Regional Heterogeneity of Perivascular Adipose Tissue: Morphology, Origin, and Secretome

Perivascular adipose tissue (PVAT) is a unique fat depot with local and systemic impacts. PVATs are anatomically, developmentally, and functionally different from classical adipose tissues and they are also different from each other. PVAT adipocytes originate from different progenitors and precursors. They can produce and secrete a wide range of autocrine and paracrine factors, many of which are vasoactive modulators. In the context of obesity-associated low-grade inflammation, these phenotypic and functional differences become more evident. In this review, we focus on the recent findings of PVAT’s heterogeneity by comparing commonly studied adipose tissues around the thoracic aorta (tPVAT), abdominal aorta (aPVAT), and mesenteric artery (mPVAT). Distinct origins and developmental trajectory of PVAT adipocyte potentially contribute to regional heterogeneity. Regional differences also exist in ways how PVAT communicates with its neighboring vasculature by producing specific adipokines, vascular tone regulators, and extracellular vesicles in a given microenvironment. These insights may inspire new therapeutic strategies targeting the PVAT.


INTRODUCTION
Obesity is becoming a substantial public health concern since it gives rise to a wide range of disorders (Olshansky et al., 2005). Obesity prevalence has tripled since 1975 and it is now one of the WHO's priorities to cease the quick rise in obesity (Jaacks et al., 2019). The burgeoning global epidemic of obesity, consequently leading to type 2 diabetes, dyslipidemias, cardiovascular disease, and even some cancers will soon be devastating if actions are not taken (Alberti et al., 2009;Grundy, 2012;Scheen and Van Gaal, 2014). Obesity occurs due to excess energy intake, dramatic fat mass accumulation caused by increased adipocyte size (hypertrophy), or increased adipocyte number (hyperplasia), or both. Given the systemic impacts, obesity cannot be seen simply as a fat accumulation, and it goes hand-in-hand with many metabolic complications. People often experienced obesity as a chronic affliction characterized by low-grade inflammation. Despite the absence of the four cardinal signs of typical inflammation, the unresolved inflammation, entwined with fibrosis and defected angiogenesis-induced hypoxia, is a common pathway in the development of various cardiometabolic diseases.
Excessive fat accumulation can happen in various adipose depots, including a kind of unique adipose tissue known as perivascular adipose tissue (PVAT). PVAT surrounds most blood vessels except capillaries, pulmonary and cerebral blood vessels. PVAT plays wide-ranging physiological roles far beyond supporting connective tissue. It is now considered a metabolically active organ that regulates vascular function in both autocrine and paracrine fashions by producing various adipokines (Rajsheker et al., 2010;Van de Voorde et al., 2013). Over the decades, our understanding of PVAT biology has increased simultaneously with a rising prevalence of obesity and related metabolic complications. Not only does PVAT actively maintain vascular homeostasis, but it also markedly modulates various inflammation-related cardiovascular diseases. Due to the proximity of PVAT to the associated blood vessels, PAVT can be a unique target to improve arterial function in the setting of obesity.
PVATs are not only different from classical adipose tissues, they vary from location to location developmentally and functionally (Chang et al., 2020). Most intriguingly, PVAT's origin largely remains unknown, despite the development of cellular fate mapping and lineage tracing in other fat depots (Sanchez-Gurmaches et al., 2016;Guimaraes-Camboa et al., 2017;Schwalie et al., 2018;Merrick et al., 2019;Shao et al., 2019;Cattaneo et al., 2020). Numerous publications have presented us with a bulk of mixed information, given the fact that these publications always include more than one PVAT, which challenges readers outside of the field. The general information on the pathophysiological functions of PVAT has been the subject of many recent comprehensive reviews (Nosalski and Guzik, 2017;Hildebrand et al., 2018;Chang et al., 2020). In this review, we aim to form a conceptual picture of PVAT heterogeneity, mainly focusing on histology, developmental origin, and briefly on secretome.

