Bioenergetic Aspects of Mitochondrial Actions of Thyroid Hormones
Functionally, the assembly of SCs has been proposed to have several biological significances, among which are: reduction of ROS generation [162,163]; optimization of the catalytic activity of the individual complexes [164]; boost of the efficiency of electron transfer through substrate channeling [145,149,165]; stabilization and assembly of Complex I [166,167]; prevention of aggregation of protein subunits of SCs themselves [168]. Some of these functions are still under debate (for recent review, see [169]).
Taking into account the pleiotropic effects which TH may evoke in the mitochondrial compartment, under both physiological and pathological states, affecting quali- and quantitatively, respiratory complexes, energy efficiency, and membrane lipid composition [170,171], it should not be surprising that a close interrelation between THs and SCs could exist (Table 1). Unfortunately, to date, very few studies have been devoted to this specific matter. During the last decade, some new insights into both the mitochondrial response mechanisms to iodothyronines and the resulting proteome alterations have been obtained by applying one-dimensional BN-PAGE to examine the abundance of mitochondrial OXPHOS complexes as well as their activity and supramolecular organization [90,128,172,173,174]. By this approach, the influence of the administration of either T2 or T3 to chemically induced hypothyroid rats on the organization of OXPHOS into SCs, in liver mitochondria, was studied [90]. The results indicated differential effects elicited by the two tested iodothyronines: T2, influences the kinetic properties of specific mitochondrial respiratory pathways, and promotes a rapid response of the organelle; T3, enhances the abundance of respiratory chain components and favors the organization of respiratory chain complexes in SCs, inducing a slower and prolonged response. Specifically, T2 enhanced both coupled and uncoupled respiration in succinate-energized liver mitochondria, but the effect on uncoupled respiration did not involve proton leak, as in the case of T3. Moreover, different from T3, T2 was not able to influence the abundance of respiratory complexes but it enhanced the activity of complex II-linked mitochondrial respiratory pathways and the “in gel activity” of complex II.
T2 also stimulated G3PDH activity, suggesting the existence of a selective effect of T2 on mitochondrial pathways producing FADH2, that could represent an iodothyroinine-specific mitochondrial thermogenic mechanism, considering that the amount of ATP generated by FADH oxidation is lower than that generated by NADH oxidation at complex I. T3, on the other hand, while sharing with T2 the ability to activate complex II- and G3PDH-linked respiratory pathways and increase complex V and G3PDH levels (being its effects more pronounced that those of T2), significantly activated proton-leak and increased ANT mitochondrial levels, both clear cut signs of its potent thermogenic action. Finally, only T3 favored the assembly of individual respiratory complexes in SCs, this likely being a mechanism to enhance the tissue’s oxidative capacity, sustain the increased metabolic rate, and limit or balance ROS generation under hyperthyroid conditions [90].
The link between form and function finds its full realization in mitochondrial architecture. Over the past decades it has universally been accepted the ideas that mitochondria are not static but very “dynamic” organelles, since they can change themselves at macro-and ultrastructural levels. They are able to respond cellular cues and energy demands by remodeling their morphology and shape, regulating directly their bioenergetic functions, allowing the cell to respond to its ever-changing physiological conditions [175,176,177,178]. Mitochondria can exist in networked phase or in small units [179,180], undergoing constant turnover (mitophagy-biosynthesis) and continuously fusing (fusion) and dividing (fission) [181,182,183,184,185].
Ultrastructurally, cristae represent the high specialized areas where OXPHOS occurs and assembles in respiratory complexes and SCs [186,187]. Cristae are membrane invaginations protruding into the mitochondrial matrix perpendicularly to the inner boundary membrane, i.e., the unfolded portion of the inner mitochondrial membrane. The cristae interior is called intracristal space and communicates with the outer intermembrane space through the cristae junction, which divides the cristae membrane from the inner boundary membrane (for review see, [188]). A major component of the cristae junction is the mitochondrial contact site and cristae organizing system (MICOS), composed of two distinct MICOS subcomplexes marked by the core components Mic60 and Mic10 (for review, see [189]).
