Effects of exercise in a cold environment on transcriptional control of PGC-1α

Abstract

Peroxisome proliferator-activated receptor-α coactivator-1α (PGC-1α) mRNA is increased with both exercise and exposure to cold temperature. However, transcriptional control has yet to be examined during exercise in the cold. Additionally, the need for environmental cold exposure after exercise may not be a practical recovery modality. The purpose of this study was to determine mitochondrial-related gene expression and transcriptional control of PGC-1α following exercise in a cold compared with room temperature environment. Eleven recreationally trained males completed two 1-h cycling bouts in a cold (7°C) or room temperature (20°C) environment, followed by 3 h of supine recovery in standard room conditions. Muscle biopsies were taken from the vastus lateralis preexercise, postexercise, and after a 3-h recovery. Gene expression and transcription factor binding to the PGC-1α promoter were analyzed. PGC-1α mRNA increased from preexercise to 3 h of recovery, but there was no difference between trials. Estrogen-related receptor-α (ERRα), myocyte enhancer factor-2 (MEF2A), and nuclear respiratory factor-1 (NRF-1) mRNA were lower in cold than at room temperature. Forkhead box class-O (FOXO1) and cAMP response element-binding protein (CREB) binding to the PGC-1α promoter were increased postexercise and at 3 h of recovery. MEF2A binding increased postexercise, and activating transcription factor 2 (ATF2) binding increased at 3 h of recovery. These data indicate no difference in PGC-1α mRNA or transcriptional control after exercise in cold versus room temperature and 3 h of recovery. However, the observed reductions in the mRNA of select transcription factors downstream of PGC-1α indicate a potential influence of exercise in the cold on the transcriptional response related to mitochondrial biogenesis.

INTRODUCTION

Mitochondrial dysfunction has been implicated in peripheral artery disease (28), aging (12), obesity (7), and other diseases (8). Cold exposure during exercise in humans has the potential of promoting mitochondrial biogenesis (45). If increased mitochondrial density can potentially be achieved through cold temperature training, this may serve as novel treatment of diseases associated with mitochondrial dysfunction and enhance exercise-training outcomes (17, 30). This training model utilized in the diseased population could possibly allow exercise to take place at a lower intensity with the same training effects. It could also be used a preventative measure in nondiseased populations by eliciting a greater response in mitochondrial density. Postexercise recovery in a cold environment can lead to higher levels of oxygen uptake and higher gene expression associated with mitochondrial biogenesis (45, 46), such as peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α). PGC-1α is a key regulator of energy metabolism and is likely involved in disorders such as obesity, diabetes, and cardiomyopathy (25). Cold exposure during exercise increases PGC-1α in rats, whereas the stimulus of cold alone is unable to produce such an increase. Thus, it appears that physical activity with cold exposure is a requirement to elicit a strong induction in the expression of PGC-1α (43). The mechanism of cold-induced increases in PGC-1α expression is via β₃-adrenergic receptor (6). When cold exposure is combined with exercise, there is an apparent synergistic effect of increasing PGC-1α expression (45). However, it is unknown whether the stimulus of cold recovery is needed. Further examination of PGC-1α transcriptional control is needed to elucidate this mechanism.

Paradoxical findings have occurred between PGC-1α mRNA and transcription factors downstream. PGC-1α coactivates the expression of genes encoding oxidative metabolism, such as nuclear respiratory factor-1 (NRF-1), nuclear respiratory factor-2 (NRF-2), myocyte enhancer factor-2 (MEF2A), and estrogen-related receptor-α (ERRα). These transcription factors play key roles in mitochondrial development (10, 42). In this, way, PGC-1α regulates many aspects of mitochondrial development. As PGC-1α mRNA increases, an increase in the downstream transcription factors would be expected. However, a paradox occurs after 1 h of cycling and 3 h of cold environmental recovery (46); that is, PGC-1α mRNA increases to a greater extent compared with room temperature conditions, but ERRα and NRF2 are lower in the cold compared with room temperature.

In addition to PGC-1α coactivating transcription factors for mitochondrial development, MEF2A, activating transcription factor 2 (ATF2), cAMP response element-binding protein (CREB), and forkhead box class-O (FOXO1) proteins regulate the transcription of PGC-1α by binding to the promoter region of PGC-1α (15). Examination of PGC-1α mRNA alone may not provide the complete story of PGC-1α and its role in mitochondrial development. The role of transcriptional regulation of PGC-1α mRNA may provide further insight to mechanisms of mitochondrial development. Examination of the binding of MEF2A, ATF2, CREB, and FOXO1 may help explain the paradoxical relationship between an increase in PGC-1α mRNA and a decrease in select transcription factors downstream of PGC-1α in the cold.