Nomenclature of Perivascular Adipose Tissues
A PVAT is generally named after its adjacent vessel's name, such as pericoronary adipose tissue, referring to the fat tissue around the coronary artery. Over the years, various non-standard names were used for the same PVAT, which can be merely confusing or even misleading. For instance, "cardiac" PVAT surrounding the coronary artery or epicardial adipose tissue was abbreviated as either "C-PVAT" (Drosos et al., 2016), or "PVAT-CA" (Lu et al., 2017), or even "epi" for shorter (Mazurek et al., 2003). It is not uncommon that these names are mistakenly referred to, for example, in the case of pericardial and epicardial fat (Iacobellis, 2009). These facts reason the need to standardize nomenclature for PVATs. Some research articles have used a combined format to define these concepts (Tran et al., 2018;Ye et al., 2019). Throughout this review, we will use their traditional names according to their locations but with some specifications. In particular, we "standardized" these names by introducing a common name, "PVAT," following a lowercase letter, indicating where it is localized. For example, pericoronary adipose tissue is termed as "cPVAT," thoracic periaortic adipose tissue is abbreviated as "tPVAT," and abdominal periaortic adipose tissue as "aPVAT." Adipose tissue around the mesenteric artery is called "mPVAT." A comparison among the four most commonly studied PVATs is presented in Table 1 and Figure 1.

Multicolored Adipose Tissue
Adipose tissues are empirically color-coded as white (WAT), brown (BAT), and beige/brown-in-white (brite) adipose tissues. The color explicitly reflects the main adipocyte population and the amount of iron-containing mitochondria in each adipocyte. The more mitochondria, the darker the color is. In addition to major depots, there are other location-specific adipose tissues, including mammary "pink" adipose tissue and bone marrow adipose tissue, which is primarily red in young individuals and turns yellow during aging. It would undoubtedly be oversimplified if we classify adipose tissues based solely on the appearing color and neglect their complexity and heterogeneity. Classical inter-scapular BAT has been previously assumed to contain a homogeneous population of brown adipocytes. A recent study has revealed a new brown adipocyte subpopulation with low thermogenic activity coexists with the classical highthermogenic brown adipocytes (Song et al., 2020). More interestingly, these two brown adipocyte subpopulations may interconvert dynamically in response to the environmental temperature (Song et al., 2020). Using AdipoChaser mouse models, Wang et al. also demonstrated that "pink" mammary adipose tissue can undergo reversible dedifferentiation during lactation, where de-differentiated fibroblasts proliferate and redifferentiate into adipocytes upon weaning (Wang et al., 2018).
PVAT's color is not visually presented. Instead, it is determined by the subcellular composition and enrichment of specific genetic biomarkers. For example, tPVAT, characterized as brown adipose tissue, is labeled with highly expressed uncoupling protein-1 (UCP-1) and cell death activator CIDE-A (Cidea) (Tran et al., 2018). Conversely, aPVAT is seemingly white but contains a mixture of beige and white adipocytes (Padilla et al., 2013). Therefore, when we talk about the color of a PVAT, we mainly use generalized terminology to reflect their cellular characteristics and corresponding genetic profiles. Generally speaking, white PVAT adipocytes share a lot of similarities with their visceral counterparts, which are morphologically classified by the appearance of large unilocular lipid droplets, fewer mitochondria, and small cytoplasmic volumes. Brown PVAT adipocytes, on the other hand, are distinguished by multilocular lipid droplets and high density of mitochondria. In rodents, arteries including the mesenteric, carotid, and femoral are surrounded by WAT, while the thoracic aorta is encircled by BAT and the abdominal aorta by beige PVAT (a mixture of white and brown adipocytes) (Brown et al., 2014).
cells (Sarantopoulos et al., 2018). Understandably, PVAT is composed of adipocytes, nerves, and a stromal vascular fraction consisting of monocytes, endothelial cells, pericytes, macrophages, T cells, and mesenchymal stem cells (Chatterjee et al., 2009). Large vessels are separated from their PVATs by a layer of adventitia composed of elastic bundles, fibroblasts, nerves, and vasa vasorum. On the other hand, PVATs around small vessels and microvessels are seen as a seamless continuum from their associated blood vessels. Different PVATs are dominated by diverse adipocytes, indicating regional phenotypic heterogeneity in a given vasculature (Fitzgibbons et al., 2011).

