Brown Adipose Tissue Recruitment in a Rodent Model of Severe Burns

Abstract

Background:

Burn trauma results in a prolonged hypermetabolic response. Brown adipose tissue (BAT), abundant in uncoupling protein 1 (UCP1), plays a key role in non-shivering thermogenesis. We set out to determine if BAT is recruited in response to severe burns.

Methods:

Male balb-c mice underwent scald burns on approximately 20–25% of their total body surface. BAT was harvested from the interscapular fat pad of sham and burned mice at 3 hours, 24 hours, 4 days, and 10 days post injury. High-resolution respirometry was used to determine mitochondrial respiratory function in BAT. BAT protein concentration, and mitochondrial enzyme activity were also determined.

Results:

Respiration increased in BAT of burned mice, peaking at 24 hours post injury (P<0.001). While UCP1 independent respiration was not significantly altered by burn, UCP1 dependent respiration increased >2-fold at 24 hours post injury when compared to the 3 hours and sham group (P<0.01). Normalized to CS activity, total uncoupled (P<0.05) and UCP1 dependent (P<0.01) respiration remained elevated at 24 hours post injury.

Conclusions:

We show a time-dependent recruitment of rodent BAT in response to severe burns. Given recent reports that humans, including patients with severe burns, have functional BAT, these data support a role for BAT in the hypermetabolic response to severe burns.

BACKGROUND

Severe burns results in a prolonged hypermetabolic stress response [1]. Increased cardiac contractility, whole body protein turnover, and substrate cycles increase ATP utilization in response to burn trauma [2]. However, greater ATP turnover cannot fully explain this hypermetabolic response, meaning that mitochondrial oxygen consumption exceeds that required for ATP synthesis following burns [1]. While mitochondria in several tissues are capable of uncoupling respiration from ATP production, brown adipose tissue (BAT) mitochondria are highly specialized for this role. Indeed, the presence of uncoupling protein 1 (UCP1) allows BAT mitochondria to generate heat directly from fuel oxidation [3]. When activated, the inner mitochondrial membrane carrier protein UCP1 allows protons to re-enter the mitochondrial matrix independently of ATP synthase, dissipating membrane potential as heat.

The relatively recent observation that humans have functional BAT [4] has rekindled interest in the metabolic role of BAT in health and disease. Indeed, in response to adrenergic stress, BAT activation in humans can increase metabolic rate and alter fuel metabolism [5–8]. Given that severe burn trauma results in a prolonged adrenergic stress response that is accompanied by profound hypermetabolism, BAT activation may contribute to increased energy expenditure post burn. In this study, we utilized a mouse model to determine the impact of burn trauma on BAT bioenergetics. We hypothesize that severe burn trauma will result in the recruitment of BAT UCP1.

MATERIALS AND METHODS

Animal procedures

The study was approved by the Institutional Animal Care and Use Committee at the University of Texas Medical Branch, Galveston, Texas. A mouse model of full-thickness scald burns was utilized [9–11]. Male balb-c mice (8–16 weeks old) were housed at ~24–26°C on a 12:12 light:dark cycle throughout the study period. Following a pre-emptive buprenorphine (0.1 mg/kg i.p.) dose, mice were anesthetized by isoflurane (3–5%) in order to create a scald burn on approximately 20–25% of the total body surface. Briefly, the dorsum and flanks were shaved with electrical clippers before the dorsum was exposed to ~95°C water for ~10 seconds. Animals were then resuscitated with 2 ml lactated ringers. Sham treated animals underwent the same procedure with the exception of exposure to scalding water. Mice were housed individually after sham/burn treatment and had unrestricted access to water and chow. Burned mice were euthanized at 3 hours, 24 hours, 4 days, and 10 days post injury and BAT collected as described below. A total of 5–6 mice were included per group. Sham animals (n=6) were pooled from sham treated mice euthanized at 3 hours, 24 hours, and 10 days post injury.

