Opposite Effect of Thyroid Hormones on Oxidative Stress and on Mitochondrial Respiration in COVID-19 Patients

Nonthyroidal illness syndrome (NTIS) is generally considered part of the adaptive host neuroendocrine and metabolic response to survive critical illness [1]. It can be diagnosed in up to 70% of critically ill patients [2,3,4,5] of all ages [6]. In the International Classification of Diseases, 11th revision, NTIS has a diagnostic code (5A06), under the old name of sick-euthyroid syndrome [7], and is classified among the disorders of the thyroid gland or of the thyroid hormones (TH)s systems. NTIS is observed in intensive care units (ICU) in association with many different conditions and diseases and has a negative prognostic impact on the course of the disease, with a relevant increased risk of death [8,9,10,11]. However, even if the prognostic relevance of reduced FT3 serum levels is clearly demonstrated, the treatment of NTIS with either synthetic liothyronine (LT3) or levothyroxine (LT4) is still debated [12]. Many of the randomized clinical trials performed so far failed to clearly demonstrate any beneficial effect of such treatments. Recently, we postulated that the lack of convincing evidence reported by many published interventional randomized clinical trials might rely on the choice of inadequate primary outcomes [13]. In the absence of any clear evidence regarding the beneficial effect of the treatment on THs in such patients, the current opinion is that such treatment should not be given unless patients show clear clinical signs of hypothyroidism. Even the assessment of thyroid function in seriously ill patients should not be performed except when there is a strong suspicion of thyroid dysfunction [14]. This is also the opinion of the experts in the thyroid field, as stated in the guidelines published by the American Thyroid Association [15].

NTIS has been detected in COVID-19 patients too and, as for other critical conditions, its occurrence in such patients is associated with a higher risk of a more severe disease [16] as well as of death [17,18]. We have recently demonstrated that in COVID-19 patients, the acute deficiency of T3, typically observed in NTIS, is responsible for the occurrence of hydroelectrolytic disequilibrium at the peripheral level, with the induction of an anasarcatic condition, similar to that observed in myxedema, due to long standing, untreated, overt hypothyroidism, which can be easily measured by bioelectrical impedance analysis (BIA) [17]. In addition, by means of the nanostring analysis, we demonstrated that reduced levels of FT3 were associated with the dysregulation of many genes coding for protein located in the mitochondria or involved in mitochondrial function [16].

Mitochondria play pivotal roles in cellular energy metabolism. Indeed, they supply most of the intracellular adenosine triphosphate (ATP), the organic compound that provides energy to drive many processes in every single living cell. For this reason, the analysis of mitochondrial dysfunctions performed in PBMCs has gained much attention. In fact, this cell population mirrors systemic changes within the body and, for this reason, provides a source of sensitive peripheral biomarkers in several diseases [19], including COVID-19. Mitochondria are thought to play a central role in the immune response to viruses and it has been suggested that the maintenance of mitochondrial integrity is essential, especially for an adequate innate immune system response to SARS-CoV-2 virus infection [20]. The inflammatory state induced by the SARS-CoV-2 virus infection, and known as “cytokine storm”, is associated with the elevated production of reactive oxygen species (ROS), responsible for the induction of oxidative stress. Both alterations point toward an altered mitochondrial activity in patients affected by COVID-19 [21]. Moreover, the debilitating condition consisting of the occurrence of redox imbalance, energy metabolic deficits, and a hypometabolic state, observed in the post-acute sequelae of severe acute respiratory syndrome coronavirus 2 infection (PASC), has been linked to mitochondrial damage [22,23] and to mitochondrial dysfunction of any cause [24]. Mitochondria represent one of the major ROS sources and the occurrence of altered mitochondrial dynamics is crucial in the development of the disease. For this reason, the analysis of these cellular organelles and of the substances that are directly involved in the regulation of oxidative stress is essential.

