Increased adipose tissue expression of IL‐18R and its ligand IL‐18 associates with inflammation and insulin resistance in obesity

Abstract Introduction The proinflammatory cytokine IL‐18 is involved in the pathogenesis of metabolic syndrome. While the changes in IL‐18 are known, IL‐18R expression and relationship with IL‐18 and other inflammatory markers in the adipose tissue in obesity/type‐2 diabetes (T2D) remain unclear. Methods We, therefore, determined the adipose tissue expression of IL‐18R and IL‐18 mRNA/protein in lean, overweight, and obese individuals with and without T2D, 15 each, using qRT‐PCR, immunohistochemistry, and confocal microscopy. Data (mean ± SEM) were analyzed using unpaired t‐test and Pearson's correlation (r); all P values ≤0.05 were considered statistically significant. Results We found the upregulated gene/protein expression of IL‐18R and IL‐18 in non‐diabetic obese/overweight as compared with lean individuals (P < 0.05). BMI correlated positively (P < 0.05) with the adipose tissue expression of IL‐18R (mRNA: r = 0.90 protein: r = 0.84) and IL‐18 (mRNA: r = 0.84 protein: r = 0.80). Similarly, in T2D individuals, gene and protein expression of IL‐18R/IL‐18 was significantly higher in obese as compared with overweight/lean individuals. The BMI was associated with the changes in both IL‐18R (mRNA: r = 0.55 protein: r = 0.50) and IL‐18 (mRNA: r = 0.53 protein: r = 0.57) expression. IL‐18R/IL‐18 gene expression in the adipose tissue was positively associated (P < 0.05) with local gene expression of other inflammatory markers including CD11c, CD86, CD68, CD163, TNF‐α, and CCL5. Homeostatic model assessment of insulin resistance (HOMA‐IR) was higher in diabetic/non‐diabetic obese and it correlated with BMI (P < 0.05). IL‐18R and IL‐18 mRNA/protein expression in obesity was associated with HOMA‐IR only in non‐diabetics. Conclusions The adipose tissue IL‐18R/IL‐18 expression is enhanced in obesity which associates with proinflammatory gene signature and insulin resistance in these individuals.


Introduction
Obesity is the leading cause of chronic low grade inflammation which results in insulin resistance and development of type-2 diabetes (T2D) and its associated metabolic complications. The changes in the expression of proinflammatory cytokines and the increased numbers of various inflammatory cell populations in the expanding adipose tissue play a critical role in the induction of metabolic inflammation. The most common cytokines/ chemokines that are associated with metabolic inflammation include tumor necrosis factor (TNF)-a, interleukin (IL)-1b, IL-6, interferon (IFN)-g, C-C chemokine ligand (CCL)-2 or macrophage chemoattractant protein (MCP)-1, C-X-C chemokine ligand (CXCL)-8/IL-8, and CCL-5 or regulated on activation normal T-cell expressed and secreted (RANTES) [1]. IL-18, also called IFNg-inducing factor, is a unique proinflammatory cytokine that belongs to the IL-1 family of cytokines and is constitutively expressed and secreted by monocytes/macrophages, Kupffer cells, dendritic cells, as well as other cell types such as epithelial/endothelial cells, and vascular smooth muscle cells [2]. IL-18 is an 18 kDa glycoprotein derived by caspase-1-mediated cleavage of its 23 kDa precursor called pro-IL18 [3]. IL-18 plays immunoregulatory role via the induction of IFN-g which exerts changes in a variety of cell types and is associated with the pathogenesis of several inflammatory diseases such as rheumatoid arthritis, Crohn's disease, and T2D. In regard with metabolic syndrome, IL-18 is associated with obesity, insulin resistance, hypertension, atherosclerotic lesions, dyslipidemia, and cardiovascular disease [4].
IL-18 mediates its bioactivities through a heterodimeric IL-18 receptor (IL-18R) comprising of aand b-chains that are expressed on a wide variety of cells including macrophages, naive T/B lymphocytes, natural killer (NK) cells, neutrophils, endothelial cells, smooth muscle cells, and chondrocytes [5]. IL-18Ra binds IL-18 with low affinity while IL-18Rb does not bind the ligand directly but rather binds with high affinity to IL-18/IL-18Ra complex and induces intracellular signal transduction including the recruitment and activation of myeloid differentiation factor 88 (MyD88) and IL-1R-associated kinase (IRAK) to the receptor complex [6,7]. The downstream activation promotes the maturation of T-lymphocytes and natural killer cells as well as the expression of IFN-g, other proinflammatory cytokines, chemokines, cell adhesion molecules, and matrix metalloproteinases [8,9].
Whereas, IL-18 expression in metabolic disease was investigated [10][11][12], modulations in the expression of IL-18R with regard to IL-18 and other inflammatory markers in the adipose tissue in obesity and/or T2D still remain unclear. Therefore, in this study, to further extend the knowledge and highlight importance in metabolic inflammation, we determined simultaneous changes in the receptor and its ligand and, herein, our data show upregulated adipose tissue expression of IL-18R and IL-18 in obese diabetic/nondiabetic individuals. This altered expression was found to be associated positively with monocyte/macrophage markers and expression of TNF-a and CCL-5. These changes also correlated with homeostatic model assessment of insulin resistance (HOMA-IR) index.

