β3-adrenergic receptor agonist causes acute thermogenic metabolic crisis in ACSS1-K635Q knock-in mice

TL;DR

This study shows that a genetic mutation in ACSS1 causes a severe metabolic crisis in mice when exposed to a β3-adrenergic receptor agonist.

Contribution

The study reveals a novel role for non-acetylated ACSS1 in thermoregulation and fatty acid metabolism.

Findings

Acss1K635Q/K635Q mice showed hypothermia and metabolic failure after CL-316243 injection.

Mitochondria from mutant mice had impaired respiration and unresponsive lipid metabolism.

Beige adipocytes from mutants displayed reduced fatty acid oxidation and altered UCP1 regulation.

Abstract

Mitochondrial Acyl-Coenzyme Synthetase Short Chain Family Member-1 (ACSS1) converts free acetate into acetyl-coenzyme A (acetyl-CoA), regulated, in part, by acetylation at lysine 635 (ACSS1-K635). We challenged our ACSS1 constitutive acetylation mimic knock-in (K635Q) mice by injecting a β3-adrenergic receptor agonist, CL-316243 (CL), to induce a thermogenic response. Strikingly, we show that Acss1K635Q/K635Q mice exhibit hypothermia and acute metabolic crisis following CL stimulus, as shown by significantly reduced oxygen consumption, carbon dioxide production, respiratory exchange ratio, and heat production. We also observed histological differences in both brown adipose tissue (BAT) and subcutaneous white adipose tissue (WAT), accompanied by altered expression and regulation of lipogenic enzymes and Uncoupling Protein 1 (UCP1) in Acss1K635Q/K635Q. In contrast to wild-type adipose tissues, Acss1K635Q/K635Q did not show changes in acetyl-CoA and acetate levels in response to CL, and mitochondria isolated from BAT displayed impaired respiration on palmitate. Lastly, beige adipocytes differentiated ex vivo from Acss1K635Q/K635Q mice showed altered response to the adenylate cyclase stimulator, forskolin, with unresponsive mitochondria and lipogenic lipid droplets, and lower fatty acid oxidation activity. These results suggest that non-acetylated ACSS1 plays an essential role in thermoregulation and the ability to metabolize free fatty acids.

Figures

Full text

1. Introduction

Acetyl-Coenzyme A (acetyl-CoA) is essential to energy metabolism and lipid synthesis, as well as epigenetic and protein regulation 1, 2. In mammalian cells, the majority of acetyl-CoA is generated through glycolysis and fatty acid oxidation (FAO). As such, acetyl-CoA functions as a small molecule metabolic sensor and regulates protein function through non-enzymatic acetylation of lysine residues 3, 4. Thus, increased FAO, as under both fasting and high-fat diet conditions, increases acetylation of ACSS1 and other mitochondrial proteins 10, 11. Acetyl-CoA Synthetase Short Chain Family Member-1 (ACSS1, previously known as AceCS2) catalyzes the ligation of acetate and coenzyme A to generate acetyl-CoA in the mitochondria, and plays a role in lipid metabolism, at least under certain nutrient conditions 5. ACSS1 enzymatic activity is regulated, in part, through acetylation at lysine 635 in mice (K642 in humans), which reduces ACSS1 activity, and by deacetylation by Sirtuin 3 (SIRT3) 6-9.

As the primary mitochondrial deacetylase, SIRT3, plays an important role in regulating metabolism, including FAO 12 and brown fat thermogenesis 13. Brown adipose tissue (BAT) dissipates energy as heat during non-shivering thermogenesis. Studies in both humans and murine models have demonstrated that higher levels of brown fat are protective against developing type 2 diabetes and cardiovascular diseases 14-16. When stimulated by cold or thyroid hormone, BAT uses glucose and free fatty acids (FFAs) released from triglycerides by lipases for thermogenesis. Fully active thermogenic adipose tissue can increase whole-body energy expenditure by more than 100% in mice 17 and 40-80% in humans 18.

