NLRP3 inflammasome mediates white adipose tissue browning after burn

Abstract

A hallmark after burn is the stress and inflammatory-induced hypermetabolic response. Recently, we and others found that browning of white adipose tissue (WAT) is a critical component of this complex detrimental response. Although browning and inflammation have been independently delineated to occur after injury, their interaction is currently not well defined. One of the master regulators of inflammation and adipose tissue remodeling after burns is nucleotide-binding and oligomerization domain, leucine rich repeat and pyrin domain containing 3 (NLRP3) inflammasome. The aim of this this study was to determine whether NLRP3 modulates and activates WAT browning after burn. To obtain molecular and mechanistic insights, we used an NLRP3 knockout (NLRP3^−/−) murine burn model. We demonstrated that genetic deletion of NLRP3 promoted persistent and augmented browning in adipocytes, evidenced by increased gene expression of peroxisome proliferator-activated receptor γ and CIDEA at 3 days (5.74 vs. 0.29, P < 0.05; 26.0 vs. 0.71, P < 0.05) and uncoupling protein 1 (UCP1) and PGC1α at 7 days (7,406 vs. 3,894, P < 0.05; 20.6 vs. 2.52, P < 0.01) and enhanced UCP1 staining and multilocularity. Additionally, the main regulator of postburn WAT browning, IL-6, was elevated in the plasma acutely after burn in NLRP3^−/− compared with wild-type counterparts (478.9 vs. 67.1 pg/mL, P < 0.05 at 3 days). These results suggest that NLRP3 has antibrowning effects and that blocking NLRP3 increases thermogenesis and augments browning via increased levels of IL-6. Our findings provide insights into targeting innate inflammatory systems for regulation of adaptive thermogenesis, a critical response after burns and other hypermetabolic conditions.

INTRODUCTION

According to the World Health Organization, more than 330,000 deaths per year worldwide are related to burn injury (5). Recent evidence indicates that hypermetabolism, a hallmark of severely burned patients, is a major contributor to these detrimental outcomes (8, 10). The hypermetabolic response to burn injury can be essentially divided into two distinct phases, the early inflammatory phase and the late hypermetabolic state, during which adipose tissue remodeling occurs.

For decades, adipose tissue has been categorized into two distinct types: white adipose tissue (WAT), which primarily stores fat, and brown adipose tissue (BAT), which utilizes UCP1 for thermogenesis (1). Rodents and humans both possess BAT; however, whereas BAT depots remain throughout life in mice, human BAT diminishes with age. Although BAT is not a critical contributor to human thermogenesis, several studies have investigated the metabolic role of cold-activated BAT in glucose uptake and whole body energy expenditure in humans (18). Importantly, it was shown that BAT is present in adult humans and varies in amount and activity (27). Given its metabolic potential, intraspecies variations in BAT or tissues with BAT characteristics could have an important role in outcomes after burns.

Recently, it was discovered that certain WAT depots could be induced to develop BAT characteristics (i.e., UCP1 and mitochondrial biogenesis), a phenomenon known as browning. This phenotypic switch from WAT to BAT has garnered significant attention for its potential to be exploited in the treatment of obesity (1). In fact, several studies have shown that activation of browning in obese mice results in elevated lipolysis, resting energy expenditure, and weight loss (28). However, the potentially beneficial aspects of browning in obesity are detrimental in burn and cancer patients, who also undergo a browning response (12, 21, 24). Indeed, the browning of WAT has been reported to contribute to the progression of cancer-associated cachexia, a condition characterized by severe weight loss and muscle catabolism (12, 21). Similarly, we recently reported that browning of WAT after burns results in increases in lipolysis and plasma lipids, which accelerates hepatic steatosis. In turn, altered fat metabolism further fuels the hypermetabolic response in these burn patients (10, 24). Although the early inflammatory and delayed browning phases of the hypermetabolic response have been characterized individually, these phases are not discrete. We previously demonstrated that this delayed browning response relies on the production of bone marrow-derived IL-6, a proinflammatory cytokine. Additionally, this response is dependent on local infiltration of immune cells, namely M2 macrophages (3). Therefore, there is a link between inflammation and browning after burns, although the cross talk between these stages has not been completely elucidated at this point.

