of whether these also occur in vitro in the organ bath, and if so whether the changes in hibernation that we are able to detect in our assay are less than those that occur in vivo. Thus, an interesting and complementary issue for further study would be to investigate mesenteric arterial responses at hibernating body temperature.

In conclusion, our results show an increase in the sensitivity of sympathetic neurotransmission in mesenteric arteries of the golden hamster at 8 wk of hibernation, which rapidly reverts to normal during arousal from hibernation. Vasodilator function of the mesenteric arterial vascular endothelium does not change during 8 wk of hibernation. These results are consistent with a role for sympathetic perivascular nerves in inducing and maintaining an increased vascular resistance in hibernation.