Four-week cold acclimation in adult humans shifts uncoupling thermogenesis from skeletal muscles to brown adipose tissue

Abstract

Key points: Muscle-derived thermogenesis during acute cold exposure in humans consists of a combination of cold-induced increases in skeletal muscle proton leak and shivering. Daily cold exposure results in an increase in brown adipose tissue oxidative capacity coupled with a decrease in the cold-induced skeletal muscle proton leak and shivering intensity. Improved coupling between electromyography-determined muscle activity and whole-body heat production following cold acclimation suggests a maintenance of ATPase-dependent thermogenesis and decrease in skeletal muscle ATPase independent thermogenesis. Although daily cold exposure did not change the fibre composition of the vastus lateralis, the fibre composition was a strong predictor of the shivering pattern evoked during acute cold exposure.

Abstract: We previously showed that 4 weeks of daily cold exposure in humans can increase brown adipose tissue (BAT) volume by 45% and oxidative metabolism by 182%. Surprisingly, we did not find a reciprocal reduction in shivering intensity when exposed to a mild cold (18°C). The present study aimed to determine whether changes in skeletal muscle oxidative metabolism or shivering activity could account for these unexpected findings. Nine men participated in a 4 week cold acclimation intervention (10°C water circulating in liquid-conditioned suit, 2 h day^-1 , 5 days week^-1 ). Shivering intensity and pattern were measured continuously during controlled cold exposure (150 min at 4 °C) before and after the acclimation. Muscle biopsies from the m. vastus lateralis were obtained to measure oxygen consumption rate and proton leak of permeabilized muscle fibres. Cold acclimation elicited a modest 21% (P < 0.05) decrease in whole-body and m. vastus lateralis shivering intensity. Furthermore, cold acclimation abolished the acute cold-induced increase in proton leak. Although daily cold exposure did not change the fibre composition of the m. vastus lateralis, fibre composition was a strong predictor of the shivering pattern evoked during acute cold. We conclude that muscle-derived thermogenesis during acute cold exposure in humans is not only limited to shivering, but also includes cold-induced increases in proton leak. The efficiency of muscle oxidative phosphorylation improves with cold acclimation, suggesting that reduced muscle thermogenesis occurs through decreased proton leak, in addition to decreased shivering intensity as BAT capacity and activity increase. These changes occur with no net difference in whole-body thermogenesis.

Keywords: cold-acclimation; energy metabolism; non-shivering thermogenesis; proton leak; shivering; uncoupling.

Figure 1. Overall study design

Four week cold acclimation preceded and followed by an acute cold experimental session protocol at 18 °C (Blondin et al., 2014) and at 4 °C (present study).

Figure 2. Thermal responses

Mean skin (A) and core temperature (B) during room temperature and cold exposure, pre‐ and post‐acclimation. C, total energy expenditure during cold exposure, pre‐ and post‐acclimation. Weighted mean of total muscle shivering activity (D), burst shivering rate (E), and continuous and burst shivering intensity (F) pre‐ and post‐acclimation. Mean shivering activity (G), burst shivering rate (H), and continuous and burst shivering intensity (I) of m. vastus lateralis, pre‐ and post‐acclimation. ^**** P < 0.0001 vs. room temperature (ANOVA with Bonferonni post hoc test). ^# P < 0.05, ^## P < 0.01 vs. pre‐acclimation (Student's t test).

Figure 3. Relationship between EMG‐determined muscle activity and whole‐body heat production

Relationship between cold‐induced changes in metabolic rate (MR) and shivering intensity in men exposed to 4 °C for 150 min (A) prior to and (B) following a 4 week cold acclimation. Values presented are from five sampling intervals during cold exposure (time = 30, 60, 90, 120 and 150 min) from all subjects also having had a muscle biopsy (n = 6).

Figure 4. Oxygen consumption in permeabilized muscle fibres

A, state 2, state 3 (complex I), state 3 (complex II) and state 4 (measured after ATP synthase inhibition by oligomycin treatment) and non‐mitochondrial (measured after complex III inhibition by antimycin A treatment) OCR at room temperature and immediately following cold exposure prior to and following a 4 week cold acclimation (n = 6). B, quantification of mitochondrial respiratory chain complexes (Complex II, III and V (ATP synthase) presented relative to pre‐acclimation room temperature conditions (n = 8). C, representative Western blots in m. vastus lateralis., with α‐tubulin used as a loading control (bottom). D, percentage of OCR as a result of proton leak at room temperature and immediately following cold exposure prior to and following a 4 week cold acclimation (n = 6). E, citrate synthase activity measured in m. vastus lateralis at room temperature and immediately following cold exposure prior to and following a 4 week cold acclimation (n = 8). F, quantification of UCP3 expression (top) presented relative to pre‐acclimation room temperature conditions (n = 8) and representative Western blots of UCP3 expression in m. vastus lateralis (bottom). α‐tubulin was used as a loading control. G, relationship between percentage of OCR as a result of proton leak and m. vastus lateralis shivering intensity pre‐acclimation (n = 6). ^* P < 0.05 vs. room temperature, ^## P < 0.01 vs. pre‐acclimation (ANOVA with Bonferonni post hoc test).

Figure 5. Quantification of fibre typing of m. vastus lateralis by immunohistochemical detection

Fibre‐type proportions in the m. vastus lateralis prior to and following a 4 week cold acclimation (n = 7). A, muscle sections were stained based on myosin heavy chain expression for type I, type IIa, type I‐IIa hybrid and type IIx (unstained fibres). Relationship between m. vastus lateralis shivering intensity and proportion of type I fibres (B) or type IIa fibres (C).

Figure 6. Whole‐body substrate utilization

Absolute (A) and relative (B) substrate utilization at room temperature and during cold exposure prior to and following a 4 week cold acclimation. Values are the mean ± SE (n = 9). ^** P < 0.01, ^**** P < 0.0001 vs. room temperature (ANOVA with Bonferonni post hoc test).