Study population
A total of 15 T2D (eight male and seven female, aged 45-64 years) and 15 non-diabetic (four male and 11 female, aged 24-60 years) individuals were recruited in the study through the clinics of Dasman Diabetes Institute (DDI), Kuwait. Those of age <18 years or with serious comorbid conditions including diseases of the lung, kidney, heart, and liver, or those with hematologic or immune disorders, pregnancy, type-1 diabetes, or malignancy were excluded. The participants were subclassified as lean, overweight, and obese based on body mass index (BMI). The T2D group comprised of one lean (BMI ¼ 23.54 kg/m 2 ); seven overweight (BMI ¼ 28.0 AE 0.5 kg/m 2 ); and seven obese (BMI ¼ 35.0 AE 0.5 kg/m 2 ) individuals. The non-diabetic group comprised of five lean (BMI ¼ 22.7 AE 1.3 kg/m 2 ); five overweight (BMI ¼ 28.3 AE 0.6 kg/m 2 ); and five obese (BMI ¼ 32.5 AE 1.3 kg/m 2 ) individuals. T2D patients had hypertension and hyperlipidemia, four each, while one non-diabetic individual had hypertension. The clinicodemographic data of the study participants are summarized in Table 1.

Ethics statement
Study protocol (RA-2010-003) was approved by the institutional ethics committee (Ethical Review Committee of Dasman Diabetes Institute, Kuwait). Written informed consent was obtained from all the participants for inclusion in this study.

Anthropometric and physio-clinical measurements
Height and weight were measured using calibrated portable electronic weighing scales and portable inflexible height measuring bars; the waist circumference was measured using constant tension tape. The whole body composition including body fat percentage, soft lean mass, and total body water were measured using IOI353 Body Composition Analyzer (Jawon Medical, South Korea). Blood pressure was measured by using Omron HEM-907XL digital automatic sphygmomanometer (Omron Healthcare Inc. Lake Forest, CA, USA). BMI was calculated by using standard formula as follows: BMI ¼ body weight (kg)/height (m 2 ).
Peripheral blood was collected from overnight-fasted individuals and analyzed for fasting glucose, glycated hemoglobin (HbA1c), fasting insulin, and lipid profile. Glucose and lipid profiles were measured using Siemens dimension RXL chemistry analyzer (Diamond Diagnostics, Holliston, MA). HbA1c was measured by using Variant TM device (BioRad, Hercules, CA). Plasma triglycerides were also measured using commercial kit (Chema Diagnostica, Monsano, Italy). All assays were carried out following instructions and guidelines from the manufacturers. HOMA-IR was calculated by using standard formula as follows: HOMA-IR ¼ [Glucose Â Insulin]/22.5 whereas, fasting plasma glucose was measured in molar units (mmol/L) and fasting insulin was measured as mIU/L.

Collection of subcutaneous adipose tissue samples
Human adipose tissue samples ($0.5 g) were collected by abdominal subcutaneous fat pad biopsy lateral to the umbilicus using standard surgical method. The biopsy tissue was further incised into small pieces, rinsed in cold phosphate buffered saline (PBS), fixed in 4% paraformaldehyde for 24 h and embedded in paraffin. At the same time, freshly collected adipose tissue samples ($50-100 mg) were preserved in RNAlater or embedded in optimal cutting temperature (OCT) and stored at À808C until use.

