Mitochondrial adaptations from heat acclimation – A narrative review

Abstract

In recent decades, global temperature and humidity levels have surged. These increases in temperature and humidity are associated with higher risk of heat-related illnesses and impaired exercise performance, thereby prompting investigation into physiological responses to heat stress. Preconditioning strategies, including heat acclimation/acclimatization (HA), elicit physiological adaptations to enhance response to future heat exposures. Within HA research, an area of growing interest is examination of subcellular adaptations that contribute to whole-body acclimation, such as changes to/within mitochondria. External heat stress alters molecular pathways involved in mitochondrial biogenesis and bioenergetic function, but the relationship between these alterations and whole-body HA adaptations remains relatively unknown. Therefore, this review provides a detailed examination of the impact of HA on mitochondria across cell models, rodent models, and humans, linking these changes to exercise performance. Based on the current evidence, we propose a HA protocol aimed at promoting mitochondrial adaptations while maximizing traditional HA benefits. Lastly, we identify key areas for future research to further explore and enhance our understanding of mitochondrial responses to HA.

1. Introduction

Climate change, particularly global warming, is an ever-growing concern characterized by increasing average temperature year-over-year and increased frequency and severity of extreme heatwaves (Buguet et al., 2023; Climate change, 2023). Understanding the physiological responses while performing physical activity in the heat is imperative as many sporting, military, and occupational populations are frequently exposed to hot conditions, resulting in an elevation of the incidence of heat-related illnesses and impaired exercise performance (American College of Sports Medicine et al., 2007; Deane and Swann, 2023; Parsons et al., 2019). Consequently, there has been a substantial increase in research centered around the impact of heat stress and how to cope with it, such as preconditioning strategies for physiological adaptation (Fenemor et al., 2023; Sumi et al., 2023; Williamson-Reisdorph et al., 2023; Périard et al., 2023; Oberholzer et al., 2019; Gagnon et al., 2023; Dunn et al., 2024; Bach et al., 2024; Ramos et al., 2024; Kajiki et al., 2024; Coehoorn et al., 2024; Brown et al., 2024).

Heat stress refers to environmental and metabolic conditions that lead to increased body temperatures and can be classified as either acute or chronic (Sawka et al., 2011). Acute heat stress refers to a single or short-term exposure to elevated temperatures, often resulting in increased thermoregulatory and cardiovascular strain that impairs health and exercise performance without allowing adequate time for adaptation (Sawka et al., 2011; Périard et al., 2021). In contrast, chronic heat stress is characterized by repeated exposures to hot temperatures across a longer time period, such as several days or weeks (Tyler et al., 2024; Daanen et al., 2018; Périard et al., 2015). These repeated exposures may lead to cellular and physiological adaptations within the body, thereby creating a heightened tolerance to heat stress (Tyler et al., 2024; Daanen et al., 2018; Périard et al., 2015).

Heat acclimation/acclimatization (HA) is a popular preconditioning strategy to induce resiliency to future heat stress via achieved physiological adaptations, whereby individuals are artificially (acclimation) or naturally (acclimatization) exposed to hot conditions repeatedly (Tyler et al., 2024). Several reviews have explored and reported the traditional thermoregulatory, cardiovascular, sudomotor, fluid regulatory, and metabolic adaptations induced by HA (Périard et al., 2021; Tyler et al., 2024; Rahimi et al., 2019; Benjamin et al., 2019; Brown et al., 2022). These physiological adaptations improve performance during exercise in the heat, such as improved work capacity, time trial performance, and maximal oxygen uptake (VO_2max) (Périard et al., 2021). Within HA research, an emerging area of interest is the exploration of cellular mechanisms and adaptations that mediate systemic responses to heat stress (Deshayes et al., 2023; Ely et al., 2014), particularly those aimed at enhancing resiliency and performance during future heat exposure. However, this focus remains underexplored compared to the broader research on heat adaptations for improving general health. Furthermore, among the few reviews that have explored this area, the majority of the discussion has focused on the relationship between heat shock proteins and heat stress. In contrast, there has been less focus on the mitochondrial mechanisms associated with heat stress and HA, which could offer valuable insights into adaptations that enhance exercise performance.

Mitochondria are paramount for basic physiological function and exercise performance (e.g., mitochondrial network volume and aerobic capacity, mitochondrial proteins and bioenergetic function, mitochondrial antioxidants and protection against oxidative damage, etc.) (Drake et al., 2016; Furrer et al., 2023; Casanova et al., 2023). As mitochondria link oxygen delivery to adenosine triphosphate (ATP) synthesis to meet the energetic demands of skeletal muscle (Huertas et al., 2019), aerobic exercise performance may be enhanced via increased mitochondrial biogenesis (Margolis and Pasiakos, 2013). The utilization of oxygen combined with the high energy demand required for skeletal muscle activation during exercise drives adaptations associated with the cardiorespiratory system (Garcia-Roves et al., 2025). Furthermore, the mitochondrial network contributes to other processes that may directly or indirectly influence cellular mechanisms of exercise performance (e. g., calcium [Ca²⁺]) regulation, lipid metabolism), and consequently lead to a delayed time to fatigue in exercising skeletal muscle (Huertas et al., 2019; Talley and Mohiuddin, 2024; Holloszy and Booth, 1976). Together, these diverse roles highlight the critical importance of mitochondrial function in supporting both cellular and systemic responses to exercise, with factors such as mitochondrial biogenesis and bioenergetic efficiency playing key roles in performance and fatigue resistance. External stressors (i.e., heat stress) influence factors associated with mitochondrial biogenesis and bioenergetic function, although this research has produced conflicting results between cell and rodent models (Liu and Brooks, 2012; Salgado et al., 2017; Tamura et al., 2014; Patton et al., 2018) compared to humans (Mang et al., 2021; Hafen et al., 2018, 2019; Maunder et al., 2024; Hesketh et al., 2019; O’Reilly et al., 2021; Slivka et al., 2012; Heesch et al., 2016; Marchant et al., 2022).

