Alterations in adipose tissue composition and function are associated with obesity and contribute to the development of type 2 diabetes. While the significance of this relationship has been cemented, our understanding of the multifaceted role of adipose tissue in metabolic heath and disease continues to evolve and expand. Heterogenous populations of cells that make up adipose tissue throughout the body generate diverse secretomes containing a mosaic of bioactive compounds with vast structural and signaling capabilities. While there are many reports highlighting the important role of adipose tissue endocrine signaling in insulin resistance and type 2 diabetes, the direct, local, paracrine effect of adipose tissue has received less attention. Recent studies have begun to underscore the importance of considering anatomically discrete adipose depots for their specific impact on local microenvironments and metabolic function in neighboring tissues as well as regulation of whole-body physiology. This article highlights the important role of adipose tissue paracrine signaling on metabolic function and insulin sensitivity in nearby tissues and organs, specifically focusing on visceral, pancreatic, subcutaneous, intermuscular, and perivascular adipose tissue depots.
Excess fat and how/where it is stored is fundamental to our understanding of metabolic health and disease. When considering an individual’s risk for type 2 diabetes, whether or not they are overweight or have obesity is typically of utmost importance. However, the connection between adipose tissue and disease progression is anything but simple. Adipose tissue has matured far beyond its original designation as a simple storage depot for excess energy. It is a dynamic tissue that is comprised of a multitude of different cell types, including fibroblasts, endothelial cells, immune cells, extracellular matrix (ECM), preadipocytes, and adipocytes (1). Each of these cell types has distinct abilities to synthesize and secrete lipid and protein mediators, and together this heterogenic population of cells that compose adipose tissue contributes to whole-body physiology in many different and important ways. Not only is adipose tissue diverse on the level of cellular composition and secretome, but its metabolic profile and functional roles vary widely depending upon anatomical location. Compartments that have received the most attention with regard to type 2 diabetes and metabolic disease are the subcutaneous adipose depot, which sits beneath the skin, and the visceral adipose depot, which lines internal organs within the abdominal cavity. In addition, adipose tissue accumulates in and around other metabolic tissues and organs such as the pancreas, skeletal muscle, and vasculature, and not one of these ectopic adipose depots is the same as the next (2). Fundamental differences in developmental origin, proliferative and metabolic capacity, vascularization, cellular composition, hormonal sensitivity, and secretion patterns all contribute to the heterogeneity between adipose tissue from different anatomical locations (3). Adipose tissue becomes even more variegated when considering the significant distinctions between white, brown, beige, and pink adipocytes as well as adipose compartments that have yet to find their place on the adipose tissue rainbow, such as bone marrow adipose tissue. The plasticity of adipose tissue results in depots with unique and specialized metabolic implications, and excellent work exploring the impact of brown, beige, pink and other classifications of adipose tissue on metabolic disease has been conducted elsewhere (4–10). This article will focus on the white adipose tissue specifically and how its anatomy translates to specific and diverse functional interactions with nearby tissues that subsequently impact whole-body physiology.
It is known that adipose tissue secretes a multitude of bioactive compounds in abundance, such as proteins, cytokines, adipokines, free fatty acids, and other lipid biomolecules with potent signaling and structural impact on metabolic tissues (Fig. 2B) (11). While we have a wealth of reports highlighting the important role of adipose tissue endocrine signaling in insulin resistance and type 2 diabetes, the direct, local, paracrine effect of white adipose tissue accumulation from distinct anatomical locations on neighboring metabolic tissues and organs has received less attention. Considering the impact adipose tissue-derived compounds are known to have on distant tissues after traveling through systemic circulation, the effect that this milieu of secreted factors has on tissues residing just next door could be profound. This article will highlight paracrine interactions between specific adipose depots and their anatomically relevant metabolic tissues, specifically focusing on the impact of visceral, pancreatic, subcutaneous, intermuscular, and perivascular adipose tissue depots on insulin sensitivity and metabolic disease.