BAT collection and handling

Immediately following euthanasia, the skin around the dorsal portion of the neck and scapula was dissected to expose the interscapular fat pad. This fat pad was removed and BAT dissected out for biochemical analysis. A piece of BAT was frozen in liquid nitrogen and stored at −80°C for future biochemical analysis. Another portion of BAT was placed in an ice-cold preservation buffer (pH of 7.1) upon collection. This buffer contained 10 mM CaK₂-EGTA; 7.23 mM K₂-EGTA; 20 mM imidazole; 20 mM taurine 50 mM K-MES; 0.5 mM dithiothreitol; 6.56 mM MgCl₂; 5.77 mM ATP and 15 mM creatine phosphate. These BAT samples were subsequently transferred to the laboratory so that mitochondrial respiration analysis could be performed.

High-resolution respirometry

Approximately 1–3 mg of BAT (wet weight) was transferred into the chamber of an Oxygraph-2K (O2K) high-resolution respirometer (Oroboros Instruments, Innsbruck, Austria) containing 2 mls of buffer (MiR05 composition: 0.5 mM EGTA; 3 mM MgCl₂; 0.5 M K-lactobionate; 20 mM taurine; 10 mM KH₂PO₄; 20 mM HEPES; 110 mM sucrose; and 1 mg/ml essential fatty acid free bovine serum albumin) for high-resolution respirometry analysis. BAT samples were permeabilized by the addition of digitonin (2 μM) to the respiration buffer. Temperature was maintained at 37°C and O₂ concentration within the range of 250–400 nmol/ml for all analyses. O₂ concentration within the O2K chamber was recorded at 2 to 4-second intervals (DatLab, Oroboros Instruments, Innsbruck, Austria) and used to calculate respiration per milligram of tissue.

Mitochondrial respiration uncoupled from ATP production was assayed in digitonin permeabilized BAT samples following the addition of saturating concentrations of substrates (1.5 mM octanoyl-l-carnitine, 5 mM pyruvate, 2 mM malate, 10 mM glutamate and 10 mM succinate). Thereafter, UCP1 was inhibited following the titration of GDP into the O2K chamber at a final concentration of 20 mM [9, 12]. The reduction in respiration following GDP titration represents UCP1-dependent uncoupled respiration. The respiratory rate remaining after GDP titration represents UCP1-independent uncoupled respiration.

Mitochondrial enzyme activity

Maximal rates of two key mitochondrial enzymes were quantified as surrogates of BAT oxidative capacity and mitochondrial content. Approximately 5–10 mg of frozen BAT was used to make a 5 mg/ml tissue lysate in a 175 mM KCl buffer containing 2 mM EDTA and 1% Triton. Cytochrome C Oxidase (COX), the terminal complex of the electron transport chain, was assayed in BAT lysates respirometrically. Briefly, 40 μl of lysate was suspended in 2 ml of respiration buffer containing 2 μM Antimycin A. The COX electron donor, tetramethylphenylenediamine (TMPD) and ascorbate were then added to the respiration buffer to support COX-driven respiration. After accounting for the chemical background, COX activity (pmol/s/mg) was calculated as the peak respiratory flux following titration of TMPD and ascorbate [12]. The activity of citrate synthase (CS), a key tricarboxyllic acid cycle enzyme, was also quantified in tissue lysates. CS activity was determined spectrophotometrically as the rate of acetyl-CoA condensation with oxaloacetate in a Tris buffer containing Ellman’s reagent [13]. CS activity (μM/min/mg) was calculated from the change in light absorbance (at 405 nm) due to the complexing of Ellman’s reagent with free Co-enzyme A produced from condensation of acetyl-CoA and oxaloacetate. Lysate protein concentration was also quantified spectrophotometrically. COX and CS activities were subsequently normalized to lysate protein.

Statistical analysis

All data is presented as group means ± the standard error. Statistically significant differences were detected between sham/burn groups by means of a one-way analysis of variance with a Tukey’s multiple comparisons test. Statistical analysis was performed using GraphPad Prism Version 7 (GraphPad, La Jolla, CA), with significance accepted at P<0.05.