Despite the significance of mitochondria in aerobic performance, the relationship between the effect of HA on potential mitochondrial adaptations still needs to be better understood. Understanding this relationship is crucial for optimizing HA protocols to improve exercise performance, as mitochondrial adaptations could play a pivotal role in the benefits of HA. Therefore, we aimed to synthesize the existing scientific literature on mitochondrial responses and adaptations to HA and connect these mitochondrial adaptations with improvements in exercise performance. This review begins with a comprehensive overview of the effects of HA on mitochondria in cell models, rodent models, and in humans, and connects these adaptations to exercise performance. Model-specific differences are emphasized. We then propose an evidence-based HA protocol designed to induce mitochondrial adaptations while optimally achieving traditional HA adaptations. Finally, we outline future research directions aimed at enhancing the understanding of mitochondrial adaptations from HA.

2. Mitochondrial adaptations from heat acclimation

2.1. Overview of mitochondrial biogenesis and oxidative phosphorylation

Mitochondrial adaptations occur with repeated heat stress, evidenced by increased markers of mitochondrial biogenesis, mitochondrial proteins, oxidative phosphorylation (OxPhos) machinery, and mitochondrial bioenergetic function (Figs. 1 and 2). However, because there is minimal research on the mechanistic insights between heat stress and its induced mitochondrial adaptations, there are some discrepancies, particularly when comparing findings across diverse models in cell culture, animals (i.e., mice vs. rat, strain), and the known variability in humans. Mitochondrial biogenesis is the physiological process of increasing the mitochondrial network within a cell’s cytoplasm from pre-existing mitochondria (Popov, 2020). In healthy, properly functioning cells, this process is primarily mediated by the transcription factor peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) (Popov, 2020), considered the master regulator of mitochondrial biogenesis. Various external stressors (e.g., physical activity, heat stress, cold stress) initiate signaling cascades that lead to the activation of PGC-1α through either phosphorylation or deacetylation, including the Ca²⁺/calmodulin-dependent protein kinase (CaMK) – p38 mitogen-activated protein kinase (MAPK) cascade, CaMK – cyclic adenosine monophosphate (cAMP) responsive element binding protein (CREB) cascade, cAMP – protein kinase A (PKA) – CREB cascade, AMP-activated protein kinase (AMPK) cascade, and sirtuin 1 (SIRT1) cascade.

Fig. 1.

Mitochondrial biogenesis and markers that are promoted through acute or continuous external heat stress. Abbreviations: AMP/ATP: ratio of adenosine monophosphate to adenosine triphosphate; AMPK: adenosine monophosphate-activated protein kinase; Ca²⁺: calcium; CaMK: calcium/calmodulin-dependent protein kinase; cAMP: cyclic adenosine monophosphate; CREB: cAMP responsive element binding protein; MAPK: mitogen-activated protein kinase; mRNA: messenger ribonucleic acid; mtDNA: mitochondrial deoxyribonucleic acid; NAD⁺/NADH: ratio of nicotinamide adenine dinucleotide (oxidized) to nicotinamide adenine dinucleotide (reduced); NRF: nuclear respiratory factor; PGC-1α: peroxisome proliferator-activated receptor γ coactivator 1α; PKA: protein kinase A; SIRT1: sirtuin 1; TFAM: mitochondrial transcription factor A. Created in https://BioRender.com.

Fig. 2.

Oxidative phosphorylation and markers that are promoted through acute or continuous external heat stress. Abbreviations: ADP: adenosine diphosphate; ATP: adenosine triphosphate; CoQ: coenzyme q; CytC: cytochrome c; ETC: electron transport chain; FAD: flavin adenine dinucleotide (oxidized); FADH₂: flavin adenine dinucleotide (reduced); I, II, III, or IV: ETC complexes I-IV; NAD⁺: nicotinamide adenine dinucleotide (oxidized); NADH: nicotinamide adenine dinucleotide (reduced); OCR: oxygen consumption rate; UCP: uncoupling protein. Created in https://BioRender.com.

OxPhos is the cellular process that occurs as electrons are passed along electron transport chain (ETC) complexes embedded within the inner mitochondrial membrane, creating an electrochemical gradient that allows for ATP synthesis (Deshpande and Mohiuddin, 2024). Briefly, this process involves electron transport from nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FADH₂) molecules to ETC complexes I and II, respectively, and subsequent transport to complexes III and IV, and then to oxygen the final electron acceptor. These complexes couple electron transport with H⁺ pumping from the mitochondrial matrix into the intermembrane space, thus generating an electrochemical gradient that is harnessed by complex V (i.e., ATP synthase) to phosphorylate adenosine diphosphate (ADP) to ATP. The abundance of OxPhos proteins increases during mitochondrial biogenesis, thereby increasing the capacity for aerobic ATP resynthesis (Popov, 2020).

HA, involving repeated exposures to heat stress, can elicit beneficial mitochondrial adaptations through mechanisms such as mitochondrial biogenesis and OxPhos capacity. However, it is important to note that acute bouts of heat stress alone may initially cause mitochondrial damage (Lu et al., 2023). In response to acute stress and damage, mitochondrial network turnover (i.e., mitochondrial biogenesis, fusion and fission, and mitophagy) plays a crucial role in maintaining mitochondrial quality via the removal of dysfunctional mitochondria and rebuilding of the network. As with other stressors to which mitochondria adapt (e.g., exercise) (Bishop et al., 2025; Craige et al., 2024) or maladapt (e.g., alcohol exposure) (DiLeo et al., 2024), these processes are potential factors in driving the adaptive response to heat stress. To shift from damage to adaptation, consistent and repeated heat stress is required (Hafen et al., 2018). While definitive thresholds separating mitochondrial damage from adaptation under heat stress are still not established, current evidence suggests they are both dose- and model-dependent, primarily influenced by temperature increases, and by duration and frequency of exposure. The following sections discuss various studies conducted across cell, animal, and human models regarding potential mitochondrial adaptations from repeated heat stress.

