282

ARTICLE Fuel selection during short-term submaximal treadmill exercise in the cold is not affected by pre-exercise low-intensity shivering Dominique D. Gagnon, Hannu Rintamäki, Sheila S. Gagnon, Juha Oksa, Katja Porvari, Stephen S. Cheung, Karl-Heinz Herzig, and Heikki Kyröläinen

Abstract: Exercise and shivering rely on different metabolic pathways and consequently, fuel selection. The present study examined the effects of a pre-exercise low-intensity shivering protocol on fuel selection during submaximal exercise in a cold environment. Nine male subjects exercised 4 times for 60 min at 50% (LOW) or 70% (MOD) of their peak oxygen consumption on a motorized treadmill in a climatic chamber set at 0 °C with (SHIV) and without (CON) a pre-exercise cooling protocol, inducing low-intensity shivering. Thermal, cardiorespiratory and metabolic responses were measured every 15 min whereas blood samples were collected every 30 min to assess serum nonesterified fatty acids (NEFA), glycerol, glucose, ␤-hydroxybutyrate (BHB) and plasma catecholamine concentrations. Rectal and skin temperatures were lower in the SHIV condition, within LOW and MOD conditions, during the first 45 min of exercise. Norepinephrine (NE) concentration was greater in SHIV vs. CON within LOW (1.39 ± 0.17 vs. 0.98 ± 0.17 ng·mL−1) and MOD (1.50 ± 0.20 vs. 1.01 ± 0.09 ng·mL−1), whereas NEFA, glycerol and BHB were greater in SHIV vs. CON (1060 ± 49 vs. 898 ± 78 ␮mol·L−1; 0.27 ± 0.02 vs. 0.22 ± 0.03 mmol·L−1; 0.39 ± 0.06 vs. 0.27 ± 0.04 mmol·L−1, respectively) within MOD only. No changes were observed in fat or carbohydrate oxidation between SHIV and CON during exercise. Despite increases in NE, NEFA, glycerol and BHB from pre-exercise low-intensity shivering, fuel selection during short-term submaximal exercise in the cold was unaltered. Key words: energy metabolism, cold, shivering, submaximal exercise, carbohydrates, fat. Résumé : L’exercice physique et le frissonnement empruntent différentes voies métaboliques et, par conséquent, sélectionnent différents carburants. Dans cette étude, on utilise un protocole de frissonnement léger préalable a` un exercice physique pour analyser son effet sur la sélection de carburant au cours d’un exercice sous-maximal dans le froid. Neuf hommes effectuent quatre fois 60 min d’exercice a` une intensité sollicitant 50% (« LOW ») ou 70% (« MOD ») de leur consommation d’oxygène de pointe sur un tapis roulant dans une chambre climatique réglée a` 0 °C incorporant (« SHIV ») ou pas (« CON ») un protocole préexercice de refroidissement suscitant un frissonnement léger. Toutes les 15 min, on enregistre des réponses thermiques, cardiorespiratoires et métaboliques et, toutes les 30 min, on prélève des échantillons de sang pour analyser la concentration sérique d’acides gras non estérifiés (NEFA), de glycérol, de glucose, de ␤-hydroxybutyrate (« BHB ») et la concentration plasmatique de catécholamines. On observe une plus faible température rectale et cutanée dans la condition SHIV (LOW et MOD) au cours des 45 premières minutes d’exercice. On observe une plus grande concentration de norépinéphrine (« NE ») en SHIV comparativement a` CON dans les conditions LOW (1,39 ± 0,17 vs 0,98 ± 0,17 ng·mL−1) et MOD (1,50 ± 0,20 vs 1,01 ± 0,09 ng·mL−1); en revanche, les concentrations de NEFA, de glycérol et de BHB sont plus élevées en SHIV comparativement a` CON (1060 ± 49 vs 898 ± 78 ␮mol·L−1; 0,27 ± 0,02 vs 0,22 ± 0,03 mmol·L−1; 0,39 ± 0,06 vs 0,27 ± 0,04 mmol·L−1, respectivement) dans la condition MOD seulement. On n’observe pas de modification de l’oxydation des gras et des sucres durant l’exercice physique en SHIV comparativement a` CON. Même si on observe une augmentation des NE, NEFA, glycérol et BHB durant le frissonnement léger préalable a` l’exercice physique, on n’observe pas de modification du choix de carburant durant un exercice sous-maximal de courte durée dans le froid. [Traduit par la Rédaction] Mots-clés : métabolisme énergétique, froid, frissonnement, exercice sous-maximal, sucres, gras.

Received 13 February 2013. Accepted 21 August 2013. D.D. Gagnon and H. Kyröläinen. Department of Biology of Physical Activity, P.O: Box 35, University of Jyväskylä, Jyväskylä FI-40014, Finland. H. Rintamäki. Finnish Institute of Occupational Health, Oulu, Finland; Institute of Biomedicine, Department of Physiology, University of Oulu, Oulu, Finland. S.S. Gagnon. Department of Health and Rehabilitation Sciences, Western University, London, ON N6G 1H1, Canada. J. Oksa. Institute of Biomedicine, Department of Physiology, University of Oulu, Oulu, Finland. K. Porvari. Institute of Diagnostics, Department of Forensic Medicine, University of Oulu, Oulu, Finland. S.S. Cheung. Department of Kinesiology, Brock University, St. Catharines, ON L2S 3A1, Canada. K.-H. Herzig. Finnish Institute of Occupational Health, Oulu, Finland; Department of Psychiatry, Kuopio University Hospital, Kuopio, Finland; Biocenter of Oulu, Oulu, Finland. Corresponding author: Dominique D. Gagnon (e-mail: dominique.gagnon@jyu.fi, [email protected]). Appl. Physiol. Nutr. Metab. 39: 282–291 (2014) dx.doi.org/10.1139/apnm-2013-0061

Published at www.nrcresearchpress.com/apnm on 16 September 2013.

Gagnon et al.

Introduction Muscle activation during shivering relies on muscle fibre recruitment pathways that greatly differ from exercise (Tipton et al. 1997; Haman et al. 2004a, 2004b, 2005). During exercise, the contribution of carbohydrates (CHO) and fat to the energy yield is approximately the same at 50% of peak oxygen consumption (V˙O2peak) and tend to rely more on CHO as exercise intensity increases (van Loon et al. 2001; Venables et al. 2005). However, with resting shivering, this phenomenon occurs at ⬃20% of V˙O2peak with a greater reliance on CHO as shivering intensifies (Tipton et al. 1997; Haman et al. 2007). Exercise and shivering may coexist depending on the degree of body cooling and exercise intensity (Nadel et al. 1973; Hong and Nadel 1979; Meigal et al. 1998; Weller et al. 1997). During aerobic exercise, the additional presence of shivering increases metabolic demand, and consequently oxygen consumption (V˙O2). Hong and Nadel (1979) observed an increase in thermoregulatory V˙O2 (i.e., total V˙O2 minus exercise-related V˙O2 obtained in 25 °C) of 250– 500 mL·min−1 from additional shivering during cycling in the cold. Weller et al. (1997) further demonstrated an increase of 6.3% in maximal oxygen consumption (V˙O2max) from shivering during walking in the cold (5 °C vs. 15 °C), which was also accompanied by higher respiratory exchange ratio (RER). In contrast, Kruk et al. (1991) observed a greater V˙O2 at rest at 5 °C compared with 24 °C because of shivering, prior to exercise bouts, but a lower RER during subsequent exercise. Compared with a thermoneutral environment, Hurley and Haymes (1982) also observed a lower RER in the cold following 30 min of passive cold exposure followed by 60 min of exercise. These previous reports have provided valuable insights on the effects of shivering on exercise metabolism but the results on energy use remain difficult to interpret as various environmental conditions (cold, cool, thermoneutral, warm) were used, which is known to influence fuel selection (Gagnon et al. 2013; Layden et al. 2002). Moreover, these studies have superimposed shiveringassociated thermoregulatory V˙O2 over the metabolic V˙O2 requirements, leading to an increase in total V˙O2 at a given workload. This would mask the precise contribution of shivering itself on fuel selection as increased V˙O2 is directly linked to a greater CHO reliance and would be unrealistic of natural exercise behavior (i.e., maintaining the same workload despite an increase in both V˙O2 and in rate of perceived exertion) (Lander et al. 2009). Finally, some studies have used a pre-determined cooling time period (e.g., 30 min cooling) prior to exercise. As fuel selection is V˙O2 dependent, it would be more useful to standardize the metabolic or thermoregulatory state of subjects prior to exercise, which can greatly vary in time due to the recognized effects of morphological characteristics of individuals on shivering onset and intensity (i.e., body fat, body surface area, resting metabolic rate, fitness level, muscle fiber distribution) (Eyolfson et al. 2001; Xu et al. 2005). The effects of shivering prior to exercise on fuel selection is therefore difficult to determine when environmental conditions, V˙O2 and shivering state are not standardized across subjects during thermal and exercise challenges as these all modulate fuel oxidation. In the present study, the effects of pre-exercise low-intensity shivering on CHO and fat utilization O determined by RER, absolute substrates oxidation and relative substrates contribution to the energy yield O were investigated during short-term submaximal treadmill exercise in the cold (0 °C), while total V˙O2 was clamped to nullify the effects of increased V˙O2. In addition, the possible effects of pre-exercise shivering, if any, on fuel selection between submaximal exercise intensities of 50% and 70% V˙O2peak were determined. Based on previous literature that indicated the preferential use of fat during low-intensity shivering, we hypothesized that (i) combined low-intensity shivering and exercise would increase lipids as an energy source as opposed to exercise