Histology of Pericoronary Adipose Tissue
Broadly, cardiac fat depots include epicardial and pericardial adipose tissues. A big portion of epicardial fat surrounding the coronary arteries is called pericoronary adipose tissue (cPVAT) (Fernandez-Alfonso et al., 2017). Human cPVAT was firstly considered as a unique subtype of WAT since the cellular morphology and gene profile were more consistent with WAT than BAT (Chatterjee et al., 2009). Differentiated pericoronary adipocytes are irregularly shaped, smaller in size, and lower in lipid accumulation relative to subcutaneous and perirenal counterparts (Chatterjee et al., 2009). Meanwhile, human cPVAT exhibits an increased capacity to attract immune cells and induce angiogenesis, contributing to coronary atherosclerosis development (Chatterjee et al., 2009). Notably, these features were observed in differentiated preadipocytes in vitro, which might differ from in vivo physiological settings.
Evidence indeed showed cPVAT was composed of small unilocular adipocytes positively stained with brown fat marker  FIGURE 1 | Anatomical, morphological, and developmental differences of the perivascular adipose tissues (PVATs). The color differences reflect the different proportions of brown adipocytes in the adipose tissue. Adipose tissues around thoracic aorta (tPVAT), abdominal aorta (aPVAT), and mesenteric artery (mPVAT) have distinct fat cell populations that are derived from different adipocyte progenitors. These progenitors could be identified by highly expressed cell markers, such as platelet-derived growth factor receptor-α (PGDFRα), myogenic factor 5 (Myf5), smooth muscle protein 22-α (SM22α), and peroxisome proliferator-activated receptor-γ (PPARγ).
Frontiers in Pharmacology | www.frontiersin.org June 2021 | Volume 12 | Article 697720 UCP-1 (Chatterjee et al., 2013;. Sacks et al. further proposed that it would be more reasonable that cPVAT might closely resemble beige adipocytes Sacks and Symonds, 2013). Consistent expression of UCP-1 in cPVAT has been repeatedly reported since then (Chechi et al., 2017), but less expression of adiponectin, an anti-inflammatory adipokine was also reported (Numaguchi et al., 2019). Thus cPVAT can be classified as beige adipose tissue, with a large proportion of white adipocytes, which may contribute to coronary atherosclerosis development.

Histology of Thoracic Periaortic Adipose Tissue
Thoracic periaortic PVAT (tPVAT) expands from the aortic arch at the T4 vertebra to the diaphragm's aortic hiatus at the T10-T11 vertebrae. The comparison of adipocyte size, inflammation, and macrophage polarization indicated that tPVAT was close to subcutaneous adipose tissue (Fitzgibbons et al., 2011). It is now widely accepted that tPVAT is morphologically and functionally like BAT ( Figure 2A). Multilocular brown adipocytes and abundant mitochondria were observed under light or electron microscopy (Fitzgibbons et al., 2011;Padilla et al., 2013), which might contribute to the decreased atherosclerotic plaque burden (Fitzgibbons et al., 2011). In tPVAT, the adipocyte phenotypic switching via specific signaling pathways is evident in both whitening and browning directions. In a diet-induced obesity mouse model, tPVAT adipocytes become "whiter", with a primarily unilocular appearance and larger lipid (Galvez-Prieto et al., 2008;Sacks and Symonds, 2013). Aging together with miR-146b-3p downregulation could decrease browning (Pan et al., 2018). Reduced brown adipogenic differentiation of resident stromal cells via loss of peroxisome proliferatoractivated receptor-γ (PPARγ) coactivator-1 α (PGC-1α) was also observed in aged animals . Cold is a capable inducer for perivascular adipocyte to re-brown and to restore protective effects on metabolism and vascular function (Boa et al., 2017;Bussey et al., 2018). This relationship between tPVAT and adiposity, employed as a predictor for cardiovascular diseases, warrants further investigation in a clinical setting.