Real-time reverse-transcription polymerase chain reaction (RT-PCR)
Total cellular RNA was purified using RNeasy kit (Qiagen, Valencia, CA) as per manufacturer's instructions. RNA samples (1 mg each) were reverse transcribed to yield cDNA using random hexamer primers and TaqMan reverse transcription reagents (High Capacity cDNA Reverse Transcription kit; Applied Biosystems, Foster City, CA, USA). For real-time RT-PCR, cDNA (50 ng) was amplified using TaqMan . Each cycle comprised of denaturation for 15 sec at 958C, annealing/extension for 1 min at 608C which started after uracil DNA glycosylase (UDG) activation (508C for 2 min) and AmpliTaq Gold (Abcam, Cambridge, MA, USA) enzyme activation (958C for 10 min). GAPDH expression was used for internal control to normalize the differences in individual samples and expression levels of test genes relative to control gene (lean adipose tissue) were calculated using À2 DDCt method. Relative mRNA expression was measured as fold change over average of control gene expression taken as one and data were expressed as mean AE SEM values.

Immunohistochemistry
Paraffin-embedded sections of subcutaneous adipose tissue (4 mm) were deparaffinized in xylene and rehydrated through descending grades of ethanol (100%, 95%, and 75%) to water. Antigen was retrieved by using target retrieval solution (pH6.0; Dako, Glostrup, Denmark) under pressure cooker samples were incubated for 1 h with secondary HRPconjugated antibody and color was developed with 3,3 0diaminobenzidine (DAB) chromogenic substrate. Specimens were washed in running tap water, lightly counterstained with Harris hematoxylin, dehydrated through ascending grades of ethanol (75%, 95%, and 100%), cleared in xylene, and finally mounted in dibutyl phthalate xylene (DPX). For protein expression quantification and data analysis, digital photomicrographs of entire sections (100Â, PannoramicScan, 3DHISTECH, Budapest, Hungary) were used to quantify staining in the three selected regions outlined using Aperio ImageScope software (Aperio Vista, CA) and the regional heterogeneity was assessed in samples. Aperio-positive pixel count algorithm version 9 was used to quantify staining intensity in tissue samples.

Confocal microscopy
Formalin-fixed, paraffin-embedded adipose tissue samples were stained for immunofluorescent microscopy. After antigen retrieval and blocking as described before, colocalization staining of samples was performed by incubating at room temperature for 1 h with adipocyte marker antibody (mouse anti-human adiponectin 1:400 diluted antibody, Abcam 1 ab22554) followed by overnight incubation with primary antibodies (rabbit anti-human 1:100 diluted IL-18R1 antibody, Abcam 1 ab124458 and rabbit anti-

Statistical analysis
The data (mean AE SEM) were analyzed using GraphPad Prism software version 7.0 (La Jolla, CA). Unpaired Student t-test was used to compare means between groups and Pearson's correlation (r) was used to assess dependence between variables. All P values 0.05 were considered as significant.

Increased adipose tissue expression of IL-18R in obesity
To test the hypothesis whether the IL-18R expression was upregulated in the adipose tissue of obese non-diabetic/T2D individuals, we determined the expression of IL-18R at both mRNA and protein levels in lean, overweight and obese adipose tissue samples from non-diabetic and diabetic subjects. The data show that in non-diabetic subjects, IL-18R adipose tissue mRNA expression was significantly upregulated in obese and overweight as compared with lean individuals (Obese: 3.7 AE 0.6 Overweight: 2.4 AE 0.5 Lean: 1.0 AE 0.02; P < 0.05) (Fig. 1A).
Similarly, IL-18R protein expression in the adipose tissue was also significantly elevated in obese and overweight as compared with lean individuals (Obese: 852 AE 35 Overweight: 673 AE 91 Lean: 377 AE 23; P < 0.05) (Fig. 1B). The BMI correlated positively with both IL-18R mRNA (r ¼ 0.90 P < 0.0001) (Fig. 1C) and IL-18R protein (r ¼ 0.84 P ¼ 0.0002) (Fig. 1D) expression. The representative immunohistochemistry images from three independent determinations depicting IL-18R expression in the adipose tissue samples from non-diabetic lean, overweight, and obese individual, three donors each, are shown in Figure 2A. The changes in IL-18R protein expression in the adipose tissue were further confirmed by confocal microscopy as shown in Figure 2B. As expected, in T2D patients as well, IL-18R adipose tissue mRNA expression was significantly upregulated in obese as compared with lean/overweight individuals (Obese: 7.0 AE 0.8 Lean/Overweight: 2.4 AE 0.5; P < 0.05) (Fig. 3A). Similarly, IL-18R protein expression was also elevated in obese as compared with lean/overweight individuals (Obese: 742 AE 50 Lean/ Overweight: 431 AE 79; P < 0.05) (Fig. 3B). The BMI tended to associate positively with IL-18R mRNA (r ¼ 0.55 P ¼ 0.08) (Fig. 3C) and IL-18R protein (r ¼ 0.50 P ¼ 0.06) (Fig. 3D) expression in the adipose tissue. The representative immunohistochemistry images from three independent determinations of IL-18R expression in the adipose tissue samples from diabetic lean (one donor), overweight (three donors), and obese (three donors) are shown in Figure 4A. The changes in IL-18R protein expression in the adipose tissue samples from T2D patients were further confirmed by confocal microscopy as shown in Figure 4B.