283

alone, and (ii) that both exercise intensities would generate similar fuel selection responses.

Materials and methods Participants Eleven male participants initially volunteered for the study, from which data from 9 were retained. One participant terminated his participation at mid-point during the study and severe vasoconstriction from the cold in another prevented sufficient blood sampling for data collection. The participants were moderately active and not cold acclimatized as the study was conducted during the months of May and June. They each provided written informed consent and were screened with a Physical Activity Readiness Questionnaire form and for cardiovascular and respiratory conditions that could be aggravated by cold air exposure or exercise. Mean (±SE) characteristics of the participants were determined during a familiarization trial and were as follows: age, 24 ± 1 years; height, 181.4 ± 3.4 cm; body mass, 83.1 ± 3.2 kg; body fat, 19.1% ± 1.6%; body surface area, 2.04 ± 0.06 m2; V˙O2peak, 52.9 ± 1.7 mL·kg−1·min−1; and shivering peak intensity (Shivpeak), 20.9 ± 0.8 mL O2·kg−1·min−1. The study was performed according to the declaration of Helsinki and was approved by the Ethical Committee of the Central Finland Health Care District. Experimental protocol Each participant took part in 4 experimental sessions, wearing shorts and t-shirt (clothing equivalent of ⬃0.2 to 0.3 clo), separated by at least 72 h and at the same time of day to control for circadian rhythms. They arrived at the laboratory between 0700 and 0800 h, preceded by a 24-h period without alcohol, caffeine, tobacco and vigorous exercise. The participants were requested to arrive at the laboratory in a fasted state to avoid loading of their glycogen stores, a determinant factor in fuel selection during passive shivering (Haman et al. 2004b). They were also instructed to record their dietary and fluid intake 24 h before the first session and to keep the same nutritional guidelines for each day preceding a session. Water was available ad libitum before all sessions. After instrumentation, the participants sat for a baseline period of 15 min in a climatic chamber set at 25.0 ± 0.2 °C, 40% relative humidity (RH) and 0.2 m·s−1 wind speed. Thereafter, the participants moved to the adjacent experimental chamber set at 0.0 ± 0.2 °C, 40% RH and 0.2 m·s−1 wind speed with cold air as a cooling medium. Then, they either immediately started exercising (control: CON), or remained seated until V˙O2 from shivering activity averaged 40% of Shivpeak (SHIV: see eq. 1 below), over a period of 5 min, and then started to exercise. The low-intensity shivering was chosen as moderate- and high-intensity shivering may impede neuromuscular function and prevent task completion such that combined exercise and shivering activity would typically only occur during low-intensity shivering only. Exercise comprised 60 min of treadmill exercise at a grade of 1.0% (Tunturi T40, Accell Group, Heerenveen, the Netherlands). After the exercise session, the subject was moved back to the instrumentation chamber for a recovery period of 30 min in a sitting position. Shivpeak was determined from the formula by Eyolfson et al. (2001), which has previously been used (Haman et al. 2004b; Xu et al. 2005): (1)

Shivpeak ⫽ 30.5 ⫹ 0.348 × V˙O2max ⫺ 0.909·BMI ⫺ 0.233 age

where Shivpeak and V˙O2max are in mL O2·kg−1·min−1, body mass index (BMI) is in kg·m−2, and age is in years. Experimental sessions followed a balanced design and involved the following conditions: (i) exercise at 50% V˙O2peak without shivering (low-intensity exercise; LOW CON); (ii) exercise at 50% V˙O2peak after shivering (LOW SHIV); (iii) exercise at 70% V˙O2peak without shivering (moderate-intensity exercise; MOD CON); and Published by NRC Research Press

284

Appl. Physiol. Nutr. Metab. Vol. 39, 2014

(iv) exercise at 70% V˙O2 peak after shivering (MOD SHIV). Lowintensity exercise was performed via walking while moderateintensity exercise was done via running. Treadmill speed for each subject corresponding to the protocol intensities was determined individually during the familiarization trial and manually adjusted if needed during experimental trials to ensure that relative %V˙O2peak was consistently at their respective target levels throughout the full 60 min of exercise. Instrumentation and measurements The participants were instrumented while standing (⬃45 min) in the instrumentation chamber. This was to ensure that together with the 15-min stabilization period, the subjects were in a thermoneutral condition prior to testing. Rectal core temperature (Tre) was measured using a rectal thermistor (YSI 401, Yellow Springs Instruments, USA) inserted 10 cm beyond the anal sphincter. Skin temperature Tsk was measured from 6 sites (face, chest, forearm, hand, thigh, and back) using thermistors (NTC DC95, Digi-Key, USA). Both core and Tsk data were recorded by a portable data logger (SmartReader Plus 8, ACR Systems Inc., Surrey, B.C., Canada). ˉ sk was subsequently calculated using the weighted Weighted mean T average of the 6 sites (Palmes and Park 1948): (2)

ˉ sk ⫽ 0.14 共Tface) ⫹ 0.19 (Tchest) ⫹ 0.11 (Tforearm) T ⫹ 0.05 共Thand兲 ⫹ 0.32 共Tthigh兲 ⫹ 0.19 (Tback)

