Editor's evaluation: Hypermetabolism in mice carrying a near-complete human chromosome 21

Abstract

Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract The consequences of aneuploidy have traditionally been studied in cell and animal models in which the extrachromosomal DNA is from the same species. Here, we explore a fundamental question concerning the impact of aneuploidy on systemic metabolism using a non-mosaic transchromosomic mouse model (TcMAC21) carrying a near-complete human chromosome 21. Independent of diets and housing temperatures, TcMAC21 mice consume more calories, are hyperactive and hypermetabolic, remain consistently lean and profoundly insulin sensitive, and have a higher body temperature. The hypermetabolism and elevated thermogenesis are likely due to a combination of increased activity level and sarcolipin overexpression in the skeletal muscle, resulting in futile sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA) activity and energy dissipation. Mitochondrial respiration is also markedly increased in skeletal muscle to meet the high ATP demand created by the futile cycle and hyperactivity. This serendipitous discovery provides proof-of-concept that sarcolipin-mediated thermogenesis via uncoupling of the SERCA pump can be harnessed to promote energy expenditure and metabolic health. Editor's evaluation This important paper provides new insight into the effect of extra-copies of a chromosome, thus aneuploidy, on body metabolisms in mammals. The authors used various solid analyses on the metabolisms and physiology of the transgenic mouse with most of human chromosome 21 and presented convincing results to support the authors' claims. The work would be of interest to researchers who work on the physiology and biochemistry of body metabolisms in mammals. https://doi.org/10.7554/eLife.86023.sa0 Decision letter Reviews on Sciety eLife's review process Introduction The presence of an extra chromosome in mammals is generally lethal during fetal development, due to widespread cellular havoc caused by misregulated gene expression arising from gene dosage imbalance (Zhu et al., 2018). Down syndrome (DS), resulting from trisomy of chromosome 21, is one of the rare aneuploidies compatible with life although as many as 80% of trisomy 21 conceptuses miscarry (Antonarakis et al., 2020). The increased expression of ~500 transcribed sequences of human chromosome 21 (Hsa21) affects many cell types and organ systems during development and in the postnatal period (Korenberg et al., 1994; Antonarakis, 2017). Humans with trisomy 21 have cognitive deficits, altered craniofacial development, and are at significantly higher risk for congenital heart defects, hearing and vision loss, leukemia, gastrointestinal disease, and early-onset dementia (Antonarakis et al., 2020). Given the significant impact of intellectual disability on the lives of individuals with DS, research emphasis has naturally focused on the neurological deficits underpinning trisomy 21 (Potier and Reeves, 2016). In addition to developmental abnormalities associated with DS, there is an increasing awareness that adolescents and adults with DS also have an increased incidence of obesity, insulin resistance, and diabetes (Van Goor et al., 1997; Bertapelli et al., 2016; Fonseca et al., 2005). Although this was first noted in the 1960s (Milunsky and Neurath, 1968), the underlying cause for these metabolic dysregulations is mostly unknown and largely underexplored. Beyond clinical observations, limited studies have been conducted to determine the physiological underpinnings of metabolic impairments seen in DS (reviewed in Dierssen et al., 2020). Our recent study on the Ts65Dn mouse model represents the most in-depth metabolic analysis, to date, of any DS mouse model (Sarver et al., 2023). However, the segmental trisomic Ts65Dn mouse contains only ~55% of the orthologous protein-coding genes (PCGs) found on Hsa21 (Gupta et al., 2016). In addition, it contains additional trisomic genes from the centromeric region of mouse chromosome 17 (Mmu17) not found in Hsa21, thus complicating the genotype-phenotype relationships (Duchon et al., 2011; Reinholdt et al., 2011). In the past two decades, more than 20 