Elevated adipose tissue expression of cognate ligand IL-18 in obesity
We further asked if the adipose tissue expression of the cognate ligand IL-18 was also increased in obese diabetic/nondiabetic subjects. The data show that in non-diabetic subjects, the adipose tissue IL-18 mRNA expression was significantly upregulated in obese and overweight subjects as compared with lean controls (Obese: 9.5 AE 1.2 Overweight: 5.9 AE 0.9 Lean: 2.1 AE 0.3; P < 0.05) (Fig. 5A). Likewise, IL-18 protein expression in the adipose tissue was also significantly elevated in obese and overweight as compared with lean individuals (Obese: 841 AE 38 Overweight: 520 AE 66 Lean: 360 AE 27; P < 0.05) (Fig. 5B). The BMI correlated positively with both IL-18 mRNA (r ¼ 0.84 P ¼ 0.0001) (Fig. 5C) and IL-18 protein (r ¼ 0.80 P ¼ 0.0004) (Fig. 5D) expression. The representative immunohistochemistry images from three independent determinations of IL-18 expression in the adipose tissue samples from non-diabetic lean, overweight, and obese individuals, three each, are shown in Figure 6A. The changes in IL-18 protein expression in the adipose tissue were also confirmed by confocal microscopy as shown in Figure 6B.
IL-18R/IL-18 adipose tissue expression changes relate mainly to immune cells Next, we wanted to know whether IL-18R/IL-18 expression changes in obesity were shown mainly by adipocytes or immune cells. To address this question, we performed co-localization studies on adipose tissue samples from nondiabetic subjects using confocal microscopy. For IL-18R colocalization study, we used Alexa Fluor (AF)-488 conjugated anti-IL-18R antibody (green fluorescence) with AF647conjugated adiponectin (adipocyte marker) antibody (red fluorescence). Conversely, for IL-18 co-localization study, we used AF647-conjugated anti-IL-18 antibody (red fluorescence) with AF488-conjugated adiponectin marker antibody (green fluorescence). The data show that both IL-18R (Fig. 9) and IL-18 ( Fig. 10) were expressed mainly by the immune cells around adipocytes, called as crown-like structures (CLS; also visible in the immunohistochemistry images).

Discussion
In this study, for the first time to our knowledge, we show that IL-18R mRNA/protein expression in the adipose tissue samples from diabetic/non-diabetic obese and overweight individuals was significantly increased than their lean counterparts. Our data further show that the adipose tissue gene/protein expression of the proinflammatory cytokine IL-18 was also significantly higher in diabetic/non-diabetic obese and overweight subjects as compared with respective lean controls. IL-18 is produced constitutively by many cell types including macrophages, dendritic cells, endothelial/ epithelial cells, and vascular smooth muscle cells [13,14]. All these cell types, especially macrophages, are found in elevated numbers in the expanding adipose tissue in obesity/T2D and a polarization shift from the antiinflammatory M2 to the inflammatory M1 type is also observed in the adipose tissue compartment [15]. IL-18 is also produced by adipocytes [16] although, non-fat cells are regarded as the main source of IL-18 in the adipose tissue [17]. In agreement with these data, we also found that IL-18R/IL-18 expression changes in obesity were manifested mainly by the immune cells (i.e., non-fat cells) visible as CLS in the adipose tissue. IL-18 mediates its immunobiological activities by interacting with the functional IL-18Ra/b heterodimer receptor complex which is expressed mainly by immune cells including macrophages, dendritic cells, and T/B

