Reduction of IL‐6 gene expression in human adipose tissue after sleeve gastrectomy surgery

Summary Objective There is increasing evidence that immune cell interactions in adipose tissue contribute to the development of metabolic dysfunction. Pro‐inflammatory cytokines have been shown to mediate insulin resistance, and the presence of macrophages is a salient feature in the development of obesity. The present study aimed to evaluate adipocyte size and macrophage activation in women before and 3 months after laparoscopic vertical sleeve gastrectomy (VSG). Methods Subcutaneous abdominal adipose tissue biopsies were obtained from women scheduled to undergo VSG. Histological evaluation of adipocytes and macrophages was performed as well as cytokine expression quantification before and after VSG‐induced weight loss. Results Weight loss following VSG resulted in a reduction in adipocyte size as well as a decrease in interleukin (IL)‐6 cytokine mRNA expression in subcutaneous adipose tissue. There was no change in the presence of crownlike structures after weight loss. Conclusions Early weight loss after VSG is associated with a reduction in adipocyte size and a decline in IL‐6 gene expression in local adipose tissue. Macrophage infiltration and crownlike density structures persist in adipose tissue from tissues impacted by excess body weight 3 months after VSG‐induced weight loss.

CNS in order to suppress food intake and limit adipocyte expansion. 11 Circulating levels of leptin increase as adipose tissue expands and as such positively correlates with the degree of obesity. 12 However, an eventual resistance to leptin develops, yet the pro-inflammatory effects of leptin persist and perpetuate a state of chronic low-grade inflammation. 3,8,13 IL-6 is a major inflammatory mediator that has been implicated in obesity in several studies. 14 Aside from its important immune function, it is a pleotropic cytokine that is secreted by various cell types including skeletal muscle and adipocytes, which can mediate distinct biologic functions depending on where it is secreted. 15 For instance, levels of IL-6 can increase systemically after exercise. However, the majority of exercise induced IL-6 arises from skeletal muscle rather than adipose tissue. 16 Indeed, it has been demonstrated that IL-6 released from skeletal muscle has beneficial effects on insulin resistance because of an increase in lipolysis and fat oxidation and an increase in glucagon like peptide-1 (GLP-1). 17 By contrast, IL-6 derived from adipose tissue in the context of obesity leads to hepatic insulin resistance via activation of suppressor of cytokine signalling-3 (SOCS3). 18,19 Levels of IL-6 acutely rise as part of an early response to inflammation or traumatic injuries, while chronic low-grade inflammation involving increased circulating levels of IL-6, and other cytokines such as TNFα and IL-1β, are a known driver of metabolic disease. 20 ATMs can comprise over half of the leukocytes present in adipose tissue. 3,21 One role of ATMs in tissues impacted by obesity is to participate in the clearance of dead adipocytes. However, ATM clusters are also seen during acute caloric restriction. 22 Thus, they may also play an important role in lipid clearance during weight loss. 23 Crownlike structures (CLSs) are formed by macrophages coalesced around dead hypertrophied adipocytes in inflamed adipose tissue, which are mainly seen in the context of obesity, and have been associated with increased insulin resistance. [24][25][26] This suggests that weight loss should lead to a reduction in inflammatory mediators, macrophages, and CLS. Numerous studies have shown that bariatric surgery reduces insulin resistance, cardiovascular disease, and mortality independent of weight loss, effects thought to be related to changes in gut hormones and a reduction in inflammation seen after bariatric surgery. 27 However, the finding that pro-inflammatory mediators decrease after bariatric surgery-induced weight loss has been inconsistent across multiple studies. [28][29][30] A few studies have shown a reduction in subcutaneous adipose tissue gene expression of IL-6 months after bariatric surgery. 31,32 After bariatric surgeryinduced weight loss, systemic IL-6 levels have been reported to be more variable: either reduced or unchanged. [33][34][35] Differences in type of bariatric surgery, tissue source, patient age, and/or follow-up duration may account for some of these inconsistencies. However, there is a limited amount of data regarding the early changes that occur in chronic low-grade inflammation after a sleeve gastrectomy.
Here, adipose tissue was evaluated from women with obesity before and early after vertical sleeve gastrectomy (VSG) to determine changes in parameters related to inflammation, such as inflammatory cytokine gene expression, systemic cytokines, ATM infiltration, and presence of CLS.

| Patient selection
This was a prospective study that included a total of 12 patients with obesity having a body mass index (BMI) above 35 kg m −2 who were previously scheduled to undergo laparoscopic VSG at the University of Chicago Center for the Surgical Treatment of Obesity, Chicago, IL (Table 1). Not all 12 patients returned for a post-VSG subcutaneous fat biopsy or laboratory assessment. Because of limited tissue availability, the number of patients included in each experiment is denoted in the respective figure legend and includes pre-VSG and post-VSG assessments of the respective patients. Only women were recruited in the study in order to ensure proper powering, since the majority of patients who obtained a VSG at the University of Chicago are women.
Patients were excluded from the study if they had pre-existing diabetes, were on drugs known to alter insulin sensitivity, or had untreated obstructive sleep apnoea, which is also known to alter insulin sensitivity. Patients who were selected for VSG met with a multidisciplinary team for assessment that included surgeons, anaesthesiologists, bariatric physicians, and dieticians. Demographic information, vital signs, bioimpedance information, and blood samples were collected before and 3 months after VSG. Bioimpedance was assessed using the Quantum X BIA analyser validated in this population as previously described. 36 Data included weight, height, and body fat mass values.