Visceral Adipose Tissue
Central to our current perspective on the link between adipose tissue and type 2 diabetes is the expansion of the visceral adipose tissue (VAT) depot. VAT is broadly understood to be intra-abdominal adipose tissue that surrounds internal organs, and this compartment can be broken down more specifically to include the omental, mesenteric, and retroperitoneal adipose depots (Fig. 1) (12). While retroperitoneal adipose tissue is within the abdominal cavity, it is exterior to the peritoneum, and, as such, secretions from this depot are released into systemic circulation rather than through the portal vein, like in the intraperitoneal adipose compartments (omental and mesenteric) (13). The direct access to the portal circulation and then the liver is central to the association between VAT and metabolic disease, and multiple studies have found that intraperitoneal adipose depots are independently associated with impaired peripheral and hepatic insulin action, glucose metabolism, and prevalence of type 2 diabetes, whereas retroperitoneal adipose tissue is not (14–17). Because of this, when investigating VAT specifically, under the lens of metabolism, it is most relevant for only the mesenteric and omental adipose depots to be included. However, adipose tissues surrounding retroperitoneal organs are receiving increasing attention for their important and independent paracrine roles on essential neighboring tissues, and those relevant to metabolic disease will be discussed separately in this review. Additionally, it is important to note that while many attributes of adipose tissue are shared across various species, there are anatomical differences that are important to consider when making cross-species comparisons. Given the abundance and usefulness of rodent models in adipose research, the degree of dissimilarity between rodent and human adiposity is of particular relevance. For instance, while the rodent mesenteric fat pad is thought to be the most similar to human VAT, it is rarely used due to low accessibility. Instead, the rodent perigonadal fat pads are often used as a surrogate for VAT despite the dissimilarity to human VAT (18).
As mentioned, one area that is widely agreed upon and has been well documented is the important role of VAT accumulation on hepatic metabolic regulation. The increased delivery of free fatty acid (FFA) to the liver in obesity and type 2 diabetes disrupts normal hepatic regulation of glucose and fatty acid/triglyceride disposal, contributing to hepatic steatosis as well as systemic hyperglycemia, insulin resistance, and type 2 diabetes (19–21). Visceral adiposity in obesity is associated with an altered stromal-vascular fraction, marked by increased abundance of immune and endothelial cells and enhanced expression of the ECM (Fig. 2A) (22–24). These and other nonadipocyte cells in VAT are estimated to account for over 33% of total cell composition and are responsible for secreting over 90% of factors comprising the VAT secretome (25,26). Consequently, in addition to FFA, which are known to decrease postprandial glucose uptake and increase gluconeogenic flux, VAT has been shown to secrete an abundance of proinflammatory cytokines, adipokines, and hormones that interact with Kupfter cells and other immune cells in the liver, such as NKT cells, which are involved in inflammation-induced insulin resistance, and these processes have been reviewed extensively elsewhere (27,28). In addition to what has been previously described, studies using proteomics analyses of the VAT secretome have found that signaling/regulatory proteins, as well as proteins involved in immune function and the modulation and regulation of the ECM, are secreted prominently and abundantly (Fig. 2B) (29,30). Altered ECM remodeling observed in obesity and type 2 diabetes is an important factor in the development of liver fibrosis and NAFLD (31,32). Direct structural effects, such as increased fibrillar collagen deposition and the development of scar tissue, are hallmarks of liver fibrosis; however, the ECM also provides an acellular storage site for multiple otherwise soluble factors that modulate fibrogenesis/fibrolysis, and ECM proteins have also been shown to interact with cell surface receptors such as integrins and discoidin domain receptors (33). These transmembrane signaling interactions with ECM proteins are vital to hepatocyte differentiation and liver homeostasis and have been shown to modulate liver fibrogenesis as well as insulin and other metabolically relevant signaling cascades (33–35). The potential paracrine contribution of VAT-derived ECM proteins on the liver has yet to be explored, and more studies are required to reveal possible mechanistic implications on hepatic dysregulation in metabolic disease.