Histology of Abdominal Periaortic Adipose Tissue
Abdominal periaortic adipose tissue (aPVAT) is the adipose segments surrounding the abdominal aorta from below the diaphragm to the femoral bifurcation. As a continuum of thoracic aorta, the abdominal aorta is encircled by 4-to 10fold more adipose tissue than the adipose tissue around the aortic arch in rats (Henrichot et al., 2005). aPVAT is more prone to accumulation when exposed to a high-fat diet and shows significant adipocyte hyperplasia and hypertrophy (Henrichot et al., 2005). A comparison with tPVAT demonstrates that aPVAT contains a mixture of cells, with a predominant proportion of unilocular white adipocytes and a small portion of brown adipocytes (Police et al., 2009). Moreover, the average aPVAT adipocyte size is much bigger than that of tPVAT ( Figure 2B). Electron microscopy demonstrated that aPVAT was composed of white adipocytes with fewer mitochondria (Padilla et al., 2013). Upon a 4 month high-fat diet feeding, mouse aPVAT showed higher gene expression of inflammatory cytokines than tPVAT. Leptin coincided with the inflammatory cytokines in aPAVT, whereas UCP-1 was elevated in tPVAT but suppressed in aPVAT throughout dietinduced obesity (Police et al., 2009). Further studies provided evidence that only aPVAT-, but not tPVAT-conditioned medium could promote inflammatory cytokine MCP-1 generation and macrophage migration (Police et al., 2009). aPVAT was found to express a higher level of inflammatory receptors, including CD11c, IL-6R, and TNFR1/2 relative to tPVAT and interscapular brown adipose tissues (Padilla et al., 2013). These convincing pieces of evidence differentiate two segments of PVAT, although they wrap along the same artery. These intriguing features remain enigmatic due to the lack of solid advances in the developmental origin, transdifferentiation, and cellular fate mapping of PVAT.

Histology of Mesenteric Perivascular Adipose Tissue
In humans and mice, the fat tissue around the resistance mesenteric arteries forms mesenteric perivascular adipose tissue (mPVAT) ( Figure 2C). Among adipose tissues collected from different locations, mPVAT was traditionally categorized as FIGURE 2 | Histological characteristics of the perivascular adipose tissues. Murine adipose tissues around thoracic aorta (tPVAT, panel (A)), abdominal aorta (aPVAT, panel (B)), and mesenteric artery (mPVAT, panel (C)) were stained by hematoxylin and eosin. Brown adipocytes in tPVAT are characterized by small-sized, multilocular lipid droplets, whereas white adipocytes in mPVAT are classified by the appearance of a larger unilocular lipid droplet and small cytoplasmic space. aPVAT has mixed adipocyte populations. Scale bar, 100 µm.

