The Metabolic Significance of Intermuscular Adipose Tissue: Is IMAT a Friend or a Foe to Metabolic Health?

White adipocytes appeared early during vertebrate evolution (1). A critical advancement in vertebrate (versus invertebrate) storage tissues included protection of other tissues against lipotoxicity via accumulation and long-term storage of excess lipids (2,3). The contemporary view of adipose tissues, however, is not simply one of an inert lipid reservoir but rather as an endocrine warehouse poised for storage and release of fuel and other factors that are specialized to contribute to local and systemic physiological regulation. Adipose tissues therefore are not homogeneous and show site-specific properties.

Anatomical organization, genes, and metabolic pathways in storage organs and adipose tissues can be traced from their invertebrate origins through lower vertebrates to mammals (4). The most thoroughly studied energy storage organ of invertebrates is the insect fat body, which develops in the abdomen and is composed primarily of adipocytes (5). Higher-order vertebrates such as mammals (inclusive of rodents, livestock animals, and humans) also have centrally located adipose depots that have been intensively investigated over the years. However, an elusive and understudied adipose tissue depot, most likely due to its limited accessibility, is the intermuscular adipose depot. As of June 2021, there were 484 publications specifically investigating intermuscular adipose tissue (IMAT) compared with the more than 24,000 publications on abdominal adipose tissues (inclusive of subcutaneous, visceral, omental, etc.). For clarity, IMAT refers to the adipose depot located in the muscle bed and interspersed among muscle fibers in humans (Fig. 1), rodents, and livestock animals. Intramuscular fat refers to adipose tissue deposited within a single muscle group; however, this is typically only reported in meat or livestock animals (reviewed in Komolka et al. [6]) and will not be discussed here. Intramyocellular lipid (IMCL) refers to lipid droplets located within muscle fibers and has been extensively reviewed elsewhere (7–11).

The first assessment of IMAT was published in 1972 and covered seven different anatomical locations in pigs (12). The majority of the early publications on IMAT were focused on rodent models and meat animals from agricultural industry research. At the turn of the 21st century the scientific community began to study IMAT in the context of human health and disease but with little to no evidence of its basic biological mechanisms to date. The first clinical trial to include human IMAT was published in 2002 and investigated the effects of diet and exercise on the sizes of different adipose depots (13). Soon afterward, a study in nearly 3,000 elderly men and women with normal body weight identified large amounts of IMAT as a significant risk factor for metabolic abnormalities, including type 2 diabetes (14). IMAT accumulation can occur in humans for a variety of reasons, such as age, injury, obesity, sedentariness, etc. (Fig. 1), and the etiology of IMAT accumulation may therefore impact the properties of IMAT. For the purposes of this Perspective, we are focusing on non-injury-related IMAT accumulation.

One of the fundamental cellular processes germane to white adipose tissue is the storage of lipids until needed for combustion by other tissues. Individual adipose depots expand and mature in response to naturally occurring environmental cues such as diet and physical activity (famine and migration for colonization in humans, as an early example), which dictate the cellular and metabolic function of the organ itself. As a consequence, the adipose organ has been and continues to be engineered for more diverse functions, begging the question of whether certain depots such as IMAT are a friend or a foe to metabolic health. Transient changes in adiposity, specifically the size of the IMAT depot, and its relationship with whole-body metabolism and energetic efficiency are major gaps in human adipose biology research in need of further exploration. Does the quantity, quality, or location of the IMAT depot in humans dictate its metabolic response to an intervention such as diet or exercise? We know from some of our own work that attenuation of age-associated increases in IMAT through physical activity in older adults is related to preservation of strength (15); however, selectively increasing and decreasing the IMAT depot is not very feasible in humans but could be addressed in rodent models. The goal of this Perspective is to review the evidence of the distinct and poorly understood intermuscular adipose depot from rodents to livestock animals to humans and discuss its potentially beneficial and harmful roles in human metabolic health.