mouse models of DS have been generated (Herault et al., 2017). Despite their utility in advancing DS research, none of these models recapitulate the full spectrum of human DS. With the exception of Tc1, all the DS mouse models are trisomic for some, but not all, of the orthologous mouse genes found in Hsa21 (Herault et al., 2017). Tc1 is the first mouse model with an independently segregating Hsa21 (O’Doherty et al., 2005). However, Tc1 mice are missing >50 of the 220 PCGs on Hsa21 due to deletion and mutations (Gribble et al., 2013). In addition, Tc1 mice show extensive mosaicism (i.e. the human chromosome is present in zygotes but is lost randomly from cells during development). As a consequence, every mouse has a unique developmental trajectory, complicating the interpretations of results obtained from Tc1 mice. To overcome the limitations of previously generated trisomic mouse models, a transchromosomic mouse model (TcMAC21) carrying an independently segregating and near-complete copy of Hsa21 was recently developed (Kazuki et al., 2020). TcMAC21 is not mosaic and contains 93% of HSA21q PCGs, and is considered the most representative mouse model of DS. Both mouse and rat that carry a non-mosaic Hsa21 recapitulate many DS phenotypes related to the central nervous system (e.g. reduced cerebellum volume, learning and memory deficit), craniofacial skeleton, and heart (Kazuki et al., 2020; Kazuki et al., 2022). The metabolic phenotype of TcMAC21, however, is unknown and has yet to be examined. The availability of the TcMAC21 mouse model has afforded a unique opportunity to address two fundamental questions: (1) what is the impact of aneuploidy on systemic metabolism; (2) what are the molecular, cellular, and physiological consequences of introducing a foreign (human) chromosome from an evolutionarily distant species into mice? Unexpectedly, we discovered that TcMAC21 mice have all the hallmarks of hypermetabolism, likely driven by a combination of hyperactivity and elevated mitochondrial respiration and futile sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA) pump activity in the skeletal muscle as a consequence of endogenous sarcolipin (SLN) overexpression. Our study has provided further evidence and proof-of-concept that endogenous SLN-mediated uncoupling of the SERCA pump can be harnessed for energy dissipation, weight loss, and metabolic health. Results Human chromosome 21 genes are differentially expressed and regulated in mouse adipose tissue, liver, and skeletal muscle TcMAC21 mice carry a non-mosaic and independently segregating mouse artificial chromosome with a near-complete copy of the long arm of human chromosome 21 (Hsa21q) (Kazuki et al., 2020). The Hsa21q in TcMAC21 is comprised of ~37 Mb and 199 PCGs (Figure 1A). RNA-sequencing showed that TcMAC21 mice are capable of expressing Hsa21-derived transcripts in each of the tissues examined, and that gene expression is regulated in a tissue-specific manner. The transcriptional activity map of Hsa21 shows regions of gene expression and repression (Figure 1B). There are three large regions of Hsa21 with little or no transcription activity: 29.5–31.1 Mb, 44.5–44.8 Mb, and 45.3–46.2 Mb. The first gap of transcriptional inactivity contains the PCGs Cldn8 and Cldn17, Girk1, and 33 distinct Krtap (keratin-associated protein) genes. The second transcriptionally inactive area contains 16 Krtap genes, and the third transcriptionally inactive area contains 8 PCGs (Col6a1, Col6a2, Col18a1, Fctd, Lss, Pcbp3, Slc19a1, and Spatc1l). By filtering the transcriptional map to display expressed genes only, we highlighted all the genes with their differential expression profiles across five major metabolic tissues—brown adipose tissue (BAT), inguinal white adipose tissue (iWAT), gonadal white adipose tissue (gWAT), liver, and skeletal muscle (Figure 1C). Figure 1 Download asset Open asset Human chromosome 21 genes are differentially expressed and regulated in mouse adipose tissue, liver, and skeletal muscle. (A) Graphical representation of human chromosome 21 (Hsa21) and the entire long arm (Hsa21q) region carried by a mouse artificial chromosome in the transchromosomic mouse model (TcMAC21). Four deletions that occurred