Thyroid hormones (TH)s are master regulators of cell metabolism [25]. They regulate a variety of pathways involved in metabolism and energy expenditure (i.e., burned calories), both in the resting metabolic rate and during physical activity [26]. Their role in stimulating cell energy has been known for long time [27]. It consists of a relative acceleration of the basal metabolism that includes an increase in the rate of both catabolic and anabolic reactions. As a consequence of such a stimulatory effect on cell metabolism, they produce an increase in ROS generation. Many of their effects are exerted on metabolically active organs, including the liver, brown adipose tissue (BAT), white adipose tissue (WAT), skeletal muscle, and the heart. In brown adipocytes, they enhance basal, maximal, ATP-linked, and proton-leak oxygen consumption rates (OCR)s. In these cells, they dissipate the generated energy via heat production by proton currents and by means of the uncoupling protein 1 (UCP-1) [28]. However, their action has also been studied in innate immune cells, and remarkable and crucial effects on neutrophils, macrophages, and dendritic cells have been reported [29]. It has been demonstrated that THs participate to the mechanism of inflammation and to oxidative stress [30]. Their role in the maintenance of an adequate level of energy in immune cells as well as the effects of acute deficiency of FT3 on mitochondrial respiration and on oxidative stress at the periphery in COVID-19 patients has not yet been examined. The aim of this study was to investigate the effects of acute deficiency of FT3 on the oxidative balance by measuring the derivatives of reactive oxygen metabolites (dROMs) and the biological antioxidant potential (BAP) in the serum and by analyzing the mitochondrial respiration in the PBMCs obtained from a cohort of COVID-19 patients with and without NTIS.

Oxidation is a process that occurs naturally in the body when oxygen combines with reduced molecules and provides energy. Any process that stimulates energy production is associated with an increase in oxidation and results in the generation of ROS. An efficient scavenger and antioxidant system is required to reduce all the excessive ROS generated. Oxidative stress is represented by the imbalance between the generation of reactive oxygen/nitrogen species and the ability to counteract them with an efficient antioxidant capacity. Oxidative stress and inflammation play key roles in the multisystem disorder caused by the SARS-CoV-2 infection, and they have been linked to many different pathologies that are predisposed to critical outcomes in COVID-19, including cardiovascular disease and diabetes mellitus type 2 [32,33]. In addition, the uncontrolled oxidative stress produced by the SARS-CoV-2 infection is associated with a pro-inflammatory state and cytokine production, which may contribute to the development of a respiratory syndrome, and in the worst cases, may lead to death. Considering the pathogenic role of oxidative stress, the use of antioxidation therapy has been recently proposed in a clinical trial to prevent organ and tissue damage triggered by the cytokine storm [34].

The clinical severity of COVID-19 correlates with a dysfunction of the immune system. Two major hallmarks of the disease are represented by the occurrence of lymphopenia together with the induction of a peculiar systemic hyper-inflammation condition, known as a “cytokine storm” or macrophage activation syndrome (MAS) [35,36]. Mitochondrial dysfunction represents a key factor in the severity of the COVID-19 disease. It contributes to all known COVID-19-severity risk factors, such as aging [37,38] and various age-related metabolic diseases, including obesity [39], metabolic syndrome [40,41], diabetes [42], hypertension [43], and coronary heart disease [44]. It also plays a relevant role in other risk factors, such as lung diseases, cancers and neurodegenerative diseases [45,46]. In agreement with these observations, an altered bioenergetic and mitochondrial dysfunction has been recently described in immunological circulating blood cells obtained from patients with COVID-19 pneumonia. Mitochondrial function and associated metabolic changes were recently analyzed in live cells from patients with COVID-19. In a study, based on a cohort of seven COVID-19 patients, it was demonstrated that their PBMCs exhibited reduced maximal respiration and reserve capacity, indicating compromised mitochondrial respiration or mitochondrial dysfunction, as compared to nine healthy controls and seven patients with other chest infections [47]. In this study, there was no mention regarding the thyroid function and the possible occurrence of NTIS in these patients. In another cohort of five COVID-19 patients and seven matching healthy controls, the same analysis was performed on CD4+ T cells. This study demonstrated an altered cell proliferation but not mitochondria functionality, measured as the mitochondrial oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) [48]. The mitochondrial functional respiration was recently evaluated in monocytes isolated from PBMCs of 13 COVID-19 patients with COVID-19 pneumonia [49]. Monocytes from COVID-19 patients displayed a reduction in basal and maximal respiration, proton leak, and spare respiratory capacity, compared to those from the healthy controls. Finally, an analysis of eight single-cells RNA-seq datasets, obtained from PBMCs, nasopharyngeal samples, and bronchoalveolar lavage fluid (N = 1,192,243 cells), revealed a significantly reduced mtDNA gene expression in immune system cells in COVID-19 patients [50]. All together, these results indicate that the mitochondrial metabolic pathway is altered in COVID-19 patients and suggest that targeting this pathway could represent a novel strategy for COVID-19 treatment.