One of the critical mediators involved in both inflammation and adipose tissue remodeling after burns is nucleotide-binding and oligomerization domain, leucine rich repeat and pyrin domain containing 3 (NLRP3) inflammasome. NLRP3 inflammasome is a multiprotein complex that regulates inflammation and macrophage activity by cleaving the interleukins IL-1β and IL-18 into their bioactive forms in response to pathogens or injury (6, 7, 14–17, 19, 23, 30, 31). We have recently shown that NLRP3 is activated in burn patients and characterizes the early phase of the hypermetabolic response to burns (25). Although the inflammatory and immune role of NLRP3 has been previously investigated, its function in relation to adipose tissue remodeling has recently become apparent. A recent study revealed that NLRP3 induces oxidative stress and mitochondrial dysfunction in adipocytes and thus inhibits WAT browning (17). However, the activation and function of the NLRP3 inflammasome in adipose tissue remodeling in the context of burns has yet to be completely elucidated.

The present study addresses the role of NLRP3-mediated inflammation in browning of WAT after burns. Here, we utilized a murine model to investigate the impact of a genetic NLRP3 knockout (NLRP3^−/−) on browning. This study highlights the importance of a more complete understanding of this biphasic stress response to severe burns. Ideally, pharmacological interventions that target NLRP3-mediated inflammation to harness browning could be a therapeutic strategy in obesity. Conversely, dampening the browning response could improve morbidity and mortality in burn patients.

METHODS

Human samples.

Patients admitted to the Ross Tilley Burn Centre at Sunnybrook Hospital (Toronto, ON, Canada) or nonburn patients undergoing elective surgery were consented preoperatively for tissue collection. Approval for our study was obtained from the Research Ethics Board at Sunnybrook Hospital. We enrolled a total of 15 patients, stratified based on days after burn. Early burn tissue was obtained <7 days postinjury, and late burn tissue was obtained ≥7 days postinjury. Average total body surface area (TBSA) was 37 ± 15% for early burns and 34 ± 11% for late burns. Adipose tissue was obtained at various time points after burn and was immediately frozen at −80°C or transferred into fixative until time of analysis.

Animals and model.

Animal experiments were conducted in accordance with and approved by the Sunnybrook Research Institute Animal Care Committee (Toronto, ON, Canada). Wild-type (WT) C57/B6 and NLRP3^−/− male mice (6–8 wk old) were purchased from Jackson Laboratories (Bar Harbor, ME), housed at ambient temperature, and cared for in accordance with the Guide for the Care and Use of Laboratory Animals. All mice were anesthetized with 2.5% isoflurane and shaved along the dorsal spine region. Ringers lactate (2–3 mL) was injected subcutaneously in all treatment mice to protect the spine, and buprenorphine (0.05–0.1 mg/kg body wt) was injected for pain management. A full-thickness, third-degree scald burn encompassing 30%–35% TBSA was induced by immersing the back of the mice in 98°C water for 10 s and the ventral region for 2 s. Mice were euthanized at 3, 7, and 14 days after burn. Sham mice (control) underwent identical experimental procedures, with the exception of the burn injury. All tissues were harvested upon euthanization and stored in −80°C until analysis.

Histology and immunohistochemistry.

Inguinal WAT (iWAT) was collected, immediately fixed in 10% formalin, and then maintained in 70% ethanol before paraffin embedding. Subsequently, tissues were sectioned and incubated with anti-UCP1 (Abcam, cat. no. 100790) and anti-NLRP3 (BioLegend, cat. no. 101202) antibody followed by diaminobenzidine staining. Imaging was performed on a laser scanning microscopy confocal microscope (Zeiss, Germany). For quantification of adipocyte size, sections were imaged at ×20 magnification. The smallest and largest diameters of each adipocyte were manually measured using Leica Application Suite Version 4.3.0 (Leica Microsystems, Switzerland), and the average of the two values was used for subsequent analyses.

