Adaptive mechanisms of intracellular calcium homeostasis in mammalian hibernators

Ca²⁺ is a universal intracellular messenger that participates in numerous biological processes from neural regulation to muscle contraction,and from gene expression to cell growth and death (for a review, see Berridge et al., 2000). This incredible versatility necessitates that intracellular Ca²⁺concentration ([Ca²⁺]_i) be tightly regulated, to ensure coordination among the multiple processes involving Ca²⁺ that underlie normal cell functioning. Impaired regulation of[Ca²⁺]_i often leads to cell dysfunctioning such as occur in life-threatening diseases. For example, a transient increase of[Ca²⁺]_i is required to initiate contraction of muscle cells; inability to elevate [Ca²⁺]_i to the required level following an excitation is a major pathogenic mechanism in heart failure(Gomez et al., 2001). In contrast, excessive elevation of resting [Ca²⁺]_i is deleterious to almost all cell types, and can be associated with either necrotic or apoptotic cell death (Trump and Berezesky, 1995). Abnormal handling of[Ca²⁺]_i in the heart may induce severe arrhythmias and ventricular fibrillation (Lakatta and Guarnieri, 1993). [Ca²⁺]_i is also a key issue in transplanted organ preservation(Kim and Southard, 1998). Although a tightly regulated and constant [Ca²⁺] is not necessarily optimal under all conditions, for example when increased energy use is not an advantage (see Bickler and Buck,1998), [Ca²⁺]_i regulation is always important for a healthy cell life. Therefore, a precise understanding of[Ca²⁺]_i regulation is tantamount not only to understanding the nature but also the development of novel therapies for treating disease.

Two different strategies for probing calcium homeostatic mechanisms are:(1) the discovery of mechanisms of [Ca²⁺] dysregulation and development of methods to prevent or reverse the defects; and (2)investigation of wild natural models that show an extraordinary capability of handling intracellular Ca²⁺.

Hibernating mammals are one such special natural model. A mammalian hibernator, like all other mammals, can maintain its body temperature(T_b) at approximately 37°C during most of its lifetime. But in winter, hibernators can actively regulate their T_b down to only a few °C, entering into a distinct state known as hibernation (for reviews, see Lyman et al., 1982; Wang, 1988). During hibernation, circulation and respiration are well maintained, although at much lower rates than normal. T_b can be periodically,temporarily restored during hibernation (the whole period of entry,maintenance and arousal from a period of hibernation is termed `hibernation bout'), indicating that neural regulation is still active, despite deep hypothermia. A complete arousal occurs either upon external stimulation or as`scheduled' by an internal clock, during which, in ground squirrels and hedgehogs, normal T_b can be restored within 30 min by internal heat production. In this hibernation—arousal cycle, hibernators have to survive a set of extreme conditions that are fatal to humans and other non-hibernating mammals, including sustained deep hypothermia, violent shifts in T_b, highly intensified sympathetic innervation (during arousal), high viscosity and hypocoagulation of blood, and oxidative stress. As an adaptation during evolution, hibernators exhibit distinct resistance to hypothermia, arrhythmias (Johansson,1996) and hypoxia (for a review, see Wang and Zhou, 1999a).

In this brief review, we summarize the major known aspects of the adaptive mechanisms of intracellular Ca²⁺ homeostasis in hibernating mammals, and discuss their general significance and possible applications.

The kinetics of Ca²⁺ cycling are reduced at low temperatures, so adaptive mechanisms to maintain intracellular Ca⁺ homeostasis are one of the keys to surviving hibernation. In resting cardiac myocytes from the rat, which does not hibernate, [Ca²⁺]_i increases from 140 nmol l^-1 at 30-35°C to 200-300 nmol l^-1 during cooling to 5-10°C (Liu et al.,1991b; Wang and Zhou,1999b; Fig. 1). This is accompanied by a reduced amplitude of Ca²⁺ transients during cell excitation (Wang et al.,2000). These changes result in increased resting tension and a reduction of contractility (Liu et al.,1990; Wang et al.,1997b). Low temperature also causes spontaneous calcium waves(Wang et al., 1999) that are arrhythmogenic (Lakatta,1992), which may be one reason why ventricular fibrillation is often encountered during hypothermia (Chao,1959).