There is much evidence in the literature, indicating that THs regulate cell bioenergetics by acting through direct effects on mitochondrial function [25,51]. Mitochondria have been shown to be the major sites of T3 accumulation in cells, where they exert a direct effect on mitochondrial activity and on energy metabolism [52]. In addition, free radical production is associated with the hypermetabolic state in hyperthyroidism, whereas the hypometabolic state induced by hypothyroidism leads to a decrease in free radical production [30,53,54].

Recently, we have demonstrated that PBMCs obtained from COVID-19 patients with NTIS showed a peculiar gene expression signature [16]. In particular, we found four genes that were deregulated in these patients, namely the IFIT3, the NLRP3, the CD38, and the CD79B. All these genes encode for proteins involved in the mitochondrial function. The IFIT3 is localized in the mitochondria and is significantly induced upon RNA virus infection. It plays a relevant role as a modulator of innate immunity [55]. The NLRP3 inflammasome is triggered by a variety of situations of host ‘danger’, including infection. It can sense mitochondrial dysfunction, and it is activated upon mitochondrial damage and destabilization [56,57]. In addition, it can sense mitochondrial dysfunction, and it is activated upon mitochondrial damage and destabilization [58,59]. The CD38 can drive mitochondrial trafficking in multiple myeloma [60], as well as after stroke [61]. In addition, it is involved in age-related NAD decline and mitochondrial dysfunction [62]. Finally, the CD79B interacts with Syk, which promotes mitochondrial superoxide generation via the modification of the mitochondrial electron transport chain in hematopoietic and nonhematopoietic cells [63], and its phosphorylation and subsequent translocation to lipid rafts appears to be involved in the generation of reactive oxygen species (ROS), a decrease in mitochondria membrane potential, and the induction of caspase-dependent apoptosis [64].

Seahorse XF technology proved to be capable of assessing, in a highly sensitive manner, the mitochondrial function in terms of several bioenergetic parameters, including basal respiration, ATP production, proton leak, maximal respiration, spare respiratory capacity, and non-mitochondrial respiration [65,66]. The assessment of the bioenergetic profile of human PBMCs has emerged as a new translational research biomarker and a new potential tool for the assessment of organ-specific mitochondrial dysfunction, relevant for the clinical outcome of critical illness [67]. Functional mitochondrial changes have been recently analyzed using Seahorse technology in ex vivo PBMCs obtained from patients with COVID-19 [47,48,49]. It has been demonstrated that SARS-CoV-2 is able to hijack host mitochondrial function and manipulate metabolic pathways for its own advantage. As a result, the PBMCs are left with a compromised mitochondrial function and an energy deficit that can be compensated for by a metabolic switch to glycolysis [49]. In another study, the Seahorse XFe24 analyzer was used to demonstrate the effect on neutrophils metabolism of COVID-19 patients, measured as ECAR, of the combined treatment with lipopolysaccharide and colchicine [68]. In disagreement with the previous study, a very recent observation, based on the analysis by high-resolution respirometry using the Oxygraph O₂k (OROBOROS Instruments, Innsbruck, Austria) and by the Cellular Oxygen METabolism (COMET) monitor, found an increase in mitochondrial oxygen tension (mitoPO₂) and mitochondrial oxygen consumption (mitoVO₂) between the isolated PBMCs from the SARS-CoV-2 patient groups, as compared to those from healthy controls and those patients without COVID-19 undergoing general anesthesia because of cardiothoracic surgery [69]. According to these results, mitochondrial respiration is increased in cases of severe COVID-19 compared to other critically ill patients, suggesting the occurrence of a relative hypermetabolic state in such patients. On the contrary, our results indicate that mitochondrial respiration is indeed hampered from COVID-19, and this effect was more evident in those with NTIS. Moreover, we demonstrated that treatment with LT3, either in vitro or in vivo, was able to stimulate mitochondrial respiration in PBMCs or our COVID-19 patients with NTIS. Treatment with LT3 in our patient 2 was also able to ameliorate the hydroelectrolytic alterations observed at admission and measured by BIA, while further reduction in serum FT3 levels, observed in patient 1, was associated with a worsening of BIA parameters and, in particular, on the hydration of fat-free mass, confirming our previous observation that T3 plays a crucial role in the control of hydroelectrolytic balance at the periphery [17]. In our patient 1, affected by severe COVID-19 with NTIS, treatment of the PBMCs in vitro with LT3 resulted in a significant increase in mitochondrial respiration, measured as basal and maximal OCR. A more evident effect was seen in the PBMCs of the patient 2 who received in vivo oral treatment with LT3 for one week, suggesting that LT3 exerts a stimulatory effect on OCR, associated to mitochondrial respiration and on ECAR, ascribed to glycolytic capacity. In addition, LT3 treatment ameliorates the spare respiratory capacity, which indicates the capability of the cells to respond to an increased energetic demand. It is interesting to note that, since mitochondrial respiration of PBMCs of our COVID-19 patients was stimulated after acute in vitro treatment with LT3 for 60 min (Figure 6A), we may speculate that LT3, besides its well-known transcriptional activity, may also act through other non-transcriptional effects. To the best of our knowledge, the stimulating effect of T3 on mitochondrial respiration in PBMCs observed in our COVID-19 patients with NTIS has never been reported before. One limit of our study relies on the fact that we did not perform sorting and analysis of the different cell populations within the PBMC fraction and, therefore, we could not confirm previous observations regarding the importance of monocytes in COVID-19 immunopathogenesis [49] or assess whether they could represent specific targets of TH action in these patients.