Gene expression using real-time PCR.

Total RNA isolated from adipose tissue was analyzed by quantitative real-time PCR. RNA was isolated from tissue and cells using TRIzol-chloroform (Life Technologies) with subsequent purification using the RNeasy Kit (Qiagen) according to the manufacturer’s instructions. RNA (2 µg) was transcribed to cDNA using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems). Real-time PCR was performed using the Applied Biosystems Step One Plus Real-Time PCR System. Gene expression was expressed relative to β-actin, and gene expression levels were determined using the following formula: 2^(-ΔCt) (4). Housekeeping gene expression was stable between groups, and samples with Ct values >25 were not included in our analyses. Primer sequences used are available upon request.

Cytokine profile.

Rodent plasma was collected, and plasma IL6 was measured using an IL-6 ELISA kit according to the manufacturer’s protocol (Abcam).

Western blotting.

Proteins from rodent adipose tissue were extracted in RIPA buffer containing phosphatase and protease inhibitor cocktails (Roche). Protein concentrations were determined by the BCA Protein Assay kit (Pierce, Mississauga, ON, Canada). Proteins were resolved by SDS-PAGE followed by Western blotting using the following antibodies at 1:1,000–1:5,000 concentration: UCP1 (Cell Signaling, MA) and α-tubulin (Cell Signaling). Species-appropriate secondary antibodies conjugated to horseradish peroxidase (BioRad, Mississauga, ON, Canada) were used and proteins visualized by enhanced chemiluminescence using the BioRad ChemiDoc MP Imaging System. Band intensities were detected, normalized, and quantified with the ChemiDoc and Image Laboratory 5.0 software (BioRad Laboratories, Hercules, CA). Antibody concentrations are expressed relative to α-tubulin.

Statistical analysis.

All data are represented as mean ± SE. Statistical analysis was performed using Student’s t test, one- and two-way ANOVA, and Mann–Whitney U test to compare groups, where appropriate. All graphs were created using GraphPad Prism 6.0 (San Diego, CA) and analyzed statistically using SPSS 20 (IBM Corporation, NY, NY), with significance accepted at *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 for comparison between time points and #P < 0.05, ##P < 0.01, ###P < 0.001, and ####P < 0.0001 for comparison between groups.

RESULTS

Upregulation of Inflammatory and Browning Responses in Human Adipose Tissue

Burn patients.

First, we compared expression of NLRP3 and its activation byproduct IL-1β in the adipose tissue of burn patients to normal (control) patients. Burn patients were classified based on how many days after burn the tissue was obtained. Early burns were deemed to be <7 days since injury, and late burns were ≥7 days (Tables 1 and 2) (25). Immunohistochemical staining for NLRP3 demonstrates increased expression in adipose tissue obtained from burn compared with normal patients, with the greatest NLRP3 positivity in early samples (Fig. 1A). Furthermore, burn patients demonstrate increased NLRP3 gene expression in adipose tissue compared with normal patients, which is in accordance with previous work (10.0 vs. 1.06, P = 0.09 early; 6.43 vs. 1.06, P < 0.05 late) (Fig. 1B). Similarly, the NLRP3 activation byproduct IL-1β is also elevated in burn compared with normal tissue (24.9 vs. 1.34, P = 0.10 early; 51.4 vs. 1.34, P < 0.05 late) (Fig. 1C).

Table 1.

Demographics for early burn patients

Demographic	Value
No. of burn patients (early)	7
Age, yr	44 ± 12
Males, n (%)	4 (57)
TBSA	37 ± 15
Days postburn	4 ± 1

Values are means ± SD for age, total body surface area (TBSA), and days postburn.