In contrast to [Ca²⁺] dysregulation in non-hibernating animals and humans during hypothermia, [Ca²⁺] regulation in hibernator cells is strikingly resistant to temperature change. At 30-10°C, resting[Ca²⁺]_i in heart cells from the ground squirrel(Spermophilus dauricus, a well-characterized hibernating rodent)changes very little (range 125±10 nmol Ca⁺ l^-1)(Wang et al., 1999; Fig. 1). The dynamic amplitude of Ca²⁺ transients following excitation is actually increased during cooling (Wang et al.,2000), which may help to retain forceful contraction despite the decreased Ca²⁺ sensitivity of myofilaments at low temperatures(Khromov et al., 1990; Liu et al., 1993). As a result, cardiac muscle from the ground squirrel and the hedgehog (another hibernating mammal) exhibited even higher contraction amplitudes at low temperatures than at normal temperatures(Liu et al., 1990; Wang et al., 1997b), which is an adaptive mechanism to ensure sufficient pumping pressure despite the consequent increased blood viscosity and peripheral resistance.

The adaptive capability to maintain intracellular Ca²⁺homeostasis in hibernator cells is associated with stable cell function despite some pathological or stressful stimulii(Johansson, 1996). It was observed that, in hedgehog hearts, the epicardial application of aconitine,administration of high concentrations of CaCl₂, injection of procaine after previous adrenaline treatment, or ligation of the left descending coronary artery, each failed to induce the ventricular fibrillation that usually occurs in guinea pig hearts in response to these perturbations(Johansson, 1996). During experimental ischemia—reperfusion paradigms, ground squirrel heart showed significantly less injury, monitored by creatine kinase leakage, than rat heart, suggesting that hibernator cells are resistant to the oxygen paradox and calcium paradox (Gao et al.,1996). Brain cells of hibernating mammals are also protected against a variety of insults that are detrimental to humans and other nonhibernating species (for a review, see Drew et al., 2001), but the relationship of the neural protection to [Ca²⁺]_iregulation still needs further study.

Intracellular Ca²⁺ homeostasis requires a dynamic balance between Ca²⁺ entry into and exclusion from the cell, and between Ca²⁺ release from and re-uptake into organelles(Fig. 2). Although Ca²⁺ entry through L-type channels shows similar properties in guinea pigs (a nonhibernator) and ground squirrel (S. richardsonii)cardiac myocytes (Herve et al.,1992), there is ample evidence that intracellular Ca²⁺cycling is different during hibernation. Action potentials of cardiac cells from hibernating chipmunks, ground squirrels and hedgehogs are characterized by the absence of a plateau at 0 mV (Fig. 3, arrow); both action potential and contraction become less sensitive to L-type Ca²⁺ channel antagonists such as nifedipine and Cd²⁺, compared with the sensitivity in non-hibernating or aroused individuals (Kondo and Shibata,1984; Kondo, 1986; Wang et al., 1995). These facts suggested that Ca²⁺ influx during excitation is reduced during hibernation. Direct measurement of whole-cell current in hibernator heart cells confirmed the downregulation of L-type Ca²⁺ currents(Alekseev et al., 1996) and suggested that the suppression is due to a reduction in cAMP-independent phosphorylation of the L-type Ca²⁺ channels(Kokoz et al., 2000). The suppression of voltage-dependent Ca²⁺ entry is also observed in other hibernator tissues. The binding site density in ileal longitudinal smooth muscle from ground squirrels (S. richardsonii) was approximately one order of magnitude less than that from guinea pigs(Wolowyk et al., 1990). In the neural system, for example, both resting [Ca²⁺]_i and depolarization-induced accumulation of Ca²⁺ in isolated synaptosomes are significantly lower in hibernating ground squirrels (S. tridecemlineatus) than in cold-adapted, non-hibernating animals, owing to decreased Q-type Ca²⁺ channel activity(Gentile et al., 1996).