Heat flux transducers (Transducers Model Ha13-18-10-P, Thermonetics Co., USA) connected to portable data loggers (SmartReader Plus 8, ACR Systems Inc.) measured dry heat loss from the skin and were positioned on 10 sites: head, chest, abdomen, lower back, arm, forearm, hand, thigh, calf and foot and were fixed by adhesive tape, which covered the rim of the transducer. Heat flux (W·m−2) values indicated the rate of heat exchange from the skin towards the environment, and therefore was considered as dry heat loss. Whole-body dry heat loss was subsequently calculated from heat flux values, the regional contribution of each site measured (face 7%; chest, abdomen and back 35%; arms and forearms 14%; hands 5%; legs 19%; calves 13%; feet 7%) (Hardy and Dubois 1938) and body surface area (BSA) as follows: (3)

whole-body dry heat loss ⫽ site × regional % contribution of site × BSA

where heat loss is in watts (W) and site is in W·m−2. Shivering was assessed by measuring surface electromyography (EMG) activity (model ME6000, Mega Electronics, Kuopio, Finland). Electrodes were placed longitudinally over the muscle belly between the centre of the innervations zone and the distal tendon (Hermens et al. 2000) of each of the following 4 muscles: pectoralis major, latissimus dorsi, rectus femoris and sternocleidomastoid. Distance between recording contacts was 2 cm with the ground electrodes placed on inactive tissues. A sampling rate of 1000 Hz was used for the EMG signals, which were amplified 2000 times with a preamplifier positioned 6 cm after the electrodes and were recorded continuously. The signal band, which was between 20– 500 Hz, was full-wave rectified and averaged with a 0.1-s time constant. A power spectrum analysis was subsequently conducted to assess median frequency (MF), and mean power frequency (MPF) (Megawin PC-Software 3.1, Mega Electronics). The combined data of the 4 established sites during the last minute of baseline and the first minute of recovery were used to determine both averaged EMG (aEMG) and frequency components. Heart rate (HR) was continuously monitored and recorded with a heart rate monitor (T6, Suunto, Vantaa, Finland). V˙O2 and RER were measured using an open circuit ergospirometer with a gas mixing chamber (Medikro 919, Medikro Oy, Kuopio, Finland). A

1-way Hans–Rudolph valve, connected to a breathing tube, was used in all trials to collect expired gases. Gas collection and mixing was done outside of the climatic chamber at thermoneutral temperature (25 °C). Oxidation of CHO and fat was calculated using stoichiometric equations (Jeukendrup and Wallis 2005) from V˙O2 and carbon dioxide output (V˙CO2) values as follows: (4)

CHO ⫽ 4.21 × V˙CO2 ⫺ 2.962 × V˙O2 ⫺ 0.4 × n

(5)

fat ⫽ 1.695 × V˙O2 ⫺ 1.701 × V˙CO2 ⫺ 1.77 × n

where CHO and fat are in g·min−1 and n represents nitrogen excretion from protein oxidation (estimated at 135 ␮g·kg−1·min−1) (Carraro et al. 1990). Protein oxidation was not directly calculated as short-term cold exposure does not tend to alter its contribution to energy expenditure (Vallerand and Jacobs 1989; Vallerand et al. 1995). Metabolic rate and relative contribution of CHO, fat and protein was subsequently calculated for the entire 60 min of exercise using the energy potentials for CHO (mixture of 20% glucose and 80% glycogen; 4.07 kcal·g−1·min−1), fat (9.75 kcal·g−1·min−1), and protein (4.09 kcal·g−1·min−1) (Jeukendrup and Wallis 2005). Protein energy contribution was estimated from the assumption that 5.57 g of protein is oxidized for each gram of nitrogen excreted (Jeukendrup and Wallis 2005). Blood sampling and analyses An OCRILON polyurethane catheter (Optivia I.V 18G, Jelco, Smith’s Medical, Ashford, UK), positioned in the antecubital vein before the start of the experiment and maintained throughout exercise, was used to collect blood samples in 3.5-mL vacuumsealed serum tubes with silicon coating (BD Vacutainer SST tubes, BD, N.J., USA) and in 3 mL K2EDTA whole blood tubes (BD Vacutainer Plus Plastic K2EDTA tubes, BD). Catheters were maintained using adhesive hypoallergenic, water-resistant tape (3M Transpore Surgical Tape, 3M Health Care, London, Ont., Canada). The blood samples in serum tubes were given 30 min to coagulate as recommended by the manufacturer. Centrifugation was then performed at 3500 rpm for 10 min (4100g) followed by isolation of plasma and serum samples in Eppendorf tubes and frozen at −80 °C for future analysis. Plasma catecholamine concentrations (epinephrine (Epi) and norepinephrine (NE)) were analyzed via a commercial ELISA kit (DRG Instruments GmbH, Germany). Coefficients of variance for intraassay precision were 15.0% for Epi and 16.1% for NE at 2.5 ng·mL−1 and 24.4 ng·mL−1 levels, respectively. Serum energy substrates (nonesterified fatty acids (NEFA), glycerol, glucose, ␤-hydroxybutyrate (BHB)) and lactate were analyzed by Konelab 20XTi (MedWOW, Nicosia, Cyprus). Their sensitivities of intra-assay coefficient of variances were 7.4%, 4.6%, 2.4%, 0.8% and 1.7%, respectively. Statistical analyses Three-way repeated measures ANOVA were used with the factors of shivering (levels: CON and SHIV), exercise intensity (levels: low intensity (LOW) and moderate intensity (MOD)), and time ˉ sk, metabolic rate, heat loss, RER, CHO oxidation, fat oxida(Tre, T tion, relative V˙O2, %V˙O2, HR, and energy contribution: baseline, 15 min, 30 min, 45 min, and 60 min of exercise, and 15 min and 30 min of recovery; levels for EMG: baseline and recovery; levels for Epi, NE, NEFA, glycerol, glucose, BHB, and lactate: baseline, 30 min, 60 min of exercise, and 30 min of recovery). Thermoregulatory and cardiorespiratory measurements and calculated fuel ˉ sk, metabolic rate, heat oxidation and energy parameters (Tre, T loss, RER, CHO oxidation, fat oxidation, absolute V˙O2, and %V˙O2) were averaged over the preceding 10-min periods for the determined time points. Treadmill speed was also analyzed via a 3-way repeated measure ANOVA but with the levels of time of 15 min, 30 min, 45 min and 60 min of exercise only. Published by NRC Research Press

Gagnon et al.

285

Fig. 1. Mean (±SE) core (A) and mean skin temperature (B), metabolic rate (C) and heat loss (D) in LOW CON (filled circles), LOW SHIV (open circles), MOD CON (filled triangles), and MOD SHIV (open triangles). *, Significant difference between SHIV and CON within both LOW and MOD (p < 0.05); †, significant difference between SHIV and CON within LOW (p < 0.05); ‡, significant difference between SHIV and CON within MOD (p < 0.05). LOW CON, lowintensity exercise; LOW SHIV, low-intensity exercise after shivering; MOD CON, moderate-intensity exercise; MOD SHIV, moderateintensity exercise after shivering; SHIV, CON, with and without a pre-exercise cooling protocol, respectively; LOW, MOD, 50% and 70% peak oxygen consumption, respectively.

A 2-way ANOVA was used to determine statistical differences in relative energy contribution from CHO and fat (%En) for the entire 60 min of exercise with the factors of shivering (levels: CON and SHIV), and exercise intensity (levels: LOW and MOD). A review by Haman (2006) demonstrated important inter-individual differences in fuel selection during passive shivering. To isolate the electromyographic and metabolic effects of pre-exercise LOW SHIV protocol, a paired t test was performed for aEMG, %En, %V˙O2 and RER with combined data of both SHIV conditions between baseline and the last 5 min of the SHIV protocol. Results are reported as means ± SE using p < 0.05 to identify statistical differences. Post hoc analysis was conducted by using independent Tukey’s HSD test. All analyses were performed with the statistical software package SPSS 20 for Windows (IBM, Armonk, N.Y., USA).