2.1.1. Cell models

The first study to investigate heat-induced mitochondrial adaptations applied 1 h of heat stress at 40 °C for five consecutive days to C2C12 myotubes compared to control cells housed at 37 °C (Liu and Brooks, 2012). This heating protocol increased Pgc-1α protein and mitochondrial DNA (mtDNA) copy number, suggesting an adaptive effect of enhanced mitochondrial biogenesis in vitro. Although no differences in Sirt1 protein expression were observed, Sirt1 activity was not assessed. Pgc-1α is considered the master regulator of mitochondrial biogenesis (Popov, 2020), while mtDNA copy number has previously been used as a marker to verify changes in mitochondrial biogenesis (Medeiros, 2008). Furthermore, OxPhos complex proteins (I-III, & V) were increased. This likely improves the rate of electron transfer across the ETC, resulting in increased ATP production due to a higher electrochemical gradient across the inner mitochondrial membrane. These findings were replicated in C2C12 myotubes in which six consecutive days of heat stress at 40 °C for 2 h per day also increased expression of proteins involved in mitochondrial biogenesis; specifically, p-ampka1, Sirt1, Pgc-1α, Nrf1, and Tfam proteins. There were also increases in markers of mitochondrial content, particularly citrate synthase (CS) and cytochrome C (CytC) (Patton et al., 2018). In another study, C2C12 myocytes were exposed to a constant temperature of 40 °C for 24 h, resulting in elevated expressions of the Ppargc1a, Nrf1, and Tfam genes (Salgado et al., 2017). Concomitantly, mitochondrial content was enhanced, as determined by Mitotracker staining and flow cytometry. The similarities in the promotion or increase of mitochondrial biogenesis and OxPhos components contribute to the potential of an adaptative effect to heat stress (5–6 consecutive days of 1–2 h per day, or constant 24 h).

Mitochondrial efficiency refers to oxygen consumption linked to ATP resynthesis versus other oxygen-consuming processes (e.g., proton leak, cyclooxygenases, etc.). Mitochondrial efficiency can be indirectly assessed via measuring uncoupling proteins (UCPs) and directly via live-cell extracellular flux analysis (Levitt et al., 2021). Findings from cell culture-based studies suggest that increased mitochondrial efficiency following HA, assessed indirectly and directly, may contribute to improved OxPhos (Salgado et al., 2017; Patton et al., 2018). UCPs allow proton leakage across the inner mitochondrial membrane without coupling proton movement with ATP resynthesis, thereby reducing mitochondrial efficiency and increasing the generation of “wasted” energy as heat (Salgado et al., 2017). From the aforementioned study conducted by Patton and colleagues, Ucp2 content was reduced following the six-day HA protocol in C2C12 myotubes (Patton et al., 2018). Similarly, Salgado et al. found that Ucp3 gene expression and H⁺ leak were reduced in C2C12 myocytes following 24 h of constant heat stress of 40 °C (Salgado et al., 2017). In addition, heat-stressed cells demonstrated reduced basal and peak oxygen consumption rate (OCR) and increased mitochondrial content compared to control cells. However, whether ATP-linked oxygen consumption and overall coupling efficiency improved with heat stress was not reported. OCR is an established indicator for mitochondrial function (Muller et al., 2019; Brand and Nicholls, 2011), and basal OCR is comprised of oxygen consumed via OxPhos and other oxygen-consuming cellular processes. Mitochondrial efficiency, indicated by reductions in UCPs, decreased proton leak, and increased proportion of OCR linked to ATP synthesis, improves after HA in C2C12 myocytes, leading to lower basal OCRs. Moreover, a decrease in UCPs likely results in lower intracellular heat production, which is suggested as a mechanism for thermal adaptation to future heat stressors (Salgado et al., 2017).

These studies collectively underscore the potential of heat stress to induce beneficial mitochondrial adaptations relevant to HA in cell models (Liu and Brooks, 2012; Salgado et al., 2017; Patton et al., 2018). Specifically, repeated heat stress has been shown to upregulate key regulators of mitochondrial biogenesis, along with increases in mtDNA copy number and OxPhos complex proteins, suggesting enhanced mitochondrial content and ATP production capacity. Additionally, reductions in uncoupling proteins (Ucp2 and Ucp3) suggest improved mitochondrial efficiency via decreased proton leak. Although mitochondrial function markers such as OCR have shown reductions in basal and peak levels following 24 h of continuous heat stress, it remains unclear whether ATP-linked oxygen consumption and overall coupling efficiency are improved.

2.1.2. Animal models

CS, which is a Krebs cycle enzyme often used as a marker of mitochondrial density (Vigelsø et al., 2014), catalyzes the reaction by which acetyl-CoA and oxaloacetate combine to form citrate (Wiegand and Remington, 1986). This is the initial reaction in the Krebs cycle and is imperative for the reduction of NAD⁺ and FAD to NADH and FADH₂, which are then shuttled to the ETC and utilized for ATP production. CS has been the most frequently used marker assessed in rodents to determine the effect of HA on mitochondrial content (Tamura et al., 2014; Tardo-Dino et al., 2021). In one of the few studies to examine these markers in vivo, Tamura et al. investigated the impact of passive HA and passive HA following exercise in a thermoneutral environment in a murine model (six-week-old male ICR mice), which consisted of five days per week for three weeks (Tamura et al., 2014). In the passive HA group, mice underwent 30 min of whole-body heat stress in a 40 °C environment, while the group that exercised immediately beforehand ran for 30 min in a temperate environment (22 °C) followed by whole-body heat stress. CS activity was increased in both plantaris and soleus muscles from passive heat stress alone and exercise training combined with passive heat stress, where the combination group experienced an increased magnitude of this adaptation. This finding may indicate that there was an increase in mitochondrial biogenesis, but a lack of assessment of other factors makes it difficult to draw this conclusion from only an increase in CS activity. In contrast, other work has observed no significant increase in CS activity in skeletal muscle of male Wistar rats following passive HA (Tardo-Dino et al., 2021). Rats completed a five day per week, six-week protocol and were divided into four groups: control, training (1 h running), HA (passive whole-body heating at 37 ± 3 °C; 1 h of heat), and training combined with passive HA. Following the six weeks, CS activity was increased in the soleus muscle in the training and training combined with passive HA groups, but there was no difference between groups, indicating that HA did not further increase CS activity beyond the effects of exercise training alone. Opposing results between these two studies may stem from differences in the models (mice vs rats), training program, and HA protocol. The contrasting findings regarding CS activity in response to HA in rodents highlight the necessity for further research into the various influencing factors and protocols for inducing mitochondrial biogenesis adaptations, particularly by incorporating additional assessed markers. Additionally, it is important to recognize that mitochondrial content markers can yield inconsistent results, as no single biomarker is without its limitations (Larsen et al., 2012; Groennebaek et al., 2020).