The downregulation of sarcolemmal Ca²⁺ channels may help to prevent excessive Ca²⁺ entry into cells during hypothermia. At low temperatures, ion transport becomes slow, and the cell tends to become depolarized owing to the loss of ionic gradients(Wang et al., 1997a). If depolarized to approximately -50 mV, a tonic, non-inactivating `window'current of voltagegated Ca²⁺ channels usually becomes activated,and may further depolarize the cell, leading to more influx of Ca²⁺. As a result, cells become arrhythmic and calcium-overloaded. Downregulation of voltage-gated Ca²⁺ channels during hibernation decreases the chance of activation of the window current. Moreover, in hibernating ground squirrels (S. undulatus) the activation threshold of L-type Ca²⁺ channels shifts towards more positive potentials(Alekseev et al., 1996); the cells in hibernators can also better maintain ionic gradients (for reviews,see Willis, 1979; Wang, 1988) and thereby maintain their membrane potential (Liu et al., 1991a; Wang et al.,1997a) independently of temperature. These adaptive mechanisms effectively prevent the excessive Ca²⁺ entry and intracellular Ca²⁺ overload that would otherwise occur during hypothermia.

Contraction of myocytes is usually initiated by their excitation. In heart cells, excitation—contraction (E—C) coupling is governed by a mechanism known as Ca²⁺-induced Ca²⁺ release(Fabiato and Fabiato, 1975), by which Ca²⁺ influx through individual L-type Ca²⁺channels triggers the Ca²⁺ release channels/ryanodine receptors(RyRs) in the sarcoplasmic reticulum (SR) to release Ca²⁺(Wang et al., 2001) and initiates cell contraction. Downregulation of L-type Ca²⁺ channels in hibernating mammals decreases the magnitude of the trigger signal for E—C coupling. Normally this would result in a reduced cardiac contractility. However, neither the calcium transient nor the contraction amplitude of cardiac muscle from hibernating ground squirrels is reduced;instead, they are markedly stronger than those from non-hibernating animals(South and Jacobs, 1973; Wang, 1988; Fig. 3).

The enhanced contractility may be due either to increased Ca²⁺transients or to an increase in myofiliment sensitivity to [Ca²⁺]. It was found that the myofiliment sensitivity to [Ca²⁺] decreases as temperatures are lowered, in both hibernating and non-hibernating mammals(Khromov et al., 1990; Liu et al., 1993). Although myofiliments from hibernating ground squirrels (S. richardsonii)exhibit a somewhat higher Ca²⁺ sensitivity at low temperature than squirrels in the non-hibernating state, this still cannot fully explain the observed enhanced contractility over a wide temperature range.

The size of the Ca²⁺ transients has not yet been compared in animals in the hibernating and non-hibernating states, but all pharmacological evidence to date supports the idea that SR Ca²⁺ release is increased during hibernation. Blocking SR Ca²⁺ release by ryanodine or caffeine caused greater inhibition of myocardial contraction in hibernating chipmunks (Kondo and Shibata,1984) and ground squirrels (S. richardsonii)(Zhou et al., 1991) than when they were non-hibernating. This implies during hibernation, cardiac E—C coupling is remodeled so that a lower Ca²⁺ influx triggers a greater Ca²⁺ release response.

Various pumps and exchangers are involved in maintaining the low[Ca²⁺]_i in cells. Plasma membrane Ca²⁺-ATPase and Na⁺—Ca²⁺ exchangers extrude Ca²⁺from the cell, whereas sarcoplasmic/endoplasmic reticulum Ca²⁺-ATPase (SERCA) and a mitochondrial uniporter return Ca²⁺ to intracellular organelles(Berridge et al., 2000), their relative contributions varying among cell types. In cardiac cells,Ca²⁺ uptake by SR is the dominant mechanism for rapid dissipation of intracellular Ca²⁺, and accounts for 70-92% of total Ca²⁺ removed during each excitation—contraction cycle(Bers, 2000).