Secretions from VAT depots have direct access to more than just the liver, as they are also in contact with intraperitoneal organs, namely, the intestines (Fig. 1). How might the interaction between these visceral adipose depots and the tissues they directly neighbor contribute to metabolic health and disease? Composition and diversity of the gut microbiome has emerged as a meaningful component in the maintenance of homeostasis among a variety of metabolic tissues. Obesity is associated with alterations to the gut microbiome as well as increased intestinal barrier permeability, or “leaky gut,” which occurs when the regulation of the tight junctions of the mucosal lining of the gut is disrupted, leading to the translocation of bacteria and bacterial byproducts into neighboring tissues as well as circulation (36,37). Increased gut permeability is associated with decreased insulin sensitivity as well as increased levels of circulating inflammatory cytokines in individuals with obesity (36), although specific mechanisms driving these relationships are still poorly understood. In response to high-fat diet and hyperglycemia, rodent models of obesity have reported significant disruption in epithelial integrity and intestinal barrier permeability, leading to increased influx of microbial products systemically (38). Interestingly, the activation of proinflammatory mediators in the small intestine have been shown to precede increases in fat mass and development of obesity and insulin resistance (39). Additionally, signals that control gut lipid absorption have been shown to originate specifically from adipose tissue, driving caloric influx and systemic metabolic regulation in the mouse model (40). The digestive tract is our first line of defense against ingested nutrients and is susceptible to chronic exposure from high-fat and inflammatory diets, which are associated with altered absorption and decreased intestinal barrier function, leading to leaky gut (41–43). Considering the proximity of our visceral adipose depots to the gut, the cross talk between these two organs, and exposure to this influx of inflammatory bacterial components, may play an important role in the proinflammatory nature of VAT.
VAT is highly immunogenic, particularly the omentum, which has been referred to as the “policeman of the abdomen” due to its diverse defensive abilities, ranging from the mitigation of peritonitis to providing factors involved in tissue healing and repair (44). This immune capacity is largely due to aggregates of leukocytes embedded between adipocytes, referred to as Milky Spots (MS), in omental adipose tissue. MS collect antigens and pathogens and support innate and adaptive immunity, initiating and responding to inflammation, recruiting macrophages, and forming populations of B and T cells (45). The mesenteric adipose depot also contains similar immune cell clusters called fat-associated lymphoid clusters, although they are not as populous as MS in the omentum (46). The immune profiles of these VAT depots are responsive to increased BMI, and obesity-associated VAT inflammation is linked to the accumulation of proinflammatory immune cells including M1 macrophages, CD8+ T cells, and neutrophils (42,47). Additionally, a reduction in anti-inflammatory immune cell populations such as M2 macrophages, regulatory T cells, and ILC2 cells has been observed in obese VAT (Fig. 2A) (48–50). When considering the influx of proinflammatory microbial agents associated with leaky gut in obesity, these sophisticated and highly reactive immune populations that are primed for an immunogenic response in VAT are of particular relevance. One can easily imagine a paracrine loop in which high-fat diet/hyperglycemia leads to increased gut permeability and bacterial translocation, which chronically activates MS and fat-associated lymphoid clusters in VAT, resulting in an onslaught of proinflammatory aggravation and tissue damage and contributing to chronic systemic inflammation and insulin resistance. The link between VAT and the liver in metabolic disease is unrefuted; however, it is evident that other significant paracrine interactions are at play that may even precede the negative impact of VAT on the liver, and this gut–adipose cross talk requires further investigation.