There are some divergent findings in the effects of HA on OxPhos proteins in rodent models. In the study by Tamura et al., mice in both the passive HA and passive HA combined with exercise training groups showed increases in the protein contents of Complexes I-IV in plantaris and soleus muscles, again with a more pronounced increase in the passive HA combined with exercise training group (Tamura et al., 2014), indicating an adaptation or enhancement of energy production capacity. In contrast, Tardo-Dino and colleagues found no effect of, or interaction with, HA in Complexes I-III and V in rats beyond exercise training-induced adaptations; however, there was an interaction between HA and training to increase Complex IV (Tardo-Dino et al., 2021). Together, the results of these two studies demonstrate some HA-related improvements in OxPhos protein abundance, but whether HA-related increases in Complex IV proteins alone are sufficient to drive improvements in function requires further examination.

Studies in non-mammalian models have also explored the effects of repeated heat stress on oxidative stress markers. For example, chronic heat stress induces ovarian damage via enhanced oxidative stress in pullets (Cao et al., 2023). However, there was an observed increase in antioxidant capacity when supplementing with N-acetyl-l-cysteine, suggesting an amelioration of chronic heat-stress induced ovarian damage (Larsen et al., 2012). While avian models differ considerably from humans, these findings underscore the potential vulnerability of female-specific tissue to heat-induced oxidative stress during development, and the potential efficacy of antioxidant supplementation to counteract these negative effects. This raises the possibility that similar oxidative stress mechanisms could contribute to female-specific heat responses in humans, particularly during certain developmental periods, however, more targeted research is required to confirm such effects and preventative measures in humans. Moreover, this work highlights the potential for targeted nutrition or supplementation strategies to minimize heat-induced oxidative damage, although caution is warranted as some oxidative stress is likely necessary to drive adaptations.

The limited studies conducted in rodents have produced mixed results regarding mitochondrial biogenesis and OxPhos in response to heat stress (Tamura et al., 2014; Tardo-Dino et al., 2021). Tamura et al. found that both passive HA and passive HA combined with exercise increased CS activity and OxPhos complexes I-IV in mouse skeletal muscle (Tamura et al., 2014), suggesting potential enhanced mitochondrial biogenesis and energy production. In contrast, a study investigating Wistar rats reported no additional CS activity or OxPhos protein increases from HA, with only Complex IV being elevated when HA was combined with exercise (Tardo-Dino et al., 2021). These conflicting findings highlight the need for further research utilizing a broader range of mitochondrial markers to clarify the effects of HA on mitochondrial adaptations in rodent models.

2.1.3. Human models

In humans, most HA interventions aimed at exploring mitochondrial adaptations have utilized passive methods, such as whole-body and localized heating. Passive whole-body heating can be achieved through various approaches, including saunas, hot water immersion, and water-perfused suits. However, there has been only one study focused on mitochondrial adaptations following HA using passive whole-body heating. Hesketh and colleagues conducted a study in which young sedentary males completed six weeks (three sessions per week, 40–50 min per session) of passive heat therapy via sitting in a heat chamber at 40 °C and ~40 % RH (Hesketh et al., 2019). From vastus lateralis (VL) muscle samples, mitochondrial density, assessed only via cytochrome C oxidase subunit 4 (COX4) protein expression, did not increase (Hesketh et al., 2019). COX4 is the largest subunit of the cytochrome c oxidase complex (Complex IV) that plays a vital role in several important functions, including stabilizing Complex IV, facilitating electron transfer, and helping maintain the proton gradient across the inner mitochondrial membrane (Li et al., 2006). In this investigation, participants did not experience high core temperature (T_c) elevations from the whole-body heating, as T_c was maintained at ~ 37–38 °C throughout the sessions (Hesketh et al., 2019). While not measured, it is possible that skeletal muscle temperature also experienced only minimal changes, thereby reducing the effect of heat stress on factors of mitochondrial biogenesis or OxPhos in skeletal muscle. However, it is difficult to surmise that passive heating did not induce any mitochondrial adaptations as only COX4 was assessed (Hesketh et al., 2019), leaving out many other markers of mitochondrial biogenesis or OxPhos that may have been affected.

Passive HA protocols via localized heating have also examined the efficacy of inducing human mitochondrial adaptations (Hafen et al., 2018, 2019; Marchant et al., 2022). Local application of heat stress to skeletal muscle is typically performed using heat pads or diathermy, demonstrating positive findings regarding mitochondrial biogenesis (Hafen et al., 2018, 2019; Marchant et al., 2022). Diathermy uses electromagnetic currents (i.e., radio and sound waves) to generate heat locally to a specific body area (Goats, 1989). Hafen et al. performed two separate studies that investigated the use of diathermy to produce repeated local heat stress on skeletal muscle and evaluated mitochondrial adaptations (Hafen et al., 2018, 2019). In their initial investigation (Hafen et al., 2018), healthy, sedentary individuals underwent six consecutive days of diathermy treatment (2 h per day) to the VL. This form of local heating resulted in a 3.9 ± 0.31 °C increase in muscle temperature by the 30th minute from ~36.0 °C to ~39.9 °C and was sustained at this elevated temperature throughout the remaining 90 min of heating. Upon completion of the ten days of treatment, there were increases in PGC-1α protein content but no changes to CS activity. The subsequent study determined that ten consecutive days of diathermy (2-h sessions) to the VL maintained, and even slightly increased from baseline, PGC-1α protein content during complete immobilization of one leg (Hafen et al., 2019). From another research group (Marchant et al., 2022), six weeks of diathermy (three days per week, 2-h sessions) did not change CS activity; however, it is possible that other mitochondrial adaptations were induced but not measured. Together, these findings suggest that applying local heat stress via diathermy may be a strong enough stimulus to induce increases in PGC-1α protein content, but not CS activity, in humans.