during generation of the transchromosomic mice eliminate 14/213 protein-coding genes (PCGs; 7%) and 105/487 predicted or known non-protein-coding genes (NPCGs; 22%) (Kazuki et al., 2020). (B) Global view of transcriptionally expressed and repressed PCG and NPCG regions over the entire Hsa21q across five tissues. Gray box denotes transcript that is not detected. (C) Transcriptional activity map showing only Hsa21 genes expressed by at least one tissue. Gray box denotes transcript that is not detected. (D) Overlap analysis showing shared expression of human PCGs and NPCGs across five tissues. Of the 235 unique human genes expressed by the TcMAC21 mice, 54% are PCGs and 46% are NPCGs. B, brown adipose tissue; iW, inguinal white adipose tissue; gW, gonadal white adipose tissue; L, liver; M, skeletal muscle (gastrocnemius); n.d., not detected. n=5 RNA samples per group per tissue type. Mice were on high-fat diet for 16 weeks at the time of tissue collection. One of the more striking differences in expression profile is between visceral (gonadal) and subcutaneous (inguinal) white adipose tissue (gWAT and iWAT respectively). gWAT expresses 115 human PCGs while iWAT expresses only 27. A similar pattern can be seen in the non-protein-coding genes (NPCGs). We were unable to detect any Hsa21-derived NPCGs in the iWAT, while in gWAT we observed 37. Overlap analysis was carried out to assess how similar expression profiles were between tissues (Figure 1D). Of the 126 Hsa21-derived PCGs expressed by at least one tissue, a majority (65 total) are shared between BAT, gWAT, liver, and skeletal muscle. Of the 109 Hsa21-derived NPCGs expressed by at least one tissue, the majority (53 total) are uniquely expressed by skeletal muscle. Of note, the liver uniquely expresses 6 PCGs and 2 NPCGs, gWAT 4 and 3, skeletal muscle 3 and 53, BAT 1 and 10, and iWAT 0 and 0. Together, these data indicate that Hsa21-derived transcripts are differentially expressed and regulated across major metabolic tissues in TcMAC21 mice. Hypermetabolism in TcMAC21 mice fed a standard chow Having established that Hsa21-derived transcripts are differentially expressed and regulated in mouse organs and tissues, we next asked the impact of the extra human genetic material and genes on systemic metabolism. As previously documented, TcMAC21 pups are born at the same weight as their euploid littermates (Kazuki et al., 2020). However, by 3 months of age TcMAC21 mice fed a standard chow weighed significantly less (~8.5 g) than euploid littermates, and this weight difference remained stable over time (Figure 2A). The size and body weight differences were not due to reduced plasma IGF-1 and growth hormone, as their circulating levels were in fact higher in TcMAC21 compared to euploid mice (Figure 2—figure supplement 1). Body composition analysis showed that TcMAC21 have significantly reduced absolute and relative (normalized to body weight) fat mass (Figure 2B). Although the absolute lean mass was reduced in TcMAC21 mice, the relative lean mass (normalized to body weight) was not different between genotypes (Figure 2B). Tissue collection at termination of the study also showed smaller visceral and subcutaneous fat mass and liver weight in TcMAC21 mice (Supplementary file 1). Figure 2 with 3 supplements see all Download asset Open asset Hypermetabolism in TcMAC21 mice fed a standard chow. (A) Body weights of mice fed a standard chow. (B) Body composition analysis of fat and lean mass (relative to body weight). (C) Absolute and relative (normalized to body weight) food intake over a 24 hr period. (D–F) Energy expenditure (EE) and physical activity level over 24 hr period in ad libitum chow-fed mice (D), during fasting (E), and refeeding after a fast (F). EE is normalized to lean mass in the 24 hr trace or analyzed by ANCOVA where body weight was used as a covariate. (G) Hematoxylin and eosin (H&E)-stained sections of inguinal white adipose tissue (iWAT) and adipocyte cross-sectional area (CSA) quantification. (H) Histology of gonadal white adipose tissue (gWAT) and adipocyte CSA quantification. (I) Histology of liver tissues with quantification of area covered by lipid droplets per focal plane. (J) Fasting serum