Since severe COVID-19 progression can be divided into phases, including early infection, host immune response, hyperinflammatory phase, and multiorgan dysfunction [70], we cannot exclude that the discrepancies between our results and those reported by Streng et al. [69] could be due to a different timing of patient inclusion, or alternatively, to the different methods used to analyze mitochondrial respiration. Our results regarding the stimulatory effects of LT3 on mitochondrial respiration are in accordance with those previously reported in other conditions and in different types of cells, including murine brown adipocytes [71] and murine, as well as human alveolar epithelial cells [72].

The two major differences we observed between PBMCs obtained from our cohort of COVID-19 patients, with and without NTIS, with regard to mitochondrial respiration, are related to the maximal respiration capacity and the proton leak. Both parameters, measured by Seahorse, were reduced in COVID-19 patients with NTIS compared to those without NTIS. The maximal OCR is measured by adding the uncoupler FCCP which acts by mimicking a physiological “energy demand” to meet an upcoming metabolic stress challenge. The induced stress stimulates the respiratory chain of the cells to reach the maximum rate of respiration they can achieve with a consequent activation of the rapid oxidation of substrates (sugars, fats, and amino acids). The proton leak is measured after the injection of an ATP synthase inhibitor (oligomycin) and reflects the remaining basal respiration not coupled to ATP production. Proton leak through uncoupling proteins (UCPs) is responsible for most of the uncoupling respiration in mitochondria, which generates heat instead of ATP, a TH effect typically observed in brown adipose tissue and muscle [28]. It has been reported that T3 is able to modulate both a basal and inducible proton leak in the mitochondria of skeletal muscle cells [73]. Proton leak has been considered as a sign of mitochondrial damage. However, according to the “uncoupling to survive” hypothesis, proton leak pathways are aimed to minimize oxidative damage by tempering electrical potential (Δp) and mitochondrial superoxide production [74]. The proportion of respiration that is used to drive the energy-dissipating “futile” proton cycle across the mitochondrial inner membrane is surprisingly high. It has been calculated, in fact, that the energy entirely devoted to driving this cycle accounts for 20–25% of the cellular basal metabolic rate. Such a proportion is remarkably constant even in species of widely different body mass, ranging from a mouse to a horse [75]. Since the mitochondrial proton cycle appears to be such an important energy drain and a wide range of organisms are prepared to pay a very high energetic price to maintain it, the functions that depends on it must be very important. Among such functions, could be several possible relevant ones, including thermogenesis, an improved ability to regulate energy metabolism, a safety valve for the avoidance of dielectric breakdown of the membrane at excessive membrane potentials, the ability to continue carbon metabolism when the ATP demand is low, the regulation of body mass, and the attenuation of free radical production [74]. Reduced proton leak observed in our COVID-19 patients with NTIS, strictly dependent on the reduction of FT3 serum levels, could be part of the adaptive response aimed to spare energy. However, it is reasonable to hypothesize that such an adaptive response may be associated with the induction of several associated cellular damages. The crucial role played by THs in such regulation is also demonstrated by experiments that were performed many years ago on liver mitochondria isolated from rats. These experiments demonstrated that the thyroid status of the animals strongly correlated with the basal proton conductance in mitochondria and, in particular, with the increase in the proton conductance of the inner mitochondrial membrane [76]. Our results agree with such observations and reinforce the hypothesis that thyroid hormones control the respiration rate required to balance the backflow of protons across the inner mitochondrial membrane.