Results The duration of the pre-exercise LOW SHIV protocol ranged from ⬃40–120 min. Compared with baseline, the SHIV protocol induced an increase in aEMG activity (29 ± 2 vs. 3 ± 1 ␮V; p < 0.001), %En of fat (63% ± 4% vs. 50% ± 4%; p = 0.002) and %V˙O2peak (14.9 ± 0.7 vs. 8.3% ± 0.5% V˙O2peak; p < 0.001), and a decrease in %En of CHO (27% ± 2% vs. 35% ± 3%; p = 0.044) and RER (0.79 ± 0.01 vs. 0.81 ± 0.01; p = 0.007). Core and skin temperature Rectal temperature increased during exercise and subsequently decreased during recovery (Fig. 1A). Rectal temperature in the SHIV conditions demonstrated lower values compared with CON within both exercise intensities in the first 45 min of exercise (F = 129.6; p < 0.001). At 60 min, Tre in the SHIV condition was also lower than in CON but in the LOW condition only. During recovery, Tre in SHIV was higher than CON in MOD only, whereas no ˉ sk (Fig. 1B), a rapid differences were observed in LOW. Concerning T decrease was observed during exercise, followed by a sharp increase during recovery. Lower values were also seen in the SHIV condition within both LOW and MOD in the first 45-min of exerˉ sk cise (F = 239.7; p < 0.001). From 60 min to the end of recovery, T was lower only during SHIV in LOW (F = 4.9; p < 0.001). Metabolic rate and heat loss The increase in metabolic rate during exercise was greater in MOD compared with LOW by 337 ± 44 W (F = 223.1; p < 0.001) (Fig. 1C). Low-intensity shivering only affected metabolic rate at 30 min in LOW where values were lower in SHIV compared with CON (F = 3.8; p < 0.018). Concerning heat loss, its response was also intensified during exercise and demonstrated a strong increase in all conditions (F = 572.8; p < 0.001) (Fig. 1D). Similarly to metabolic rate, heat loss was greater during exercise in MOD compared with LOW (695 ± 28 W vs. 577 ± 24 W) (F = 72.4; p < 0.001). Additionally, heat loss in SHIV was lower than CON in both LOW and MOD at 15 and 30 min of exercise, and in MOD only at 45 and 60 min (F = 5.0; p < 0.001). No differences were seen during recovery.

EMG Baseline and recovery aEMG data are presented in Fig. 2. Preexercise SHIV protocol induced low-intensity shivering compared with baseline which was carried in recovery with greater aEMG Published by NRC Research Press

286

Fig. 2. Mean (±SE) differences between baseline (25 °C) (white columns) and recovery (25 °C) (dashed columns) averaged EMG (aEMG) activity of combined pectoralis major, latissimus dorsi, rectus femoris and sternocleidomastoid in LOW CON, LOW SHIV, MOD CON and MOD SHIV. *, Significant difference between baseline and recovery (p < 0.05). See Fig. 1 for definitions of abbreviations.

Appl. Physiol. Nutr. Metab. Vol. 39, 2014

Table 1. Heart rate, oxygen consumption and treadmill speed during baseline, 60 min of low- (LOW) and moderate- (MOD) intensity exercise and 30 min of recovery, with (SHIV) and without pre-exercise shivering (CON). LOW CON

activity in the LOW SHIV condition (3 ± 1 ␮V vs. 9 ± 2 ␮V) (F = 3.5; p < 0.01). Regarding MPF, the main effect of time demonstrated lower values in recovery compared with baseline (73 ± 5 Hz vs. 95 ± 6 Hz, respectively) (F = 7.5; p < 0.011). The main effect of time also modulated MF with lower values during recovery compared with baseline (45 ± 4 Hz vs. 56 ± 3 Hz) (F = 19.8; p < 0.001). Pre-exercise shivering had no effect on neither MPF nor MF. V˙O2, RER and treadmill speed V˙O2 (Table 1) and RER (Fig. 3A) showed no differences between CON and SHIV conditions during exercise and recovery (p > 0.154), although time affected both V˙O2 (F = 3452.2; p < 0.001) and RER (F = 81.1; p < 0.001). Treadmill speed data are presented in Table 1. Fuel oxidation and relative energy contribution CHO and fat oxidation (Fig. 3B and 3C, respectively) both demonstrated increases in LOW and MOD exercise intensities, although no statistically significant differences were seen because of shivering (p > 0.358). Relative contributions of CHO, fat, and protein to the energy yield are presented in Fig. 4. A shift towards greater reliance in CHO to the energy yield was seen in MOD compared with LOW but no significant differences were observed because of shivering (F = 0.18; p > 0.696). Plasma catecholamines Epi and NE data are presented in Fig. 5. In both LOW and MOD exercise intensities, Epi was affected by time (F = 19.5; p < 0.001) but not by shivering (F = 0.5; p > 0.487) or by exercise intensity (F = 1.6; p > 0.246). Epi concentrations were greater at 30 and 60 min of exercise compared with baseline and recovery (p < 0.01). An interaction between time and shivering indicated that in the SHIV condition, NE concentration values at 30 min was significantly greater compared with CON (p < 0.001). SHIV was also greater than CON across times (F = 5.0; p < 0.009). Finally, NE values at 30-min was greater than all other times (F = 26.7; p < 0.001). NEFA, glycerol, glucose, BHB, and lactate Data for NEFA, glycerol, glucose, and BHB at baseline, during exercise and recovery are presented in Fig. 5–7. The main effect of time influenced NEFA (Fig. 6A) as values increased during exercise before decreasing during recovery (F = 75.8; p < 0.001). The NEFA response in SHIV was greater than in CON within MOD across

MOD SHIV

CON

Heart rate (beats·min−1) Baseline 67±5 69±3 Exercise 15 min 121±6 111±4* 30 min 123±5 112±5 45 min 125±4 119±6 60 min 128±4 130±7 Recovery 15 min 77±8 71±5 30 min 74±5 72±4 Oxygen consumption (%) Baseline 8.7±0.7 9.3±0.7 Exercise 15 min 48.5±2.7 47.0±0.9 30 min 46.8±1.3 44.5±1.3 45 min 47.3±0.6 45.8±1.0 60 min 48.2±0.8 45.7±1.0 Recovery 15 min 9.6±0.8 9.4±0.6 30 min 8.7±0.6 9.6±0.4 Oxygen consumption (L·min−1) Baseline 0.36±0.03 0.39±0.02 Exercise 15 min 2.14±0.11 2.00±0.13 30 min 2.01±0.12 1.95±0.10 45 min 2.09±0.10 1.99±0.09 60 min 2.12±0.11 2.00±0.10 Recovery 15 min 0.38±0.03 0.44±0.02 30 min 0.39±0.03 0.42±0.02 Treadmill speed (km·h−1) Exercise 15 min 6.07±0.14 5.39±0.25* 30 min 6.53±0.14 5.95±0.27* 45 min 6.60±0.14 6.31±0.28 60 min 6.64±0.14 6.51±0.26

SHIV

68±4

71±5

154±3 158±5 160±5 161±5

148±5 159±6 163±5 165±5

86±5 80±5

91±4 87±4

8.2±0.5

8.0±0.6

70.0±1.8 70.9±1.1 70.9±1.8 70.6±1.1

66.8±1.6 68.1±0.2 69.1±1.0 69.7±1.1

9.4±0.4 8.6±0.4

9.7±0.6 9.1±0.6

0.34±0.03

0.33±0.02

2.99±0.12 3.04±0.14 3.04±0.16 3.04±0.12

2.85±0.10 2.93±0.14 2.96±0.13 2.98±0.14

0.40±0.02 0.37±0.02

0.39±0.03 0.37±0.02

8.14±0.28 8.94±0.50 8.86±0.53 8.85±0.56

7.64±0.30* 9.11±0.60 9.14±0.60 9.20±0.63

Note: Values are represented as mean (±SE). *Significantly lower in SHIV compared with CON (p < 0.05).

time (F = 7.1; p < 0.028). Similarly to NEFA, glycerol concentration was affected by the main effect of time (F = 46.2; p < 0.001) as it increased during exercise and decreased during recovery (Fig. 6B). The glycerol response was also greater in SHIV compared with CON within MOD across time (F = 5.5; p < 0.046). In addition, the main factor of shivering interacted with time as SHIV was greater than CON at 30 and 90 min (F = 3.4; p < 0.035). The glucose response showed no significant differences based on either shivering (F = 0.0; p > 0.950) or exercise intensity (F = 1.8; p > 0.222) but was affected by time (F = 10.3; p < 0.001) (Fig. 7B). Recovery values were lower than 30 and 60 min of exercise. The BHB response was modulated by shivering as values in SHIV were greater than those in CON within MOD (F = 7.9; p < 0.022) (Fig. 7C). Moreover, BHB concentrations were greater in MOD compared with LOW and at 60 and 90 min compared with other times (F = 9.3; p < 0.033). Finally, lactate increased during exercise (F = 35.9; p < 0.001) before returning to baseline values (Fig. 7A). Shivering affected lactate values (F = 9.3; p < 0.021), which were greater in SHIV (2.03 ± 0.15 mmol·L−1) compared with CON (1.69 ± 0.15 mmol·L−1). Exercise intensity, however, did not modulate the response (F = 4.6; p < 0.063). Published by NRC Research Press

Gagnon et al.