Finally, it is important to note the unique and specific anatomy and potentially different functions of the two subcompartments of VAT. Omental adipose tissue is often described as a sheath that hangs off the stomach and covers the intestine (“omentum” is derived from the Latin word for apron), whereas mesenteric adipose tissue is more deeply burrowed and surrounds the intestines (Fig. 1) (51). Because of this, studies concerning VAT in humans are much more commonly conducted on omental adipose tissue, which is more easily accessed. Very little attention has been given to the differences between the omental and mesenteric compartments of VAT in humans; however, one study comparing the two from individuals with type 2 diabetes found that basal lipolytic rate is significantly higher in mesenteric adipose tissue than in omental adipose tissue (52). Additionally, compared with omental adipose tissue, mesenteric adipose tissue has significantly higher expression of important genes involved in lipid and hormone metabolism, such as leptin, peroxisome proliferator-activated receptor-γ (PPARγ), fatty acid translocase (FAT/CD36), and 11β-hydroxysteroid dehydrogenase type 1, while lower expression of adiponectin is observed (52). These findings suggest that mesenteric adipose tissue is more metabolically active than omental adipose tissue; however, more studies are required to further elucidate the functional relevance of the differences between these two depots as it pertains to metabolic disease in humans.
Pancreatic Adipose Tissue
In addition to expanding our understanding of the intraperitoneal (visceral) adipose depots as they relate to metabolic disease, adipose tissues associated with retroperitoneal organs are beginning to receive more attention for their specific and unique functional roles on the physiology of neighboring tissues. Of particular relevance to the development of type 2 diabetes is pancreatic adipose tissue (PAT), which surrounds the pancreas and establishes direct contact with pancreatic islets and insulin-producing β-cells (Fig. 1) (53). PAT accumulation is positively associated with BMI and type 2 diabetes risk, and many reports have found a negative relationship between PAT and insulin secretion (54–57), although some report no association (58,59). In addition to adipocytes, PAT has an abundant stromal vascular fraction made up of fibroblasts, macrophages, T and B cells, and lymphocytes (Fig. 2A) (60). When measured in normal-weight mice, PAT displayed 10 times more stromal vascular cells per gram of tissue than other abdominal adipose depots (60). These cell types, in addition to adipocytes and preadipoctyes, are known to synthesize and secrete a variety of factors that could directly impact islet function (Fig. 2B). In fact, in human pancreatic resections from tumor-free regions, there was greater macrophage and monocyte accumulation in islets in close proximity to PAT, possibly mediated through paracrine secretion of interleukin-8 (IL-8) and monocyte chemoattractant protein-1 (MCP-1), suggesting that PAT creates an inflammatory microenvironment that could impact islet function (54). Additionally, adipose-derived factors, such as resistin, have been shown to impact islet blood flow, which could introduce an indirect role for PAT on pancreatic insulin release and nutrient sensing (61). However, due to the high risk involved with surgically intervening or operating on the pancreas, mechanistic studies aimed at elucidating the functional relationship between PAT and the pancreas do not exist to date in humans, and our understanding of the association between PAT and type 2 diabetes comes from noninvasive imaging techniques.
However, the potentially significant paracrine impact of PAT on pancreatic hormone regulation has not been lost on investigators, and limited studies in rodent and cell culture models have interrogated the cellularity of PAT as well as the composition of the PAT secretome and its effects on insulin-producing cells in vitro. In a mouse model of obesity, compared with gonadal VAT, PAT is associated with the aggregation of larger and more abundant inflammatory foci and higher baseline expression of immune-modulatory cytokines (62). Additionally, in comparisons of the PAT from lean versus obese mice, obese PAT has significantly increased expression of specific cytokines such as interferon-γ (62). The paracrine potential of this altered PAT immune profile in obesity is underscored by the observation that PAT conditioned medium generated from obese mice causes significantly increased proliferation of pancreatic β-cells in vitro, an effect not observed in response to VAT conditioned medium from the same obese mice (63). Additionally, obese PAT conditioned medium has been shown to increase β-cell insulin secretion in vitro during the development of type 2 diabetes (53). However, adipocytes have also been shown to decrease insulin production in response to stimulation in vitro, as has been seen in a coculture model including insulin-producing MIN6 cells and 3T3-L1 adipocytes (64). These effects on insulin-secreting cells suggest a role for adipose tissue paracrine signaling in the regulation of pancreatic insulin production; however, the degree to which this may happen in humans is unknown. Taken together, these studies suggest that PAT is a unique adipose depot that undergoes significant and specific modifications in response to diet-induced obesity and that these alterations in tissue dynamics and secretory profile may transform the pancreatic microenvironment and directly impact islet function. This could be an important aspect of the etiology of type 2 diabetes, and more studies are needed to fill the current gaps in knowledge regarding the functional relationship between PAT and the pancreas in humans.
