ABSTRACT

Thyroid hormone (TH) plays a critical role in cells and whole body physiology, regulating the expression of developmental programs as well as specific cell functions. In homeothermic species, TH acquires another function, which is to regulate thermogenesis, playing a critical role in temperature homeostasis and thereby influencing the rate of metabolism and energy expenditure. I here review some basic concepts and progress made recently on the regulation of thermogenesis by TH. TH augments obligatory thermogenesis (heat produced as result of processes inherent to life) in homeothermic species basically by increasing ATP consumption, hence forcing the cell to produce more, with the attendant heat liberation, as well as by reducing the efficiency of ATP synthesis for the sake of producing heat. TH utilizes several mechanisms, such as increasing the calcium exchange between cytosol and sarcoplasmic reticulum and stimulating glycerol-3-phosphate shuttle in skeletal muscle as well as lowering the proton motive force in mitochondria. It is so far questionable whether UCP3 could mediate this latter effect. In addition, TH is essential for facultative thermogenesis (additional heat produced on demand in cold environments) for which it interacts synergistically with the SNS. The effects on facultative and obligatory thermogenesis are coordinated to ensure temperature homeostasis and avoiding hyperthermia in hyperthyroid states.

INTRODUCTION

The thyroid gland is present in all vertebrates, and its hormones, thyroxine (T₄) and triiodothyronine (T₃), collectively called here thyroid hormone (TH), play an essential role controlling the developmental programs and specific functions in various organs of the body. Only in homeothermic species TH seems to control thermogenesis [reviewed in (1)].

Whether one looks at whole body or separate organs, homeotherms have a faster metabolism than poikilothermic species (2;3). This means that the energy investment to sustain life is higher in homeotherms in order to produce more heat, that is, the homeothermic machine is thermodynamically less efficient for the sake of temperature homeostasis. Homeotherms indeed need more ATP to sustain life and produce ATP with more dissipation of energy as heat. Skeletal muscle, where it is easier to measure “work”, is thermodynamically less efficient in homeothermic species (4). It appears that TH is responsible for a large fraction of the difference between poikilothermic and homeothermic species. Physiologically, that is in the transition from the hypothyroid to the euthyroid status, TH recapitulates the differences between the cold-blooded and warm-blooded species. Thus, for any amount of mechanical work, TH increases heat production by muscle (5), as it also increases the energy cost of gluconeogenesis and ureogenesis in hepatocytes [(6) and references therein]. This article reviews the possible mechanisms that may mediate the thermogenic effect of TH.

TH-INDUCED INCREASE IN ATP TURNOVER

Cells attempt to keep a critical level of ATP as an immediate source of energy to support activities inherent to life. Mitochondrial respiration is tightly regulated by the amount of ATP available. Increases in the ADP/ATP ratio as a consequence of increased ATP utilization cause proportional increments of mitochondrial respiration. As in any process of energy capture from exergonic reactions, ATP synthesis captures only a fraction of the free energy of substrates, the difference being dissipated as heat. Thus, changes in ATP turnover are associated with proportional changes in heat production. The stimulation of numerous processes requiring energy i.e. ATP, is one of the mechanisms whereby TH increases heat production.

The maintenance of sodium and potassium gradients across cell membranes by Na/K ATPase consumes a substantial fraction of cell ATP. Indeed, the requirement of ATP needed to maintain these gradients explains a significant fraction of the difference in energy expenditure (oxygen consumption, QO₂) between poikilothermic and homeothermic species (2). Initially, there was excitement about this enzyme being a major thermogenic mechanism for thyroid hormone (7). Even though this enzyme activity could explain about 20% of resting metabolic rate in mammals, about 2/3 of this is accounted for by brain and kidney, and in these tissues the enzyme is not responsive to TH (8), so that the fraction that is TH-dependent is small. It is possible that the energy (ATP) required to maintain these gradients is more important in the hyperthyroid state, but not physiologically because the difference in Na/K ATPase activity between the hypothyroid and the euthyroid state is small (9).

