Defining the vascular niche of human adipose tissue across metabolic conditions

White adipose tissue requires a well-maintained vascular network to function properly. Although advances in single-cell transcriptomics have allowed the development of comprehensive human white adipose tissue atlases, there has been a little focus on deciphering the heterogeneity and the functional states of adipose vascular cells, including blood adipose endothelial cells (AdECs), adipose lymphatic endothelial cells (AdLECs), and mural cells (pericytes and vascular smooth muscle cells). Using single nucleus RNA sequencing (snRNA-seq), we mapped human subcutaneous adipose tissue cellular clusters, and focused on cells comprising the vascular niche. We further integrated our in-house generated cohort with seven publicly available datasets of human SAT single cell RNA-seq (scRNA-seq) and snRNA-seq. The generated atlas comprised nearly 70,000 vascular cells from 65 donors, which allowed us to explore the adipose vascular niche. By mining this rich resource, we identify eight endothelial populations, comprising seven canonical subtypes and one additional heterogeneous population of sub-endothelial AdECs. We also provide detailed analyses comparing the transcriptomes of vascular cells from individuals living with or without obesity and type 2 diabetes. These data provide an extensive atlas for the exploration of the adipose vascular niche and its modulation of adipose tissue biology in health and disease. Overall design: Human deep subcutaneous adipose tissue samples were obtained from lean donors undergoing hemicolectomy, laparoscopic transabdominal preperitoneal approach, fundoplication with mesh hiatoplasty, or band-reinforced gastric bypass and obese donors undergoing elective cosmetic abdominoplasty, gastric sleeve resection, or gastric bypass surgeries at the Clinic for General, Visceral, and Pediatric Surgery of the University Medical Center Göttingen in Göttingen, Germany. Given the wide range of obesity and type 2 diabetes phenotypes, we used all the available clinical parameters to classify the donors. Obesity was defined based on body mass index (BMI) (kg/m2): Lean donors (BMI < 30) and donors with obesity (BMI > 30). Donors were classified as having diabetes based on the presence of a previous history of type 2 diabetes and antidiabetic medication. This yielded three groups in total matched for age and BMI; lean (2 females/ 2 males, age: 45.75±13.27, BMI: 23.27±0.72), obese (3 females/ 4 males, age: 38±5.78, BMI: 53.22±8.94), and obese diabetic (2 females/ 1 male, age: 42±6.16, BMI: 51.3±6.16). The human samples used in the present study were obtained based on the presence of a signed consent according to the University Medical Center Göttingen institutional review board ethical approval (38/4/21).