Cristae shape and composition change under physiological stimuli, impacting profoundly on mitochondrial performance and survival [190]. From the first observation of the cristae dynamism [191], over the years, many more studies have confirmed inner mitochondrial membrane plasticity [192]. As recently shown, individual cristae differ not only in their composition [193] but also in their functionality [194]. Cristae fold increases under energy pressure, to augment oxidative phosphorylation areas [195], under the orchestration of several classical (i.e., TH) and newly discovered “remodeling” master regulators, namely: (i) Optic atrophy protein (OPA1), [196], (ii) MICOS [197], (iii) Complex V [198,199].
At the macroscopic levels, under high energy demands, the shift towards fusion favors the generation of interconnected mitochondria, for maintaining normal mitochondria health, by mixing healthy and damaged organelle as complementation [200], and for increasing respiratory capacity. Indeed, mitochondria “hyperfusion” is accompanied by increased mitochondrial ATP production, to optimize mitochondrial function [201]. In contrast, a shift towards fission produces small mitochondria, functionally distinct small spheres or short rods.
Recent evidences support the idea of a tight interplay between mitochondrial dynamics and autophagy as collaborating actors in the same quality control system. Mitochondrial division or loss of fusion membrane protein (i.e., mitofusin 2 (Mfn2) or OPA1), can drive the removal of damaged organelles from the mitochondrial network to allow a specific form of autophagy, referred as mitophagy [202]. Although mitophagy shares many of characteristic aspects with autophagy [203], the selectivity of mitochondrial cargo by the autophagosome is driven by a specific protein set and stimuli [202].
The modulation of cristae organization, mitochondrial dynamics and mitophagy al together deeply impact mitochondrial activity. These processes are among those tightly regulated by thyroidal state. Early studies by electron-microscopy showed that T4 treatment increases cristae in the number and length, by inducing a hypermetabolic state in skeletal muscle of both normal and thyroidectomized rats [204]. Recently, electron microscope analyses indicate that increased respiratory enzyme activity produced by TH in the liver is, in large part, the result of enhanced development of the cristae folding membrane [205]. Conversely, the depression of respiration found in thyroidectomized or hypothyroid animals is correlated to a decreased cristae area [206]. In the developing brain, T3 deprivation determines highly fractured and degenerated mitochondrial cristae with low membrane potential, due to the decreased membrane surface area rather than of enzymes of the respiratory chain, with compromised OXPHOS [207].
Until now, there are no mechanistic studies that correlate directly TH administration with fusion/fission processes. Very recently, it has been shown how T2 modulates mitochondrial dynamics in skeletal muscle of HFD-fed rats. In particular, T2 treatment inhibits diet-dependent mitochondrial shift toward fission, protecting muscle cells against mitochondrial dysfunction [208].
In regard to quality control system, the ability of TH to activate several forms of autophagy (from mitophagy to lipophagy) has been finely investigated [209,210]. To counteract ROS-inflicted cell damage and death, T3 induces mitophagy via Beclin1, PTEN Induced Kinase (Pink)/Parkin (Park), and Microtubule-associated protein 1A/1B-light chain 3 (LC3) pathways [211,212,213,214]. However, the coupling of mitophagy and OXPHOS [213,214] occurring during TH-regulated metabolism, can be enrolled not only to repair mitochondrial damage, but also to prevent it and maintain high energy demand. In liver of mice, T3 induces mitophagy via the ROS-AMP activated protein kinase (AMPK)-unc-51-like autophagy activating kinase 1 (ULK1) pathway [210] that primarily reduces ROS damage accumulation, promoting mitochondria renewal. Sinha and co-workers demonstrated, in vivo/in vitro, that mitophagy is necessary also for sustaining efficient oxidative phosphorylation induced by T3, during times of high-energy demand [210]. Moreover, both T3 and T2 have been shown to be able to induce in liver autophagy and increase the mobilization of lipids (lipophagy) from lipids droplets towards beta oxidation to provide substrates for OXPHOS, maintaining a hypermetabolic state [210,213]. In soleus of hyperthyroid mice, T3-mediated mitochondrial activity and biogenesis are also linked to autophagy activation via ROS-AMPK-ULK1 in a sort of positive loop that improves mitochondrial function [214]. A similar effect has been recently found also in brown adipose tissues and brown primary preadipocytes from hyperthyroid mice, where T3 has been reported to act on mitochondrial respiration, OXPHOS as well as autophagic flux, mitophagy, and mitochondrial biogenesis, the trigger of mitophagy likely being the mTOR inhibition [215].