Importantly, white adipose tissue (WAT) can go through a browning process, with “beige” adipocytes appearing in WAT in response to increased thermogenic demand from cold stimulus or pharmaceuticals. Both beige and brown adipocytes contain multilocular lipid droplets and a high density of mitochondria that express Uncoupling Protein 1 (UCP1). When activated, UCP1 creates an alternate portal for protons back across the inner mitochondrial membrane, thereby uncoupling oxidative phosphorylation and generation of the proton gradient from ATP synthesis and dissipating the energy potential as heat instead 19. Adipocyte thermogenic capacity thus depends primarily on catabolism of FFAs through oxidative metabolism 20. Classically, adipose-tissue thermogenesis is driven by sympathetic nerve-mediated adrenergic signaling, which stimulates lipolysis. FFAs serve both as fuel for thermogenesis and as allosteric activators of UCP1 function. Indeed, long-chain fatty acids are uniquely required for UCP1-induced uncoupling 21. WAT is key in this process by releasing FFAs into circulation through lipogenesis to support BAT thermogenesis 22.

While acetylation has been shown to decrease ACSS1 ligase activity 7, the physiological role of ACSS1 acetylation status in thermogenesis has not been characterized. In this regard, we generated an ACSS1-acetylation (Ac) mimic knock-in mouse, where lysine 635 was mutated to glutamine (K635Q), which structurally and biochemically mimics acetylated lysine 5, 9. In metabolic conditions requiring increased reliance on FAO, including fasting 5 or on a ketogenic diet 23, these mice exhibited hepatic steatosis and altered lipid profile compared to control mice under the same conditions. These data suggested that Acss1^K635Q/K635Q^ mice have difficulty with increases in metabolic demand that depend on FAO. For this reason, we investigated the role of ACSS1 acetylation in the thermogenic response, which is highly dependent on fatty acid metabolism, by injecting Acss1^K635Q/K635Q^ mice with the specific β3-adrenergic receptor agonist, CL-316243 (CL). Strikingly, we found Acss1^K635Q/K635Q^ mice to be intolerant of CL injection, exhibiting acute metabolic decompensation, hypothermia, and in some cases, death 24 hours after injection.

2. Methods

2.1. Animal model

All animal experiments were conducted in compliance with the National Institutes of Health (NIH) Guidelines for Humane Care and the Use of Laboratory Animals, and all murine studies were approved by the University of Texas Health San Antonio (UTHSA) Institutional Animal Care and Use Committee. Our Acss1^K635Q^ mouse was generated in a C57BL/6 background and has been described in detail previously 5. Mice were maintained in specific pathogen-free conditions at 21 ± 2 °C with 12h:12h light:dark cycle in the Animal Facility at UTHSA. Mice had free access to food and water, and were given pelleted Inotiv Teklad Rodent Diet traditional formula LM-485. For all studies, we used four-month-old male and female mice, as indicated for each experiment. Confounders such as order of measurements and cage location were not controlled.

Thermogenic stimulation: CL-316243 (CL) was prepared in saline at a solution of 0.2 mg/mL and then injected in mice intraperitoneally at 1 mg/kg. Then mice were monitored every hour for measurements and health status hourly. Temperature was measured using a mouse rectal thermometer. A Countour Next EZ glucometer and glucose strips were used to measure blood glucose from a small tail snip. Rectal temperature was monitored at the time points indicated with a rodent rectal temperature probe Ret 3 (Kent scientific).

Evaluation of circulating lipids: Whole blood was collected from the mice and serum was obtained by letting the blood to clot for one hour and then spun 10 minutes at 2000xg. Supernatants were then stored minus 80. Serum was sent to IDEXX to analyze free fatty acids, triglycerides and cholesterol based on a rodent panel.

Respirometry measurements: Mice were individually housed in metabolic cages, with free access to food and water. Mice were acclimated for two days before recording baseline parameters with PhenoMaster (Sable Systems) and following injection with CL (1 mg/kg) the mice are used for measurements of respiration for the following 24 hours.

2.2. Histology

Tissue blocks of formalin-fixed, paraffin-embedded mouse tissues were processed by the Laboratory Medicine core, UT Health San Antonio using standard methods. When adequate color development was seen, slides were washed in water to stop the reaction, counterstained with Meyer's hematoxylin (Dako), and covered with a Permount mounting medium (Richard-Allan Scientific). The micrographs were taken under a light microscope (Leica). Representative images are shown.