Marbling of meat is highly valued in livestock; thus, a considerable volume of literature has explored IMAT in cattle, pigs, sheep, and even broiler chickens. Interestingly, cell number per gram of IMAT is a better predictor of marbling score than fat cell size (16). Studies of IMAT in pigs and cattle are typically invasive and sampled from either multiple sites or the entire IMAT depot. Early work has shown that adipocytes from the IMAT depot in growing pigs, cattle, and sheep exhibit hypertrophy much less than adipocytes from the subcutaneous and perirenal adipose depots and increase in number more than perirenal adipocytes in sheep (17). Furthermore, the morphology, rather than the number, of IMAT adipocytes in cattle and sheep is least affected by a restricted diet compared with adipocytes from the subcutaneous and perirenal adipose depots (18,19). More contemporary work in pigs has revealed that overall lipid content of the IMAT depot is lower and adipocytes are smaller than those in the subcutaneous and perirenal depots (20). Growth rates of intermuscular and subcutaneous adipose tissues (SAT) are similar among many livestock animals; however, in some locations, like the belly, the IMAT expands more rapidly and at an earlier stage in development (20). Similarly, in cattle, the IMAT depot is the largest adipose depot until slaughter, with the amount of subcutaneous depots representing only 60% of the total IMAT fat mass throughout growth (16). Several studies in meat animals have also demonstrated that the interfascicular adipocytes appear very early in development and that adipocyte number is critical to the total quantity of IMAT (17).

In general, in livestock animals, it seems that adipocytes from the IMAT depot are greater in number than size compared with those of other adipose depots, particularly the subcutaneous depot. It is important to note that due to their general physiological similarity to humans, similar fat cell size, body fat distribution, and genetic code, studies in pigs are very relevant for humans (21,22). With this in mind, studies on adipocyte size in humans have overwhelmingly associated larger adipocyte size with obesity-related comorbidities and cardiovascular disease risk (23); thus, the prevailing theory is that a larger adipocyte has more deleterious effects on whole-body metabolism than a smaller adipocyte, and that adipose tissue expansion via hypertrophy (increase in cell size) is more detrimental to whole-body metabolic health than expansion via hyperplasia (increase in cell number). Adipose tissue lipid sequestration via adipocyte proliferation and differentiation, the “expandability hypothesis” observed primarily in abdominal subcutaneous adipose tissue (24), constitutes a critical defense mechanism against lipotoxicity and metabolic disease (25,26). Based on the evidence from livestock research, it seems that IMAT adipocytes can be less metabolically adverse or perhaps even beneficial if they can provide a means of safe expansion of adipose tissue in times of need. Adipocyte size, however, is a static measurement that provides only one aspect of the metabolic landscape. In the last decade, adipocyte turnover (inclusive of cell number and triglyceride pool) has become a critical biomarker of metabolic health in humans (27). Adipocyte size and turnover of the IMAT depot are completely unknown and unexplored in humans. Taking advantage of certain technologies, such as incorporation of atmospheric ¹⁴C or deuterium-labeled water (²H₂O) into adipocyte DNA and triglyceride (28), requires a relatively small amount of tissue to be sampled and could provide a better understanding of the ability of this depot to expand and turn over in humans across a span of age and metabolic health.

Recent advancements in molecular biology techniques have permitted deeper characterization of transcriptional and epigenomic landscapes of IMAT, particularly in livestock animals. Comprehensive genomic assessments of IMAT in livestock animals have revealed large numbers of genes and signaling pathways related to adipogenesis, lipogenesis, glucose metabolism, and cholesterol and bile acid homeostasis (29–31). Studies in beef cattle and pigs found that mRNA expression patterns were different among the visceral, subcutaneous, and intermuscular adipose depots, suggesting that each is a uniquely regulated tissue (32,33). Additionally, in those same studies, IMAT exhibited low transcript levels of genes related to oxidative metabolism and high transcript levels of genes related to inflammatory cytokines, suggesting that IMAT is primed for storage and inflammation. A contemporary study of adipose tissue epigenomics in pigs showed that visceral adipose tissue (VAT) and IMAT have DNA methylation patterns similar to those of SAT. Therefore, it appears that IMAT is a uniquely regulated adipose tissue having a molecular profile similar to that of VAT and potentially capable of secreting cytokines known to adversely influence surrounding tissues. In humans, obesity-related inflammation of the SAT and VAT depots has been directly linked with adverse metabolic outcomes and cardiovascular disease risk (34); thus, IMAT in humans could also play a significant role in the pathophysiology of metabolic disease. As discussed previously, adipose tissue itself is a pliable organ with the ability to adapt to its physiological context. Whether that adaptation is harmful or beneficial remains to be explored in IMAT in human health and disease.