All these data corroborate the universal recognition of the intimate connection between TH and mitochondrial dynamism to improve bioenergetic performance. However, more efforts have to be made to well-characterize the direct impact of THs on such specific mechanisms, since the literature is still poor.
Several thyromimetics, besides the natural occurring TH derivatives, have been designed to specifically avoid binding to TRα while maintaining their affinity to TRβ isoforms [16] and references therein, thus uncoupling beneficial actions on liver and central nervous system from deleterious effects on the heart, muscle and bone. Among these, GC-1 (also known as sobetirome), a halogen-free thyroid hormone agonist, was the first TRβ selective thyromimetic to be developed. It has been reported to increase mitochondrial O2 consumption and H2O2 production but to a lesser extent compared to T3 putatively due to the higher increases induced by T3 in NRF-1 and 2 levels [216]. Even though showing encouraging actions against hypercholesterolemia, non-alcoholic steatohepatitis and in the stimulation of hepatocytes proliferation, it was stopped after phase 1 clinical trials [217]. Another agent being studied is VK2809 (formerly known as MB07811) which exhibits increased TR activation in liver relative to other tissues ([16] and references therein) and is now considered one of the most promising lipid-lowering agent, currently under phase 2–3 clinical trials. VK2809 has been reported to be able to augment metabolic rate in the liver and, specifically, increase the rate of mitochondrial FA oxidation, of mitochondrial respiration as well as the activity of hepatic mitochondrial glycerol-3-phosphate dehydrogenase [218]. Furthemore, VK2809 treatment has been reported to decreased hepatic triglyceride levels in glycogen storage disease type Ia mice, through a simultaneous restoration of autophagy, mitochondrial biogenesis, and FA oxidation [219].
Mechanistically, most of the effects caused by these thyromimetics depend on their binding to TRβ receptors, however, at least for CG1, non-genomic effects have also been reported, the latter requiring its binding to the TH membrane receptor integrin alfavbeta3 [220].
In recent times, also functional rather than structural TH analogs, with lower affinities for TRs, were designed and synthesized. Among these, substituted pyrazoles such as TRC150094 (TRC), which has attracted particular attention emerging as a novel functional analog of iodothyronines linking fat consumption with the pathogenesis of hepatic steatosis [221]. In particular, in a preclinical animal study, TRC has been demonstrated to stimulate mitochondrial FA uptake and oxidation as well as OXPHOS activity [221]. For these reasons it is now considered a novel mitochondrial modulator, although its mechanism of action as well as its cellular receptors have not yet been identified. Recently its safety, tolerability, and pharmacokinetics after oral administration to overweight/obese adult and elderly subjects have been studied in a randomized, Phase-I clinical trial [222]. Moreover, a methyl-substituted dicarboxylic acid thyromimetic and its analogs have been reported to directly target mitochondrial PTP gating, thus showing a thyromimetic/calorigenic activity in vivo [76].