2.3. Mitochondria isolation and oxygen consumption assays

Approximately 10 mg BAT was collected on ice for Seahorse analysis. Mitochondria were isolated from tissue as previously described. In brief, the tissue was placed in a mitochondrial isolation buffer [70 mM sucrose, 210 mM mannitol, 1 mM ethylene glycol tetraacetic acid (EGTA), 5 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), pH 7.2] and homogenized using a glass Teflon Dounce homogenizer. This solution was then centrifuged twice at 1000 g for 5 min at 4 °C to remove cell debris. The mitochondrial pellet was resuspended in a mitochondria assay solution (MAS) (70 mM sucrose, 220 mM mannitol, 5 mM phosphate, 5 mM magnesium chloride, 1 mM EGTA, 2 mM HEPES, pH 7.4). Protein was quantified by bicinchoninic acid (BCA) assay (Pierce) and concentrations adjusted with MAS to 3 μg of protein. The Seahorse plate containing 100 μL/well of the mitochondrial suspensions was centrifuged at 2000 g for 5 min at 4 °C with free deceleration. Finally, the well volume was brought up to 180 μL by addition of MAS supplemented with malate. The oxygen consumption rate (OCR) was measured in the presence of 10 mM BSA-conjugated palmitate, using a Seahorse XFe96 analyzer at 37 °C, and 5 mM glycerol-3-phosphate (G3P) and 3 mM guanosine diphosphate (GDP) were added sequentially at controlled time points during the OCR measurements. Data were collected and processed using Agilent Seahorse Wave software.

2.4. Western Blotting and Imaging

Total protein lysate preparation from tissues for immunoblot have been described 24-27. Protein concentrations were determined by Pierce BCA Protein Assay, and 20 µg from each sample was separated by NuPAGE^TM^ Bis-Tris 4-12% gradient sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Invitrogen), transferred onto polyvinylidene difluoride (PVDF) membrane (Bio-Rad), probed with primary antibody (Table 1) and then with horseradish peroxidase (HRP) conjugated to either a mouse or rabbit secondary antibody (Cell Signaling), and visualized with enhanced chemiluminescence (Fisher Scientific; SuperSignal^TM^ West Femto Maximum Sensitivity Substrate).

Immunoblots were imaged were obtained using ProteinSimple FluorChem M, using the Auto Exposure feature to takes successively longer exposures of the membrane until an optimum exposure time is achieved. The subsequent images are then acquired in standard resolution, using 4x4 pixel binning (832x626-pixel images).

2.5. Quantitative real-time polymerase chain reaction analysis

Total RNA was extracted and purified from iWAT tissue and cells using the RNeasy Mini Kit (Qiagen), following the manufacturer's instructions. mRNA concentrations were determined with a Take3 Micro-Volume Plate (BioTek) and normalized to 150 ng prior to cDNA synthesis. Complementary DNA was generated using random hexamer primers and the iScript cDNA synthesis kit (Bio-Rad). Quantitative RT-PCR was carried out using the SsoAdvanced™ Universal SYBR Green Supermix (Bio-Rad) on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad). Relative gene expression was quantified using the 2^-∆∆Ct^ method, with transcript levels normalized to 18S rRNA. Primer specificity was confirmed via melting curve analysis (see Table 2 for sequences).

2.6. Acetyl-Coenzyme A and Acetate Quantification

We used frozen BAT and iWAT to quantify acetyl-CoA and acetate using the Abcam fluorometric PicoProbe^TM^ acetyl-CoA (ab87546) and colorimetric acetate (ab204719) assay kits, respectively. Results were normalized to protein, as measured by Pierce BCA assay.

2.7. Cell Culture

Isolation: Subcutaneous adipose depots were harvested from euthanized WT and Acss1^K635Q/K635Q^ mice and digested in collagenase type I. The adipose-derived stem cells were enzymatically isolated following a previously published protocol 28. Briefly, digestion was performed in an orbital shaker at 37 °C for 60 min. The digest was then centrifuged and the floating adipocyte layer discarded. The pelleted stromal vascular fraction was washed two times with complete media [Dulbecco's Modified Eagle Medium (DMEM) with Ham's F12 nutrient mixture, supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic] and plated on tissue culture plastic. Cells at passages 2 to 5 were used for experiments.