Subcutaneous Adipose Tissue
The subcutaneous adipose tissue (SAT) depot is commonly regarded as a healthy or benign compartment for the storage of excess energy as fat and has even been described as protective against type 2 diabetes and coronary artery disease (65). It resides immediately beneath the skin, in the hypodermis, and can be found virtually wherever skin is found, making SAT not only the most abundant adipose tissue in the human body but also the most anatomically diverse (66). As the heterogeneity of SAT has been investigated more closely, it has been shown, not surprisingly, that there are significant differences in the metabolic and physiological influence of SAT depending upon anatomical location. For instance, abdominal SAT accumulation has been found to be more metabolically detrimental than lower body (gluteal-femoral) SAT, which is actually associated with a reduced incidence of type 2 diabetes when total adiposity is equivalent (67,68). In an attempt to explain these important physiological disparities, studies have found that compared with gluteal-femoral SAT, abdominal SAT is characterized by smaller adipocyte size, greater rates of lipolysis, and altered gene and protein expression, although the underlying functional explanation for the differences in disease risk associated with bodily SAT distribution remains unclear (69,70).
One observation that has emerged as having significant involvement in SAT heterogeneity is the distinction between superficial and deep adipose compartments within the SAT depot. A thin membranous layer of connective tissue, referred to as the superficial fascia (Scarpa’s fascia in the abdominal region), divides SAT into two compartments, with the superficial SAT superior and the deep SAT inferior (Fig. 1) (71,72). This fascial division of superficial versus deep exists throughout the human body’s SAT, including the lower body and extremities; however, most attention to this division has been focused on the abdominal region. This is because multiple studies have shown that this deep SAT in the abdomen is strongly related to insulin resistance and type 2 diabetes in a manner similar to that of VAT, whereas the superficial SAT is not related and seems to be more consistent with lower-body SAT (73,74). Among individuals with type 2 diabetes, this association between deep SAT and metabolic risk factors is higher in males than in females, a finding that is also in line with the trends commonly seen with VAT (75). Additionally, the two subcompartments differ histologically, with the superficial layer containing cuboidal fat lobules that are organized and compact, whereas the deep layer contains flat irregular fat lobules (76,77). While it is clear that the subcompartments of SAT are biologically and physiologically distinct tissues, we still do not know why they relate to metabolic health in such different ways.
While SAT is known to synthesize and secrete an abundance of proteins, adipokines, and lipid mediators (Fig. 2B), the impact of the SAT secretome is generally considered in a strictly endocrine manner. However, the recognition of the significant physiological differences between the deep and superficial compartments of SAT may introduce the need to reexamine the anatomy of the two compartments and the potential for differing paracrine influences between the two. The anatomy of this region traveling from outside to inside can be identified generally as skin, superficial SAT, superficial fascia (described above), deep SAT, deep fascia, and finally skeletal muscle (Fig. 1) (78). Although fascia is often oversimplified as an encasing for muscle and other tissues, it is also a dynamic structure that not only supports but also interacts with and responds to a variety of different stimuli (79). This fascial membrane, which is entirely capable of synthesizing and transporting molecules as well as transmitting signals, is the only thing separating deep SAT from one of the most important tissues in regulating glucose homeostasis, skeletal muscle. This is an important difference between the deep and superficial SAT compartments. It has been shown that expression of genes involved in the regulation of lipolysis (lipoprotein lipase [LPL] and hormone-sensitive lipase [LIPE]) and lipid biogenesis (FA synthase [FASN] and stearoyl–coenzyme A desaturase-1 [SCD1]) as well as the rate of lipolysis itself are increased in deep compared with superficial SAT (73,80). A higher expression of inflammatory genes (interleukin-6 [IL6] and MCP-1) has also been reported in the deep versus superficial SAT compartment (73). This altered metabolic and immune cell profile could result in increased secretion of proinflammatory cytokines, FFA, and eicosanoids that are well-known to cause inflammation and decreased insulin sensitivity in skeletal muscle. Additionally, when measuring secreted proteins from primary human adipocytes, in contrast to superficial SAT, multiple parallels can be drawn between adipocytes from deep SAT and VAT, such as increased resisitin and decreased leptin and 11β-hydroxysteroid dehydrogenase type 1 secretion (77,81). These factors have been shown to regulate insulin signaling, glucose uptake, and fatty acid metabolism in skeletal muscle and are associated with obesity and type 2 diabetes (82–84). Therefore, the proximity of deep SAT and its secretome to skeletal muscle may be an important factor in the association between this adipose depot specifically and insulin resistance. However, mechanistic studies are limited, as there has been no evidence of multiple subcutaneous layers in the rodent, as has been described in humans (18). To date, there have been no studies comparing the composition of the deep and superficial SAT secretome or the functional paracrine interactions with nearby skeletal muscle.