The fundamental mechanism of a healthy white adipocyte is to store lipid under basal conditions, release lipid (in the form of fatty acids and glycerol) under stimulated conditions, and then re-esterify unused fatty acids back into its lipid pool; this process is known as fatty acid/triacylglycerol (FA/TAG) cycling. In livestock or meat animals, particularly in pigs, the IMAT depot seems to be engineered for lipogenesis and lipid storage based on high activity levels of lipogenic enzymes, such as malic enzyme, glucose-6-phosphate dehydrogenase, and acetyl-CoA-carboxylase, compared with other adipose depots (35). Much less is known about lipolytic activity and FA/TAG cycling of these adipocytes in pigs and other livestock animals; however, this process of FA/TAG cycling has been investigated in rodent models.

In rodents, the popliteal fat pad is known as the “intermuscular depot,” and it is located in the popliteal fossa in the posterior knee and is not typically visible in very young animals (36). The popliteal fat pad, or IMAT depot, in mice has been shown to convert glucose to fatty acids an order of magnitude faster than the epididymal fat pad or skeletal muscle during radiolabeled glucose test meals. These results indicate that adipocytes within the skeletal muscle (i.e., IMAT adipocytes) play a major role in de novo synthesis of fat from dietary carbohydrate in mice (37). In guinea pigs, a 3-month exercise training study demonstrated that the capacity for glucose utilization was reduced following training in all adipose depots studied except the popliteal (IMAT) depot. In addition, the amount of lipid decreased proportionally to the increase in protein in this IMAT depot following training (38), suggesting that certain intermuscular adipose depots indeed act as a local energy source for adjacent muscles. Following an hour of exercise in a hamster ball, the rates of FA/TAG cycling rose significantly in the intermuscular (IMAT) and some superficial adipose depots of Dwarf hamsters and correlated closely with the site-specific activities of hexokinase and phosphofructokinase. These data are consistent with the notion that adipose depots are depleted selectively during exercise (39). A study in isolated adipocytes from nine different adipose depots of guinea pigs found that the two small IMAT depots had the lowest rates of basal lipolysis but exhibited the fastest and greatest maximum response to stimulation by adenosine deaminase and noradrenaline. Furthermore, noradrenaline-stimulated lipolysis was most effectively inhibited by small quantities of insulin in these IMAT depots (40). IMAT may therefore buffer lipids locally to adjacent muscle via adrenergic stimulation and insulin inhibition of release (i.e., lipolysis). Such metabolic flexibility of adipose tissue is central to maintenance of whole-body metabolic homeostasis (41).

Agricultural research on IMAT is focused on enhancing meat quality for commercial purposes rather than metabolism, and the rodent literature is limited by the very small number of studies on metabolic function; thus, metabolic phenotypic data related to IMAT quantity and quality are virtually absent from these animal models. For example, is a cow with more marbled meat (i.e., more IMAT) also more insulin resistant than a cow with less marbled meat (i.e., less IMAT)?

Adult human adipose tissue can be broadly categorized into upper- and lower-body adiposity. Within these two categories, the adipose tissues can be further divided into specific depots depending on location (e.g., visceral, subcutaneous, intermuscular, gluteal-femoral, etc.). In 1953, Vague identified central adiposity (upper body adiposity) as a key determinant of several common chronic diseases, including type 2 diabetes, myocardial infarction, hypertension, and stroke. In contrast, he proposed that gluteal-femoral adiposity (lower body adiposity) was metabolically beneficial (42). This is consistent with the concept of partial lipodystrophy as an etiology of insulin resistance and type 2 diabetes (43) in which a failure to properly accumulate adipose tissue in the lower body is hypothesized to lead to ectopic fat in other locations, including the abdominal cavity, liver, muscle, β-cell, and brain. Understanding why gluteal-femoral adipose tissue can expand in some individuals is essential to explain the overall variation in the susceptibility to developing insulin resistance and/or obesity-related comorbidities. Much like SAT (e.g., abdominal versus gluteal-femoral), IMAT quantity and quality likely differ by anatomical location (e.g., abdomen versus thigh). However, IMAT from different muscle beds and anatomical locations have not been investigated. To date, IMAT measured “below the waist” has been related to metabolic abnormalities and cardiovascular disease risk (44–46), but whether there are beneficial aspects of lower-body IMAT is an exciting and unexplored concept in human adipose biology research. We will therefore explore what is currently known about IMAT quantity and quality in humans and discuss its potential role as a friend or a foe to metabolic health and disease.