Exercise followed by a 3-h recovery period in a cold environment of 7°C increases PGC-1α mRNA above that of exercise and recovery at a normal room temperature environment of 20°C (45). Transcription factors downstream of PGC-1α associated with mitochondrial biogenesis are lower in cold environmental temperature compared with a room temperature (20°C) environment when coupled with exercise (45, 46). However, prolonged cold exposure implemented in these studies after the exercise bout may not be practical. Sitting in a cold environment for 3 h is associated with shivering and an increase in oxygen utilization, which may further elicit an increase in PGC-1α. This may not be a preferred method of treatment due to the associated thermal discomfort. In order for a temperature-optimized exercise protocol to be practical, recovery from exercise needs to occur in an environment conducive to activities of daily living, such as normal room temperature. It is not clear whether recovery in a normal room temperature environment after exercise in the cold will stimulate a similar response in the transcriptional response associated with mitochondrial development. Recovery in the cold following exercise in the cold may be necessary to elicit the previously observed cold-induced increases in the expression of PGC-1α due to increased oxygen consumption observed in the cold (45). The increase in total body oxygen consumption during cold exposure is small relative to the increase during exercise, and thus other mechanisms inherent to cold may be responsible.

The purpose of this study was to determine 1) PGC-1α transcriptional control and 2) the human mitochondrial-related gene expression response during exercise in a cold environment with recovery at room temperature. This study may provide the rationale behind the development of a temperature-optimized protocol for enhancing the exercise training response and for treating mitochondrial dysfunction. Additionally, utilizing a room temperature recovery period after cold exercise may provide insight in the response of transcription factors downstream of PGC-1α. During exercise, a cold-induced increase in PGC-1α mRNA may be possible without a decrease in downstream transcription factors. Furthermore, data from this study together with other previous results from our laboratory (45, 46) may establish necessity of recovery in a cold environment after exercise in the cold to alter markers of mitochondrial development.

METHODS

Study participants.

All testing and protocols were approved by the Institutional Review Board at the University of Nebraska at Omaha. A total of 11 recreationally active male subjects were recruited for this study. Recreationally active was determined by the risk stratification form from participating in at least moderate physical activity for 30 min/day and 3 days/wk. Participants ranged from 21 to 29 yr of age with no prior history of medical issues and were free from medications or health conditions that would prevent their completion of the research protocol. Participants signed the Institutional Review Board-approved informed consent form before participation in this study.

Descriptive data.

During the first testing session, descriptive data were recorded for all participants before the experimental protocol began. Height, weight, body composition, and aerobic capacity via peak volume of oxygen (V̇o_2peak) were assessed. Body density was measured by hydrodensitometry with correction for estimated residual lung volume and gastrointestinal air volume (36). Participants were instructed to completely submerge themselves in the hydrostatic weighing tank, expel the air in their lungs, and remain still for ∼3 s while their underwater weight was assessed using a calibrated electronic load cell-based system (Exertech, Dresbach, MN). Six to 10 trials were performed, and the three best trials were averaged and used to estimate percentage of body fat. Body density was converted to percent fat using the Siri equation (44).

Participants then performed a graded exercise test to assess volitional fatigue-limited peak aerobic capacity measured on an electronically braked cycle ergometer (Velotron; RacerMate, Seattle, WA) using a gas- and flow-calibrated metabolic cart (ParvoMedics TrueOne Metabolic System, Sandy, UT) to measure expired gases. The graded cycle test started at 95 W and increased by 35 W after every 3-min stage. The participants cycled until volitional fatigue, and the highest V̇o₂ obtained during the test was recorded as V̇o_2peak. Heart rate was monitored continuously during the course of the trial with a chest strap (Polar Electronic, Lake Success, NY). The results from the graded cycle protocol were used to establish intensity for the experimental trial sessions. Maximum workload associated with V̇o_2peak (W_max) was determined by adding the highest completed stage (in W) and the proportion of time in the last stage multiplied by the 35 W/stage increment. The intensity set for each participant during the experimental trials was 60% W_max. This intensity was selected so that the participants would be able to safely complete the 1-h cycling bout and to compare these results with previous investigations (45).

Experimental trials.