Adipose derived stem cells beige differentiation: Cells (passage 2-4) were cultured in 24-well tissue culture plates and incubated in complete media. When cells reached 95% confluence, differentiation was initiated. For beige differentiation, cells were incubated for 4 days in induction media (complete media with 5 μg/mL insulin, 10 μM forskolin, 2 μg/ml dexamethasone, 125 μM indomethacin, 0.5 μM rosiglitazone and 1 nM triiodothyronine) and then maintained in complete media with 5 μg/ml insulin, 10 μM forskolin, 1 μM rosiglitazone and 1 nM triiodothyronine. Differentiation media was changed every other day until harvest.

Evaluation of fatty acid oxidation, lipid droplets, and mitochondrial membrane potential in mature adipocytes in vitro: FAO Blue was used at 1 μM concentration in DMEM without serum or phenol red. Beige differentiated stromal vascular cells were incubated in this media for 15 minutes in a humidified incubator and then directly imaged with a Leica Stellaris Sp5 confocal with the 405 laser line. Images were acquired at 63x and then fluorescence intensities were analyzed with the LAX software. To assay lipid droplets and mitochondrial membrane potential, we used bodipy at 1 μM and tetramethylrhodamine methyl ester (TMRM) ratiometric dye at 10 nM, and incubated cells for 10 minutes in a humidified incubator before imaging at 488 and 563 excitation. The lipid droplets and mitochondria were analyzed with LAX software by binning for size, count, and fluorescence intensity.

2.8. Statistical Analysis

Densitometric analysis of protein quantitation was determined by ImageJ software v0.5.5. (National Institutes of Health; http://imagej.nih.gov/ij Java 1.8.0- internal). Statistical differences were calculated as indicated in the figure legends by one- or two-way ANOVA, multivariate analysis, or unpaired Student's t-test in Microsoft Excel v2501 or GraphPad Prism 10.4.1. P-values were two-sided, and tests were considered significant at P < 0.05. All measurements were taken from distinct samples and replicates are biological. No data points were excluded. The number of replicates (mice) for each experimental group is given in the figure legends in the main text and supplement.

3. Results

3.1. Acss1K635Q/K635Q

mice are intolerant to β3-adrenergic receptor agonist

As we have previously demonstrated, Acss1^K635Q/K635Q^ mice are phenotypically normal, except for a notably smaller body mass in males 5. Here, we show this size difference continues to be significant at three months on standard chow fed ad libitum (Figure 1A). We found no difference in body temperature or blood glucose levels in either sex (Figure 1B,C). Our previous studies showed that Acss1^K635Q/K635Q^ mice seem to have impaired lipid metabolism and develop fatty liver under nutrient conditions demanding greater reliance on FAO, such as fasting 5 and a ketogenic diet 24. Here, we turn our focus from the liver to adipose tissue, and test the thermogenic response in Acss1^K635Q/K635Q^ mice.

To induce a thermogenic response, three-month-old mice were administered the potent selective β3-adrenergic receptor agonist, CL-316243 (CL), by intraperitoneal (IP) injection at 1 mg/kg. Rectal temperature was monitored every 2 hours and blood glucose hourly. Both male and female Acss1^K635Q/K635Q^ mice exhibited a drop in body temperature and blood glucose levels that was greatest after 24 hours, and more significant in female mice (Figure 1D-G), compared to no change in WT mice. In addition, roughly 25% of Acss1^K635Q/K635Q^ mice died 24 hours after injection. These results demonstrate Acss1^K635Q/K635Q^ mice are acutely intolerant to β3-adrenergic stimulation by CL and unable to maintain blood glucose and proper thermogenic function. In addition, these data further suggest that these mice are vulnerable to conditions that require increased fatty acid metabolism.

We further characterized the response to CL in female mice since they displayed more severe effects. In this regard, serum lipids were measured in mice given saline versus 4- and 24-hours post CL injection. We saw an increase in serum triglycerides 4 hours post-injection in both wild-type (WT) and Acss1^K635Q/K635Q^ mice, with minimal difference between genotypes (Figure 1H). In contrast, Acss1^K635Q/K635Q^ mice showed a significant increase in circulating FFA post-injection (Figure 1I), suggesting that CL stimulates lipolysis but that Acss1^K635Q/K635Q^ mice are unable to effectively utilize circulating FFA. While cholesterol levels followed a similar trend in both genotypes, the dip and subsequent increase in WT post-injection appeared to be of somewhat greater magnitude (Figure 1J).