Historical and current published evidence in IMAT from humans have been derived from IMAT collected as “samples of convenience.” In other words, IMAT is sampled randomly during a standard vastus lateralis muscle biopsy with a Bergstom needle (47) (Supplementary Material). In our experience, IMAT is sampled roughly 50% of the time using this technique; however, sampling of convenience makes it extremely difficult to perform prospective and interventional clinical research studies to interrogate IMAT with such low success rates. We have therefore developed an ultrasound-guided muscle biopsy technique to specifically target IMAT (Fig. 2). In B mode, ultrasound adipose tissue and fascia appear white, while muscle is dark gray to black. Using a linear transducer probe and a single-handed biopsy instrument (Monopty Disposable Core Biopsy Instrument; Becton, Dickinson and Company, NJ), a biopsy operator can triangulate IMAT in the muscle of interest. By inserting the biopsy needle in the same plane as the ultrasound measurement, one can visualize the needle placement for sampling in real time to maximize the probability of IMAT collection. Using this technique, we have achieved closer to 90% success rate with IMAT biopsy sampling in our ongoing studies. In addition to consistent sampling of IMAT from the vastus lateralis, this technique may also permit IMAT sampling from other muscle depots in humans. Another limitation to the current sampling during a Bergstrom muscle biopsy (without ultrasound) is the small amount of tissue obtained, which prohibits the use of numerous platforms such as single-cell/single-nucleus transcriptomics and epigenomics, lipidomics, metabolomics, etc., without pooling numerous samples. These ultrasound-guided biopsies may improve the yield of IMAT collected and, thus, create opportunities for deeper molecular investigations of this enigmatic tissue.

Accumulations of visceral (VAT) and abdominal deep subcutaneous (SAT) adipose tissues have been inextricably linked with the pathogenesis of insulin resistance and type 2 diabetes (48,49). Whole-body imaging studies demonstrate that IMAT is a novel depot similar in size to VAT (50) and therefore hypothesized to impact insulin responsiveness and metabolic function to an extent similar to that of VAT or deep SAT (51,52). Potential reasons for IMAT accumulation in humans being deleterious or beneficial include 1) sex and related circulating hormone levels, 2) race and ethnicity, and 3) lifestyle factors (e.g., diet and physical activity level) (Fig. 3).

Circulating sex steroid hormones dramatically impact adipose biology (53). For example, reducing testosterone concentrations below baseline increases SAT stores in the appendices and abdomen. Conversely, elevating testosterone concentrations above baseline induces a greater loss of adipose tissue from the smaller, deeper intermuscular (IMAT) depot of the thigh (54). In postmenopausal women with abdominal obesity, pulsatile growth hormone secretion demonstrated an independent and negative association with thigh IMAT, whereas basal growth hormone secretion showed an independent and negative association with VAT. These data suggest that the neuroendocrine associations between fat mass and the somatotropic axis are depot dependent and that thigh IMAT is important in this interplay (55).