The combination of localized heating with exercise training has also been investigated for the potential to induce mitochondrial adaptations and elicit an ergogenic effect for endurance performance (Maunder et al., 2024). Maunder and colleagues found that combining a cycling exercise protocol (three weeks, five sessions per week) with concurrent local heat application via a heating pad to one upper leg did not promote further increases in CS activity (Maunder et al., 2024). Based on this finding, along with the work conducted in local heating via diathermy, it seems that CS activity is not affected by external heat stress in humans. This does not necessarily mean that localized external heat stress concomitant with exercise does not elicit additional mitochondrial biogenesis adaptations. Indeed, these studies have not measured many of the regularly assessed markers of mitochondrial biogenesis (e.g., PGC-1α, voltage dependent anion-selective channel [VDAC], mtDNA).

Active HA involves exercising in hot environmental conditions and typically elicits greater physiological adaptations compared to passive HA, but few such studies have been conducted in humans that have examined markers of mitochondrial biogenesis (Mang et al., 2021). Mang and colleagues subjected active males (VO_2max: 56.4 ± 4.4 mL kg·min⁻¹) and females (VO_2max: 42.3 ± 3.4 mL kg·min⁻¹) to a 10-day active HA protocol across 14 days, which consisted of 90 min of walking at a relatively low intensity (30–40 % maximal velocity at 3 % incline) in environmental conditions of ~42 °C with 30–50 % relative humidity (Mang et al., 2021). Upon completion of HA, there were no alterations to PGC-1α, TFAM, or CaMK in muscle samples collected from the VL. Similarly, no changes were observed in PGC-1α, NRF1, TFAM, CS, or COX4 mRNA expression in untrained females after a 22-day active HA protocol, which involved 60 min of cycling at an intensity corresponding to a rating of perceived exertion of 15 (hard; Borg 20-point scale) in 33 °C and 40 % RH (McGlynn et al., 2022). In another study, Slivka et al. also found no changes to PGC-1α gene expression or mitochondrial content in untrained males following a 16-day active HA protocol consisting of 60 min of cycling at 50 % peak power in 33 ° C, 40 % RH (Slivka et al., 2021). Together, these three studies indicate that active HA, aimed primarily at increasing T_c rather than skeletal muscle temperature, does not induce mitochondrial adaptations. These findings are consistent for both trained (Mang et al., 2021) and untrained males (Slivka et al., 2021) and females (McGlynn et al., 2022), regardless of the duration of the HA protocol (10–22 days) or the exercise intensity. This may be due to inadequate increases in skeletal muscle temperature caused by the lack of localized heating. Additionally, the active HA protocols discussed may not have involved an exercise intensity high enough to adequately elevate skeletal muscle temperature and effectively stimulate mitochondrial adaptations. However, this cannot be fully elucidated due to the limitation of not measuring intramuscular temperature in these studies.

Collectively, human studies investigating mitochondrial adaptations to HA have primarily utilized passive heating protocols, such as whole-body heating and localized heating. From whole-body heating, no changes were observed in mitochondrial marker COX4 (Hesketh et al., 2019), potentially due to insufficient elevation in core and muscle temperatures. With localized heating via diathermy, increases have been observed in PGC-1α protein, however CS remained unchanged (Hafen et al., 2018, 2019; Marchant et al., 2022). When combining localized heating with concomitant exercise, CS activity similarly remained unchanged (Maunder et al., 2024). Active HA protocols have generally not induced changes in mitochondrial markers (Mang et al., 2021; McGlynn et al., 2022; Slivka et al., 2021). Overall, while localized heating shows some promise for promoting mitochondrial adaptations in humans, use of both passive and active HA require further investigation with broader marker panels and direct muscle temperature measurements to best elucidate findings.

2.2. HIF-1α regulation of mitochondria

Hypoxia-inducible factor 1 alpha (HIF-1α) is a transcription factor that mediates the cellular hypoxic response and assists in the regulation of mitochondrial biogenesis and respiration (Huang et al., 2022; Mialet-Perez and Belaidi, 2024). Under external stressors, such as heat, an intracellular accumulation of HIF-1α occurs which combines with HIF-1β to form HIF-1 (Maloyan et al., 2005). HIF-1 activates various downstream pathways associated with stress responses, including glucose metabolism, mitochondrial biogenesis, and oxidative stress responses (Ely et al., 2014; Huang et al., 2022; Weidemann and Johnson, 2008a). Considering this relationship between HIF-1 and heat stress, it is possible that changes in HIF-1 expression may reflect improvements in heat tolerance, partly through mitochondrial biogenesis and respiration. Additionally, acclimatory HIF-1 adaptations may result in elevated erythropoietin levels (leading to increased blood volume), vascular endothelial growth factor (leading to enhanced skin blood flow for heat dissipation), and inducible nitric oxide synthase (leading to increased splanchnic blood flow), all of which provide further benefits during exercise under heat stress (Ely et al., 2014). Surprisingly, only a few studies have investigated HIF-1α adaptations in response to HA (Maloyan et al., 2005; Horowitz and Assadi, 2010; Alexander-Shani et al., 2017; Lee et al., 2016), and even fewer have explored the connection between HA-induced adaptations of HIF-1α and their effects on mitochondrial biogenesis and respiration (Horowitz and Assadi, 2010; Alexander-Shani et al., 2017).