Interestingly, the deep fascia separating adipose tissue from muscle in the limbs is comprised of two or three fibrous sublayers and is four to five times thicker than the deep fascia of the trunk/abdomen, which contains just one layer (Fig. 1) (78). This thin deep fascia of the trunk penetrates the muscle and the two cannot be separated, whereas the multilayered deep fascia of the limbs is easily separable from the muscle, which slides around freely beneath it due to the abundance of ECM and layers of hyaluronic acid (85). It is possible that this singular thin and infiltrating divide between deep SAT and muscle in the abdomen allows for easy transport of (potentially proinflammatory) molecules and signals from neighboring adipose tissue in a way that the more profound partition in the extremities does not. This difference could help to explain the disparity in association between abdominal and gluteal-femoral SAT and insulin resistance. However, the specific regional paracrine relationships between muscle, fascia, and adipose tissue in metabolic disease is an area that has yet to be explored.
Intermuscular Adipose Tissue
While various degrees of fascial layers separate deep SAT from skeletal muscle, there is also an entirely distinct adipose tissue that accumulates within the fascial compartment, surrounding skeletal muscle fibers, called intermuscular adipose tissue (IMAT). While IMAT accounts for a very small percentage of total body fat content, it has repeatedly been shown to be predictive of insulin resistance and positively associated with the pathogenesis of type 2 diabetes, cardiovascular disease, and sarcopenia (86–90). Just like any other adipose tissue, IMAT has a diverse and heterogenous cellular composition that synthesizes and secretes an abundance of bioactive mediators (Fig. 2A and B) (1). As skeletal muscle is constantly bathing in these many factors that comprise the IMAT secretome, there is the potential for very direct and potent functional effects of IMAT on muscle metabolism. Skeletal muscle is the largest consumer of postprandial circulating glucose and therefore is essential to systemic metabolic regulation, and it is likely that the paracrine communication between IMAT and muscle contributes to the correlation between IMAT and insulin resistance. However, the anatomical location of IMAT makes it especially challenging to access in humans, and mechanistic studies aimed at addressing the functional relationship between these two tissues are extremely limited.
Due to its influence on meat palatability, IMAT has been studied more extensively in livestock, which allows for the use of invasive techniques that are not as feasible in humans. Molecular studies aimed at identifying differentially expressed genes in porcine SAT, VAT, and IMAT revealed that VAT and IMAT cluster and are mainly associated with inflammatory and immune responses, whereas SAT associates with mediators of glucose and fatty acid metabolism (91). These findings were corroborated by a microRNA transcriptome study, also done in pigs, which found that SAT-enriched miRNAs were related mainly to lipid metabolic homeostasis, while IMAT-enriched miRNAs were related mainly to inflammation and diabetes, targeting, for example, genes regulating the IL-6/STAT3 pathway as well as canonical PPARγ target genes ADIPOQ, UCP1, and FABP4 (92–94). Additionally, in mice, it has been found that the skeletal muscle of obese mice had higher expression of macrophages, T cells, and inflammatory cytokines, which were localized to IMAT (95). Collectively, these studies in animal models suggest that IMAT is highly immunogenic and shares a composition of cellular and genetic distribution similar to those of VAT, which may correlate to a functionally analogous influence on insulin sensitivity.