Table 2.

Demographics for late burn patients

Demographic	Value
No. of burn patients (late)	8
Age, yr	52 ± 22
Males, n (%)	4 (50)
TBSA	34 ± 11
Days postburn	14 ± 8

Values are means ± SD for age, total body surface area (TBSA), and days postburn.

Fig. 1.

Nucleotide-binding and oligomerization domain, leucine rich repeat and pyrin domain containing 3 (NLRP3) and the browning markers uncoupling protein 1 (UCP1) and PRDM16 are upregulated in human adipose tissue after burn. Immunohistochemical staining for NLRP3 and UCP1 at the early (<7 days) and late (≥7 days) phase after burn indicates greater NLRP3 positivity at early time points and greater UCP1 positivity at late time points (A). NLRP3 (B) and IL1β (C) gene expression in white adipose tissue (WAT) of healthy controls (n = 4) and early (n = 6) and late (n = 7) adult burn patients. UCP1 gene expression in WAT of healthy controls (n = 4) and early (n = 4) and late (n = 5) adult burn patients (D). PRDM16 gene expression in WAT of healthy controls (n = 4) and early (n = 7) and late (n = 7) adult burn patients (E). Average adipocyte diameter in controls (n = 34) and early (n = 39) and late (n = 50) adult burn patients (F). Values are presented as mean of burn or control ± standard error. Student’s t test and one-way ANOVA, burn vs. normal skin *P < 0.05, ****P < 0.0001; burn at different time points #P < 0.05, ####P < 0.0001.

Previous in vitro studies indicate that NLRP3-mediated inflammation attenuates UCP1 induction via IL-1β signaling (17). Since we demonstrated both increased NLRP3 and IL-1β expression in human burn tissue, we were interested in studying if NLRP3 inflammasome activation is implicated in browning in vivo. To that end, we subsequently examined the immunohistochemical and gene expression of UCP1 as a marker for human adipose tissue browning. UCP1 staining was prominent in burn adipose tissue compared with normal tissue similar to previous reports. In particular, staining intensity was greatest in late burns (Fig. 1A). In addition to UCP1, we examined gene expression of the browning marker PRDM16 (Fig. 1, D and E). Our PCR data corroborated our previous results, indicating higher expression of browning genes in late burns compared with normal adipose tissue (12.9 vs. 1.12, P < 0.05 for UCP1; 5.44 vs. 0.72, P < 0.05 for PRDM16). Furthermore, quantification of adipocyte diameter in our normal compared with burn tissue demonstrated a significant decrease in droplet size in the late burn group (83.9 vs. 47.0 µM, P < 0.0001), an important feature of browning (Fig. 1F). These results indicate that NLRP3 and its activation byproducts are expressed early after burn in adipose tissue. Interestingly, the expression time course for NLRP3 and IL-1β overlap with the key browning markers UCP1 and PRDM16, suggesting a potential link between inflammation and browning in humans.

Upregulation of inflammatory responses in murine adipose tissue after burn.

Previous studies have demonstrated that murine hypermetabolic burn models are clinically relevant and demonstrate similar findings to burn patients (3). To assess the inflammatory mechanisms underlying the response after burns, we subjected WT mice to a severe (TBSA 30%–35%) dorsal and ventral scald burn. We assessed gene expression of NLRP3 after burn and confirmed NLRP3 inflammasome activation in iWAT (Fig. 2). NLRP3, IL-1β, and IL-18 gene expression increase acutely after a burn injury (3 days) (Fig. 2, A–C). NLRP3 and IL-1β gene expression peaks at 7 days after burn and is substantially elevated relative to expression at 3 days (14.4 vs. 3.39, P < 0.05 for NLRP3, 28.6 vs. 7.20, P < 0.01 for IL-1β) (Fig. 2, A and B).

Fig. 2.