triglyceride, cholesterol, non-esterified fatty acids (NEFA), β-hydroxybutyrate (ketone) levels. (K) Fasting blood glucose and insulin levels. (L) Insulin resistance index (homeostatic model assessment for insulin resistance [HOMA-IR]). (M) Glucose tolerance tests. (N) Insulin tolerance tests. Sample size for all data: euploid (n=8) and TcMAC21 (n=9). Differences in body weight were not due to reduced caloric intake, as TcMAC21 mice actually consumed the same or slightly higher amount of food despite being markedly leaner (Figure 2C and Supplementary file 2). Thus, relative to their body weight, TcMAC21 actually consumed a significantly higher amount of food than euploid controls (Figure 2C). Indirect calorimetry analysis indicated that TcMAC21 mice—regardless of the photocycle (light or dark phase) and metabolic states (ad libitum fed, fasting, refeed)—were expending ~25% more energy and were significantly more active compared to euploid controls (Figure 2D–F and Supplementary file 2). It is known that normalization to lean mass can lead to an overestimation of energy expenditure (EE) (Tschöp et al., 2012). For this reason, ANCOVA (using body weight as a covariate of EE) (Tschöp et al., 2012) were also performed. Both types of analyses suggested that TcMAC21 mice had significantly higher EE relative to euploid controls (Figure 2D–F). Despite much higher caloric intake per gram body mass, TcMAC21 mice were much leaner due to substantially elevated physical activity and EE. Hyperactivity and elevated EE were not due to altered circulating thyroid hormones, as serum triiodothyronine (T3, the active form of TH) levels were not different between chow fed TcMAC21 and euploid mice (Figure 2—figure supplement 1). Serum level of thyroxine (T4), the precursor of T3, were modestly elevated in TcMAC21 relative to euploid mice. In accordance with the lean phenotype, TcMAC21 mice had significantly smaller adipocyte cell size in both subcutaneous (inguinal) and visceral (gonadal) fat depots (Figure 2G–H), as well as significantly reduced fat accumulation in liver (Figure 2I). Fasting triglyceride, non-esterified fatty acid (NEFA), and β-hydroxybutyrate levels were not different between genotypes; fasting cholesterol, however, was higher in TcMAC21 mice (Figure 2J). Although fasting insulin levels were not different between groups, fasting blood glucose was significantly lower in TcMAC21 mice (Figure 2K). The insulin resistance index (homeostatic model assessment for insulin resistance [HOMA-IR]) along with GTT and insulin tolerance test (ITT) suggested modest improvements in insulin sensitivity in TcMAC21 relative to euploid mice (Figure 2L–N). Assessment of the pancreas showed that TcMAC21 mice have similar β-islet cross-sectional area (CSA), insulin and somatostatin (SST) content, and insulin granule and vesicle size compared to euploid controls (Figure 2—figure supplement 2). Taken together, these data indicate that chow-fed TcMAC21 mice at baseline are lean despite increased caloric intake, and this is largely due to elevated physical activity and EE. TcMAC21 mice are resistant to diet-induced obesity and metabolic dysfunction The hypermetabolic phenotypes seen in chow-fed TcMAC21 predicted that these mice would be resistant to diet-induced obesity and metabolic dysfunction. Indeed, after 8 weeks on high-fat diet (HFD), TcMAC21 mice gained only ~3 g of body weight, whereas the euploid controls gained >15 g of body weight over the same period. Consequently, TcMAC21 mice weighed ~50% less than euploid controls (Figure 3A). Consistent with the lean phenotype, the absolute and relative (normalized to body weight) fat mass were markedly reduced compared to euploid controls (Figure 3B). The weights of other organs (liver, kidney, BAT) at time of termination were also lower in TcMAC21 mice, but tibia length was not different between genotypes (Figure 3—figure supplement 1 and Supplementary file 3). Complete blood count revealed no differences in erythroid, lymphoid, and myeloid cell numbers between genotypes (Supplementary file 4). Because relative lean mass was higher in TcMAC21 compared to euploid mice (Figure 3B), the lean phenotype seen in HFD-fed TcMAC21 is largely due to reduced