The idea that IMAT is a metabolically active and immunomodulatory tissue has been supported in humans by a recent study from Sachs et al. (96) that harvested IMAT from the vastus lateralis muscle of individuals across the spectrum of insulin sensitivity and found that IMAT gene expression of macrophage markers and proinflammatory cytokines was negatively associated with insulin sensitivity. Additionally, a strong correlation was found between IMAT expression of genes associated with the ECM and donor insulin sensitivity. As previously mentioned, it has been suggested that ECM remodeling is associated with diet-induced obesity and insulin resistance due to increased deposition of proteins such as collagens as well as direct interactions of secreted ECM proteins with cell surface receptors that modulate signaling cascades involved in glucose and fatty acid metabolism (34,35). These findings suggests that in humans, similar to what has been seen in animal models, the association between IMAT and insulin sensitivity is characterized by increased immune and inflammatory cell populations capable of secreting proinflammatory cytokines, lipids, and proteins that are known to contribute to obesity-associated chronic inflammation and metabolic dysregulation.
Considering the immediate proximity of IMAT to skeletal muscle, the potential functional impact of the IMAT secretome on metabolic regulation in skeletal muscle is quite astonishing. In fact, in the same study by Sachs et al. (96), human IMAT and VAT explants had higher rates of basal lipolysis than SAT and conditioned media generated from these explants significantly increased muscle cell accumulation of 1,2-diacylglycerol, which is known to decrease skeletal muscle insulin signaling. Indeed, IMAT conditioned medium caused decreased insulin sensitivity in human primary skeletal muscles in vitro, with a potency similar to that of conditioned medium generated from VAT. To date, this is the only report on the functional impact of IMAT secretions on skeletal muscle in humans and provides evidence that factors secreted from IMAT contribute to the development of insulin resistance. However, until the composition of the IMAT secretome as well as the specific mechanistic effects of its components on skeletal muscle metabolism become more clarified, the relevance of this paracrine relationship to metabolic disease development will remain unknown.
Perivascular Adipose Tissue
Arteries and veins throughout the body are surrounded by perivascular adipose tissue (PVAT), which, similar to other adipose tissues described previously, contains a heterogenous population of cells with diverse secretions and paracrine signaling capabilities (97). PVAT comes into contact with the majority of vessels that are greater than 100 μm, with the exception of neural and pulmonary vasculature, and therefore can be found in the microvascular beds of metabolic tissues such as muscle and adipose tissue (97). In a normal or healthy state, PVAT has beneficial vasodilating effects on vascular function through the release of vasoactive substances, such as adipose-derived relaxing factor, angiotensin, adiponectin, and nitric oxide, that are essential for the regulation of vascular resistance (98). With the onset of obesity, PVAT undergoes a structural and functional transition to a more immunogenic cellular profile, resulting in the loss of the bioavailability of dilating factors and a release of more vasoconstrictive and proinflammatory substances (98). This obesity-associated alteration in PVAT and secretome composition is associated with cardiovascular and metabolic disease, and many excellent reviews describe this relationship (97–103).