Participants each performed two experimental trials in a randomized and counterbalanced order, with a minimum of 4 days between trials. Participants kept a food log 24 h before and an exercise log 48 h before the first trial and replicated the same diet and exercise before their second trial. The participants arrived in a similar hydration state for each trial, as indicated by urine-specific gravity (cold trial: 1.021 ± 0.006; room temperature trial: 1.021 ± 0.006; P = 0.929). A core temperature capsule was ingested, and a skeletal muscle biopsy sample was obtained before the participants cycled on a Velotron cycle ergometer at 60% W_max for 1 h in a temperature- and humidity-controlled environmental chamber (Darwin Chambers, St. Louis, MO). The Velotron cycle ergometer maintains power within 1% and without drift during constant power trials (1). The temperatures for the two trials were set to 7°C and 60% humidity for the cold temperature trials and 20°C and 60% humidity for the room temperature trials. These temperatures have been determined to be safe for the participants to withstand for an extended time without unsafe increases or decreases in core body temperature at the given exercise intensity (45). Participants dressed in the same clothing for both trials and were blinded to the temperature until they were in the chamber to prevent a psychological stress-induced increase in core temperature (33). V̇o₂ was monitored with a calibrated metabolic cart during four time intervals each lasting 5 min. Rating of perceived exertion (RPE) was recorded at minute 3 of each respective gas collection stage using the Borg 20-point RPE scale (5). Heart rate was monitored continuously with an Eqn 02 LifeMonitor Sensor Belt and transmitted to an Eqn 02 SE every 5 s throughout the 1-h cycling trial. Participants were instructed to consume a total of 500 ml of water during the cycling trial. Approximately one-fourth (∼125 ml) of the water was consumed before each gas analysis. After the 1-h cycling bout, a second muscle biopsy was taken, and then the participants recovered in a room temperature environment for 3 h. The 3-h recovery was chosen because it is within the time course of peak PGC-1α expression (2–6 h), and peak expression of downstream transcription factors occurs shortly after PGC-1α (24, 51). After the 3-h recovery period, a final muscle biopsy sample was obtained.

Core temperature and skin temperature.

Core body temperature was monitored continuously and recorded during exercise and recovery. Participants ingested a Jonah Core Body Temperature Capsule (Hidalgo, Cambridge, UK) along with 125 ml of water and a granola bar (Fiber One; General Mills, Minneapolis, MN) to assist with moving the sensor through the stomach and into the small intestine (14). Exercise began 55 ± 2 min after ingestion of the capsule. The thermistor capsule transmitted information to the EQO2 LifeMonitor Sensor Electronics Module (Hidalgo) and recorded the core temperature every 5 s throughout the trial. Skin temperature was monitored using an infrared sensor on the Eqn 02 LifeMonitor Sensor Belt (Hidalgo). The skin temperature was transmitted to the EQO2 SE every 15 s throughout the experimental trial.

Whole body metabolism.

Expired gases were analyzed during exercise and recovery to determine V̇o₂. Gases were collected in 5-min increments four times during exercise. The collection occurred in 5-min increments at minutes 10–15, 25–30, 40–45, and 55–60 of exercise. During recovery, gases were collected during minutes 25–30, 85–90, and 145–150. The last 3 min of each of the four collections during exercise and three collections during recovery were averaged to represent exercise and recovery, respectively.

Muscle biopsies.

Three muscle biopsies were taken during each experimental trial. The biopsies were obtained from the vastus lateralis using the percutaneous needle biopsy technique with suction (4) preexercise, immediately postexercise, and 3 h after recovery. Each serial sample for a given trial was taken from the same leg 2 cm proximal to the previous site. The order of the leg biopsied was randomized, and the opposite leg was biopsied in the subsequent trial. Samples were cleared of excess blood, separated from connective tissue, and portioned into RNAlater (Qiagen, Valencia, CA) and flash-frozen samples. The RNAlater samples were stored for 24 h at 4°C and then stored at −80°C for later analysis of mRNA. The flash-frozen samples were stored immediately at −80°C for later analysis of chromatin immunoprecipitation.

Chromatin immunoprecipitation.