RESULTS

Mass specific respiration

The total uncoupled respiratory capacity of BAT was significantly greater after burn trauma (Figure 1A, P<0.01, one-way ANOVA). Specifically, total uncoupled respiration was higher in the 24-hour post-burn group when compared to the sham (139.0±14.9 vs. 223.0±15.1 pmol/s/mg, P<0.01) and 3-hour post burn (139.0±14.9 vs. 223.0±15.1 pmol/s/mg, P<0.05) groups. Further, respiration was lower in the 10 days post burn when compared to the 24-hour post-burn group (223.0±15.1 vs. 156.2±8.5 pmol/s/mg, P<0.05). The UCP1 independent (GDP insensitive) component of total uncoupled mitochondrial respiration was not significantly altered in BAT by burn trauma (Figure 1B). In contrast, burn trauma significantly altered the UCP1 dependent (GDP sensitive) component of total uncoupled mitochondrial respiration (Figure 1C, P<0.001, one-way ANOVA). Indeed, UCP1 dependent respiration was greater at 24 hours post burn (105.3±9.5 pmol/s/mg) versus the sham (51.2±8.2 pmol/s/mg, P<0.001), 3-hour post-burn (47.8±5.2 pmol/s/mg, P<0.001), 4-day post-burn (52.8±8.0 pmol/s/mg, P<0.01), and 10-day post-burn (52.0±8.0 pmol/s/mg, P<0.01) groups. In order to account for potential differences in the mitochondrial content, the respiratory control ratio for GDP (RCR_GDP) was calculated to assess mitochondrial respiratory control in response to GDP. Burn trauma significantly altered the RCR_GDP (Figure 1D, P<0.05, one-way ANOVA). The RCR_GDP was 0.62±0.05 and 0.69±0.05 in the sham and 3-hour post-burn groups, meaning that 38% and 31% of respiration was sensitive to GDP in these two groups, respectively. The RCR_GDP was lowest at 24 hours post burn (0.53±0.04), where 47% of respiration was sensitive to GDP. Compared to the 24-hour post-burn group, the RCR_GDP was higher at 4 (0.71±0.03, P<0.05) and 10 days post burn (0.67±0.05).

Figure 1.

Mitochondrial respiratory capacity in brown adipose tissue collected from sham-treated mice and burned mice euthanized at 3 hours, 24 hours, 4 days, and 10 days post injury (n=5–6 per group). Total uncoupled mitochondrial respiration was significantly different between groups (panel A, P<0.01, one-way ANOVA). UCP1 independent mitochondrial respiration was not significantly different between groups (panel B). UCP1 dependent mitochondrial respiration was significantly different between groups (panel C, P<0.001, one-way ANOVA). The respiratory control ratio of GDP (RCR_GDP) was significantly different between groups (panel D, P<0.05, one-way ANOVA). *P<0.05, **P<0.01 and ***P<0.001 denote significant differences between specific groups detected by a Tukey’s post hoc test.

BAT Protein Content and Mitochondrial Enzyme Activity

We quantified the protein content and activities of CS and COX in BAT lysates. BAT protein content increased after burn (Figure 2A, P<0.05, one-way ANOVA), where the protein content was greater in BAT taken from mice 24 hours post burn compared to BAT from the sham group (549±68 vs. 709±84 μg/ml, P<0.05). BAT protein content peaked at 24 hours post burn, decreasing to levels comparable to the sham group in 4- and 10-day post-burn groups. Indeed, protein levels were significantly lower in the 10-day post-burn group compared to the 24-hour post-burn group (709±84 vs. 413±29 μg/ml, P<0.05). CS represents a key tricarboxylic cycle enzyme, catalyzing the condensation of oxaloacetate and acetyl-CoA, forming citrate. CS activity is frequently used as a biochemical measure of tissue oxidative. Compared to sham, CS activity remained relatively stable at 3 hours, 24 hours, and 4 days post burn (Figure 2B). However, CS activity was significantly greater at 10 days post burn when compared to the sham, 3-hour and 24-hour groups (P<0.05) (Figure 2B, P<0.05, one-way ANOVA). We also assayed the activity of COX in BAT lysates as an additional measure of tissue oxidative capacity (Figure 2C). While COX activity appeared to be numerically reduced in the initial 4 days post burn, this difference did not reach statistical difference.