Another postulated mechanism of increased ATP consumption is the stimulation of Ca2+ transfer from cytosol to sarcoplasmic reticulum (8). By stimulating the activity of the sarcoplasmic-endoplasmic reticulum Ca2+-dependent ATPase (SERCA) in skeletal muscle [reviewed in (10)], TH increases the sarcoplasmic calcium pool, the release of calcium during contraction and ultimately ATP demands to return the calcium back into the sarcoplasmic reticulum. TH administration to hypothyroid rats increases the SERCA activity and the amount of sarcoplasmic reticulum, which is associated with a substantial increase in the amount of energy spent in Ca2+ pumping (11). Slow twitch muscle is more responsive to T₃ than fast muscle, largely owing to the predominant stimulatory effect TH on SERCA1 gene expression, the baseline expression of which is low in this type of muscle (10;12;13). In addition to increasing calcium fluxes during activity, it is possible that TH enhances the “leak” of calcium at rest by stimulating the number and activity of ryanodine receptors (14;15), thus enhancing ATP expenditure also in the resting condition. Interestingly, for any level of mechanical work (tension) in both fast-twitch and slow-twitch muscle heat production is significantly greater in the euthyroid than in the hypothyroid state (5). Moreover, extrapolating the heat vs. tension curves to zero tension i.e. rest, the difference between the two thyroid states is maintained (5), providing independent support to the idea that resting muscle heat production is also greater in euthyroidism than in hypothyroidism. Since skeletal muscle normally contributes about 20-30% of resting metabolic rate [(16) and references therein], Ca2+ transport in muscle could explain as much as 10% of the effect of TH on resting thermogenesis. Lastly, it is important to recall that slow-twitch muscles are relatively more abundant in larger mammals, such as humans, who in addition have less brown fat, so that these mechanisms are quantitatively more important in larger species. The presence of type II iodothyronine 5’deiodinase (D2) in human but not in rodent skeletal muscle (17) may also be an indication that muscle TH-dependent thermogenesis is more important in humans.

Yet another mechanism whereby TH can stimulate the expenditure of ATP is by accelerating metabolic cycles, such as lipolysis-lipogenesis, glycolysis-gluconeogenesis. Increased consumption of substrates is coupled to synthesis, and TH stimulates the expression of key enzyme genes such as gluconeogenic and lipogenic enzymes. However, in animals fed ad libitum, synthetic process do not seem to account for a major fraction of ATP consumption. Lipogenesis can account for no more than 10% of difference in energy expenditure between hypo- and hyperthyroid rats (18). There are possibly multiple other mechanisms whereby thyroid hormone may increase ATP consumption, such as increased non-exercise physical activity, increased heart activity, etc. Although there are no rigorous measurements of ATP turnover in whole body in different thyroid states, the TH-induced increase in ATP demands probably does not account for more than 50% of differences in thermogenesis between the euthyroid and the hypothyroid state.

TH-INDUCED RESPIRATION UNCOUPLED OF ATP SYNTHESIS

Brown adipose tissue (BAT) was until recently the only site where there was controlled reduction in coupling between oxidation and ATP synthesis, which is mediated by a 32,000 Mr protein, thermogenin or uncoupling protein (UCP), now called uncoupling protein-1 (UCP1), in view of the cloning of several new homologues [see (19;20) for recent reviews]. The idea that TH could uncouple phosphorylation received attention in the 1950’s but was then abandoned in the late 1960’s following the demonstration that the earliest effect of TH was to increase nuclear RNA synthesis (21). In recent years, work by Brand’s and by Berry’s groups have provided evidence that a good fraction of the increase in QO2 following the administration of TH cannot be explained by increased ATP synthesis [see (6;22) and references therein]. Brand’s group demonstrated the presence of proton leak across the mitochondrial membrane that could account for 20-40% of basal liver respiration (23). This was later shown to occur in other tissues, notably in muscle and to be regulated by TH (22). Supporting the idea expressed above regarding the contribution of ATP turnover stimulation to TH thermogenesis, the analysis of hepatocytes indicates that the increase in QO2 in the transition from the hypothyroid to the euthyroid state is mostly due to increased proton leak and non-mitochondrial sources, with little being due to increased ATP synthesis, whereas this is as important as the proton leak in thyrotoxic hepatocytes (24).

The cloning of UCP1 homologues more ubiquitously expressed, most notably UCP2 and UCP3 raised the expectation that they would explain the proton leak and that they would be stimulated by TH, but so far the evidence is missing (20). UCP3 is clearly stimulated by T₃ (25), but transgenic UCP3-deficienct mice do not seem to have a thermogenic deficiency and their body temperature and QO₂ respond normally to T₃ (26). Such negative findings should be interpreted cautiously, though, because studies in transgenic models of gene ablation have revealed redundancy of mechanisms and recruitment of compensatory mechanisms that can conceal the effect of the ablation of a given gene.