Binding of CREB, ATF2, MEF2A, and FOXO1 to the PGC-1α promoter region was assessed by chromatin immunoprecipitation (ChIP) assay. Because of limitations to the amount of sample available from muscle biopsies, the samples from nine participants for ChIP assay were analyzed using a commercially available EpiQuick Tissue Chromatin Immunoprecipitation Kit (Epigentek, Farmingdale, NY) according to the manufacturer’s instructions. Briefly, protein-DNA interactions were fixed by cross-linking ∼20 mg of skeletal muscle with 1% formaldehyde. Samples were then homogenized with an electric tissue disruptor (PowerGen 125; Fisher Scientific, Hampton, NH), and DNA was sheared by sonication. Samples were sonicated with four pulses of 15–20 s with a Q55 sonicator (Qsonica, Newtown, CT) set to an amplitude of 40% output with ≥1 min on ice between pulses. This protocol resulted in average DNA fragment sizes of 400–1,200 base pairs. Fragment sizes were analyzed using an Agilent DNA 7500 Kit (Agilent Technologies, Santa Clara, CA) according to the manufacturer instructions and quantified on a 2100 Bioanalyzer (Agilent Technologies). Protein-bound DNA was immunoprecipitated using normal mouse IgG (negative control), anti-RNA polymerase II (positive control), or anti-CREB (cat. no. sc-377154X), -ATF2 (cat. no. sc-187x), -MEF2A (cat. no. sc-313X), or -FOXO1 (cat. no. sc-11350X) antibodies (all antibodies from Santa Cruz Biotechnology). All antibodies used for ChIP have been validated previously (47, 50). The antibody-protein-DNA complex was affinity purified using a protein A/G agarose. Cross-linking was reversed using proteinase K, and DNA was purified using spin columns with 90% ethanol. Purified, immunoprecipitated DNA was quantified using real-time reverse transcriptase polymerase chain reaction [quantitative (q)RT-PCR] as outlined below. Custom probe/primer pairs were designed for putative binding sites of each transcription factor within the PGC-1α promoter region (PrimeTime qPCR assay; Integrated DNA Technologies, Coralville, IA) as well as GAPDH to quantify input DNA. Raw PCR data were normalized using the percent input method (35), where recovered immunoprecipitated DNA is expressed as a percentage of total DNA input into the reaction.

Gene expression.

Muscle samples were analyzed using qRT-PCR to quantify the mRNA content. Approximately 20 mg of skeletal muscle was homogenized in 800 µl of TRIzol reagent (Invitrogen, Carlsbad CA) using an electric homogenizer (PowerGen 125, Fisher Scientific). Samples were incubated at room temperature for 5 min, and 160 µl of chloroform per 800 µl of TRIzol was added and shaken by hand. After another incubation at room temperature for 2–3 min, the samples were centrifuged at 12,000 g for 15 min, and the aqueous phase was transferred to a fresh tube. Then, 400 µl of isopropyl alcohol were added and incubated overnight at −20°C to precipitate mRNA. The samples were then centrifuged at 12,000 g for 10 min at 4°C, and the mRNA was washed by removing the supernatant and adding 800 µl of 75% ethanol. Samples were vortexed and then centrifuged at 7,500 g for 5 min at 4°C. The supernatant was removed and the mRNA pellet dried. Next, mRNA was dissolved in 100 µl of RNase-free water after RNA was further purified using the RNeasy mini kit (Qiagen) according to the manufacturer’s protocol, using the additional DNase digestion step (RNase-free DNase set, Qiagen). RNA was quantified using a spectrophotometer (Nanodrop ND-2000; Thermo Scientific, Wilmington, DE). First-strand cDNA synthesis was achieved using a Superscript-first-strand synthesis system kit for RT-PCR (Invitrogen) according to the manufacturer’s protocol. Each sample within a given subject was adjusted to contain the same concentration of cDNA (5 ng/μl). For real-time RT-PCR, each 20-µl reaction volume contained 1 μl of probe and primer mix (PrimeTime qPCR assay; Integrated DNA Technologies), 10 μl of Brilliant III Ultra-Fast QPCR Master Mix (Agilent Technologies), 0.3 μl of dye mixture, 6.2 μl of water, and 2.5 μl of sample cDNA. PCR was then run in triplicate using the Agilent Technologies Stratagene Mx3005p real-time PCR detection system (Agilent Technologies) with a fast two-step protocol (1 cycle at 95°C for 3 min, followed by 40 cycles of 95°C for 5 s and 60°C for 20 s).

Quantification of mRNA for genes of interest was performed for preexercise and 3-h recovery of muscle samples using the $2^{{}^{–\Delta \Delta C_T}}$ method (27). The optimal number and most stable reference genes were analyzed using geNorm software (Biogazalle, Ghent, Belgium) for each participant. If multiple reference genes were selected, the geometric mean of these stable reference genes was calculated. The candidate reference genes used were ribosomal protein 18 (RPS18), cyclophilin (CYC), β₂-microglobulin, or a combination of these genes. Genes of interest for this study were PGC-1α, ERRα, NRF-1, NRF-2, mitochrondrial transcription factor A (TFAM), VEGF, and MEF2A. Probes and primers were designed and obtained from Integrated DNA Technologies. Genes of interest were measured at 3 h of recovery and normalized to the reference genes as well as to the preexercise condition.

Statistical analysis.

Transcription factor binding to DNA, mRNA, and protein quantification was analyzed with a repeated-measures two-way ANOVA (time × trial). If significance was found, a Fisher’s protected least significant difference post hoc test was used to evaluate where significance occurred. Expired gases, heart rate, core temperature, skin temperature, and RPE were analyzed using a paired t-test. A probability of type I error of <5% was considered significant (P < 0.05). All statistical data were analyzed using the Statistical Package for Social Sciences (Chicago, IL) software (SPSS 24.0). It was calculated that a sample size of 14 total subjects would be sufficient to detect a mean difference in PGC-1α content assuming a standard deviation of three, a statistical power of 80%, and a medium effect size.