adiposity. Accordingly, TcMAC21 had significantly smaller adipocyte cell size in both subcutaneous (inguinal) and visceral (gonadal) fat depots, and a marked reduction in lipid accumulation in the liver (Figure 3C–E). Figure 3 with 1 supplement see all Download asset Open asset TcMAC21 mice are resistant to diet-induced obesity and metabolic dysfunction. (A) Body weights over time on a high-fat diet and representative mouse images. (B) Body composition analysis of fat and lean mass. (C) Histology of inguinal white adipose tissue (iWAT) and quantification of adipocyte cross-sectional area (CSA). (D) Histology of gonadal white adipose tissue (gWAT) and quantification of adipocyte CSA. (E) Histology of liver tissue and quantification of area covered by lipid droplets per focal plane. (F) Fasting serum triglyceride, cholesterol, non-esterified fatty acids (NEFA), β-hydroxybutyrate (ketone) levels. (G) Fasting blood glucose and insulin levels. (H) Insulin resistance index (homeostatic model assessment for insulin resistance [HOMA-IR]). (I) Glucose tolerance tests (GTTs). (J) Serum insulin levels during GTT. (K) Insulin tolerance tests (ITTs). (L) Blood glucose levels after an overnight (16 hr) fast and 1, 2, and 3 hr of food reintroduction. (M) Serum insulin levels after a 16 hr fast and 2 hr of refeeding. (N) Pancreas histology and quantification of β-islet CSA. (O–P) Pancreatic insulin and somatostatin (SST) contents (normalized to pancreatic protein input). (Q) Electron micrographs (EM) of pancreatic β-cells showing dense insulin granules and their surrounding vesicles, and quantification of insulin granule CSA, insulin vesicle CSA, and the ratio of insulin granule to insulin vesicle. n=8 euploid and 8–9 TcMAC21 mice for all graphs from A to P. n=3 euploid and 3 TcMAC21 used for EM quantification; each data point represents 1200 insulin granules and 1200 insulin vesicles quantified across six unique locations per mouse, graphs Q. Although fasting serum triglyceride, NEFA, and β-hydroxybutyrate levels were not different between genotypes, serum cholesterol was significantly lower in TcMAC21 mice (Figure 3F). Fasting glucose and insulin levels, and the insulin resistance index (HOMA-IR), were markedly lower in TcMAC21 mice relative to euploid controls (Figure 3G–H), indicative of enhanced insulin sensitivity. In glucose tolerance tests (GTTs), even though the rate of glucose disposal was similar between TcMAC21 and euploid mice, the amount of serum insulin present during GTT (time 0, 15, and 30 min) was dramatically lower in TcMAC21 (Figure 3I–J). This indicates that a substantially lower amount of insulin is sufficient to promote glucose clearance in TcMAC21 at a rate comparable to euploid mice, consistent with elevated insulin sensitivity in the peripheral tissues. Indeed, when we directly assessed insulin action via ITTs, TcMAC21 mice clearly exhibited higher insulin sensitivity as indicated by the significant differences in insulin-stimulated glucose disposal (Figure 3K). To independently confirm TcMAC21 mice are more insulin sensitive, we fasted the mice overnight (16 hr) then reintroduced them to food. Under this fasting-refeeding condition, we could clearly see the resumption of food intake was successful at increasing blood glucose in TcMAC21 (Figure 3L); however, the insulin response to food intake in TcMAC21 mice was strikingly smaller in magnitude compared to euploid controls (Figure 3M). Again, these data indicate that TcMAC21 mice are significantly more insulin sensitive since a substantially lower insulin response during fasting-refeeding is sufficient for glucose clearance at a rate comparable to euploid mice. These results prompted us to determine if there were developmental changes in the pancreas leading to reduced insulin secretion in response to glucose administration or food intake, independent of elevated insulin sensitivity in peripheral tissues. Quantification of β-islet size, pancreatic insulin and SST content, as well as insulin granule and vesicle size did not reveal any intrinsic differences between TcMAC21 and euploid mice (Figure 3N–Q), thus ruling out a developmental cause and arguing in favor of enhanced insulin action. Pancreatic acinar zymogen granule