A reduction in temperature leads to a reduction in the rate of Ca²⁺ removal from the cytosol. Although the relaxation velocity of myocardial contraction decreases monotonically as the temperature is lowered in both rats and ground squirrels (S. dauricus), ground squirrel myocardium shows a higher relaxation velocity at any temperature between 35°C and 10°C (Wang et al.,1997b). [Ca²⁺]_i measurements indicated that the Ca²⁺ transient decays faster in ground squirrel cells than in rat cells owing to a faster Ca²⁺ uptake rate by the SR(Wang et al., 2000). This implies that the quantity and/or quality of SR and SERCA2 may be adaptively modified in hibernators.

Indeed, ultrastructural analysis has revealed that the proportional volume of SR in myocardium from hibernating ground squirrel(Rosenquist, 1970; Tang et al., 1995) and hamster(hibernator) (Skepper and Navaratnam,1995) is double or triple that of individuals that are non-hibernating, and is mainly due to an increase in longitudinal SR, which contains abundant Ca²⁺-ATPase and is responsible for Ca²⁺ uptake. By contrast, the content of junctional SR, where Ca²⁺-release channels are located, changes little(Tang et al., 1995; see Table 1; Skepper and Navaratnam,1995).

The increase in quantity of SR during hibernation is paralleled by an increase in the Ca²⁺ uptake capacity of the SR. SR vesicles isolated from winter-hibernating ground squirrels exhibit a faster rate of Ca²⁺ uptake and a greater level of Ca²⁺ accumulation than those from non-hibernating individuals either in winter or in other seasons (Belke et al., 1991; Tang et al., 1995; Fig. 4). Even in non-hibernating ground squirrels, the Ca²⁺ uptake rate by the SR is still higher than those in rats at temperatures between 35°C and 5°C(Liu et al., 1997).

SERCA is the central protein involved in active SR/ER Ca²⁺uptake against an electrochemical gradient. Since Ca²⁺ regulation in the SR during hibernation is enhanced, it is surprising that the enzymatic activity of SERCA is unchanged (Belke et al., 1991). Although SERCA from ground squirrels (S. richardsonii) is less temperature-sensitive than that from rats(Liu et al., 1997), the major mechanism for enhanced SR Ca²⁺ uptake is reliant upon the increased volume of SR present during hibernation.

Calsequestrin is a Ca²⁺-binding protein in the SR that binds Ca²⁺ at a ratio of 40-50 Ca²⁺ per molecule, and greatly increases the Ca²⁺ storage capacity of the SR, facilitating further Ca²⁺ uptake by decreasing the free Ca²⁺ concentration in the SR lumina. Calsequestrin also directly regulates the leakage and release of Ca²⁺via RyRs by the structural link between them(Sitsapesan and Williams,1997). In an early electron microscopic study, Rosenquist(1970) noticed that the terminal cisternae of myocardial SR in hibernating ground squirrels exhibits a higher electron density than those in non-hibernating animals, which suggests an increased expression of calsequestrin during hibernation. A novel isoform of calsequestrin has been identified in isolated cardiac SR from two species of ground squirrels, with a molecular mass about 7% greater than that of cardiac calsequestrin isolated from other mammals(Milner et al., 1991). The increased molecular mass is partially due to its distinct glycosylation, which appears to include an additional carbohydrate chain that is not present in other isoforms. This molecular modification improves the binding of Ca²⁺ to calsequestrin (Milner et al., 1991) and would thus be helpful in facilitating Ca²⁺ uptake, suppressing Ca²⁺ leakage and increasing the amount of Ca²⁺ available for release; therefore, this novel isoform of calsequestrin enhances Ca²⁺-induced Ca²⁺ release in cardiac cells during hibernation.