RESULTS

Characteristics of the 11 recreationally trained males who participated in this study are shown in Table 1. During exercise, both V̇o₂ (P = 0.012) and skin temperature (P < 0.001) were lower in the cold than in the room temperature trial, whereas core temperature was not different (P = 0.550). Respiratory exchange ratio (RER), heart rate (HR), and rating of perceived exertion (RPE) during exercise were not different between the two conditions (P > 0.05; Table 2). During recovery, V̇o₂, RER, HR, skin temperature, and core temperature were not different between cold and normal room temperature trials (P > 0.05; Table 2).

Table 1.

Subject characteristics

Characteristic	Value
Age, yr	24 ± 1
Height, cm	178 ± 1
Weight, kg	80.3 ± 3.7
Body fat, %	14.6 ± 1.0
V̇o2peak, l/min	4.34 ± 0.24
V̇o2peak ml·kg−1·min−1	54.6 ± 3.3
Power at V̇o2peak, W	275 ± 12

Data are means ± SE (n = 11). V̇o_2peak, peak volume of oxygen consumption.

Table 2.

Study parameters measured during exercise

	RT	Cold
Exercise
V̇o2, l/min	2.80 ± 0.11	2.67 ± 0.36*
V̇o2, ml·kg−1·min−1	35.8 ± 1.5	34.4 ± 1.5*
RER	0.94 ± 0.01	0.94 ± 0.01
Heart rate, beats/min	155 ± 3	152 ± 2
RPE	12 ± 0.3	12 ± 0.2
Core temperature, °C	37.8 ± 0.1	37.9 ± 0.1
Skin temperature, °C	32.3 ± 0.1	27.4 ± 0.1*
Recovery
V̇o2, l/min	0.03 ± 0.01	0.03 ± 0.02
V̇o2, ml·kg−1·min−1	4.4 ± 0.1	4.3 ± 0.1
RER	0.81 ± 0.3	0.78 ± 0.2
HR, beats/min	76 ± 3	74 ± 3
Core temperature, °C	37.2 ± 0.1	37.0 ± 0.1
Skin temperature, °C	34.4 ± 0.3	33.8 ± 0.2

Data are means ± SE (n = 11). Values represent the 1-h average during exercise and 3-h average during supine recovery. HR, heart rate; RER, respiratory exchange ratio; RPE, rating of perceived exertion; RT, room temperature; V̇o₂, volume of oxygen consumption.

^*P < 0.05 from RT.

Transcription factor binding.

The binding of CREB, ATF2, MEF2A, and FOXO1 to the promoter region of PGC-1α was not different between conditions (Fig. 1). CREB, MEF2A, and FOXO1 DNA binding increased postexercise compared with preexercise conditions (P = 0.021, P = 0.018, and P = 0.034, respectively). ATF2 binding to DNA did not significantly increase postexercise compared with preexercise (P = 0.068), although a trend was noted. CREB, ATF2, and FOXO1 binding were also higher at 3 h of recovery compared with preexercise (P = 0.007, P = 0.019, and P = 0.013, respectively). MEF2A binding at 3 h of recovery was not significantly higher than preexercise (P = 0.052). CREB binding continued to increase from postexercise to 3 h into recovery (P = 0.029). ATF2, MEF2A, and FOXO1 binding to DNA did not change from postexercise to 3 h of recovery (Fig. 1).

Fig. 1.

Transcription factor binding at the promoter region of peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α). Chromatin immunoprecipitation assay was performed to determine transcription factor binding to the PGC-1α promoter region. Data are expressed as means ± SE. *P < 0.05 from preexercise; †P < 0.05 from postexercise. CREB, cAMP response element-binding protein; ATF2, activating transcription factor 2; MEF2A, myocyte enhancer factor-2; FOXO1, forkhead box class O1. Pre, preexercise; post, postexercise.

Gene expression.

PGC-1α and VEGF increased from preexercise to 3 h of recovery (P < 0.001 and P < 0.001, respectively) but were not different between the two temperature conditions (Fig. 2). ERRα, NRF2, and MEF2A were lower in the cold trial than in the room temperature trial at 3 h of recovery (P = 0.003, P = 0.037, and P = 0.034, respectively). However, they were unaffected from preexercise to 3 h of recovery. NRF-1 and TFAM did not change with from preexercise to 3 h of recovery and were not different between the cold and room temperature conditions (Fig. 2). The trials were conducted in a randomized, counterbalanced order, and no order effect was observed between the first and second trial for any gene (P > 0.05).