| Subcutaneous fat biopsy
For the subcutaneous abdominal fat biopsy, an area of 12 to 15 cm in diameter was prepped with betadine solution. Lidocaine field block was performed using 20 cm 3 of 1% lidocaine and anaesthetized for 5 to 10 minutes. An incision was made approximately 0.5 cm parallel to the waistline, and prefilled 60-cm 3 syringes containing 10-cm 3 saline solution were used to aspirate adipose tissue using up to six syringes with a 13-gauge 3-in. hypodermic needle, and approximately 8 to 12 mL of adipose tissue was obtained. Samples were removed from the syringes under sterile conditions and processed for further analysis as described below. Subcutaneous adipose tissue biopsies were performed on 10 paired patients 1 to 2 weeks before and 3 months after undergoing VSG.

| Histological analysis of adipose tissue
Subcutaneous adipose tissue was fixed with formalin and embedded in paraffin for serial sectioning. Adipocyte morphology was evaluated by immunohistochemistry (IHC) and immunofluorescence (IF) techniques using subcutaneous adipose tissue sections from six patients before and after VSG as previously described. 37 Stored 5-μm sections were stained with antibodies against perilipin (PLIN2, 1:100, Abcam), as both PLPN1 and PLPN2 surround the adipocyte membrane. 38 Appropriate secondary antibodies conjugated to a fluorophore

| Gene expression
Adipose tissue was finely minced, and samples were allowed to recover for 12 hours in a 37 C 5% CO 2 incubator in M199 media with 1nM of insulin and 40nM of dexamethasone as previously described for adipose tissue cell culture. 40  Master Mix (Agilent). PCRs were performed in triplicate in 96-well plates, The data were analysed with the ddCt method as previously described, 41 after normalization to an internal control, tyrosine 3monooxygenase/tryptophan 5-monooxygenase activation protein, zeta (YWHAZ), a routinely used housekeeping gene for studies in adipose tissue. 42 Levels of the YWHAZ transcript did not change after the intervention. The results were expressed as fold change for each target gene from baseline. Real-time PCR analyses were performed using 10 paired patients before and after surgery.

| Systemic cytokine quantification
Human cytokine quantification was performed using a Milliplex MAP kit for Human Cytokine/Chemokine magnetic bead panel, which measured five analytes in serum from nine paired patients before and 3 months after VSG. Six of the nine patients from this cohort also obtained a fat biopsy for mRNA analysis of adipose tissue. Serum samples were obtained in the morning after an overnight fast.
Analytes included in the kit were IL-6, IL-10, IL-β, and TNF-α. All samples were acquired on a Luminex 200 instrument (Millipore) and were performed in duplicate in 96-well plates. The serum concentration of leptin was determined by radioimmunoassay using the human leptin RIA kit (Millipore). The absorbance values were used to calculate the leptin concentration and were carried out in duplicate, and the average of the two values was used for data analysis.

| Statistical analysis
Fat processing for immunohistochemical studies was performed in six paired patients before and after VSG. Because of limited sample availability, only 10 paired patients were used for mRNA gene expression studies and nine paired patients for the serum cytokine assays.
Descriptive statistics were used to summarize demographic and clinical information. Linear regression was used to determine correlations between IL-6 levels and body fat using Pearson correlation coefficients. Analyses utilized paired t tests for comparison of the change in each study endpoint as a result of VSG-induced weight loss after 3 months. Data are presented in mean ± SD, and a value < .05 was considered statistically significant. Comparison and statistical tests were performed using GraphPad Prism (Version 8.2.1 for Windows 64-bit).