Figure 2.

Protein concentration and mitochondrial enzyme activities in brown adipose tissue (BAT) protein collected from sham treated mice and burned mice euthanized at 3 hours, 24 hours, 4 days, and 10 days post injury (n=5–6 per group). BAT lysate protein concentration was significantly different between groups (panel A, P<0.05, one-way ANOVA). BAT citrate synthase (CS) activity was significantly different between groups (panel B, P<0.05, one-way ANOVA). BAT cytochrome C oxidase (COX) activity was not significantly different between groups (panel C). *P<0.05 denotes significant differences between specific groups detected by a Tukey’s post hoc test.

Mitochondrial specific respiration

To further assess mitochondrial quality, we calculated mitochondrial specific respiration by normalizing the respiratory rates presented in Figures 1A–C by CS activity data presented in Figure 2B. Total uncoupled respiration per unit of CS activity was significantly altered in BAT following burn trauma (Figure 3A, P<0.01, one-way ANOVA). Specifically, total uncoupled respiration per unit of CS activity was 2-fold higher in the 24-hour post-burn group versus the sham group (44.9±9.3 vs. 83.3±12.0 pmol/s/mg/U CS activity, P<0.05). Whereas total uncoupled respiration per unit of CS activity was significantly lower in the 10 day post burn group versus the 24-hour post-burn group (83.3±12.0 vs. 28.3±1.8 pmol/s/mg/U CS activity, P<0.01). While the UCP1 independent (GDP insensitive) component of total uncoupled mitochondrial respiration was numerically greater in burned animals when normalized to CS activity, this did not reach statistical significance (Figure 3B). When normalized to CS activity, the UCP1 dependent (GDP sensitive) component of total uncoupled mitochondrial respiration was significantly altered by burn injury (Figure 3C, P<0.001, one-way ANOVA). Specifically, UCP1 dependent respiration was 3-fold greater in the 24-hour post-injury group when compared to the sham (15.4±2.7 vs. 39.8±6.7 pmol/s/mg/U CS activity, P<0.01). UCP1 dependent respiration normalized to CS activity at was also lower in the 3-hour post-burn (P<0.01), 4-day post-burn (P<0.01), and 10-day post-burn (P<0.001) groups.

Figure 3.

Mitochondrial respiratory function in brown adipose tissue collected from sham treated mice and burned mice euthanized at 3 hours, 24 hours, 4 days, and 10 days post injury (n=5–6 per group). Total uncoupled mitochondrial respiration normalized to citrate synthase (CS) activity was significantly different between groups (panel A, P<0.01, one-way ANOVA). UCP1 independent mitochondrial respiration normalized to CS activity was not significantly different between groups (panel B). UCP1 dependent mitochondrial respiration normalized to CS activity was significantly different between groups (panel C, P<0.001, one-way ANOVA). *P<0.05, **P<0.01 and ***P<0.001 denote significant differences between specific groups detected by a Tukey’s post hoc test.

DISCUSSION

Prolonged hypermetabolism is a hallmark of the stress response to trauma, especially severe burns [14]. While greater ATP turnover explains much of this hypermetabolic response, mitochondrial respiration out-paces ADP phosphorylation in burned individuals [1]. Mitochondrial respiration that is not coupled to ATP production may represent non-shivering thermogenesis, a process that typically supports the maintenance of core temperature. Non-shivering thermogenesis occurs predominantly in BAT, which is endowed with highly specialized mitochondria that permit fuel oxidation to be transduced directly to heat. Originally named thermogenin [15], the inner membrane carrier protein UCP1 allows protons that are pumped into the inter-membrane space of mitochondria to re-enter the matrix independently of ATP synthase, producing heat. Adrenergic stress and subsequent lipolysis is required for UCP1 activation [16]. In this study, we set out to determine the impact of the adrenergic stress response to burns on BAT bioenergetics. We provide direct evidence of BAT UCP1 activation in response to severe burn trauma. Specifically, we found that an acute increase in BAT mitochondrial respiration was driven by increased UCP1-dependent respiration. This response peaked at 24 hours post injury, suggesting an acute and transient recruitment of BAT thermogenesis following burn trauma.