As stated in a previous paper: “It would be convenient, not to say desirable, to explain the various actions of TH at the cellular level by a unitary molecular model. However, while this is an attractive idea, the bulk of available evidences indicates that the situation is far more complicated” [223] (p. 119) and still a lot of work has to be performed to fully understand the complex mode of action of THs, in terms of involved mechanisms and cellular targets. It seems universally accepted that iodothyronines may affect mitochondrial activity by two pathways, the first way is ‘nuclear-mediated’ (genomic) and involves a direct interaction of TH with nuclear receptors, while the second way requires a direct interaction of the hormones with the cellular organelles or the molecular components. The first pathway, however, does not exclude the possibility that, although the nuclear pathway is not directly involved, the latter can be recruited through signals starting from the primary site of interaction of the TH. As previously described, almost all components of the respiratory chain are directly or indirectly affected by iodothyronines (see, Table 1). In some cases, the actions result in an activation of specific biochemical pathways while in others the effects would result into an increase in mRNA or protein levels of specific components of respiratory chain. Another obstacle for a clear and unambiguous explanation of the mechanisms activated by iodothyronines is that they can exert both short-term (in hours) and long-term (days) effects. So, given this, in animal studies the duration of the hormonal treatments is crucial to observe some effects. In addition, as often isolation of mitochondria is required, it is crucial to compare the methods of separation of the organelles.
In fact, in various studies, mitochondria are often isolated at different gravitational forces (ranging from 3000 g to 12,000 g). In these cases, each different preparation results in mitochodria endowed of biochemical and functional differences. This was evident in a study performed on rat liver mitochondria [224]. Finally, it is not always possible to compare results obtained in different animal models such as, for example, mice and rats, or human and rodents. In fact, it has been seen that effects of TH on some parameters are differently influenced and specie-specific, such as (for example) the activation of the UCP3 gene [225].
In conclusion, much progress has been made in recent years on the actions that iodothyronines exert on the respiratory chain, but much remains to be discovered. For example, what precisely allows mitochondria from animals in different thyroid states to show different morphological characteristics (signature, cristae density, etc.), which may be the basis of some functional characteristics of organelles, is crucial.
What is the goal that lies in the short- and long-term actions? In our opinion, a twofold objective could be achieved: (a) the activation of short-term mechanism would be useful for a rapid response to sudden physiological variations (such as, for example (in animals) a systemic and cellular metabolic adaptation following a sudden and momentary drop in the environmental temperatures); (b) the activation of the long-term one could be functional to a long-lasting cellular metabolic adaptation (for example during the winter or chronic metabolic/physiologic changes of the organism). In the last case the changes may be useful to: (i) produce organelles with higher density in respiratory components; (ii) produce organelles more or less efficient in the ATP synthesis (e.g., with higher or lower UCPs concentration); (iii) regulate the mitochondrial turnover. T2 could be more functional in the first case while T3 in the second.
Elucidation of the mitochondrial molecular mechanisms and downstream targets of THs would provide not only new basic knowledge in the field of thyroid endocrinology and metabolism but also new perspectives to develop therapeutic strategies for a number of high incidence public health issues, such as dysthyroidisms (hypo- and hyperthyroidism), ageing and associated comorbidities (e.g., type 2 diabetes, obesity). Indeed, altered mitochondrial bioenergetics due to pathological and even subclinical changes in circulating TH levels or tissue-specific local action have been associated to the pathogenesis of degenerative diseases such as sarcopenia, liver steatosis, cardiomyopathy, and neurodegeneration, in all of which a strict interconnection has been described between cellular ATP levels, oxidative stress, mitochondrial dynamics, autophagy and substrate metabolism.
Writing—original draft preparation, F.C. and A.G.; writing—review and editing, F.G. and E.S. All authors have read and agreed to the published version of the manuscript.
This research was funded by “Progetto di Ricerca di rilevante Interesse Nazionale (PRIN)” 2017, grant number 2017J92TM5_003, and by University of Sannio Research Grants (no grant number available).