| RESULTS
The mean weight loss in women who underwent VSG was 21.51 ± 4.2 kg at 3 months (Table 2), which translated to a mean of 16.45 ± 2.94% of total body weight loss. Figure 1A Figure 1B). Figure 1C shows the mean adipocyte area averaged for each patient before and after VSG-induced weight loss at 3 months (n = 6). Figure 2 depicts the individual change in adipocyte area before and after VSG for each patient. There was a significant reduction in mean adipocyte area 3 months after VSG-induced weight loss (−915.7 ± 636.5μM, P = .017; Figure 1C).
Gene expression was analysed from mRNA using adipose tissue obtained from 10 patients who underwent subcutaneous fat biopsy before and 3 months after VSG. There was a statistically significant  Note. Values are presented as means ± standard deviation, n = 12. Abbreviation: BMI, body mass index.
F I G U R E 1 A, Adipocyte size. Immunofluorescence demonstrating adipocyte size from one patient, 1 week before vertical sleeve gastrectomy (VSG). B, Adipocytes 3 months after vertical sleeve gastrectomy surgery (red, antiperilipin surrounds the adipocyte; green, anti-Mac2 Ab staining for macrophages; scale bar, 100 μm; images taken at 40× magnification). C, Mean adipocyte area from individual patients 1-2 weeks before (black circles) and 3 months after (white circles) VSG (cell size quantified using CellProfiler), n = 6, *P < .05 F I G U R E 2 Adipocyte size of individual patients presented in size bins before and after vertical sleeve gastrectomy (VSG) induced weight loss (adipocyte area on the x-axis is presented in bins of 250 or 500 μm 2 as depicted in each figure; percent frequency in each size bin is presented on the y-axis; black bars, presurgery; white bars, postsurgery) F I G U R E 3 Adipose tissue cytokine gene expression. Expression of IL-6, IL-1β, TNF-α, and IL-10 in subcutaneous adipose tissue of subjects before (1-2 weeks) and 3 months after bariatric surgery (y-axis denotes relative gene expression expressed as fold change, n = 10, *P < .05; ns, not significant) before surgery compared with after VSG-induced weight loss at 3 months ( Figure 6B, representative field shown in Figure 6C).

| DISCUSSION
The present study showed that VSG results in significant weight loss and reduction of body fat mass 3 months after bariatric surgery.
Levels of the pro-inflammatory cytokine IL-6 decreased significantly at the mRNA level in subcutaneous abdominal adipose tissue from patients 3 months after VSG-induced weight loss. This is in line with other reports that show that IL-6 expression is associated with obesity. Furthermore, a positive correlation trend was observed between obese adipose tissue and mRNA IL-6 levels, which may suggest that adipose tissue is the source of IL-6 production in the state of obesity. CLSs have been associated with obesity and insulin resistance.
However, most of the studies that have investigated CLS have been performed in mice, and only a handful of studies have looked at CLS density after rapid weight loss in humans. In one study of patients who underwent RYGB-induced weight loss, there was a reduction in CLS in patients with type 2 diabetes but not in euglycaemic patients. 53 Interestingly, a study looking at postmenopausal women undergoing rapid weight loss by caloric restriction found an increase in CLS density in subcutaneous adipose tissue. 54 This study found that F I G U R E 4 Correlation between levels of IL-6 mRNA levels of female patients with obesity before vertical sleeve gastrectomy (VSG) and percent body fat (P = .069, R 2 = .354, 95% CI, −0.0549 to 0.891, n = 10) F I G U R E 5 Systemic cytokine and adipokine levels from patients 1-2 wk before vertical sleeve gastrectomy (VSG) and 3 mo after VSGinduced weight loss (***P < .001, *P < .05; ns, not significant, n = 9) there was no difference in CLS density from a mean of six patients before surgery compared with 3 months after VSG-induced weight loss. The finding that increased CLS density occurs in the setting of obesity supports that CLS formation may represent increased cell death from increased adipocyte hypertrophy. 24 In the setting of weight loss, it may represent increased macrophage accumulation because of increased lipolysis. 23 Alternatively, the development of CLS in obese adipose tissues may signify that once they are formed, they persist in adipose tissue even after weight loss. Taken together, these data suggest CLS accumulation is a dynamic process in humans.
Early weight loss after VSG is associated with a reduction in adipocyte size and a decline in IL-6 gene expression levels at the local adipose tissue level but an increase of IL-6 at the systemic level. The findings presented here suggest that while IL-6 produced by adipose tissue is associated with obesity, systemic levels of IL-6 may be associated with weight loss and improved insulin sensitivity early after VSG-induced weight loss. Macrophage infiltration and CLS density are similar in adipose tissue derived from patients with obesity after VSGinduced weight loss, indicating that ATMs may have a different function in the setting of weight loss.
The major limitation of this study was the small sample size; however, these findings add to the growing body of literature demonstrating that adipocyte cell hypertrophy is associated with obesity and that adipocyte hypertrophy decreases after VSG-induced weight loss. 55 Importantly, this study was conducted in mostly African American women, a currently understudied population. The contribution of inflammation or other factors may differ between patients because of various factors, including genetics. Given that obesity and type 2 diabetes affect non-Hispanic black women with higher prevalence, 56 inclusion of these patients in prospective studies is paramount to increase the understanding of mechanisms involved in metabolic dysregulation.
F I G U R E 6 A, Macrophage infiltration in subcutaneous adipose tissue from patients 1-2 weeks before and 3 months after vertical sleeve gastrectomy (VSG) (n = 6; ns, not significant). B, Crownlike structure (CLS) density from subcutaneous adipose tissue sections collected from patients 1-2 wk before and 3 mo after VSG (n = 6; ns, not significant). C, Representative field from patient showing CLS density 3 months after VSG (red, anti-perilipin surrounds the adipocyte; green, anti-Mac2 Ab staining for macrophages; scale bar: 80 μm; image taken at 20× magnification)