size was also not different between genotypes (Figure 2—figure supplement 2), suggesting normal development of the exocrine pancreas. Taken together, these data indicate that TcMAC21 mice are remarkably resistant to HFD-induced obesity and insulin resistance. Hypermetabolism in TcMAC21 mice fed an HFD Next, we sought to uncover the physiological mechanisms responsible for TcMAC21 resistance to weight gain and developing insulin resistance when fed an HFD. First, we wanted to rule out whether there is a change in caloric intake. TcMAC21 mice actually consumed the same or slightly higher amount of food as the euploid controls despite markedly lower body weight (Figure 4A and Supplementary file 5). Thus, relative to their body weight, HFD-fed TcMAC21 mice consumed a significantly higher amount of food than the euploid controls (Figure 4A). To rule out any potential dysfunction of the gut that might adversely affect nutrient absorption, we collected, counted, weighed, and subjected fecal samples of each mouse to fecal bomb calorimetry. Neither fecal frequency, average fecal pellet weight, nor fecal energy content was different between TcMAC21 and euploid mice (Figure 4B). Fecal energy content trended lower in TcMAC21 (Figure 4B), implying a modest increase in the efficiency of nutrient absorption. Since energy input (caloric intake per gram body weight) was significantly higher in TcMAC21 and outputs (left-over fecal energy) were similar across genotypes, the data suggest hypermetabolism being a cause of the lean phenotype seen in TcMAC21 mice. Indeed, when we measured EE and physical activity of both groups, we found that TcMAC21 mice have markedly higher EE and physical activity irrespective of circadian cycle and metabolic states (Figure 4D–F and Supplementary file 5). Aware of the potential overestimation of EE when normalized to lean mass, we also included ANCOVA where body weight was used a covariate (Tschöp et al., 2012). Both types of analyses suggested that TcMAC21 mice fed an HFD had significantly higher EE relative to euploid controls (Figure 4D–F). The striking difference in EE was very similar to TcMAC21 fed a standard chow (Figure 2D–F), but to an even greater extent when mice were fed an HFD, presumably due to the greater availability of calorie-dense lipid substrates for oxidation. Figure 4 Download asset Open asset Hypermetabolism in TcMAC21 mice fed a high-fat diet (HFD). (A) Absolute and relative (normalized to body weight) food intake in mice fed an HFD. (B) Fecal energy content, fecal weight, and fecal frequency of TcMAC21 mice and euploid controls. (C) Serum triiodothyronine (T3) and thyroxine (T4) levels. (D–F) Energy expenditure (EE) and activity level over 24 hr period in ad libitum HFD-fed mice (D), during fasting (E), and refeeding after a fast (F). EE is normalized to lean mass in the 24 hr trace or analyzed by ANCOVA where body weight was used as a covariate. (G) Deep colonic and tail temperature measured over 3 days in both the light and dark cycle. (H) Representative infrared images of mice. (I) Body weights of euploid, TcMAC21, and weight-matched (WM) control C57BL/6 mice. (J) Interscapular skin temperature of euploid, TcMAC21, and WM control mice. (K) Representative histology of brown adipose tissue (BAT). (L) Quantification of percent total lipid area coverage per focal plane in BAT of euploid and TcMAC21. (M) Expression of mouse genes (by qPCR) known to play major metabolic roles in BAT. Sample size for all data: euploid (n=8) and TcMAC21 (n=9). Because TcMAC21 mice burned a large excess of energy, we measured the circulating levels of thyroid hormones as they are known to increase metabolic rate and EE (Mullur et al., 2014). Both serum T3 (the active form) and T4 (precursor of T3) levels were significantly higher in TcMAC21 relative to euploid mice (Figure 4C). T3 hormone, however, was not elevated in HFD-fed mice housed at thermoneutrality (30°C) (Figure 2—figure supplement 1). If EE was elevated, body temperature of TcMAC21 mice would likely increase. Indeed, deep colon temperatures of TcMAC21 were elevated, most notably in the dark cycle when mice are active (Figure 4G). Assessment with thermal imaging showed an elevated skin temperature