2.2.1. Cell and rodent models

Demonstration of HIF-1’s role in HA was first revealed in a Caenorhabditis elegans model whereby HIF-1 knockout prevented acclimation to heat (Treinin et al., 2003). Following this novel discovery, Maloyan et al. investigated HIF-1α adaptations from a 30-day passive HA protocol in myocardial cells from rodent hearts (Maloyan et al., 2005). Upon completion of HA, constitutive Hif-1α protein levels increased, and there was an observed augmented induction of Hif-1α with acute heat stress (Maloyan et al., 2005). From the same study, this research group also saw elevations in basal levels in transcripts and proteins of metabolic enzymes as downstream targets of HIF-1 linked to mitochondria (i.e., pyruvate dehydrogenase kinase [Pdk1], mitochondrial Lon protease [Lon], and Cox4.2) (Horowitz and Assadi, 2010). In a follow-up analysis, Alexander-Shani et al. sought to examine disruptions to HIF-1α transcriptional activation on its mitochondrial gene targets (i.e., Pdk1, Lon, Cox4 isoforms) upon HA completion in rat hearts (Alexander-Shani et al., 2017). Following the 30-day HA protocol combined with HIF-1α: HIF-1β dimerization blockade, there was an upregulation of Pdk1 and a remodeling of the Cox4 isoform ratio, specifically a greater dominance of Cox4.2 isoform. The upregulated Pdk1 increases glycolytic lactate production as Pdk1 phosphorylates and inactivates the pyruvate dehydrogenase complex, reducing pyruvate entry into the mitochondria (Kawano et al., 2022). The greater expression of Cox4.2 improves efficiency of electron transfer and possibly ATP production (Alexander-Shani et al., 2017). With a reduction of pyruvate entering mitochondria and thereby reducing tricarboxylic acid (TCA) cycle activity (Papandreou et al., 2006; Kim et al., 2006a; McCommis and Finck, 2015), there is a downregulation of mitochondrial oxygen consumption and an attenuation of reactive oxygen species generation (Weidemann and Johnson, 2008b; Arnold and Finley, 2023). Alterations in PDK1 and COX4.2 may contribute to optimizing the balance between ATP production and redox homeostasis by restricting excessive ROS generation while sustaining adequate energy supply through more efficient substrate utilization. COX4.2 induction enhances electron transfer efficiency (Fukuda et al., 2007), supporting ATP synthesis at a lower oxygen cost, whereas PDK1 phosphorylates and inhibits pyruvate dehydrogenase, preventing pyruvate conversion to acetyl CoA (Kim et al., 2006b; Laird et al., 2024). Due to an upregulation in PDK1, there may be reduced oxygen demand, mitochondrial ROS production, and therefore greater protection of the cell against stress damage (Laird et al., 2024; Samanta and Semenza, 2017; Basheeruddin and Qausain, 2024). Furthermore, as Lon serves as a mitochondrial ATP-dependent protease that assists to degrade misfolded proteins, an upregulation of Lon will support the maintenance of mitochondrial protein quality (Laird et al., 2024; Voos and Pollecker, 2020). In the context of whole-body application, these downstream HIF-1α targets could possibly contribute to improved exercise economy by reducing the oxygen cost of ATP resynthesis, stabilizing oxidative phosphorylation efficiency, and mitigating fatigue-inducing oxidative stress during prolonged exertion. Altogether, these previous studies in cell and rodent models demonstrate the importance of HIF-1 to HA adaptation, including its role in certain mitochondrial adaptations.

2.2.2. Human models

HA-induced changes in HIF-1 have also been explored in humans, however there is only one study, and it did not include mitochondrial markers. Lee et al. employed 21 male participants (VO_2peak: 51.7 ± 7 mL kg·min⁻¹) to perform a 10-day HA protocol (40 °C, 20 % RH) consisting of 1 h cycling at 50 % VO_2peak (Lee et al., 2016). Extracellular (circulating) HIF-1α was assessed before and after exercise on days 1 and 10 of HA. Following HA, resting extracellular levels of HIF-1α were increased, coupled with a blunted post-exercise rise in extracellular HIF-1α (Lee et al., 2016). While these results suggest altered resting and exercise-induced extracellular HIF-1α in humans following an active HA protocol, caution is warranted as the functional effects of HIF-1α are intracellular. Whether changes in extracellular HIF-1α reflect changes in intracellular HIF-1α levels and function in humans is unknown. In addition, a lack of mitochondrial markers measured from this study limits the interpretation of HIF-1α role in mitochondrial adaptation to HA. Despite these limitations, aerobic performance, assessed via a 16.1 km cycling time trial, improved after HA (Lee et al., 2016), which could be related to HIF-1α and its effect on mitochondria. Future studies should seek to determine the effect of HA on intracellular HIF-1α and its subsequent effect on mitochondrial biogenesis and function in humans. This would potentially allow for extrapolation of findings to a whole-body aerobic performance adaptation due to improvements in mitochondrial function.

In cell and rodent models, HA has been shown to increase Hif-1α protein levels and its responsiveness to acute heat stress, alongside elevated expression of mitochondrial-associated target markers (i.e., Pdk1, Lon, and Cox4.2) (Maloyan et al., 2005; Horowitz and Assadi, 2010; Alexander-Shani et al., 2017). These findings highlight the role of HIF-1α in driving mitochondrial adaptations during HA. When translating this research to humans, it has been observed that HA induces increased resting and blunted post-exercise extracellular plasma HIF-1α levels (Lee et al., 2016). However, the study lacked intracellular and mitochondrial markers, limiting conclusions about mitochondrial adaptation.

3. Link between HA-induced mitochondrial adaptations and exercise performance

A primary performance adaptation from active HA is enhanced exercise economy, which is defined as the amount of energy spent per unit of distance (di Prampero, 2003). Exercise economy, along with VO_2max and lactate threshold, is highlighted as one of the three main pillars that constitute endurance performance (Joyner and Coyle, 2008). Previous literature has speculated that HA’s adaptation of enhanced exercise economy may derive from increases in mitochondrial biogenesis, mitochondrial efficiency, and improved OxPhos (Mang et al., 2021). In addition, a ‘transfer effect’ of HA adaptations for improving exercise performance in temperate environments has been contemplated, which may be partially explained by enhanced exercise economy via mitochondrial adaptations (Mang et al., 2021).