3.2. Acss1K635Q/K635Q

mice are unable to adapt to the metabolic demand of CL stimulation

To further characterize the metabolic response of three-month-old females to CL stimulation, we performed respirometry measurements in individualized cages. While acclimating the mice for two days, we did not see major metabolic differences as we had previously in male mice 5. However, following CL injection, we saw dramatic effects in the metabolic parameters of Acss1^K635Q/K635Q^ mice. While WT mice responded to CL with increased O_2_ consumption (Figure 2A), Acss1^K635Q/K635Q^ mice showed a dramatic decrease in O_2_ consumption, CO_2_ production, respiratory exchange ratio, and heat (Figure 2A-D), which may play a mechanistic role in the 25% lethality observed 24 hours post-injection. These data clearly demonstrate that Acss1^K635Q/K635Q^ mice are intolerant to the thermogenic stimulus elicited by CL, suggesting a potential acute disruption of metabolism.

We collected subcutaneous inguinal WAT (iWAT) and BAT at 0, 4, and 24 hours after CL injection to evaluate tissue histology and protein expression levels. We observed that, following injection, the suprascapular subcutaneous WAT depot became less apparent in WT mice but remained unchanged in* Acss1^K635Q/K635Q^* (Suppl. Figure 1A), suggesting that Acss1^K635Q/K635Q^ mice were unable to utilize lipids from WAT depots during thermogenic stimulation.

Hematoxolin and eosin (H&E) staining showed a homogenous distribution of smaller lipid droplets in BAT from WT mice 24 hours after CL injection (Figure 2E, bottom left panel, and Suppl. Figure 1B). In contrast, BAT from Acss1^K635Q/K635Q^ mice exhibited disorganized lipid droplet distribution with some cells containing large lipid droplets while others appeared completely exhausted with areas of dark staining (Figure 2E, bottom right panel, and Suppl. Figure 1B). H&E staining revealed a similar trend in iWAT, with Acss1^K635Q/K635Q^ showing a heterogenous distribution of lipid droplets (more cells with smaller lipid droplets) as well as multilocular cells, which are not normally found in this tissue under basal conditions (Figure 2F, upper panels). After CL injection, iWAT from WT mice displayed smaller adipocytes and some multilocular cells (Figure 2F, lower left), while Acss1^K635Q/K635Q^ showed numerous multilocular cells and areas of lipid exhaustion (Figure 2F, lower right panel, and Suppl. Figure 1C). These results imply a disruption of the FFA mobilization required for the thermogenic response and suggests that Acss1^K635Q/K635Q^ mice are unable to metabolically adapt to the acute thermogenic response elicited by CL.

We also performed seahorse oxygen consumption assays with mitochondria isolated from WT and Acss1^K635Q/K635Q^ BAT. In the presence of 10 mM BSA-conjugated palmitate, Acss1^K635Q/K635Q^ BAT mitochondria showed lower basal respiration and reduced response to glycerol-3-phosphate (G3P), while inhibiting UCP1 with guanidine diphosphate (GDP) reduced oxygen consumption rate (OCR) by approximately 64% in both genotypes (Figure 2G). These results further suggest that mitochondria from Acss1^K635Q/K635Q^ BAT show deficient respiration in the presence of palmitate, pointing out to deficient FAO metabolism which may impair thermogenesis.

3.3. Acss1K635Q adipose tissue displays dysregulated lipogenic signaling

Immunoblots were done to assess relative levels of major lipogenic, lipolytic, mitochondrial biogenesis, and thermogenic response proteins in BAT and iWAT at baseline and following CL injection (Figure 3A,B, quantified in Suppl. Figures 2 and 3). These experiments showed that the mitochondrial biogenesis marker, Peroxisome Proliferator-activated Receptor Gamma Coactivator 1-alpha (PGC1α), was significantly increased in both BAT and iWAT from Acss1^K635Q/K635Q^ mice following CL treatment as compared to WT. Lipolytic activity in Acss1^K635Q/K635Q^, as assessed by Hormone Sensitive Lipase (HSL) phosphorylated at serine 563 versus total HSL, was decreased in BAT following CL treatment, while in iWAT it was significantly increased. The lipolytic proteins, Adipose Triglyceride Lipase (ATGL) and Perilipin 5 (PLIN5) did not show major changes.