DIFFERENT ORIGIN OF PERIVASCULAR ADIPOSE TISSUES
Typical mature adipocytes originate from progenitor cells, which are committed preadipocytes derived from stem cells of multiple sources (Rodeheffer et al., 2008). The distinction between WAT and BAT reasonably leads to a simple classification of two precursor populations, giving rise to white and brown adipocytes, respectively. A commonly known classifier is myf5, which encodes myogenic factor 5 (Myf5) (Timmons et al., 2007), whereas PRDM16 may be a controller for brown adipocyte generation (Seale et al., 2008). However, using mTmG reporter, a study revealed that white adipocytes in the subcutaneous and retroperitoneal WAT also originate from Myf5-expressing precursors. This approach also demonstrated that Myf5 expression could not track many brown adipocytes (Sanchez-Gurmaches and Guertin, 2014a). The adipogenic capacity of vasculature-residing mural cells (e.g., pericytes) in the adipose tissue was well documented in many papers (Tang et al., 2008;Gupta et al., 2012;Tran et al., 2012;Berry and Rodeheffer, 2013). This statement is challenged by a study where pericytes do not contribute to adipocytes' generation, although they seem to act as progenitors in vitro (Guimaraes-Camboa et al., 2017). Merrick et al. further indicated that reticular interstitium rather than the vasculature is the residing site for interstitial progenitor cells, which give rise to the preadipocytes expressing intercellular adhesion molecule-1 and another group of cells expressing protein CD142 (Merrick et al., 2019). Most recent results further revealed that only fibroblasts, neither mural nor endothelial cells, are cells of the vascular wall with significant adipogenic potential in vivo in both WAT and BAT (Cattaneo et al., 2020). Thus developmental origins of adipose tissue and the mechanisms controlling its expansion are just beginning and more intriguing findings are expected to come soon. Adipocytes of different PVATs may originate from distinct precursors (Hepler et al., 2017;Tran et al., 2018). Despite the aforementioned development of cellular fate mapping and lineage tracing in other adipose depots, the origins of PVAT adipocytes, in general, are barely known. This session presents distinct adipocyte development in Table 1 and Figure 1. A clear definition of adipocyte origin can reveal the determined precursors and the regulatory mechanisms.

Origin of Epicardial and Periaortic Arch Perivascular Adipose Tissue
Epicardial fat originates from the splanchnic mesoderm in human  and in mouse (Walden et al., 2012). Periaortic adipose tissue is potentially derived from Myf5 + progenitors (Sanchez-Gurmaches and Guertin, 2014a). A more recent study showed that periaortic adipocyte progenitors expressed smooth muscle protein 22-alpha (SM22α) during development. Besides, knockout of PPARγ in neural crest cells leads to developmental delay of the periaortic arch PVAT . This evidence indicates that periaortic arch PVAT adipocytes have multiple lineages, mainly from ectoderm-derived neural crest cells, rather than the mesoderm-derived Myf5 + progenitors . A possible explanation is that ectoderm-derived neural crest cells have a broad differentiation potential and give rise to a diverse range of cell types. For example, neural crest cells were once identified as one of the progenitors of white adipocytes (Sanchez-Gurmaches and Guertin, 2014b). Another study revealed that neural crest-derived cells resided along the vessels within the subcutaneous adipose tissue. These results demonstrate that neural crest-derived adipocyte-committed progenitors contribute to adipogenesis (Sowa et al., 2013).

Origin of Abdominal Periaortic Adipose Tissue
Abdominal PVAT preadipocytes demonstrate lower brown adipocyte developmental transcription factors relative to tPVAT (Tran et al., 2018). Abdominal PVAT lacks Zic1 gene, encoding zinc finger proteins that are critical for early BAT development (Contreras et al., 2016). In the absence of the adipogenic transcription factor PPARγ in the VSMCs, failure of aPVAT development was observed (Chang et al., 2012). These findings demonstrate that aPVAT shares, at least to some extent, similar developmental origins with SM22α + and PPARγ + VSMCs.