Nucleotide-binding and oligomerization domain, leucine rich repeat and pyrin domain containing 3 (NLRP3) and its byproducts are upregulated in murine inguinal white adipose tissue (iWAT) after burn. Gene expression of NLRP3 (A; n = 4, 5, 4, 4), IL1β (B; n = 5, 4, 4, 4), and IL18 (C; n = 4, 4, 6, 5) in murine iWAT in sham and 3, 7, and 14 days after burn. Values are presented as mean burn or sham ± standard error. One-way ANOVA, burn vs. sham *P < 0.05, ***P < 0.001; burn at different time points #P < 0.05, ##P < 0.01, ###P < 0.001.

Although NLRP3 is an acute-phase mediator, elevated gene expression of NLRP3 and its activation byproduct, IL-1β, beyond the acute phase indicates a potential overlap with the initiation of browning responses. Therefore, to assess the mechanistic and functional role of NLRP3 in the browning response after burns, we utilized an NLRP3^−/− model. First, we confirmed that there was no increase in gene expression of NLRP3, IL-1β, or IL-18 in our NLRP3^−/− (Fig. 3). After verifying nonexistent NLRP3 signaling in NLRP3^−/− mice postinjury, we subsequently assessed burn-induced iWAT browning in these mice compared with WT.

Fig. 3.

Nucleotide-binding and oligomerization domain, leucine rich repeat and pyrin domain containing 3 (NLRP3)^−/− have diminished gene expression of NLRP3 (A; n = 4, 4, 5), IL1β (B; 4, 5, 4), and IL18 (C; n = 6, 6, 6) compared with wild-type (WT). Values are presented as mean burn ± SE. One-way ANOVA and Student’s t test, WT at different time points *P < 0.05, **P < 0.01, ***P < 0.001; WT vs. NLRP3^−/− burn #P < 0.05, ##P < 0.01. KO, knockout.

Genetic deletion of NLRP3 results in a persistent and augmented browning response.

First, we assessed the timeline for the browning response in our murine burn model. Interestingly, expression of UCP1, peroxisome proliferator-activated receptor γ (PPARγ), CIDEA, and PGC1α peaked at 7 days, mirroring the expression profile of NLRP3 and IL-1β (Fig. 4, A–D). In a similar fashion to the inflammatory markers in Fig. 2, UCP1, PPARγ, CIDEA, and PGC1α gene expression trended down at 14 days. Immunohistochemical staining for UCP1 corroborated these results, with greater UCP1 positivity at 3 days compared with shams and 14 days, corresponding with “early” and “late” UCP1 expression in our human adipose tissue (Fig. 4E) (1, 20). Additionally, adipocyte diameter decreased significantly at both 3 and 14 days after burn compared with shams, analogous to our human data (Fig. 4F).

Fig. 4.

Browning markers are upregulated in murine adipose tissue after burn. Gene expression of uncoupling protein 1 (UCP1) (A; n = 4, 4, 10, 4), peroxisome proliferator-activated receptor γ (PPARγ) (B; n = 4, 4, 6, 5), CIDEA (C; n = 4, 4, 5, 5), and PGC1α (D; n = 4,4,5,5) in murine inguinal white adipose tissue in sham and 3, 7, and 14 days after burn. Immunohistochemical staining for UCP1 indicates positivity as early as 3 days after burn (E). Average adipocyte diameter in wild-type (WT) sham (n = 105), WT 3 days after burn (n = 156), and WT 14 days after burn (n = 194) (F). Values are presented as mean burn or sham ± SE. One-way ANOVA, burn vs. sham *P < 0.05, **P < 0.01, ****P < 0.0001; burn at different time points #P < 0.05, ##P < 0.01.