While baseline levels of UCP1 were higher in Acss1^K635Q/K635Q^ BAT, no increase was observed following CL treatment. As expected, UCP1 was barely detectable in iWAT at baseline. However, UCP1 levels were significantly increased only in WT and not in Acss1^K635Q/K635Q^ iWAT after CL treatment. Gene expression analysis by qPCR confirmed that, in contrast to WT, Acss1^K635Q/K635Q^ iWAT did not significantly increase Ucp1 expression following CL treatment (Suppl. Figure 4A). The fact that iWAT was unable to upregulate UCP1 is striking and suggests an aberrant beiging response, which may contribute to the failure of Acss1^K635Q/K635Q^ mice to maintain body temperature.

Interestingly, ACSS1 levels tended to be higher in Acss1^K635Q/K635Q^ adipose tissues than in WT, which was not previously observed in other tissues 5, but may be an attempt to compensate for reduced ACSS1 activity, although Acss1 levels were not significantly different by qPCR analysis, except in Acss1^K635Q/K635Q^ BAT 24 hours post CL injection (Suppl. Figure 4B). Finally, levels of the Hydroxyacyl-CoA Dehydrogenase Alpha subunit (HADHA), part of the trifunctional multienzyme protein for FAO, was slightly higher at basal levels in BAT, with a small decrease in Acss1^K635Q/K635Q^ iWAT 24 hours post CL injection.

We also measured acetyl-CoA and acetate in BAT and iWAT from WT and Acss1^K635Q/K635Q^ mice to assess the substrate and product levels that result from altered ACSS1 enzymatic activity. While we found no significant differences in BAT acetyl-CoA levels with CL treatment (Figure 3C), we observed consistently lower acetyl-CoA levels in iWAT from the Acss1^K635Q/K635Q^ mice (Figure 3D). Acetate levels were increased in both tissues in Acss1^K635Q/K635Q^ mice, as expected (Figure 3E and 3F). These data suggest that Acss1^K635Q/K635Q^ iWAT specifically is not able to meet the metabolic demand elicited by CL and fails to produce adequate acetyl-CoA.

3.4. Acss1K635Q/K635Q ex vivo beige adipocytes display aberrant response to forskolin

Stromal vascular cells were isolated from Acss1^K635Q/K635Q^ and WT mouse iWAT and differentiated into beige adipocytes to study lipid droplet utilization and mitochondrial function. Cells were exposed to 10 μM forskolin for 24 hours in order to stimulate lipolysis and thermogenesis 27. Live cell imaging with bodipy was used to stain lipid droplets and tetramethylrhodamine methyl ester (TMRM) to assess mitochondrial membrane potential. WT beige adipocytes showed a significant reduction in lipid droplet size following forskolin treatment (Figure 4A, left panels, quantified in 4B). In contrast, Acss1^K635Q/K635Q^ beige adipocytes showed a large range in the size of lipid droplets at baseline, which did not change with forskolin (Figure 4A, right panels, quantified in 4B). Mitochondrial membrane potential was decreased after forskolin treatment in WT cells as expected, due to UCP1 respiration uncoupling. However, Acss1^K635Q/K635Q^ cells displayed a much higher membrane potential at baseline that was decreased following exposure to forskolin as they appeared to excessively depolarize suggesting a decrease in electron transport chain activity and the proton motive gradient (Figure 4A,C), pointing to a defect in mitochondrial function.

We used FAO Blue to measured FAO at baseline and after 4- and 24-hours forskolin treatment and found that FAO in Acss1^K635Q/K635Q^ beige adipocytes tended to be lower than in WT at all timepoints (Figure 4D,E). All these data suggest that Acss1^K635Q/K635Q^ beige adipocytes exhibit dysregulated lipolytic activity, FAO, and mitochondrial function.

4. Discussion

In this study, we demonstrate that Acss1^K635Q/K635Q^ mice are intolerant to the thermogenic response elicited by CL being unable to maintain their body temperature with a dysregulated metabolic response. While these mice respond by increasing serum FFA, possibly increasing lipolysis to compensate for decreased FAO 21, fatty acids do not seem to be effectively oxidized. Although PGC1α is elevated in Acss1^K635Q/K635Q^ adipose tissues and BAT showed high UCP1 levels, Acss1^K635Q/K635Q^ iWAT failed to increase UCP1 in response to CL, suggesting impaired beiging. Mitochondria isolated from Acss1^K635Q/K635Q^ BAT also showed impaired respiration. Although some compensatory responses appear to be present, such as higher FAO enzymatic expression and increased ACSS1, they are insufficient to compensate for ACSS1 functional deficiency in response to the metabolic demands induced by CL.