Our data demonstrating greater UCP1 function in response to burn are in line with a previous report demonstrating increased glucose (~6-fold), fatty acid (~2.7-fold) and acetate (~2-fold) uptake in rodent BAT 24 hours following a severe burn [17]. Further, BAT lipid content was reduced by around 50% in these burned mice [17], suggesting that within the first day after a severe burn, BAT activation results in a depletion of local BAT lipid stores as well as driving the uptake of substrates from the circulation. Our mitochondrial respirometry data support the notion that an acute increase in BAT UCP1 function underlies altered BAT fuel oxidation in the first 24 hours after a severe burn.

Interestingly, Carter et al. [17] demonstrated that the acute response of BAT to burn trauma was similar to that of mice housed at 4°C for 24 hours, and mice that underwent a cutaneous excision. These three conditions resulted in an acute increase in UCP1 mRNA levels in BAT. This response was greater in the cold exposed mice and mice that underwent a cutaneous excision when compared to burned mice. Similarly, while there was a ~6-fold increase in BAT glucose uptake in burned mice, there was approximately a 15- and 13-fold increase in BAT glucose clearance in the cold-exposed and cutaneous excision groups, respectively [17]. These data have important physiological implications, as they suggest that while adrenergic stress is perhaps critical for BAT activation, the drive to thermoregulate and maintain core temperature is likely the principal mediator for this response. Accordingly, these data indicate that the loss of a skin barrier and the ability to thermoregulate efficiently likely drives BAT activation in response to burn trauma.

Previous reports have demonstrated a marked increase in UCP1 mRNA at 24 hours post burn, a response that is subsequently diminished at 48 and 72 hours post injury [18]. These data suggest a transient signal to increase UCP1 transcription following burn trauma. Indeed, while we found that UCP1-dependent mitochondrial respiration was increased at 24 hours post burn, UCP1 function had returned to values comparable to sham treated animals at 4 and 10 days post burn. However, while we did not find evidence of greater BAT activation at 4 or 10 days post burn, others have shown that glucose uptake remains elevated in the BAT of burned rats at one week post injury [19]. Similarly, these authors reported that there was marked lipid depletion accompanied by increased mitochondrial number in BAT of burned rats [19]. Thus, while we report changes in BAT UCP1 function only at 24 hours post burn, this does not necessarily mean that there is not greater BAT UCP1 activation in vivo over a more prolonged period after burn trauma, particularly when considering the prolonged nature of the adrenergic stress response to burns.

Given that we observed an acute increase in BAT UCP1 function following burn trauma, we quantified BAT protein content and the activities of key mitochondrial enzymes in order to determine the role of altered BAT oxidative capacity in UCP1 activation in response to burn trauma. Similar to mitochondrial respiration, there was a transient increase in BAT protein concentration. This is in line with BAT’s response to cold exposure, where an increase in BAT total protein precedes an increase in BAT UCP1 content [20]. Of interest though, the maximal activities of both CS and COX, often used as surrogates of mitochondrial protein levels, were largely unaltered in BAT following burn trauma. With that said, we did detect a significant increase in BAT CS activity at 10 days post burn. This supports previous data where BAT mitochondrial number was shown to be increased in rats 7 days after a severe burns [19]. Nevertheless, while there may be a chronic change in BAT mitochondrial protein levels in response to burn trauma, our data do not support the notion that increased mitochondria protein levels drive the acute increase in UCP1 function that accompanies a severe burn. Accordingly, when normalizing mitochondrial respiratory fluxes to CS activity, UCP1 dependent respiration is only increased in burned animals at 24 hours post injury. These data suggest that the acute increase in UCP1 function in response to burn trauma is likely the result of a qualitative shift in BAT mitochondrial function.