Fig. 3. Mean (±SE) respiratory exchange ratio (RER; A), carbohydrates (CHO) oxidation (B), and fat oxidation (C) in LOW CON (filled circles), LOW SHIV (open circles), MOD CON (filled triangles), and MOD SHIV (open triangles). See Fig. 1 for definitions of abbreviations.

Discussion The main finding of this study was that during short-term submaximal treadmill exercise in the cold, pre-exercise low-intensity shivering did not influence fuel selection at both 50% and 70% of V˙O2peak. Importantly, neither the increased fat utilization because of pre-exercise shivering nor the greater concentrations of NE, NEFA, glycerol, and BHB in the SHIV condition at 70% of V˙O2peak were associated with a concomitant change in the fuel selection response during exercise. These findings were determined by

287

Fig. 4. Relative energy contribution of carbohydrates (CHO) and fat for 60 min of exercise in LOW CON, LOW SHIV, MOD CON and MOD SHIV. Data are means (±SE). See Fig. 1 for definitions of abbreviations.

(i) using the same ambient temperature in all trials (0 °C), (ii) V˙O2 clamping between trials to avoid the interference of a change in total V˙O2 originating from additional thermal stress, and (iii) standardizing the pre-exercise shivering state across subjects at 40% Shivpeak (equivalent to 15% V˙O2peak). The pre-exercise SHIV protocol lasted between 40 and 120 min depending on subjects’ characteristics (i.e., age, % body fat, BMI, fitness) (Eyolfson et al. 2001; Xu et al. 2005). A recent study indicated that steady shivering states require approximately 60 min to be established upon cold exposure (Blondin et al. 2010). The large difference in times to reach 40% of Shivpeak among participants can be attributed to the variations in peak V˙O2 (52.9 ± 1.7 mL·kg·min−1) but also to initial body heat content, although body surface area was similar in most participants (2.04 ± 0.04 m2). During exercise, treadmill speed, which was adjusted during the trials to maintain constant V˙O2, was expected to be lower during the pre-exercise shivering trials because of pre-exercise muscle cooling and reduced neuromuscular performance expressed, for example, as decreased force and power production of the acting muscles and increased activity of antagonist muscles. This was observed at both exercise intensities: during the first 30 min in LOW and the first 15 min in MOD. Kruk et al. (1991) observed a lower respiratory quotient during exercise in the cold following resting whole-body cooling (5 °C) that was terminated upon shivering onset. Although calculations indicated that their pre-exercise cooling protocol induced 35%– 40% of Shivpeak, steady-state shivering was likely unachieved as resting periods were only ⬃20 min and core temperature did not decrease below 37.9 °C. Hurley and Haymes (1982) also observed a lower RER with exercise in the cold in 10 °C vs. 25 °C during cycling at 50 W following a 30-min cooling period in 10 °C. The short pre-exercise cold exposure in combination with a cool (10 °C) but not cold (0 °C) ambient temperature likely induced very limited pre-exercise shivering activity. In addition, progressive cooling was possible during exercise because of the very low cycling power output as core temperature and RER were lower at 90 min of exercise only. The present results indicated no significant changes in RER, CHO and fat oxidation or relative contribution to the energy yield when comparing control to the SHIV condition within both low and moderate exercise intensities. This was interesting as Tre was lower in the first 45 min of exercise during SHIV compared with CON in both exercise intensities and reduced core temperature has been linked to increased fat oxidation (Clavert Published by NRC Research Press

288

Fig. 5. Mean (±SE) epinephrine (A), norepinephrine (B) concentrations in LOW CON (closed circles), LOW SHIV (open circles), MOD CON (filled triangles), and MOD SHIV (open triangles). *, Significantly different between SHIV and CON across exercise intensities and times (p < 0.05); †, significantly different between SHIV and CON across exercise intensities (p < 0.05). See Fig. 1 for definitions of abbreviations.

et al. 1972; Hurley and Haymes 1982). Moreover, Tipton et al. (1997) demonstrated higher plasma concentrations of free fatty acids (FFA) and BHB and lower glucose disposal during increasing resting shivering activity compared with exercise when matched for V˙O2. In the present study, the pre-exercise cooling protocol induced an increase in sympathetic activity, as demonstrated by higher NE concentrations (Weller et al. 1997), increased wholebody lipolysis by higher glycerol concentrations (Romijn et al. 1993), increased NEFA availability, and higher BHB concentrations inhibiting glucose uptake (Balassee et al. 1978) during exercise. These changes were expected to increase fat reliance during exercise but surprisingly did not. Previous studies have proposed different energy metabolic pathways in cold environments compared with thermoneutral conditions at rest and during exercise (Vallerand et al. 1999; Layden et al. 2004a). In contrast to thermoneutral condition (see the review by Hawley 2002), the present study provides further evidence that point towards an uncoupling or dissociation between substrate availability and oxidation in the cold. Physiological mechanisms The cardiovascular adjustments associated with cold exposure (peripheral vasoconstriction and decreased HR) were observed as HR in the SHIV condition was lower compared with control dur-

Appl. Physiol. Nutr. Metab. Vol. 39, 2014

Fig. 6. Mean (±SE) nonesterified free fatty acids (NEFA; A) and glycerol (B) concentrations in LOW CON (filled circles), LOW SHIV (open circles), MOD CON (filled triangles), and MOD SHIV (open triangles). *, Significant difference between SHIV and CON within MOD across times (p < 0.05); †, significantly different between SHIV and CON across exercise intensities (p < 0.05). See Fig. 1 for definitions of abbreviations.

ing low-intensity exercise at 15 min (111 ± 4 vs. 121 ± 6 beats·min−1). Peripheral vasoconstriction affected heat loss, which tended to be greater in the CON condition as subjects in the SHIV condition proceeded to exercise with already vasoconstricted limbs and cooled skin, limiting heat loss during exercise. This difference was temporary in the LOW condition, until the thermoregulatory response of the CON condition would match the SHIV condition, but was carried on for 60 min by the MOD condition. In this condition, running required greater limb movements, inducing greater convective heat loss. Nonetheless, the metabolic response associated in the MOD condition with greater heat production compared with the LOW condition likely limited peripheral vasoconstriction because of increased core temperature. Both skin and core temperatures are linearly correlated to peripheral vasoconstriction (Cheng et al. 1995). Efferent sympathetic outflow in vascular beds is well understood as a cotransmitter system involving the action of NE, neuropeptide Y and ATP (Stephens et al. 2004) but central upstream signaling pathways from core cooling remain complex. The preoptic area of the hypothalamus, the thermoregulatory centre of the brain, and viscera have by themselves thermoreceptive fibres but the integration of central and peripheral thermal signals on vasoconstriction remains unknown. Finally, we expected no difference in metabolic rate as V˙O2 was clamped Published by NRC Research Press

Gagnon et al.