Following CL injection, we observed an immediate drop in glucose, which was recovered over the following 24 hours in wild-type mice but continued to fall in the Acss1^K635Q/K635Q^ mice. It is interesting that some of the other metabolic changes were apparent at 4 hours while others are seen at 24 hours. While we don't have a definitive answer for this, one possible explanation is that the changes in metabolism following CL injection may be in a pathway and that the effects at 4 hours are early events and those later are downstream.

Others have demonstrated the essential role of mitochondrial fatty acid import and oxidation for BAT thermogenesis. For instance, carnitine palmitoyltransferase II (CPT2) is an integral protein of the inner mitochondrial membrane, required for the import of long-chain fatty acids and essential for thermogenesis and maintenance of BAT. Mice with adipose-specific knockout of Cpt2 fail to induce Ucp1 in BAT upon acute cold exposure or administration of CL, forskolin, or the general adrenergic agonist, isoproterenol 29, and showed structural deterioration of BAT in response to agonist treatment 20.

Loss of Sirt3 has been previously shown to impair lipid utilization, thermoregulation, and respiration in BAT mitochondria, without affecting UCP1 expression suggesting that UCP1-mediated thermogenesis is indirectly regulated by deacetylation 13, although this study did not identify a specific SIRT3 deacetylation target for thermogenesis. This data also suggests that SIRT3, and its deacetylation targets, may be part of an axis that regulates how cells utilize fatty acids under differing metabolic states 12, 30, 31. In this regard, we clearly show that a constitutive ACSS1 acetylation mimic disrupts BAT lipid utilization and impairs the thermogenic response, as demonstrated by the increase in serum FFA and metabolic crisis in Acss1^K635Q/K635Q^ mice after CL injection. Further studies are needed to delineate the molecular mechanisms underlying the role of ACSS1 in lipid metabolism and the thermogenic response.

De novo lipogenesis, the process of converting glucose into fatty acids, plays a crucial role in thermogenesis within brown adipocytes 32. This is unusual, as fatty acid synthesis and oxidation typically counteract each other in other tissues 33. To maintain homeostasis and thermogenesis in BAT, a tightly regulated balance exists between lipolysis and lipogenesis. De novo lipogenesis relies on acetyl-CoA as a substrate. Unlike other members of the acyl-CoA synthetase family, ACSS1-3 enzymes specifically catalyze the conversion of acetate and CoA into acetyl-CoA, playing a unique role in lipid metabolism. Furthermore, ACSS1 is more highly abundant in BAT as compared to WAT, suggesting a possible involvement in thermogenesis 34. In this regard, the reduced enzymatic activity of acetylated ACSS1 may lead to lower levels of acetyl-CoA and intermediates feeding the TCA cycle and ultimately to lower FAO 35. This may underlie the apparent metabolic crisis in* Acss1^K635Q/K635Q^* mice, which are unable to meet the acute energy demand elicited by CL.

Lower acetyl-CoA levels should signal for increased FAO activity and lipolysis 36, which fits our observations of increased levels of HADHA and lipolytic activity in Acss1^K635Q/K635Q^. Another plausible mechanism for the lower FAO activity observed in these mice might be that lower acetyl-CoA levels leads to lower fatty acid esterification with acetyl-CoA which is needed in order to activate them into acyl-CoA molecules and enter FAO cycles 37. Additionally, it has been described before that excess FFA in BAT upregulate both expression of UCP1 38. We conclude that full ACSS1 functionality is needed during the thermogenic response elicited by CL, which it is unable to meet in an acetylated state. Acss1^K635Q/K635Q^ mice represent a constitutively acetylated enzyme that cannot be further regulated by this post-translational modification. This might lead to lower acetyl-CoA intermediates that feed the TCA cycle, causing an acute energetic deficit and failure to maintain body temperature. However, further mechanistic studies are needed to dissect the exact molecular mechanism.

Supplementary Material

Supplementary figures.