Several active HA studies have investigated the potential ergogenic adaptation of improved aerobic performance; however, they did not assess mitochondrial markers and could not attribute findings to changes in mitochondrial biogenesis or OxPhos. From the few active HA protocols previously discussed in section 2.1.3 that assessed both aerobic performance and mitochondrial adaptations, discrepancies exist, with some showing positive adaptations (i.e., improved max power, VO_2max, max aerobic capacity, max aerobic power), but others not (i.e., no changes in VO_2peak, first or second ventilatory thresholds). Even when positive improvements in aerobic performance were observed, it has not been accompanied by increases in mitochondrial biogenesis or OxPhos. However, only a limited number of markers have been investigated, thus a holistic analysis has not yet been performed to completely elucidate possible connections between mitochondrial adaptations and improved aerobic performance from HA. In the absence of measurable increases in mitochondrial content or improved mitochondrial efficiency, VO_2max can still improve through enhanced perfusion (capillarization) (Hellsten and Gliemann, 2024), which may have contributed to these findings. In addition, the use of different testing environments (thermoneutral vs. heat) to assess these aerobic performance adaptations likely contributes to the variability in findings.

In the one study that has combined localized passive heat application during exercise, improvements were noted in VO_2peak and the first and second ventilatory thresholds (Maunder et al., 2024). Although it is possible that adaptations in Complexes I and II contributed to improved aerobic performance, these results should be interpreted with caution. In this study, a contralateral leg design was employed and thus only one muscle (i.e., the dominant VL muscle) was heated, meaning all participants performed exercise training with only one leg heated (Maunder et al., 2024). As a result, a variety of other training adaptations could have influenced the observed improvements in aerobic performance. Still, this study demonstrates the potential impact of localized heat stress in increasing selected OxPhos components, which may contribute to the observed aerobic improvements.

From the studies that utilized passive heat application localized to the quadriceps and found positive adaptations in mitochondrial markers, no aerobic exercise tests were performed to transfer these findings to potential exercise improvements (Hafen et al., 2018, 2019). Therefore, it is difficult to determine if localized heating without exercise would improve aerobic performance. However, given the findings from Maunder and colleagues, it is likely that localized heat application combined with exercise training stimulates these improvements (Maunder et al., 2024). Many endurance athletes already have a high training load, so utilizing passive methods for HA to stimulate positive adaptations could be beneficial as no additional training load is added.

4. Suggested HA protocol for achievement of mitochondrial adaptations

We recommend the following evidence-based HA protocols aimed at effectively promoting mitochondrial adaptations, in addition to the traditional physiological HA adaptations, for exercising populations (Fig. 3). Generally, longer HA protocols have been associated with a greater magnitude of adaptations, and prior research indicates that extended protocols can lead to positive mitochondrial adaptations (Maunder et al., 2024). We have also developed two HA protocols, one active and one passive, to guide individuals with different training programs on effectively incorporating this additional time. Given that athletes already have a congested training schedule, it is crucial to carefully plan how to incorporate additional training load when using the active HA protocol. In contrast, the passive protocol can be implemented without the additional training load, but the magnitude of adaptations may be lesser (Heathcote et al., 2018).

Fig. 3.

Two suggested heat acclimation protocols designed to target both traditional physiological adaptations to heat acclimation with an emphasis on skeletal muscle mitochondrial adaptations. The active protocol (left) combines cycling exercise with simultaneous heat pad application to further increase muscle temperature. Previous literature [40] utilizing a similar combined approach demonstrated that 40 °C heat pad application during cycling exercise was sufficient to raise muscle temperature to induce mitochondrial adaptations to the vastus lateralis muscle. The 60 min cycling at 60 % VO_2max/W_max is a common time and intensity employed for traditional heat acclimation protocols. The passive protocol (right) also combines whole-body heating with additional heat application to lower body skeletal musculature, either via heat pads or diathermy. Individuals should evaluate their current training programs to determine which protocol is most suitable for implementation. Created in https://BioRender.com.

Whether active or passive HA, both protocols employ a combined strategy of whole-body heating and localized heat stress (i.e., heat pads or diathermy) targeted at the leg musculature. This combined approach aims to raise T_c, a key factor in driving resiliency to subsequent exercise in the heat, while also increasing skeletal muscle temperature, which is likely necessary for intramuscular mitochondrial adaptations (Marchant et al., 2023). In the active protocol, heat pads should be applied to the quadriceps muscles during the 60 min of cycling. Previous literature (Maunder et al., 2024) using a similar combined approach showed that applying a 40 °C heat pad during cycling exercise was sufficient to raise muscle temperature and induce mitochondrial adaptations in the VL muscle. Additionally, although many HA protocols differ in environmental conditions (35–40 ° C, 20–80 % RH), duration (60–90 min), and intensity (50–65 %), the 60 min cycling at 60 % VO_2max/W_max under 35 °C and 40 % RH environmental conditions fall within the range of traditional HA for inducing physiological, perceptual, and performance adaptations (Périard et al., 2021). Furthermore, the exercise intensity can be adjusted for less trained individuals to ensure protocol completion. In the passive protocol, heat pads or diathermy should be applied to the quadriceps muscles concurrent with 60 min of passive sitting in the sauna. After the whole-body heating, localized heating should be maintained on the quadriceps muscles for 60 min in a thermoneutral environment. Diathermy sessions 120 min in length have previously demonstrated positive mitochondrial adaptations (Hafen et al., 2018, 2019).

These proposed protocols are based on positive findings from the studies discussed in this review. However, although evidence-based, these protocols may not elicit the expected results across all populations. Notably, most previous studies in the general field of heat adaptation have predominantly involved male participants, with a significant lack of female inclusion (Kelly et al., 2024). This underrepresentation creates challenges in applying heat adaptation findings universally. Anatomical, physiological, and endocrinological differences between sexes likely contribute to variations in thermoregulatory responses to heat stress (Kelly et al., 2023). Furthermore, the majority of the mitochondrial adaptations discussed in this review were conducted with male participants, with only one of the two active HA studies including females (Mang et al., 2021). Understanding sex-based differences in thermoregulatory and mitochondrial responses to heat stress is crucial for developing more inclusive and universally applicable heat adaptation strategies.