Fig. 2.

mRNA response at 3 h of recovery normalized to preexercise conditions. To determine mRNA response, quantitative RT-PCR was performed with preexercise and 3-h recovery samples. Data are expressed as means ± SE. *P < 0.05 from preexercise; †P < 0.05 from room temperature (RT) trial. PGC-1α, peroxisome proliferator-activated receptor-γ coactivator 1α; ERRα, estrogen-related receptor-α; NRF2, nuclear respiratory factor-2; MEF2A, myocyte enhancer factor-2; NRF-1, nuclear respiratory factor-1; TFAM, mitochondrial transcription factor A; VEGF, vascular endothelial growth factor.

DISCUSSION

Mitochondrial biogenesis is stimulated with exercise and cold exposure in animal models (12, 43). Developing a temperature-optimized exercise protocol in humans is of great interest, because it could provide an approach to stimulate mitochondrial biogenesis and thus treat and prevent mitochondrial dysfunction. The potential additive or synergistic effect of exercise and cold exposure may allow for exercise to take place at a lower intensity but elicit the effects of higher-intensity exercise. The main finding of this study was that exercise in the cold with subsequent recovery in moderate room temperature led to a decreased response in mRNA expression of select mitochondrial-related genes. Specifically, ERRα, NRF2, and MEF2A decreased compared with exercise in a room temperature environment, whereas transcription mechanisms were unchanged. The findings of ERRα, NRF2, and MEF2A being lower in the cold seem contradictory in gene expression and may indicate an alternative to normal PGC-1α signaling that promotes mitochondrial development with exercise.

PGC-1α.

PGC-1α is a transcriptional coactivator responsible for activating a myriad of transcription factors. PGC-1α transcription is regulated by MEF2A, FOXO1, CREB, and ATF2 binding to the PGC-1α promoter region on DNA. No difference in the binding of these four transcription factors occurred between the two trials, which would suggest that the mRNA of PGC-1α would also not be different. Indeed, PGC-1α mRNA was not different in the cold trial compared with the room temperature trial. When this finding is considered with our previous studies that have shown an enhanced PGC-1α mRNA response to exercise in the cold when a 3-h recovery period also takes place in the cold, it is likely that exercise-induced PGC-1α gene expression is dependent on cold exposure beyond 1 h of exercise (45, 46). In essence, recovery in the cold may be the stimulus needed after exercise to elicit a further increase in PGC-1α mRNA. The two previous studies utilized exercise, one at room temperature and the other in the cold, and had a 3- to 4-h cold recovery period to further drive PGC-1α gene expression. The current study explored the possibility that cold recovery is not required to further drive an increase in PGC-1α, but our data indicate that this is not the case. The use of a similar time period of room temperature recovery did not increase PGC-1α mRNA in the cold exercise trial compared with the room temperature trial. Additionally, exposure to cold environmental temperature without physical activity is not an adequate stimulus for increasing PGC-1α mRNA (43, 52). Exposure to 3 h of cold temperature without exercise does not increase PGC-1α compared with a 3-h exposure to standard room temperature. Thus, physical activity appears to be a prerequisite stimulus to cold exposure.

MEF2.

MEF2 is coactivated by PGC-1α and is responsible for binding to the promoter region of PGC-1α for the expression of PGC-1α mRNA. In this way, PGC-1α is responsible for transcribing itself in a feed-forward loop (11, 13, 15, 41). MEF2 has a role in the utilization of glycogen, muscle fiber type determination, and muscle size and growth. MEF2-knockout mice have increased amounts of glycogen in skeletal muscle after exercise compared with control mice and smaller overall body size (2). MEF2-knockout mice also have disruptions in sarcomere assembly and function, reduction in slow-twitch muscle fibers, and perinatal lethality (35). In the current study, the decrease in MEF2A mRNA in the cold, whereas PGC-1α was not different from room temperature, is counterintuitive. The blunting of MEF2A in the cold would be expected to produce decreased use of glycogen during exercise and to reduce the conversion to slow-twitch muscle fibers that would indicate a reduction in adaptation to endurance exercise. However, glycogen utilization is increased in the cold at rest (19), but during exercise in cold temperature there are conflicting reports with glycogen utilization (19, 21, 22). Additionally, there is an increased amount of slow-twitch muscle fibers found in cold water fish compared with the same species in warmer water (32). This may further indicate that PCG-1α is not the primary regulator of this transcription factor during exercise or recovery from exercise in a cold environment. Additionally, the half-life of MEF2 protein can be up to 8 h or longer, so a longer time point may need examination to the overall effect of the cold response on this marker to be determined (3).