Racial differences in obesity and type 2 diabetes are profound, with African Americans having a higher prevalence than Caucasians. Higher insulin resistance and hyperinsulinemia have also been reported in adult African Americans than Caucasians. Racial differences in adipose tissue distribution extend to IMAT, a depot that may influence the race-ethnicity differences observed in dysglycemia (56). In a whole-body imaging study of nearly 350 healthy sedentary African American, Asian, and Caucasian adults, IMAT depots were not significantly different in size among race groups at low levels of adiposity; however, with increasing adiposity, African Americans had a significantly greater increment in the proportion of IMAT to total adipose tissue than did the Caucasians and Asians. VAT depots were not significantly different in size at low levels of adiposity, but, with increasing adiposity, VAT accumulation was greater than IMAT accumulation in the Asians and Caucasians with no significant differences observed in the African Americans (50). These data suggest that the IMAT and VAT depots in African Americans are poised for adipose expansion.

Like most adipose tissue depots, IMAT content is influenced by habitual diet and physical activity. Unlike IMCL (57), IMAT content is low in insulin-sensitive endurance-trained athletes (58). The first clinical trial to prospectively interrogate impact of lifestyle interventions on the IMAT depot was published in 2002 (13). Significant reductions in body weight (approximately 10 kg, or 10%) and in total, abdominal subcutaneous, visceral, and intermuscular (IMAT) fat were observed with 16 weeks of diet alone, diet plus endurance training, and diet plus resistance training. The changes in the body fat and metabolic variables were not different across treatment. Since that study, endurance exercise training is one lifestyle intervention that has been shown to consistently reduce IMAT content or prevents the age-associated increases in IMAT (15,59–61). Resistance exercise training has also been reported to reduce IMAT content following a 12-week intervention (62). In fact, compared with energy restriction alone, there is a greater loss of IMAT with exercise training interventions (63,64). Following a 1-year intervention, the changes in IMAT and VAT contents with exercise training were similar, with twice the decrease compared with caloric restriction alone (64); thus, exercise training appears to preferentially reduce IMAT content. An acute exercise bout of intervals at differing intensities, however, did not impact IMAT content. The impact of exercise training therefore occurs from repeated bouts (i.e., chronic exercise training) rather than as an acute response (65). Conversely, detraining or a period of chronic physical inactivity significantly increases IMAT content. Four weeks of unilateral lower limb suspension increased IMAT content by 15–20% in both the thigh and the calf of healthy young adults (59). In a cohort of healthy young men, 12 months of detraining (after completing a 12-week exercise protocol on an isoacceleration dynamometer) significantly increased IMAT depots in the upper arm by 14% and demonstrated that long-term detraining leads to unfavorable muscle adaptations, with decreases in endurance strength, volume return to pretraining size, and significant IMAT accumulation. These findings align with observations in patients with muscular dystrophies and spinal cord injuries in which extreme cases of muscle disuse and muscle wasting are accompanied by aggressive fatty infiltration (i.e., high IMAT accumulation) and often what is referred to as “fatty replacement” of the muscle. In fact, IMAT content is used as a marker of the severity or stage of the disease in muscular dystrophies (66), and, in many cases, the overabundance of IMAT is also associated with insulin resistance (67).

Insulin sensitivity is commonly expressed per kilogram of body mass and lean mass. If lean mass is assessed by hydrostatic weighing, IMAT is taken into account in making the distinction between lean mass and fat mass; however, with most dual-energy X-ray absorptiometry (or magnetic resonance imaging–based) segmentation algorithms, IMAT is most likely not taken into account. Segmentation of IMAT within individual muscle groups would allow for better normalization of insulin sensitivity data to lean mass. Based on the above evidence, the quantity of IMAT in humans is strongly linked to metabolic homeostasis, with greater amounts of IMAT being associated with adverse metabolic consequences. It is important to note that the metabolic impact of IMAT might depend on adjacent muscle quality and vice versa, as this is likely a bidirectional relationship. Thus far, associations from some of our own clinical studies have shown that force/cross-sectional area decreases in men and women as muscle attenuation decreases (15). Some of our recent in vitro studies have demonstrated that IMAT is a causal force in promoting sarcopenia and low muscle quality (68). A question that is yet to be explored, however, is whether IMAT quantity per se is the driving factor in the etiology of insulin resistance and type 2 diabetes or rather is the quality of the IMAT the true puppeteer. In general, adipose tissue quality likely influences its secreted factors, which are also likely to influence the metabolism of surrounding tissues. Investigations into the secretory function of IMAT in humans are limited in number and warrant deeper exploration.