Importantly, heterogeneous PVAT depots exert diverse paracrine effects on vascular structure and function depending on the specific anatomical location (104). When investigating the paracrine role of adipose depots on local atherogenesis, it has been shown that the number of inflammatory mediators (IL-6, MCP-1, TNF-α, and IL-1β) secreted by adipose tissue locally does not correlate with plasma concentrations of these same cytokines, demonstrating the potential impact of PVAT on specific local microenvironments and tissue function (105). Of particular relevance to metabolic disease is the PVAT surrounding resistance arteries that directly communicates with proximal and distal muscle microvessels to regulate perfusion and glucose delivery to muscle. Postprandial insulin-mediated increases in vasodilation and microvascular blood volume are essential to systemic glucose disposal, and these endothelial effects of insulin have been shown to control 30–50% of muscle insulin sensitivity (97,106). Obesity is associated with impairment in this microvascular function, which affects insulin-mediated glucose disposal and leads to insulin resistance (107). Indeed, PVAT has been found to negatively correlate with insulin sensitivity in humans independently of age, sex, VAT, liver fat, BMI, and other cardiovascular risk factors (108). PVAT secretion of the insulin-sensitizing adipokine adiponectin is decreased in the muscle of obese db/db mice compared with lean mice, which is associated with decreased insulin-induced vasodilation and regulation in muscle (109). Additionally, the surgical removal of PVAT from healthy mouse hind limb muscle resulted in the failure of resistance arteries to dilate in response to insulin and blunted muscle glucose uptake in vivo, displaying the importance of paracrine signaling by intramuscular PVAT in the regulation of muscle insulin sensitivity (106).
In addition to skeletal muscle, it is important to consider local PVAT depots impacting other critical metabolic organs, such as the liver and pancreas, as both have complex vascular systems that likely contain PVAT. However, with the exception of mesenteric PVAT generally, studies specifically investigating the relationship between PVAT accumulation around vessels feeding and draining the liver (e.g., hepatic artery/vein and portal vein) or pancreas (e.g., splenic artery/vein and superior mesenteric artery/vein) and metabolic function are lacking. Considering the association between PVAT and metabolic disease, in addition to the anatomically specific and potent effects of local PVAT on neighboring tissues, investigations into the PVAT surrounding the vasculature of metabolic organs such as the liver and pancreas could be of substantial relevance to the development of type 2 diabetes and other metabolic disorders.
Expanding perspectives on adipose tissue composition and function have highlighted the centrality of this organ, not only to energy maintenance but also to regulating a multitude of homeostatic as well as disease processes. It seems that the more we zoom in on the diverse anatomy of adipose tissue, the more we realize the regional and local heterogeneity and the importance of considering each depot for its specific and discrete cellularity and secretome. What has been highlighted here is the significance of this anatomical specificity on adipose tissue cross talk with recipient tissues and organs that are involved in the development of metabolic disease. Until recently, the distinction between VAT and SAT depots dominated the attention of investigators interested in the association between adipose tissue and metabolic dysfunction. Now, it is clear that the link is much more complicated; within the visceral and subcutaneous adipose depots are subcompartments with unique and divergent associations with metabolic health, and a variety of other depots have come into focus for their roles in disease development, such as PAT, IMAT, and PVAT. As research on these specific adipose compartments is emerging, the gaps in our understanding of the specific cellular composition, secretome, and intricate paracrine relationships with neighboring tissues are apparent. We still do not have a comprehensive picture of the differences in secretome between human omental and mesenteric visceral compartments or deep and superficial subcutaneous tissue, and there are even larger holes in our knowledge of the secretory profiles of IMAT and PAT. Furthermore, while there are strong correlations between the specific adipose tissues discussed and systemic obesity or type 2 diabetes, the mechanistic relevance of their paracrine relationships that could help explain reported associations with metabolic disease are not well described. Connections that could be further investigated are between omental/mesenteric VAT and the intestines, deep subcutaneous adipose, and neighboring skeletal muscle, pancreatic adipose, and pancreatic islets, IMAT, and skeletal muscle as well as perivascular adipose associated with the hepatic and pancreatic vasculature. A more in-depth appreciation of the modification of tissue microenvironments by local adipose tissues may be an important future step in treating the diversity of clinical complications that are observed in metabolic disease.
Funding. This work was partially supported by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grants R01DK118149 to B.C.B. and F31DK126393 to D.E.K.
Duality of Interest. No potential conflicts of interest relevant to this article were reported.