ERRα and VEGF.

PGC-1α protein induces ERRα and as levels of PGC-1α mRNA increase it would be expected that ERRα mRNA would increase as well (42). The observed reduction in ERRα mRNA after exercise in the cold does not follow the pattern of PGC-1α mRNA induction, which may suggest a different mechanism of action during cold exposure. The observed ERRα mRNA reduction is in agreement with previous research that exercise and recovery in standard room temperature conditions increased ERRα gene expression and exercise followed by recovery in a cold environment decreased ERRα (46). Decreased ERRα levels as a result of cold exposure may be detrimental to mitochondrial biogenesis and the expected overall exercise response. For instance, mice lacking ERRα protein display decreased mitochondrial mass, reduced oxidative capacity, and impaired ability in regulating temperature in the cold (48). However, reduced ERRα gene expression may not be indicative of the intracellular ERRα protein demand required to maintain mitochondrial biogenesis. In cardiomyocytes, ERRα and PGC-1α protein regulate many of the same genes and PGC-1α expression is dependent on ERRα (37). The paradoxical findings between mRNA levels of PGC-1α and ERRα should be studied further as well as protein levels of each to determine the impact on mitochondrial biogenesis beyond transcription.

VEGF is an angiogenic factor whose induction is mediated by ERRα (10). VEGF also has a role in regulating several genes related to mitochondrial biogenesis, particularly stimulation of mitochondrial import proteins (49). The present data show an increase in VEGF mRNA during both the cold and the room temperature trials. VEGF activates Akt3, which then controls the nuclear localization of PGC-1α to stimulate mitochondrial biogenesis (49). Therefore, a similar increase in VEGF mRNA from both trials in the current study suggests that angiogenesis and mitochondrial biogenesis are stimulated from a cell-signaling standpoint independent of PGC-1α transcription. Additionally, a paradox between ERRα mRNA and VEGF mRNA is apparent. ERRα mRNA decreased in the cold compared with room temperature, whereas VEGF mRNA was not different. This suggests that other potential mechanisms are regulating VEGF mRNA. In addition to stimulating mitochondrial biogenesis, VEGF has a role in angiogenesis through the regulation of hypoxia-inducible factor (HIF-1) during hypoxic conditions (20). In the present study, VEGF may have been upregulated by HIF-1 during the cycling protocol due to a decreased oxygen tension within the skeletal muscle (18, 38, 39).

NRFs and TFAM.

NRF-1 and NRF-2 are coactivated by PGC-1α and are responsible for promoting the transcription of TFAM. TFAM is highly involved in mitochondrial genome transcription and replication. NRF-2 mRNA was lower in the cold trial than in the room temperature trial after 3 h of recovery. This NRF-2 response is similar to previous studies concerning differing environmental temperatures (46). In NRF-2-knockout mice, mitochondrial respiration and fatty acid oxidation are depressed, whereas these activities are high in transgenic mice with overexpressed NRF-2 (29). Thus, the decrease in NRF-2 mRNA in the cold observed in the present study suggests that cold may not be a viable intervention in promoting mitochondrial biogenesis. Additionally, NRF-2 may not be cold dependent but rather intensity dependent, which would explain the decreased NRF-2, since V̇o₂ was lower in the cold. NRF-1 mRNA was not different between the cold and room temperature trials. This is similar to what has been shown in previous studies (31, 34, 46). TFAM mRNA did not change at the 3-h recovery time point and may require a longer time course of 6 h before expression in human skeletal muscle occurs (34), although an even longer time may be required for peak expression (31). NRF-1 is coactivated by PGC-1α and is a transcriptional activator of TFAM, both of which have key roles in mitochondrial biogenesis (40). It has been suggested that some downstream transcription factors such as NRF-1 lack increased mRNA gene expression because the basal levels of the transcription factors proteins are adequate and no further increase in these proteins is needed (9). The relationship of NRF protein and the gene expression should be examined further to determine how their respective levels influence mitochondrial biogenesis.

Oxygen consumption.