Our view of adipose structure and function has radically changed over the past 25 years. The discovery of leptin in 1994, a critical hormone in energy balance produced principally by white adipose tissue (69), deemed that adipose tissue is an endocrine organ with the potential to communicate with other organs via paracrine or endocrine signaling mechanisms. In humans, adipose tissue secretory profiles differ by metabolic phenotype (i.e., lean versus obese) (70). In addition to releasing free fatty acids, adipose tissue secretes over 300 proteins, including hormones, cytokines, adipokines, lipids such as eicosanoids and oxidized phospholipids (71,72), and miRNAs (73), that act both locally and systemically. Typically, secretion of proinflammatory cytokines is greater from VAT than SAT, and mRNA expression of the insulin-sensitizing hormone adiponectin is lower in VAT than SAT (74–77).

We know from human studies that lipid droplets resident within skeletal muscle (so-called IMCL) can provide an energy source for adjacent working skeletal muscle (11,78), but whether adipocytes in the IMAT depot can also serve as a fuel source for adjacent skeletal muscle has not been addressed in humans. A study from our group observed an attenuation in the age-related increase in IMAT (by CT scan) paralleled by preservation of muscle strength in older adults following 12 months of training (15); one interpretation of this finding is that the demand of physical activity on the skeletal muscle required the neighboring IMAT depot to provide energy as fuel to the adjacent working muscle, thereby maintaining the overall size of the depot and making use of the existing lipid pools rather than continuing to expand the IMAT depot and store more lipid. The anatomical relations of adipose tissue to skeletal muscle therefore may be an integral part of its physiological function (38,79).

Adipose tissue to skeletal muscle cross-talk has been examined in vitro primarily in two ways: 1) administration of conditioned medium collected from adipose tissue explants or cultured adipocytes directly onto myotubes and 2) an actual coculturing of adipocytes with myotubes, allowing for the natural passage of secreted factors in real time from one cell to the other. Using a three-dimensional coculture of human myotubes and adipocytes from obese individuals, an increased secretion of cytokines and chemokines by the VAT (versus SAT) adipocytes decreased expression of genes related to skeletal muscle contractility and myogenesis (70). Conditioned medium from human VAT adipocytes has also been shown to reduce insulin-stimulated glucose uptake in L6 myotubes via secretion of high levels of IL-6 (80). A landmark study in 2011 provided an exciting alternative view of adipocyte communication with skeletal muscle (81). Using a coculture system with the intention of mimicking the communication between adipocytes and skeletal muscle, a few key observations were made: 1) adipocytes in the basal state sequestered free fatty acids, thereby forcing neighboring myotubes to rely on glucose; 2) in the basal state, insulin action was enhanced in myotubes from lean but not obese donors; 3) when exposed to lipolytically active adipocytes, cocultured myotubes shifted substrate use in favor of fatty acids, which was accompanied by intracellular accumulation of triacylglycerol and even-chain acylcarnitines, reduced glucose oxidation, and modest attenuation of insulin signaling. Collectively, these findings illustrated that the metabolic (in this case, lipolytic) state of the adipocytes dictated the metabolic responses and state of the neighboring muscle cells. One caveat to this study is that the adipocytes used were from the abdominal subcutaneous adipose depot and not actually from the IMAT depot. Considering the proximity of IMAT adipocytes to muscle, it is possible that IMAT can significantly impact muscle metabolism. Recent work from our group found that the rates of basal lipolysis were similar between IMAT and VAT explants from individuals with severe obesity, both of which were greater than the rate of basal lipolysis in SAT. Conditioned medium from IMAT and VAT explants increased 1,2-diacylglycerol accumulation in myotubes, suggesting that the high rates of lipolysis in IMAT and VAT promoted bioactive lipid formation in the muscle. In the same study, IMAT and VAT explant conditioned media decreased insulin sensitivity in myotubes, while SAT conditioned medium had no effect. mRNA expression of macrophage markers and inflammatory cytokines in IMAT was correlated with whole-body insulin resistance. Collectively, these are the first data in humans to show that IMAT has a potency similar to that of VAT to promote metabolic dysfunction and insulin resistance in human skeletal muscle (82).