Fig. 7. Mean (±SE) lactate (A), glucose (B), and ␤-hydroxybutyrate (BHB; C) concentrations in LOW CON (filled circles), LOW SHIV (open circles), MOD CON (filled triangles), and MOD SHIV (open triangles). *, Significantly different between SHIV and CON across exercise intensities (p < 0.05); †, significant difference between SHIV and CON within MOD across times (p < 0.05). See Fig. 1 for definitions of abbreviations.

in all trials. Interestingly, a small but significant difference was observed at 30 min in metabolic rate, which can be attributed to the small nonsignificant difference in V˙O2 at this period influencing subsequent calculations.

289

Previous studies have indicated increased V˙O2 because of the additional presence of shivering when performing exercise in certain cold settings (e.g., swimming or walking with high wind velocity) (Nadel et al. 1973; Weller et al. 1997). Nadel et al. (1973) concluded that exercise and shivering metabolic outcomes are exerted in an additive fashion. Their finding was based on increased V˙O2 during brief bouts of cycling (150 and 200 W) and could have partially related to the increased metabolic load from cooled muscles (i.e., increased stiffness and rigidity) and not solely on shivering. V˙O2 was maintained constant during the present trials, so that potential metabolic and fuel selection changes would be linked to pre-exercise shivering and (or) the potential combined presence of exercise and shivering without the interference of a change in V˙O2 between conditions. The pre-exercise LOW SHIV protocol induced an increase in aEMG (29 vs. 3 ␮V), V˙O2 (15% vs. 8%V˙O2peak) and fat contribution to the energy yield (63% vs. 50%) and significantly lowered RER (0.79 vs. 0.81) and CHO contribution to the energy yield (27% vs. 35%). These changes indicated a preferential use of lipids prior to exercise. The presence of shivering activity immediately after exercise in the LOW SHIV conditions in combination with no change in RER, fuel oxidation and %En data during exercise itself therefore likely suggests that shivering drive and its associated metabolic outcomes were suppressed by voluntary muscle activation during exercise. It is, however, difficult to explain the basis of these findings as muscle motor unit recruitment during combined shivering and exercise is poorly understood. Whether motor units of different types or different motor units of the same type would be involved in the response is unknown (Meigal 2002). Based on the thermogenic potential between LOW SHIV and exercise, the larger contribution of exercise towards V˙O2 and heat production could have been the main fuel selection modulator. Previous findings, nonetheless, have hypothesized that core temperature and not the thermogenic effect of exercise is the main modulator of shivering activity during exercise (Meigal 2002). Core temperature data in the present study was lower in the SHIV condition at 15 min of exercise, but returned to near baseline values at 30 min. Consequently, the combined presence of both shivering and exercise, early during MOD and LOW conditions, likely occurred. Treadmill speed was lower in the early stages of exercise in the SHIV condition. This further implies the superimposed presence of shivering (Hong and Nadel 1979; Nadel et al. 1973), at least in the beginning of exercise, and possibly reduced muscle mechanical efficiency because of muscle cooling (Oksa 2000; Oksa et al. 2002). Reduced muscular efficiency, including a decrease in contraction velocity during dynamic exercise (Oksa 2000), will increase V˙O2 to perform the same work at a given workload (Weller et al. 1997). To compensate for an increase in oxygen requirements for shivering thermoregulatory purposes and reduced muscle efficiency, reducing treadmill speed was necessary to maintain stable V˙O2. As exercise progressed with increases in heat production and concomitant core temperature, speed became similar between CON and SHIV. Despite changes in treadmill speed because of shivering and reduced muscle mechanical efficiency early in exercise, fuel selection remained the same between conditions, which further suggested that the presence or absence of shivering was not a determinant factor in fuel selection as exercise itself contributed more importantly to V˙O2 and heat production. Direct shivering analyses during exercise could not be performed as exercise-based EMG signals interfere with shivering EMG signals, blunting its detection during dynamic aerobic exercise (Linnamo et al. 2003). The increase in lipolytic markers (NE, NEFA, glycerol and BHB) in the MOD condition may have originated from a change in adipose tissue blood flow (ATBF). As exercise intensity increased (from 50% to 70% V˙O2peak), the influence of greater physical stress likely increased NE concentrations. In combination from an increased blood flow from greater exercise intensity in MOD vs. Published by NRC Research Press

290

LOW, the effects of greater circulating NE on well-perfused adipose tissue would increase NEFA and glycerol release and transport. In the LOW condition, despite greater NE concentrations at 30 min, lower ATBF compared with MOD could explain the lack of changes in NEFA or glycerol. Whole-body lipolysis (indicated by increased glycerol and catecholamine concentrations) and consequently greater lipid availability (from increased NEFA concentrations) in the MOD SHIV condition did not translate into a greater use of fat during exercise. An increase in substrate availability generally equates to an increase in its utilization during exercise in a thermoneutral environment (see review by Hawley 2002) but not in the cold (Vallerand et al. 1999; Layden et al. 2002, 2004a). Although the current data may not fully explain some of our results, other energy metabolism studies have provided insights on substrate availability and utilization discrepancy. The 4 energy sources during exercise comprise of circulating glucose and NEFA and intramuscular glycogen and triglycerides (Romijn et al. 1993). The implication of intramuscular triglycerides (IMTG) has been a proposed mechanism to explain some discrepancy in previous results. Layden et al. (2004b) administered acipimox, an inhibitor of lipolysis (targeting HM74 receptors expressed in adipose tissue but not in skeletal muscles) (Tunaru et al. 2003) and a known suppressor of fat oxidation from circulating NEFA, in exercising subjects in the cold. They found no difference in total fat or CHO oxidation in 0 °C compared with 20 °C despite the significant reduction in NEFA availability because of acipimox administration and consequently concluded that IMTG contribution to energy remained unaffected by cold exposure. Unfortunately, IMTG concentrations in the muscles were not assessed and thus it is impossible to draw conclusions as to its accurate contribution. Interestingly, work performed in a thermoneutral environment has indicated a decrease in IMTG contribution to fat oxidation following 2 and 4 h of cycling at 55% V˙O2max with a concomitant increase in plasma FFA (Watt et al. 2002). In addition, a study by van Loon et al. (2005) used lipolysis-inhibiting acipimox to decrease FFA availability during cycling at 50% V˙O2max for 120 min. As opposed to a decrease in IMTG contribution from prolonged moderate exercise as seen with Watt et al. (2002), the decrease in FFA plasma availability towards total fat oxidation seen by van Loon et al. (2005) was compensated by an increase in IMTG oxidation. These results altogether seem to indicate that in a thermoneutral environment, an inverse relationship between IMTG and plasma FFA availability/oxidation works as an energetic buffer during exercise. Although intramuscular hormone sensitive lipase (HSL) has been brought forward as a likely candidate in modulating IMTG activity during exercise (Langforst et al. 1998; van Loon 2004), the interaction of cold exposure and its possible effects of HSL are beyond the scope of the current study. Nonetheless, based on these previous findings, our measured increase in circulating NEFA would systematically reduce IMTG contribution to energy expenditure, as fat oxidation was similar between conditions, and therefore, in some ways suppress HSL activity to rely mainly on circulating fat sources to fuel exercise; although accumulation of unused circulating NEFA could also translate into a greater use of IMTG and needs to be further examined. Furthermore, another enzyme known to modulate substrates transport across the plasma and mitochondrial membrane could also shed some light on our results. Upregulation of AMPactivated protein kinase (AMPK) expression, an intracellular enzyme found in skeletal muscles, from exercise is well established but can also originate from prolonged cold exposure (Mulligan et al. 2007; Weber 2011). AMPK is known to modulate glucose and NEFA muscle uptake by stimulating glucose transporter type 4 (GLUT4) and FAT/CD36, respectively, but also lipids ␤-oxidation (Hardie and Sakamoto 2006), among other energy pathways. The increase in NEFA availability and the lack of change in fat oxidation in our results may have been caused by changes in skeletal

Appl. Physiol. Nutr. Metab. Vol. 39, 2014

muscle NEFA uptake capability, namely suppression of FAT/CD36 activity. The potentially limiting rate of NEFA uptake in the muscle could also be linked to a decrease in enzyme reaction rate from lower muscle temperature. We unfortunately did not measure surface or deep tissue muscle temperature but reduced metabolism in transporter sites could have induced a reduced reaction rate of muscle membrane transport proteins, including circulating fatty acids transporter FAT/CD36. The supporting argument and potential roles of IMTG, HSL, AMPK and FAT/CD36 in our study remain, nonetheless, theoretical. The present protocol included cold air whole-body cooling until steady-state continuous low-intensity shivering appeared (i.e., 40% of Shivpeak). This was performed in the perspective that outdoor workers or winter activity practitioners may be exposed to low temperatures that may induce body cooling while they are performing tasks. Cooling to a point where task completion would be compromised (e.g., high-intensity shivering), seemed less suitable from a practical perspective. Thus, examining the effects of shivering in combination to lower exercise intensities would broaden the perspective and usefulness of our results.