Another mechanism whereby TH could reduce the efficiency of ATP production is by favoring the use of the glycerol-3-phosphate (G3P) shuttle to dump cytoplasm-generated reducing equivalents in the mitochondria, as this shuttle produces only 2 instead of 3 ATPs per pair of electrons or per molecule of generated water. The NADH-G3P shuttle constitutes a rapid way to generate ATP aerobically out of oxidative processes occurring in the cytoplasm (27), and the rate-limiting enzyme of the shuttle, the FAD-linked mitochondrial glycerol phosphate dehydrogenase (EC 1.1.99.5, mGPD), has been known for long to be stimulated by TH in several tissues (28) in proportion to the stimulation by TH of QO2 (29). Furthermore, TH does not stimulate mGPD activity either in homeothermic tissues where if does not stimulate QO2, such as the brain (28;29), or in poikilothermic species (30). We have carefully studied a transgenic mouse lacking mGPD and found the genotype associated with reduced obligatory thermogenesis partially compensated by increased BAT activity and serum TH levels (31). The tissue most affected by the lack of mGPD was skeletal muscle, as revealed by an increase in G3P and lactate/pyruvate ratio (32). While these studies show that mGPD is involved in TH-stimulated thermogenesis, they also indicate it is not essential because hypothyroid mGPD-/- mice increase normally their QO2 when given T3 over the physiological replacement (31). Interestingly, male mGPD-/- mice have an increase in UCP3 (31;32).

FACULTATIVE THERMOGENESIS AND TH

While the well-known cold intolerance of hypothyroid animals and humans may be explained by the reduced obligatory thermogenesis, studies in rats suggest that it is largely due to failing facultative thermogenesis caused by the lack of T₃ in BAT. Prompted by the presence of type II iodothyronine 5’deiodinase (D2) in BAT and its activation by the sympathetic nervous system (33) as well as by hypothyroxinemia (34), we treated hypothyroid rats acutely with replacement doses of T₄ and found that within 48 h they restored their cold tolerance and UCP1 response to cold, whereas such brief and small regimen did not affect the thyroid status of the liver nor restored plasma levels of T₃ (35). Such effect of T₄ was abolished by the blockade of D2 with iopanoic acid, even though T₃ was concomitantly given in these studies to maintain plasma T₃ levels (35). It is unclear, though, how relevant these findings are to humans. BAT is in a quiescent state in adult humans and it is unlikely to be an important site of facultative thermogenesis. It is likely that in humans and other large mammals skeletal muscle play an important role. Interestingly, humans but not rodents express D2 in skeletal muscle (17).

TH is essential for the full response of the UCP1 gene to adrenergic stimulation. A complex TH-response element (TRE) is found in a critical upstream enhancer of the gene. This TRE acts synergistically with cAMP to maximize the transcription of the gene [reviewed in (36)]. In addition, TH is needed for integrity of the norepinephrine-signaling pathway both in BAT [see (37) and references therein] and in other tissues. We recently reported that mice lacking the TH receptor α (TRα0/0) have an impaired thermal response of BAT to exogenous norepinephrine, even though UCP1 and other relevant genes responded well to cold, revealing the existence of yet another T₃-dependent step essential for the ultimate thermogenic response of BAT to adrenergic stimulation, in this case specifically requiring TRα for T₃ stimulation (38).

The multiple mechanisms whereby TH contributes to thermogenesis, temperature homeostasis and energy balance are schematically summarized in Figure 1.

Figure 1. Heat production as a function of ambient temperature and food intake. Homeothermic species produce more heat than poikilothermic species. This excess heat produced is largely TH-dependent and it is quite variable in humans, probably reflecting selections pressures. In addition, homeothermic species produce heat on demand, in a facultative or adaptive manner (facultative thermogenesis). TN represents the point of equilibrium between obligatory thermogenesis and ambient temperature or food intake. Below and over this point, adaptive responses are necessary. Facultative thermogenesis, one of these responses, is stimulated both by cold and excess food intake and the magnitude of its response is also TH-dependent.

CONCLUDING REMARKS

The stimulation of thermogenesis is a new role acquired by TH with the advent of homeothermy. The thermogenic mechanisms stimulated by TH antecede homeothermy but they evolved into being TH-dependent in warm-blooded species. Part of the stimulation of obligatory thermogenesis by TH results from TH increasing ATP utilization, forcing the cells to produce more ATP with the attendant production of heat. A significant fraction, possibly 50% or more, is due to stimulation of respiration without ATP production, that is, pure heat production, but the biochemical mechanisms mediating this effect have not been defined, but are likely related to a reduction in the proton motive force in the mitochondria in the form of a proton leak. Lastly TH is essential for facultative thermogenesis interacting with the sympathetic nervous system in a synergistic manner, well exemplified in rodents BAT, but less understood in humans and larger mammals.