around the interscapular region of TcMAC21 mice, whereas the tail skin temperature was not different between groups (Figure 4G–H). Importantly, the differences in interscapular skin temperatures persisted in TcMAC21 even when compared to weight-matched wild-type mice (Figure 4I–J), thus ruling out body weight (and hence surface area/volume ratio) as the cause of greater heat generation to compensate for greater heat loss. Elevated body temperature, however, was not observed in chow-fed mice (Figure 2—figure supplement 3), even though chow-fed TcMAC21 were also hyperactive and had higher EE. Consistent with the thermal imaging data, histological analysis of the interscapular BAT revealed a marked reduction in fat accumulation and a ‘healthy’ brown appearance in TcMAC21 mice when compared to euploid controls (Figure 4K–L), presumably due to excess lipids being utilized. We therefore expected several key metabolic genes in BAT to be upregulated in TcMAC21. Surprisingly, we found minimal differences in key thermogenic and fat oxidation genes between the two groups of mice (Figure 4M, Figure 5—figure supplement 1). Hypermetabolism of TcMAC21 mice is uncoupled from changes in adipose and liver transcriptomes The lack of changes observed in key thermogenic and fat oxidation genes mentioned above led us to assume the differences in gene expression must be due to non-canonical and potentially novel pathways. To test this, we conducted an unbiased RNA-sequencing analysis of BAT, liver, gWAT, and iWAT. Again, to our surprise and contrary to expectation, we found the transcriptomes of BAT, liver, gWAT, and iWAT in TcMAC21 to be remarkably similar to euploid controls, with only a limited number of genes whose expression was significantly altered (Figure 5A–D). Figure 5 with 5 supplements see all Download asset Open asset Hypermetabolism of TcMAC21 mice is uncoupled from changes in adipose and liver transcriptomes. (A–D) Differentially expressed mouse genes (DEGs), both protein-coding (PCG) and non-protein-coding genes (NPCG) in brown adipose tissue (BAT), liver, gonadal white adipose tissue (gWAT), and inguinal white adipose tissue (iWAT). All data is relative to euploid and presented as Log2(FC). The list of genes represents all up- and down-regulated mouse genes (significant by adjusted p-value cut-off) for all four tissues. The red bars indicate upregulated genes and the blue bars indicate downregulated genes. (E) General view and summary of transcriptional changes in BAT, liver, gWAT, and iWAT to highlight the strikingly minimal changes in the mouse transcriptome across the four tissues. There are only a combined total of 114 differentially expressed genes (DEGs) across four tissues, with the relative percentage (out of the 114 DEGs) shown for each tissue. Of the 114 DEGs, 105 are protein-coding genes (PCGs; dark yellow bar) and 9 are non-protein-coding genes (NPCGs; light yellow bar). Of the 114 DEGs, 46 are upregulated (red bar) and 68 are downregulated (blue bar). In total, only a combined 0.53% change is noted in the transcriptome of all four tissues (out of the 21,752 RNAs detected). Sample size for RNA-sequencing: euploid (n=5 per tissue) and TcMAC21 (n=5 per tissue). There was not a single tissue that had more than 0.3% of its transcriptome (mouse) significantly altered (Figure 5E and Supplementary files 6-9). None of the significant changes in gene expression—small in number—could readily account for the striking differences in phenotypes between TcMAC21 and euploid mice. Corroborating the RNA-sequencing results, quantitative real-time PCR analyses of select metabolic genes in BAT, liver, gWAT, and iWAT also showed minimal changes in TcMAC21 regardless of diet (chow or HFD) and temperature (22°C or 30°C) (Figure 5—figure supplements 1–5). Together, these data indicate that TcMAC21 mice are hyperactive and hypermetabolic with elevated thermogenesis, but these phenotypes are largely uncoupled from transcriptomic changes in BAT, liver, gWAT, and iWAT. SLN overexpression in skeletal muscle drives TcMAC21 hypermetabolism Given the remarkable differences seen in body weight, adiposity, tissue histology and lipid contents, insulin sensitivity, body temperature, physical