Oxygen utilization is elevated during recovery in cold temperatures compared with recovery in room temperature after exercise (45, 46). It was postulated that this increased V̇o₂ during recovery provided an extra stimulus for increased gene expression of PGC-1α through shivering and nonshivering thermogenesis. In the current study, V̇o₂ during exercise was higher in the room temperature trial despite a constant external workload during the trial. Thus, a greater increase in PGC-1α mRNA in the room temperature trial may be expected due to the increase in energy expenditure, but this was not the case. This indicates that factors other than exercise intensity and V̇o₂ may contribute to specific transcription factor gene expression. Alternatively, the lower V̇o₂ during the cold may provide insight on the PGC-1α mRNA expression in the current study. PGC-1α mRNA expression may be explained by the contractile activity from the intensity of exercise (26). Because the V̇o₂ was lower in the cold, indicating a lower metabolic rate, the intensity may not have been the same between the trials. This may have occurred if pedal cadence was not the same between trials, and thus a greater stimulus during the room temperature trial may have been provided (16). The lower V̇o₂ in the cold may also provide an explanation as to why ERRα, NRF2, and MEF2A were decreased in the cold trial compared with the room trial, although this lowered gene expression has also been observed in cold temperature after moderate room temperature exercise with ERRα and NRF2 (46).

During recovery, no differences were noted in heart rate, core temperature, or V̇o₂ between the two trials. The similarities between trials in these markers during recovery suggest that factors other than enhanced metabolic rate during recovery inherent to cold elicit the decreased mRNA response of ERRα, NRF2, and MEF2A compared with the room temperature trial. Furthermore, the difference in gene expression occurred despite no difference in core temperature between the trials. Because of the discomfort of intramuscular probes during cycling, intramuscular temperature was not taken during the trials, which may be a limitation in explaining differences in gene expression. To observe a difference in core temperature while exercising, it may be necessary to decrease the environmental temperature to a greater extent. Factors other than core temperature appear to have influenced the alteration in gene expression between trials. The observed difference in skin temperature between trials may play a role in the change in blood flow from the periphery to the core as well as from skin surface to deeper tissues such as skeletal muscle (23). The change in skin temperature, blood flow distribution, and subsequent metabolic activity may be a contributing factor to the altered PGC-1α mRNA response that has been observed in select mitochondrial temperature studies.

Considerations.

In the current research project, protein levels were not measured and should be examined in future studies to explore the impact that cold-induced changes have on the overall impact of mitochondrial development. Additionally, this study consisted of an acute bout of aerobic exercise with a 3-h time point. The effects of training in room temperature and cold should be considered in future studies. MEF2 has several isoforms that should be examined in future studies for a more comprehensive approach on the impact of cold exposure to mitochondrial development. This study was a proof of concept in an attempt to provide insight on the potential changes of cold on mitochondrial development. Additionally, the 1-h cycling protocol could be too difficult for those in a diseased condition. Therefore, healthy participants were used as opposed to older individuals and those with mitochondrial dysfunction. This study provides preliminary insight in cold treatment related to mitochondrial dysfunction. Only male participants were used in this study, and potential differences in PGC-1α expression with temperature and exercise should be examined between males and females.

Perspectives and Significance

The present study demonstrates an altered gene response in skeletal muscle after an acute bout of exercise in a cold environmental temperature compared with room temperature in humans. Specifically, our data show that exercise in the cold decreases gene expression of select transcription factors related to mitochondrial biogenesis without a difference in PGC-1α signaling or response. Although these findings provide a framework for optimizing exercise protocols in humans, more research is needed to determine how these alterations positively or negatively affect mitochondrial development as opposed to signaling or markers related to mitochondrial development. The cumulative responses of the selected transcription factors in the current study show that cold may not be a viable option for therapy in stimulating mitochondrial biogenesis and may be detrimental. Additionally, when taken with previous data from our laboratory, cold environmental recovery after exercise may be needed to further increase PGC-1α mRNA response. We still do not fully understand the mechanism of the altered response in a cold condition when most parameters (i.e., HR, core temperature, and workload) are held constant, and therefore, further research should be conducted to explore the impact and mechanism of the altered mRNA response on mitochondrial biogenesis.

GRANTS

This publication was made possible by grants from the National Institute for General Medical Science (NIGMS; P20GM103427), a component of the National Institutes of Health (NIH), and its contents are the sole responsibility of the authors and do not necessarily represent the official views of NIGMS or NIH. Additional funding for this project was provided by the University of Nebraska at Omaha Graduate Research and Creative Activity Grant (UNO GRACA).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.J.S., M.W.H., J.L.K., and D.R.S. conceived and designed research; R.J.S., M.W.H., J.L.K., and D.R.S. performed experiments; R.J.S., M.W.H., J.L.K., and D.R.S. analyzed data; R.J.S., M.W.H., R.B.Z., J.L.K., and D.R.S. interpreted results of experiments; R.J.S., M.W.H., R.B.Z., and D.R.S. prepared figures; R.J.S., M.W.H., R.B.Z., and D.R.S. drafted manuscript; R.J.S., M.W.H., R.B.Z., J.L.K., and D.R.S. edited and revised manuscript; R.J.S., M.W.H., R.B.Z., J.L.K., and D.R.S. approved final version of manuscript.