In NLRP3^−/−, expression of browning genes UCP1, PPARγ, CIDEA, and PGC1α demonstrated similar gene upregulation patterns to WT at 3 and 7 days after burn (Fig. 5, A–D). However, our PCR results surprisingly revealed that lack of NLRP3 significantly heightened the browning response in the iWAT following burn injury. Gene expression of PPARγ and CIDEA at 3 days in NLRP3^−/− was significantly higher than WT (5.74 vs. 0.29, P < 0.05 for PPARγ; 26.0 vs. 0.71, P < 0.05 for CIDEA). Additionally, gene expression of UCP1 and PGC1α at 7 days in NLRP3^−/− remained significantly higher than WT (7,406 vs. 3,894, P < 0.05 for UCP1; 20.6 vs. 2.52, P < 0.01 for PGC1α).

Fig. 5.

Amplified and persistent increases in expression of browning markers in nucleotide-binding and oligomerization domain, leucine rich repeat and pyrin domain containing 3 (NLRP3)^−/−. Increased gene expression of uncoupling protein 1 (UCP1) (A; n = 6, 6, 5), peroxisome proliferator-activated receptor γ (PPARγ) (B; n = 6, 6, 5), CIDEA (C; n = 6, 4, 4), and PGC1α (D; n = 6, 6, 4) in murine inguinal white adipose tissue at 3, 7, and 14 days after burn in NLRP3^−/− compared with wild-type (WT). Immunohistochemistry indicates prominent UCP1 staining even at 14 days as opposed to WT (E). Average adipocyte diameter in knockout (KO) sham (n = 48), KO 3 days after burn (n = 221), and KO 14 days after burn (n = 163) compared with WT (F). Increased circulating IL6 (burn) (G) and IL6 (burn/sham) (H) in NLRP3^−/− (n = 5, 5, 4) that is significantly elevated compared with WT (n = 6, 4, 4) acutely. Values are presented as mean burn ± SE. Two-way ANOVA with Bonferroni posttest, WT vs. other time points *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; WT vs. NLRP3^−/− burn #P < 0.05, ##P < 0.01, ####P < 0.0001.

Notably, in addition to an amplified response, NLRP3^−/− demonstrated a persistent elevation in all of the key aforementioned browning markers compared with WT. Although WT iWAT demonstrated diminished gene expression of all 4 browning markers at 14 days relative to 7 days, gene expression at 14 days in NLRP3^−/− steadily increased for all markers (13,582 vs. 1,324, P < 0.01 for UCP1; 16.6 vs. 0.28, P < 0.05 for PPARγ; 86.4 vs. 0.47, P < 0.01 for CIDEA; 47.2 vs. 0.20, P < 0.01 for PCG1α) (Fig. 5, A–D). In accordance with an exaggerated browning response, immunohistological analysis confirmed more prominent remodeling of iWAT depots in NLRP3^−/−. Abundant levels of multilocular, UCP1+ adipocytes were detected in NLRP3^−/− iWAT compared with WT, with the highest intensity of UCP1 positivity at 14 days after burn (Fig. 5E). Furthermore, average adipocyte diameter was significantly smaller in NLRP3^−/− compared with WT (16.0 vs. 18.1 at 3 days, P < 0.01; 11.3 vs. 17.4 at 14 days, P < 0.0001) (Fig. 5F).

Subsequently, we assessed IL-6 levels in NLRP3^−/− compared with WT, since IL-6 is a critical inflammatory cytokine that is necessary for browning of WAT in response to burns (3). We show here that circulating plasma IL-6 levels are elevated acutely (3 days) in NLRP3^−/− compared with WT (478.9 vs. 67.2 pg/mL, P < 0.05) (Fig. 5G). To correct for interstrain variability, we subsequently assessed IL-6 levels expressed as a ratio of burn to sham. Similar to browning marker expression, circulating plasma IL-6 levels expressed as a ratio of burn to sham was consistently elevated at all time points in NLRP3^−/− compared with their WT counterparts (Fig. 5H). Circulating IL-6 levels were significantly higher after burn (40.6 vs. 1.55, P < 0.05 at 3 days; 10.8 vs. 5.41, P < 0.05 at 14 days) but trended down by 14 days. Potentially, exaggerated circulating IL-6 levels are a compensatory response for diminished NLRP3-mediated inflammation. This in turn promotes and sustains the amplified browning response seen in NLRP3^−/−. These findings provide compelling evidence that NLRP3 inflammasome activation is critical for the attenuation of beige adipocyte formation in response to burn injury.