Conclusion Pre-exercise low-intensity shivering did not influence fuel selection during submaximal treadmill exercise in the cold at 50% and 70% V˙O2peak. This was observed despite a decreased RER and increased contribution of fat to the energy yield from the preexercise shivering protocol, and important increases in NE, NEFA, glycerol and BHB at 70% V˙O2peak. Although we cannot completely elucidate the mechanisms behind our results, some elements that seem to guide the present findings include (i) the greater contribution of exercise over shivering on energy production and consequently fuel selection, and (ii) the known dissociation between substrate availability and utilization seen during exercise in cold environments.

Acknowledgements We would like to thank all our subjects for participating in this challenging study. In addition, we would also like to express our sincere gratitude to Niina Nikolajev, Tiina Takalokastari, Krista Rahunen, Sirkka Rissanen, Tero Mäkinen (Post-mortem) and Risto Puurtinen for their most valuable assistance. Dominique D. Gagnon is supported by the Jenny and Antti Wihurin Foundation. Sheila S. Gagnon is supported in part by the Joint Motion Program – A CIHR training program in musculoskeletal health research and leadership. Stephen S. Cheung is supported by a Canada Research Chair.

References Balassee, E., Fery, F., and Neef, M. 1978. Changes induced by exercise in rates of turnover and oxidation of ketone bodies in fasting man. J. Appl. Physiol. 44: 5–11. PMID:627499. Blondin, D.P., Péronnet, F., and Haman, F. 2010. Effects of ingesting [13C] glucose early or late into cold exposure on substrate utilization. J. Appl. Physiol. 109: 654–662. doi:10.1152/japplphysiol.00440.2010. PMID:20651221. Carraro, F., Stuart, C.A., Hartl, W.H., Rosenblatt, J., and Wolfe, R.R. 1990. Effect of exercise and recovery on muscle protein synthesis in human subjects. Am. J. Physiol. Endocrinol. Metab. 259: E470–E476. PMID:2221048. Cheng, C., Matsukawa, T., Sessler, D.I., Ozaki, M., Kurz, A., Merrifield, B., et al. 1995. Increasing mean skin temperature linearly reduces the core-temperature threshold for vasoconstriction and shivering in humans. Anesthsiology, 82: 1160–1168. doi:10.1097/00000542-199505000-00011. PMID:7741291. Clavert, D.T., Clough, D.P., Findlay, J.D., and Thompson, G.E. 1972. Hypothalamic cooling, heat production and plasma lipids in the ox. Life Sci. 11(5): 223–229. doi:10.1016/0024-3205(72)90007-0. PMID:4569323. Eyolfson, D. A, Tikuisis, P., Xu, X., Weseen, G., and Giesbrecht, G.G. 2001. Measurement and prediction of peak shivering intensity in humans. J. Appl. Physiol. 84: 100–106. doi:10.1007/s004210000329. PMID:11394237. Gagnon, D.D., Rintamäki, H., Gagnon, S.S., Cheung, S.S., Herzig, K.H., Porvari, K., and Kyröläinen, H. 2013. Cold exposure enhances fat utilization but not non-esterified fatty acids, glycerol or catecholamines availability during submaximal walking and running. Front. Physiol. doi:10.3389/fphys.2013.00099. PMID:23675353. Haman, F. 2006. Shivering in the cold: from mechanism of fuel selection to Published by NRC Research Press

Gagnon et al.

survival. J. Appl. Physiol. 100(5): 1702–1708. doi:10.1152/japplphysiol.01088. 2005. PMID:16614367. Haman, F., Legault, S.R., Rakobowchuck, M., Ducharme, M.B., and Weber, J.M. 2004a. Effects of carbohydrate availability on sustained shivering: II. Relating muscle recruitement to fuel selection. J. Appl. Physiol. 96: 41–49. doi:10.1152/ japplphysiol.00428.2003. PMID:12949017. Haman, F., Legault, S.R., and Weber, J.M. 2004b. Fuel selection during intense shivering in humans: EMG pattern reflects carbohydrate oxidation. J. Physiol. 556(1): 305–313. doi:10.1113/jphysiol.2003.055152. PMID:14742724. Haman, F., Péronnet, F., Kenny, G.P., Massicotte, D., Lavoie, C., and Weber, J.M. 2005. Partitioning oxidative fuels during cold exposure in humans: muscle glycogen becomes dominant as shivering intensifies. J. Physiol. 566(1): 247– 256. doi:10.1113/jphysiol.2005.086272. PMID:15831534. Haman, F., Scott, C.G., and Kenny, G.P. 2007. Fueling shivering thermogenesis during passive hypothermic recovery. J. Appl. Physiol. 103: 1346–1351. doi:10. 1152/japplphysiol.00931.2006. PMID:17641212. Hardie, D.G., and Sakamoto, K. 2006. AMPK: A key sensor of fuel and energy status in skeletal muscle. Physiology, 21: 48–60. doi:10.1152/physiol.00044. 2005. PMID:16443822. Hardy, J.D., and Dubois, E.F. 1938. The technique of measuring radiation and convection. J. Nutr. 15(5): 461–475. Hawley, J.A. 2002. Effect of increased fat availability on metabolism and exercise capacity. Med. Sci. Sports Exerc. 34: 1485–1491. doi:10.1097/00005768200209000-00014. PMID:12218743. Hermens, H.J., Freriks, B., Merletti, R., Stegeman, D., Blok, J., Rau, G., et al. 2000. European Recommendations for Surface ElectroMyography. Roessingh Research and Development. ISBN 90-75452-15-2. Hong, S., and Nadel, E.R. 1979. Thermogenic control during exercise in a cold environment. J. Appl. Physiol. 5: 1084–1089. PMID:511710. Hurley, B., and Haymes, E.M. 1982. The Effects of Rest and Exercise in the Cold on Substrate Mobilization and Utilization. Aviat. Space Environ. Med. 53(12): 1193–1197. PMID:7159340. Jeukendrup, A.E., and Wallis, G.A. 2005. Measurement of Substrate Oxidation During Exercise by Means of Gas Exchange Measurements. Int. J. Sports Med. 28: 528–537. doi:10.1055/s-2004-830512. PMID:15702454. Kruk, B., Pakkarinen, H., Harri, M., Manninen, K., and Hanninen, O. 1991. Comparison in men of physiological response to exercise of increasing intensity at low and moderate ambient tempratures. Eur. J. Appl. Physiol. 62: 353–357. doi:10.1007/BF00634972. PMID:1874242. Lander, P.J., Butterly, R.J., and Edwards, A.M. 2009. Self-paced exercise is less physically challenging than enforced pace exercise of the same intensity: influence of complex central metabolic control. Br. J. Sports Med. 43: 789– 795. doi:10.1136/bjsm.2008.056085. PMID:19196729. Langforst, J., Ploug, T., Ihlemann, J., Enevoldsen, L.H., Stallknecht, B., Saldo, M., et al. 1998. Hormone sensitive lipase (HSL) expression and regulation in skeletal muscle. Adv. Exp. Med. Biol. 441: 219–228. PMID:9781328. Layden, J.D., Patterson, M.J., and Nimmo, M.A. 2002. Effects of reduced ambient temperature on fat utilization during submaximal exercise. Med. Sci. Sports Exerc. 34(5): 774–779. doi:10.1097/00005768-200205000-00008. PMID:11984294. Layden, J.D., Malkova, D., and Nimmo, M.A. 2004a. During exercise in the cold increased availability of plasma nonesterified fatty acids does not affect the pattern of substrate oxidation. Metabolism. 53(2):203–208. doi:10.1016/j. metabol.2003.09.018. PMID:14767872. Layden, J.D., Malkova, D., and Nimmo, M.A. 2004b. Fat Oxidation After Acipimox-Induced Reduction in Plasma Nonesterified Fatty Acids During Exercise at 0°C and 20°C. Metabolism, 53(9): 1131–1135. doi:10.1016/j.metabol. 2004.03.016. PMID:15334373. Linnamo, V., Moritani, T., Nicol, C., and Komi, P.V. 1993. Motor unit activation patterns during isometric, concentric, and eccentric actions at different force levels. J. Electromyogr. Kinesiol. 13: 93–101. doi:10.1016/S1050-6411(02) 00063-9. PMID:12488091. Meigal, A.Y. 2002. Gross and fine neuromuscular performance at cold shivering. Int. J. Circumpolar Health, 61: 163–172. doi:10.3402/ijch.v61i2.17449. PMID: 12078964. Meigal, A.Y., Oksa, J., Hohtola, E., Lupandin, Y.V., and Rintamäki, H. 1998. Influ-