Adipose tissue secretes many factors that influence whole-body metabolism, and potency varies by depot; however, our knowledge of adipocyte signaling molecules responsible for these effects is far from complete. When interpreting the above findings, it is critical to note that intense cell-to-cell communication occurs within adipose tissue, whereby different cell populations promote secretion of various signaling molecules and adipokines. When human adipose tissue is collagenase digested into adipocytes, stromal vascular fraction, and undigested tissue matrix, the sum of adipokines and signaling molecules from each fraction does not equal what is secreted by the tissue as a whole (74). Interestingly, 90% of what is secreted from adipose tissue comes from nonadipocytes, with the majority of adipokines released by the cells of the tissue matrix rather than the stromal vascular fraction that includes macrophages and preadipocytes (77). Thus, the experimental model directly influences the outcomes such that secreted factors from adipocytes in culture do not completely reflect the secreted factors from the adipose tissue itself (i.e., tissue explants). The metabolic state of the adipocyte, as well as of the tissue milieu, determines the secretome.

Accessibility of IMAT in humans presents many challenges that can be overcome with emerging technologies. We have unpublished microarray data from paired samples of IMAT and abdominal SAT in premenopausal females with obesity (Fig. 4 and Supplementary Material). While there is considerable overlap in the transcript profiles of the two tissues, these data show favorable metabolic signatures unique to IMAT, such as upregulated pathways related to glycolysis, but are confounded by an overwhelming myogenic signature. Due to its proximity to muscle and the fact that IMAT is typically a sample of convenience during skeletal muscle biopsies, contamination from muscle is nearly unavoidable. These samples in particular were dissected free of visible skeletal muscle tissue under a dissection scope, followed by targeted gene expression analyses to ensure no expression of myogenic muscle-specific markers. As evidenced by the pathway analyses, however, muscle contamination clearly remains in these IMAT samples. The advent of molecular assessments, such as transcriptome sequencing (RNAseq), assay for transposase-accessible chromatin using sequencing (ATACseq), and DNA methylation at the single-cell and single-nucleus levels, as well as the potential for ultrasound-guided biopsies specifically for IMAT, will permit more sophisticated investigations of human IMAT to eliminate the contamination from skeletal muscle cells and dramatically advance our understanding of this enigmatic adipose tissue.

IMAT, being pliable like other adipose depots, has developed its anatomical location and function over time in numerous species in response to environmental cues. To what extent the etiology, quantity, and quality of the intermuscular (IMAT) depot impacts metabolic disease in humans is a significant knowledge gap. Data from lower vertebrates and mammals clearly demonstrate the ability of IMAT to meet the metabolic demands of the whole animal without physical or metabolic adversity through mechanisms such as transient depot expansion, increased antioxidant and anti-inflammatory stores, and FA/TAG cycling, highlighting the significance of both quantity and quality. In humans, however, IMAT quantity has been negatively associated with metabolic health (14) and shown to have proinflammatory transcript profiles of the tissue and the secretome (68); however, associations do not directly assess underlying biological mechanisms, and the influence of intrinsic and extrinsic factors must also be considered. For example, assessments of IMAT content alone do not capture the metabolic flux of the tissue whereby functions such as adipocyte turnover and FA/TAG cycling are absent. Furthermore, the impact of age, sex, race, obesity, and lifestyle on IMAT quality and its consequent influence on the secretory profile and IMAT communication with other tissues are virtually unexplored (Fig. 3). The gluteal-femoral adipose depot, in the absence of excessive adipose accumulation in the central depots (i.e., pear shaped), is enriched with an open chromatin landscape that maintains genes in a poised state for driving lipid uptake and depot expansion, supporting its hypothesized role in reducing metabolic disease risk through greater tissue expandability and lipid uptake (83). Deeper investigation into the molecular regulation of depot expansion and metabolic flux in human IMAT is one of the next horizons in adipose biology research.

Acknowledgments. The authors thank John Hill (University of Colorado Anschutz Medical Center, Aurora) for his help in setting up the ultrasound-guided biopsies for the sampling of IMAT.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.