Origin of Mesenteric Perivascular Adipose Tissue
Few studies have focused on the developmental origin of mPVAT, and some comparative data may provide limited clues. Given the proximity, mPVAT's developmental origins are thought to be similar to the visceral adipose tissue. Indeed, both mPVAT and perigonadal adipose tissue expressed comparable levels of white adipocyte signature gene Tcf21 and the brite adipocyte-specific genes Tbx1 and Tmem26 (Contreras et al., 2016). Mesenteric and perigonadal adipose tissue are also found to share the same mesothelial origin in lineage tracing experiments, and preadipocytes of mPVAT have a transcriptional profile closer to that of subcutaneous, but not omental preadipocytes (Chau et al., 2014). Also, in this cell lineage analysis, 28% of mature mesenteric adipocytes were Wt1+ positive, suggesting the source of mesenchymal stem cells in mesenteric adipocytes is different from that of BAT, where Wt1 expression is undetectable (Chau et al., 2014). Chang et al. have suggested that mPVAT shares a developmental origin with VSMCs because the deletion of the PPAR γ in VSMCs resulted in a dramatic loss of mPVAT (Chang et al., 2012).

DIFFERENT SECRETOME OF PERIVASCULAR ADIPOSE TISSUES
Adipose tissue is capable to synthesize and secrete various substances just as endocrine cells do. In this sense, adipose tissue is the largest endocrine organ, and more than 600 hundred identified factors produced by adipocytes are collectively termed adipokines or adipocytokines (Halberg et al., 2008;Lehr et al., 2012). Paracrine crosstalk between PVAT and its neighboring artery, also known as "vasocrine" communication, actively regulates vascular inflammation and arterial remodeling (Yudkin et al., 2005). Anatomically distinct PVAT depots can release a quite different range of adipokines. Previous literatures have deeply explored and frequently revisited this topic (Omar et al., 2014;Owen et al., 2014;Akoumianakis et al., 2017;Nosalski and Guzik, 2017;Xia and Li, 2017;Oikonomou and Antoniades, 2019;Chang et al., 2020;Kim et al., 2020). We briefly summarized PVAT-derived adipokines, vascular tone regulators, and newly discovered exosomes/extracellular vesicles in this part. Unlike previous reviews, these factors are categorized by specific PVAT depots where detectable levels are reported (e.g. PCR, Western-blot, or immunostaining).

Pro-and Anti-Inflammatory Adipokines of Perivascular Adipose Tissues
Most PVAT-generated cytokines/chemokines such as tumor necrosis factor (TNF)-α, interleukin-6 (IL-6), plasminogen activator inhibitor-1 (PAI-1), and monocyte chemoattractant protein-1 (MCP-1), are pro-inflammatory and proatherosclerotic. Adiponectin is one of the few antiinflammatory factors that possess multiple salutary effects for cardiovascular disease prevention (Xu and Vanhoutte, 2012). Phenotypic differences between tPVAT and aPVAT are evident that tPVAT generates much less pro-inflammatory cytokines, and is thus resistant to diet-induced inflammation (Police et al., 2009;Fitzgibbons et al., 2011). mPVAT and aPVAT are similarly prone to the expansion of adipocytes and dietinduced inflammation . However, mPVAT is more sensitive to the high-fat diet challenge, where the adipose "browning" genes are dramatically down-regulated (Hou et al., 2016). Previous reviews in this field have elaborated on PVATderived adipokines and their interactions with other vascular cells (Brown et al., 2014;Gil-Ortega et al., 2015;Akoumianakis et al., 2017;Kim et al., 2019;Chang et al., 2020). The PVAT-generated representative pro-and anti-inflammatory factors are briefly updated in Table 2. Note that one type of adipokine can be generated from multiple sites and at the same time one PVAT can release a variety of adipokines. Members of the same adipokine family (i.e., pro-inflammatory or constricting) all share the same predicted vascular function.