In further support of the notion that altered mitochondrial quality underlies the acute augmentation of BAT UCP1 function following burn trauma, we also found that mitochondrial coupling control in response to the UCP1 inhibitor GDP was greater in the BAT of burned animals at 24 hours post injury. As both the numerator and denominator used to calculate this ratio are derived from the same respirometry protocol, means that coupling control ratios account for inherent differences in mitochondria protein abundance across samples. Accordingly, mitochondrial coupling control in response to GDP provides a good measure of BAT mitochondrial function, and specifically, UCP1 function. Further, in the context of severe cold stress, while BAT UCP1 mRNA and total protein concentration increase significantly within the first 24 hours, UCP1 protein levels do not measurably increase until after approximately 3 days of cold stress [20]. Collectively, our mitochondrial enzyme activity and coupling control data suggest that the acute recruitment of UCP1-dependent respiration in response to burn trauma is driven by alterations in BAT mitochondrial function as opposed to increased BAT oxidative capacity or UCP1 content.

While these data suggest that non-shivering thermogenesis in BAT may play a role in the hypermetabolic response to burns, the transient nature of this response suggests that BAT alone is unlikely to fully account for the thermogenic component of burn-induced hypermetabolism. As noted above, the fact that BAT mitochondrial respiratory function is only altered at 24 hours post injury does not rule out the possibility that BAT may be activated in vivo over a much greater period following burn trauma. However, there is also emerging evidence that BAT may not be the only site of UCP1 recruitment in response to stress. Indeed, chronic cold exposure results in thermogenically competent UCP1 positive white adipocytes [21]. This browning of white adipose tissue (WAT) has been touted as a means to alter energy metabolism in humans [22]. Like prolonged cold stress, burn trauma represents a model of chronic adrenergic stress [14]. Indeed, we previously published data showing that rodent WAT undergoes a marked browning response to burns [9]. However, we could only detect an increase in WAT UCP1 function from 4 days post-burn injury, where this browning response peaked at 20 days post injury. While our current data suggest that UCP1 recruitment in BAT may occur in the initial few days post burn, our previous data indicate that there may be a more chronic recruitment of UCP1 in WAT [9]. Indeed, we have been able to demonstrate a time dependent induction of UCP1 mRNA in WAT of burned patients, a response accompanied by altered WAT morphology and increased oxidative capacity [23]. Thus, it may be that acute BAT UCP1 activation supports thermal regulation in the initial few days after a severe burn, whereas WAT browning might play a more prominent role in thermal regulation at a later stage in the hypermetabolic response to burns. However, it is important to note that convincing data demonstrating that WAT mitochondria can develop functional UCP1 in humans in response to burns (or any other condition) are currently lacking. Moreover, WAT browning in burned rodents endows white adipocytes with thermogenic capacity (~8 pmol/s/mg [9]) that represents only a fraction of the thermogenic capacity of BAT following burn trauma (~110 pmol/s/mg). This suggests that UCP1 activation in BAT may still be the main driver of non-shivering thermogenesis in response to burn trauma, even in the face of WAT browning.

CONCLUSIONS

In summary, our data demonstrate an acute but transient recruitment of BAT in response to burn trauma. Importantly, we show that UCP1-dependent respiration is augmented in the first 24 hours post burn. Interestingly, our data indicate that neither BAT oxidative capacity nor UCP1 abundance underlay this acute increase in UCP1-dependent respiration, suggesting that altered mitochondrial quality drives this response. Collectively, these data indicate a role for BAT recruitment and UCP1 mediated thermogenesis in the hypermetabolic stress response to burn trauma. Accordingly, targeting BAT UCP1 may be a means to control burn-induced hypermetabolism.

Abbreviations:

BAT

Brown adipose tissue

WAT

White adipose tissue

UCP1

Uncoupling protein 1

CS activity

Citrate synthase activity

COX activity

Cytochrome C Oxidase activity

GDP

Guanosine diphosphate

ADP

Adenosine diphosphate

ATP

Adenosine triphosphate

ANOVA

Analysis of variance