5. Conclusions

In this review, we discussed the effects of both passive and active heat acclimation on mitochondrial adaptations, primarily mitochondrial biogenesis and OxPhos, which are summarized and presented in Table 1. HA protocols are becoming increasingly utilized in both general and exercising populations to combat future occurrences of heat stress. Moreover, investigating the cellular responses associated with mitochondria leads to an improved understanding of physiological and performance-related HA adaptations. This narrative review of existing literature demonstrates that certain methods of heat exposure, particularly localized heating, may induce adaptations of improved mitochondrial biogenesis and OxPhos. Most active HA protocols are designed to elevate T_c to induce whole-body physiological adaptations. However, muscle temperature is not usually assessed in these studies. As it seems that localized heating to leads to the greatest extent of skeletal muscle mitochondrial adaptations, passive HA protocols using heat pads or diathermy may be the desired avenue of heating. However, this emerging topic within HA research is still limited, and discrepancies exist between cells, rodents, and humans. In conclusion, it seems that repeated heat exposures associated with HA protocols have a beneficial effect on mitochondrial biogenesis and OxPhos, but it remains unclear how these adaptations translate to a whole-body ergogenic effect and the required heat dosage to induce these adaptations.

Table 1.

Summary of investigations examining the effects of heat acclimation on markers of mitochondrial biogenesis and oxidative phosphorylation.

Study	Model	Type of Activity	Heat Dosage	Protocol Length	Mitochondrial Biogenesis	Oxidative Phosphorylation
Liu and Brooks (Liu and Brooks, 2012)	Cell; C2C12 myotubes	Passive	1 h; 40 °C	5 continuous days	↑ PGC-1α protein, ↑ mtDNA copy number, ↔ SIRT1 protein	↑ Complexes I – III & V protein, ↔ Complex IV protein
Salgado et al. (Salgado et al., 2017)	Cell; C2C12 myocytes	Passive	40 °C	Continuous 24 h	↑ Ppargc1a gene, ↑ NRF1 gene, ↑ TFAM gene	↓ basal & peak OCR, ↓ OCR per relative mitochondria, ↓ H+ leak, ↓ UCP3 gene, ↑ mitochondria content
Tamura et al. (Tamura et al., 2014)	6-week-old male ICR mice; n = 6–7 mice/group	Passive and Active + Passive; treadmill running 25 m/min for 30 min	Passive: 30 min; 40 °C	3 weeks (5 days/week)	Passive: ↑ CS activity	Passive: Complexes I – V
			Active + Passive: 30 min running + 30 min; 40 °C		Active + Passive: ↑ CS activity	Active + Passive: Complexes I – V
Tardo-Dino et al. (Tardo-Dino et al., 2021)	8-week-old male Wistar rats; n = 8 rats/group	Passive and Active + Passive; 25.8–30 m/min for 25–55 min	Passive: 1 h; 35–40 °C	6 weeks (5 days/week)	Passive: ↔ CS activity	Passive: ↔ Complexes I – V
			Active + Passive: 25–55 min running + 1 h; 35–40 °C		Active + Passive: ↑ CS activity	Active + Passive: ↔ Complexes I – III, V; ↑ Complex IV
Mang et al. (Mang et al., 2021)	Humans; active males (n = 7) and females (n = 6)	Active; walking @ 30–40 % maximal velocity with 3 % inclination grade	90 min; ~42 °C	10 days	↔ CaMK protein, ↔ PGC-1α protein, ↔ TFAM protein	↔ COX4, ↔ Complexes I – IV proteins
Maunder et al. (Maunder et al., 2024)	Humans; active males (n = 8)	Active + Passive; heat pad to one exercising leg while cycling, various moderate to heavy exercise sessions	~40 °C	3 weeks (5 days/week)	↔ CS activity	↑ Complexes I – II, ↔ Complexes III – IV
Hafen et al. (Hafen et al., 2018)	Humans; healthy, sedentary males (n = 10) and females (n = 10)	Passive; diathermy	2 h	6 consecutive days	↑ PGC-1α expression, ↔ CS activity	↑ Complexes I & V, ↔ Complexes II - IV
Hafen et al. (Hafen et al., 2019)	Humans; healthy, young males (n = 12) and females (n = 11)	Passive; diathermy	2 h	10 consecutive days	↑ PGC-1α expression	↔ Complexes I – V, ↔ oxidative phosphorylation coupling efficiency, ↔ electron transfer system coupling efficiency
Marchant et al. (Marchant et al., 2022)	Humans; healthy, males (n = 5) and females (n = 7)	Passive; diathermy	2 h	6 weeks (3 days/week)	↔ CS activity	↑ maximal coupled respiration, ↔ CytC, ↔ Complexes I – V
Hesketh et al. (Hesketh et al., 2019)	Humans; young, sedentary males (n = 10)	Passive; whole-body	40–50 min; 40 °C	6 weeks (3 days/week)	–	↔ COX4 expression

Adenosine monophosphate (AMP)-activated protein kinase, AMPK; Ca²⁺/calmodulin-dependent protein kinase, CaMK; Cytochrome C oxidase subunit 2, COX2; Cytochrome C oxidase subunit 4, COX4; cyclic AMP response element binding protein, CREB; Citrate synthase, CS; Cytochrome C, CytC; Mitogen-activated protein kinase, MAPK; Mitochondrial DNA, mtDNA; Nuclear respiratory factor 1, NRF1; Nuclear respiratory factor 2, NRF2; Oxygen consumption rate, OCR; proliferator-activated receptor γ coactivator-1α, PGC-1α; Sirtuin 1, SIRT1; Mitochondrial transcription factor A, TFAM; Uncoupling protein 3, UCP3; maximal oxygen uptake, VO_2max.

6. Future research directions

We propose that future research should investigate the specific markers of mitochondrial biogenesis and OxPhos that experience adaptations from HA. In particular, protocols of both active and passive methodologies should be utilized with muscle temperature included as a primary measure to determine if there is an association between muscle temperature increases during heating with adaptations to mitochondrial markers, as well as determining an association between systemic T_c increases and muscle temperature increases. Additionally, there is a significant lack of mitochondrial life cycle (mitophagy/autophagy) assessment in the limited number of completed studies. These molecules may also experience adaptations from HA that may be related to mitochondrial function. Thus, future research should look to include these markers in their analyses. Translating cellular adaptations associated with mitochondria from HA to ergogenic benefits (e.g., improved aerobic performance) should be elucidated to provide practical applications.

Data accessibility statement

Not applicable.

Data availability

No data was used for the research described in the article.