Perivascular Adipose Tissue-Derived Extracellular Vesicle
Inter-cell and inter-organ signaling within PVAT remain a mystery.
Recently, many studies, including our own, have delved deep to identify the messengers conveying the communication between a blood vessel and its surrounding PVAT . Apart from the secretory cytokines and chemokines factors, adipocytes also secrete many types of extracellular vesicles (EVs) (Deng et al., 2009;Ogawa et al., 2010), typically including exosomes and microvesicles. EVs play important roles in intercellular communication by selective packaging of lipids, proteins, and microRNAs (miRNAs) (Valadi et al., 2007;van Niel et al., 2018). Adipose tissue was proved to constitute an essential source of circulating exosomal miRNAs, as a form of adipokine that acts locally or distantly (Thomou et al., 2017). In addition, these EVs could be taken up by neighboring or distant cells to modulate these recipient's functions (Bang et al., 2014;Sun et al., 2016). For instance, adipocyte-derived exosomal miRNAs enable the metabolic regulation of neighboring macrophages (Ogawa et al., 2010;Zhang et al., 2016). Vice versa, macrophages can secrete miRNA-containing exosomes to modulate local adipocyte function (Ying et al., 2017). Similarly, the endothelial-adipocyte interplay was the result of EV-mediated reciprocal trafficking of caveolin 1 (Crewe et al., 2018). These observations led us to hypothesize that whether PVATadipocytes secrete exosomal miRNAs, if so, how they regulate vascular function in the context of obesity. Indeed, we have demonstrated that perivascular adipocytes produce and secrete miRNA-containing EVs, which can be taken up in neighboring VSMCs . One of the most enriched miRNAs in PVAT and PVAT-derived EVs, miR-221-3p, is transported into adjacent VSMCs. The study further provided an EV-miR-221-3pmediated mechanism by which PVAT triggers an early vascular remodeling in vascular inflammation . In another study, increased miR-221/222 in the arteries promoted neointimal hyperplasia in the femoral artery following wire injury (Lightell et al., 2018).

CONCLUSION
Anatomically distinct PVATs vary in developmental origin, cellular composition, and secretome. The farther away a PVAT is from the heart, the more white-like the adipocytes are. This is true even in the same stem aorta but wrapped with distinct PVATs in the chest cage and the abdominal cavity. PVAT progenitor cells include but may not be limited to mural cells (pericyte or smooth muscle cells) and fibroblasts. These potential progenitors give rise to committed preadipocytes and contribute to adipogenesis. Adipogenesis and angiogenesis appear to coexist, and preadipocytes and pericytes may co-develop as well, which should be further studied. The developmental trajectory of PVAT adipocytes is somehow still a "bloody mess" (Rosen and Spiegelman, 2014).
Adipocytes and their neighboring vascular cells constitute a perivascular microenvironment. In an inflammatory setting, a family of intercellular message-conveying machinery is involved in these cells' interplay. Perivascular adipocytes, partially via alteration of their secretome, modulate the nearby VSMCs (Miao and Li, 2012) and endothelial cells (Sena et al., 2017). For obesity, secretion of anti-inflammatory adiponectin is markedly reduced, whereas the generation of proinflammatory cytokines is dramatically elevated. Besides, the contiguity with adventitia makes it plausible that paracrine or vasocrine crosstalks between PVAT and the encircled blood vessel are reciprocal and bidirectional.
The continuing worldwide upsurge in obesity prompts us to unveil PVAT's significant role in vascular function. A better understanding of regional heterogeneity among PVATs is just a start. PVAT's function and dysfunction in vascular homeostasis and cardiovascular pathogeny remain our long-term tasks to pursue. Developmental fate mapping is an essential technique for answering some of these questions. Single-cell techniques have empowered this process by helping draw sophisticated cellular atlas. Some researchers have explored the heterogeneity of PVAT at a single-cell level and uncovered distinct clusters with specific signature markers and signaling pathways (Angueira et al., 2021).
Creating a more precise map of such a complex tissue from single-cell sequencing data is, therefore, a challenging task, which on the other hand, opens an opportunity for us to dive into this unknown.

AUTHOR CONTRIBUTIONS
XL and YZ conceived the design and concepts. XL and YZ wrote the manuscript. ZM contributed key information for tables and figures. All authors contributed to the article and approved the submitted version.