In summary, our results indicate that thyroid hormones, in particular the active form LT3, play a relevant role in controlling mitochondrial respiration, energy production, and oxidative stress in circulating the immune cells of our COVID-19 patients. Based on the reduced serum levels of dROMs, and on the positive balance between oxidation and anti-oxidation, we support the hypothesis that reduced levels of serum FT3, observed in COVID-19 patients with NTIS, may exert a beneficial effect on the course of the disease. The reduced generation of ROS and reduced induction of oxidative stress exerts, in fact, a protective role in the pathogenesis of the acute immunological damages caused by SARS-CoV-2 infection. Since the reduced levels of FT3 are associated with reduced oxidation, it is possible that the use of T3 treatment, proposed for the management of severe COVID-19 [77], could be responsible for an increase in oxidative stress. The use of antioxidants, such as N-acetylcysteine, proved to be effective against severe COVID-19 [78] and could be useful to mitigate such effects.

However, the beneficial effect observed in COVID-19 patients with NTIS comes with a counterpart, represented by the consequence of reduced mitochondrial respiration. Such a negative effect is detrimental to the cells, since it is responsible for a markedly reduced production of energy generated by the circulating immune cells, thus dampening their response to viral infection.

It has been reported that exogenous nitric oxide (NO) can reduce systemic hyperinflammation and oxidative stress in COVID-19 patients [79]. Moreover, thyroid status could influence both the formation of and response to NO by the arterial vessels in rats [80]. It has been speculated that, since it was reported that NO administration is also able to improve arterial oxygenation, and to restore pulmonary alveolar cellular integrity, such administration, especially in the early stages of the disease, could have a preventive effect on lung damage in the management of COVID-19 patients [80]. However, we were not able to measure NO levels or perform other antioxidant assays besides the BAP test in our samples because of an insufficient amount of blood left after the measurements performed for the study. Another limit of our study was the small number of COVID-19 patients recruited for the study. It was designed as a pilot study and the results obtained will be useful to plan the clinical trial that we will perform and that will be focused on the potential use of THs, and especially T3, in critically ill patients admitted to the ICU.

Our results indicate that reduced serum FT3 levels, observed in COVID-19 patients with NTIS, have an opposite effect on oxidative balance and on mitochondrial respiration. On one side, the acute reduction of serum FT3 levels is responsible for the reduced generation of ROS, with a consequential reduction in the activation of the pathophysiological processes involved in the progression of COVID-19, including cytokine production, inflammation, and cell death. However, this beneficial adaptative effect comes with a counterpart that is represented by reduced mitochondrial respiration and a consequential reduction in energy production. This hypoenergetic state of immune cells induced by NTIS may compromise their reparative capacity and their ability to react to the damages induced by SARS-CoV-2 infection. Our pilot study, performed on two COVID-19 patients with NTIS, reinforced the role of T3 in regulating mitochondrial respiration and suggests that treatment with T3 could be beneficial in ameliorating mitochondrial respiration in circulating immune cells. However, the demonstration of the benefit of such treatment in NTIS-complicating COVID-19, as well as other critical illness, needs to be confirmed in future interventional, randomized clinical trials, with specific outcomes focused on mitochondrial function.