The Na⁺—Ca²⁺ exchanger (NCX), an important cation antiporter in the surface membrane (and some organelles) in most cells,transports three Na⁺ in exchange for one Ca²⁺. The transport is not coupled to ATP hydrolysis directly; instead, it is driven by the total Na⁺ electrochemical gradients across the cell membrane,and the Na⁺ gradient is maintained by Na⁺/K⁺-ATPase. Depending on the transmembrane Na⁺ and Ca²⁺ concentrations and membrane potential, NCX can operate in either an inward or an outward mode. In cardiac cells, NCX contributes to Ca²⁺ entry only during the early stages of cell excitation, when the membrane potential exceeds the reverse potential for NCX. In response to Ca²⁺ release from SR, NCX becomes the major driving force in extruding Ca²⁺ from cells(Bers, 2000).

Because the NCX does not consume ATP directly, it was hypothesized to be a preferred Ca²⁺ transporting system during hibernation, when ATP production is limited owing to the low temperature. The major evidence for this idea is the observation that the cardiac action potential in hibernating chipmunks (Kondo, 1987) and ground squirrels (Wang et al.,1995) is characterized by a prolonged low-level plateau that is sensitive to extracellular Ca²⁺ and Na⁺ concentrations. However, this phenomenon may also be attributed to the enhanced amplitude or prolonged duration of Ca²⁺ transient at low temperatures(Wang et al., 2000), which drive NCX more quickly, but passively. Direct analysis of Ca²⁺removal mechanisms in both ground squirrels (S. dauricus) and rats has failed to establish a dominant role of NCX in removing Ca²⁺. On the contrary, the fractional contribution of NCX to total Ca²⁺removal is significantly less in ground squirrels than in rats(Wang et al., 2000). Moreover,NCX contributes even less at lower temperatures in both species, consistent with the observation that the temperature coefficient for NCX is even greater than that for SERCA (Marengo et al.,1997). In addition, NCX is not efficient in terms of energy used. By contrast to SERCA, which transports two Ca²⁺ per ATP molecule consumed, NCX extrudes only one Ca²⁺ per ATP molecule (the three Na⁺ that enter during the exchange are then pumped out by Na⁺/K⁺-ATPase/Na⁺-pump at the expense of one ATP molecule) (Bers, 2000). Therefore, the role of NCX in Ca²⁺ removal is decreased in hibernator cells, possibly as an adaptive regulation; consequently,SERCA-based Ca²⁺ uptake becomes more important during hibernation.

Although the role of NCX is secondary in rapid Ca²⁺ removal from the cell, it is still important because it brings Ca²⁺ regulation within a global ionic homeostasis. Owing to the existence of NCX,Ca²⁺ homeostasis is tightly linked to the homeostasis of Na⁺ and K⁺ and to the ATP supply. An enhanced ability of hibernating animals to produce ATP and to maintain Na⁺ and K⁺ gradients by Na⁺/K⁺-ATPase (for reviews,see Willis, 1979; Lymann et al., 1982; Wang, 1988; Willis et al., 1992) is thus an essential part of intracellular Ca²⁺ homeostasis.

We have summarized the major aspects of the adaptive Ca²⁺regulation in hibernating mammals. Most of these studies described focused on the cardiovascular system. While Ca²⁺ homeostasis is a key issue of cardiovascular functions, Ca²⁺ regulation in other types of cells,such as neurons and endocrine cells, is also important for mammals to survive hibernation and therefore merits more study.

Based on current knowledge, the strategy used by hibernators to maintain intracellular Ca²⁺ homeostasis is to reduce Ca²⁺ entry into the cell while enhancing Ca²⁺ removal. If the SR function in ground squirrel cardiomyocytes is partially inhibited by caffeine, its resistance to hypothermia is lost, and the cells exhibit the same manifestations of Ca²⁺ overload as those observed in nonhibernator cells. Conversely, in the rat, which does not hibernate, partial blockade of Ca²⁺ entry through L-type Ca²⁺ channels prevents many effects of hypothermic Ca²⁺ overload in heart muscle(Wang et al., 1997b). These observations suggest that the Ca²⁺ homeostasis occurring under hypothermia or other similar stressful conditions is not `patented' by hibernators. The underlying regulatory mechanisms may rather also be employed by non-hibernator and human cells to improve stress resistance. In this way,extending our studies on the mechanisms of hibernation may provide strategies for developing new therapies or designing new drugs, and thereby contribute to human health.