Inhibition of NLRP3 augments browning in mice following burn injury.

Glyburide is a widely prescribed treatment for blood glucose regulation in patients with type 2 diabetes that also functions as an NLRP3 activation inhibitor (22). Here, we utilized glyburide as a means to compare inhibition to a genetic deletion of NLRP3. We compared parameters between untreated WT burn mice and burn mice administered daily injections of glyburide for 7 days (13). We chose 7 days for this component of our study because the peak of UCP1 gene expression in murine iWAT occurs at this time point. Subsequently, we determined whether glyburide treatment has an analogous effect on burn-induced browning to NLRP3^−/−. Consistent with previous results, Western blotting for UCP1 in iWAT from WT mice indicated increased protein expression relative to sham (11.01 vs. 1.00, P < 0.05) (Fig. 6, A and B). These results indicate that glyburide treatment and by extension NLRP3 blockade decreases mitochondrial coupling following burn injury. Our findings provide strong evidence that the innate inflammatory response can be selectively and pharmacologically targeted to enhance adaptive thermogenesis. Although this strategy may prove to be beneficial in models of obesity and diabetes, it may be detrimental in hypermetabolic conditions such as burns, massive trauma, and cancer.

Fig. 6.

Greater uncoupling protein 1 (UCP1) expression in inguinal white adipose tissue of glyburide (Gly)-treated mice compared with wild-type. A: representative cropped Western blot of UCP1 and α-tubulin at 7 days after burn. B: protein expression of UCP1. Sham (n = 5), burn (n = 6), Gly treated (n = 5). Values are presented as mean ± SE. One-way ANOVA, burn vs. sham *P < 0.05, ***P < 0.001; treated vs. untreated ##P < 0.01.

DISCUSSION

The hypermetabolic response to burns increases lipolysis, whole body catabolism, and multiorgan dysfunction, ultimately augmenting patient morbidity and mortality (8, 11). The two phases of this burn-induced hypermetabolic response are the early inflammatory phase, mediated by the NLRP3 inflammasome, and the late metabolic phase, consisting of WAT browning (8, 9, 11). However, the link between these two postburn stages needs to be further elucidated. In this study, we addressed the question of whether 1) the temporal expression profile of the NLRP3 inflammasome and WAT browning in the adipose tissue of burn patients and mice are distinct or identical, 2) whether the absence of NLRP3 expression after burns would interfere or augment burn-induced WAT browning, and 3) whether inhibiting NLRP3 inflammasome activation after burns would attenuate or augment WAT browning.

Our results demonstrate that NLRP3 activation in the adipose tissue of mice and burn patients was enhanced in the early phase (3 days in mice, <7 days in humans) and reduced in the late phase (14 days in mice, >7 days in humans). Conversely, WAT browning was more significantly pronounced in the late phase after burns in both mice and patients. Considering that NLRP3 is an acute-phase mediator, a notable finding of our study was its persistent downstream effects on adipose tissue remodeling. We propose that the alteration in WAT browning is related to lack of NLRP3 activation, although there is an associative rather than causative relationship between NLRP3, IL-1β, UCP1, and PRDM16 based on the current data. However, previous work indicates that NLRP3 activation is a critical attenuator of UCP1 induction in vitro (17). Importantly, the inhibitory effect of NLRP3 activation on browning is mediated by IL-1β, and blocking IL-1β protects thermogenesis (17). Therefore, we hypothesize that genetic deletion of NLRP3 has a similar effect on UCP1-mediated browning, suggesting that NLRP3 negatively regulates browning after burns.