291

ence of cold shivering on fine motor control in the upper limb. Acta Physiol. Scand. 163: 41–47. doi:10.1046/j.1365-201x.1998.00333.x. PMID:9648622. Mulligan, J.D., Gonzalez, A.A., Stewart, A.M., Carey, H.V., and Saupe, K.W. 2007. Upregulation of AMPK during cold exposure occurs via distinct mechanisms in brown and white adipose tissue of the mouse. J. Appl. Physiol. 580(2): 677–684. doi:10.1113/jphysiol.2007.128652. PMID:17272339. Nadel, E.R., Holmér, I., Bergh, U., Åstrand, P.O., and Stolwick, J.A.J. 1973. Thermoregulatory shivering during exercise. Life Sci. 13(7): 983–989. doi:10.1016/ 0024-3205(73)90089-1. PMID:4766273. Oksa, J. 2000. Neuromuscular performance limitations in cold. Int. J. Circumpolar Health, 61(2): 154–162. doi:10.3402/ijch.v61i2.17448. PMID:12078963. Oksa, J., Ducharme, M.B., and Rintamäki, H. 2002. Combined effect of repetitive work and cold on muscle function and fatigue. J. Appl. Physiol. 92: 354–361. PMID:11744678. Palmes, E.D., and Park, C.R. 1948. An improved mounting for thermocouples for the measurement of the surface temperature of the body. J. Lab. Clin. Med. 33(8): 1044–1046. PMID:1887705. Romijn, J.A., Coyle, E.F., Sidossis, L.S., Gastaldelli, A., Horowitz, J.F., Endert, E., and Wolfe, R.R. 1993. Regulation of endogenous fat and carbohydrate metabolism in relation to exercise intensity and duration. Am. J. Physiol. Endo. 265(3): 380–391. PMID:8214047. Stephens, D.P., Saad, A., Bennett, L.A., Kosiba, W.A., and Johnson, J.M. 2004. Neuropeptide Y antagonism reduces reflex cutaneous vasoconstriction in humans. Am. J. Heart Circ. Physiol. 287: H1401–H1409. doi:10.1152/ajpheart. 00061.2004. PMID:15165988. Tipton, M.J., Franks, G.M., Meneilly, G.S., and Mekjavic, I.B. 1997. Substrate utilization during exercise and shivering. Eur. J. Appl. Physiol. 76: 103–108. doi:10.1007/s004210050220. PMID:9243178. Tunaru, S., Kero, J., Schaub, A., Wufka, C., Blaukat, A., Pfeffer, K., and Offermanns, S. 2003. PUMA-G and HM74 are receptors for nicotinic acid and mediate its anti-lipolytic effect. Nat. Med. 9: 352–355. doi:10.1038/nm824. PMID:12563315. Vallerand, A.L., and Jacobs, I. 1989. rates of energy substrates utilization during human cold exposure. Eur. J. Appl. Physiol. 58: 873–878. PMID:2767069. Vallerand, A., Zamenik, J., and Jacobs, I. 1995. Plasma glucose turnover during cold stress in humans. J. Appl. Physiol. 78: 1296–1302. PMID:7615436. Vallerand, A.L., Zamecnick, J., Jones, P.J.H., and Jacobs, I. 1999. Cold stress increases lipolysis, FFA, Ra and TG/FFA cycling in humans. Aviat. Space Environ. Med. 70: 42–50. PMID:9895020. van Loon, L.J.C., Greenhaff, P.L., Constantin-Teodosiu, D., Saris, W.H.M., and Wagenmakers, A.J.M. 2001. The effects of increasing exercise intensity on muscle fuel utilisation in humans. J. Physiol. 536(1): 295–304. PMID:11579177. van Loon, L.J.C. 2004. Use of intramuscular triacylglycerol as a substrate source during exercise in humans. J. Appl. Physiol. 97: 1170–1187. doi:10.1152/ japplphysiol.00368.2004. PMID:15358749. van Loon, L.J.C., Thomason-Hughes, M., Constantin-Teodosiu, D., Koopman, R., Greenhaff, P.L., Hardie, D.G., et al. 2005. Inhibition of adipose tissue lipolysis increases intramuscular lipid and glycogen use in vivo in humans. Am. J. Physiol. Endocrinol. Metab. 289: E482–E493. doi:10.1152/ajpendo.00092.2005. PMID:15886227. Venables, A.C., Achten, J., and Jeukendrup, A.E. 2005. Determinants of fat oxidation during exercise in healthy men and women: a cross-sectional study. J. Appl. Physiol. 98: 160–167. doi:10.1152/japplphysiol.00662.2003. PMID:15333616. Watt, M.J., Heigenhauser, G.J.F., Dyck, D.J., and Spriet, L.L. 2002. Intramuscular triacyglycerol, glycogen and acetyl group metabolism during 4 h of moderate exercise in man. J. Physiol. 541(3): 969–978. doi:10.1113/jphysiol.2002.018820. PMID:12068055. Weber, J.M. 2011. Metabolic fuels: regulating fluxes to select mix. J. Exp. Biol. 214: 286–294. doi:10.1242/jeb.047050. PMID:21177948. Weller, A.S., Millard, C.E., Stroud, M.A., Greenhaff, P.L., and MacDonald, I.A. 1997. Physiological responses to cold, wet, and windy environment during prolonged intermittent walking. Am. J. Physiol. 272: R226–R233. PMID: 9039013. Xu, X., Tikuisis, P., Gonzalez, R., and Giesbrecht, G.G. 2005. Thermoregulatory model for prediction of long-term cold exposure. Comput. Biol. Med. 35: 287–298. doi:10.1016/j.compbiomed.2004.01.004. PMID:15749090.

Published by NRC Research Press

Copyright of Applied Physiology, Nutrition & Metabolism is the property of Canadian Science Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Copyright of Applied Physiology, Nutrition & Metabolism is the property of Canadian Science Publishing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.