Recently, we demonstrated that NLRP3 gene status impacts macrophage polarization and migration to critical immunometabolic organs (29). Additionally, we identified that M2 anti-inflammatory macrophages and IL-6 signaling are critical regulators of WAT browning after burns (2, 3). Although several studies have pointed to the role of macrophages and proinflammatory cytokines in browning, little is known about the role of NLRP3-mediated macrophage recruitment and polarization to adipose tissue. Potentially, lack of NLRP3 results in a greater M2 macrophage profile, promoting browning. Alternatively, blocking NLRP3-mediated inflammation could promote a compensatory, persistent NLRP3-independent inflammatory response. Indeed, we demonstrate persistently elevated levels of circulating IL-6 in our NLRP3^−/−, highlighting potential mechanisms by which NLRP3 expression and activation can regulate the browning response.

A limitation is that although we hypothesize that alterations in IL-6 between NLRP3^−/− and WT mice could account for differing browning responses, this is still speculative. Previous work using WT mice transplanted with IL-6^−/− hematopoietic cells indicated a reversal of the browning phenotype, whereas IL-6^−/− mice transplanted with WT hematopoietic cells were shown to recover WAT browning (2). This indicates that IL-6 is necessary for WAT browning; however, further work is needed to determine if IL-6 administration is sufficient. In humans, IL-6 infusion alone causes an increase in net skeletal muscle protein breakdown, similar to cachexia (26). Since browning of WAT contributes to the progression of cancer-associated cachexia, this potentially indicates that IL-6 could be a main regulator of browning-related catabolic responses.

An important consideration in this study is the differing timeline for WAT browning and NLRP3 activation in mice and men. Here, we show evidence of increased UCP1 staining and alterations in adipocyte droplet size at our early time point (3 days). On the other hand, although Patsouris et al. (20) demonstrated morphological changes in human tissue at 0–3 days after burn, significant changes in adipocyte size and expression of browning markers occurs later during the course of admission (10–21 days). Although we have not conducted a time course study for UCP1 expression in human tissue in this study, we anticipate that WAT browning in humans might exhibit a similar pattern of browning but is a delayed response compared with mice.

Our findings suggest that there is a link between inflammation and WAT browning after burns. Potentially, there is a trade-off between inflammation and browning. As NLRP3 expression decreases after the acute phase, UCP1 expression begins to rise, culminating in the increased WAT browning seen histologically in the late phase in humans and mice. We hypothesize that NLRP3 activation serves as a dampening mechanism to counteract browning, which could account for that fact that NLRP3^−/− lacks NLRP3 and IL-1β expression acutely (3 days) and demonstrate acute upregulation of browning genes (e.g., CIDEA, PPARγ, PGC1α) and a persistent, augmented browning response. Conversely, WT mice have significantly elevated NLRP3 and IL1β at 7 days and IL18 at 3 days, which is followed by a decrease in browning responses at 14 days. However, mechanisms underlying the inflammasome-adipose tissue cross talk in response to trauma will need to be further investigated at this point. Interestingly, this is one of the first studies demonstrating a clinically approved mechanism to enhance adaptive thermogenesis. These results lay the foundation for the development of pharmacologic interventions to modulate hypermetabolism and metabolic complications after burns.

GRANTS

This work was supported by grants from the Canadian Institutes of Health Research (Grant no. 123336), the Canada Foundation for Innovation Leader’s Opportunity Fund (Grant no. 25407), and the National Institutes of Health (2-R01-GM-087285-05A1). R. Vinaik is a recipient of the Frederick Banting and Charles Best Canada Graduate Scholarship (CGS-D).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

R.V. and D.B. conceived and designed research; R.V. and D.B. performed experiments; R.V. and D.B. analyzed data; R.V., D.B., and A.A. interpreted results of experiments; R.V. and D.B. prepared figures; R.V., D.B., and A.A. drafted manuscript; R.V., D.B., A.A., and M.G.J. edited and revised manuscript; D.B., A.A., and M.G.J. approved final version of manuscript.