328
lymphocytes in addition to endothelial and smooth muscle cells that constitutively express the IL-18R [13]. Thus, the proinflammatory IL-18 can stimulate all these cell types in autocrine/paracrine manner. Our data showing enhanced IL-18R/IL-18 global expression in the adipose tissues of diabetic/non-diabetic obese and overweight as compared with lean subjects point to obesity as a positive modulator of the IL-18R/IL-18 axis in this compartment. Importantly, we also found a positive correlation between BMI and IL-18R/ IL-18 global expression in the adipose tissue in diabetic/nondiabetic individuals. In agreement with these data, at least in part, increased IL-18 levels have been consistently reported in obesity, T2D, and other metabolic conditions [12,[18][19][20].
Similarly, obesity was found to be associated with increased IL-18 gene expression in the adipose tissue as well as elevated IL-18 plasma levels [21]. We speculate that the increased IL-18R/IL-18 expression in the adipose tissue in obesity/T2D may play a role to induce or augment the inflammatory responses. The emerging evidence points to a key role of chronic inflammation in the pathogenesis of metabolic diseases [22][23][24] and, notably, a consistent relationship was found between IL-18 and various components of metabolic syndrome such as obesity [12,21], insulin resistance [18,25], lipid metabolism and dyslipidemia [20,26], hypertension [27,28], atherosclerosis [19,29], cardiovascular disease [30,31], and T2D-associated nephropathy [32,33].
Our data further show that HOMA-IR was significantly higher in diabetic/non-diabetic obese as compared with Figure 9. Co-localization expression analysis of IL-18R in the adipose tissue in obesity. Co-localization staining for IL-18R protein expression (green fluorescence) was performed using adiponectin as marker for adipocytes staining (red fluorescence) in the adipose tissue samples obtained from nondiabetic lean, overweight, and obese individuals for confocal microscopy as described in Materials and Methods section. The representative microscopy images (40Â magnification) from three independent determinations are shown. In the merged images, IL-18R protein expression can be detected predominantly on immune cells known to form crown-like structures (CLS) around the depleting adipocytes in obese adipose tissue.

330
lean/overweight individuals and a positive association was found between BMI and HOMA-IR. More importantly, the data also show that the adipose tissue expression of IL-18R/ IL-18 in non-diabetic individuals was associated with HOMA-IR which is a robust clinical tool and a standard method for assessing b-cell function and insulin resistance calculated from fasting glucose and insulin or C-peptide concentrations. HOMA-IR is a structural model that has steady state insulin and glucose domains constructed from the physiological dose response data of insulin production and glucose uptake, while it has also been rigorously validated against different physiological methods used for the assessment of insulin resistance [34,35]. In agreement with our data showing a correlation between increased IL-18 gene expression in the adipose tissue in obesity and insulin resistance, Leick et al. also found that the enhanced IL-18 mRNA expression in the adipose tissue in obesity was associated with insulin resistance [21]. Dinarello et al. also showed that the elevated plasma IL-18 levels were a marker of insulin resistance in non-diabetic/diabetic individuals [18]. Nonetheless, we did not find an association between IL-18R/IL-18 expression and insulin resistance in T2D patients which may be due, in part, to confounding effects of diabetes-related immunometabolic complications or comorbidities that are often found in these subjects; and which also explains as to why HOMA-IR is a better indicator of insulin resistance in non-diabetic rather than diabetic subjects. A few studies reported that the changes in circulatory IL-18 levels were associated with altered insulin sensitivity [25,36]. We did not, however, measure systemic levels of IL-18 in the present study as it was designed to determine changes in the adipose tissue expression of IL-18R and IL-18 in obese/T2D individuals. Our data represent the IL-18R/IL-18 axis changes with regard to obesity/T2D that are globally expressed at the gene and protein levels in the adipose tissue. To this effect, we found a strong agreement not only between gene and protein expression of IL-18R (r ¼ 0.69 P ¼ 0.0001) and IL-18 (r ¼ 0.92 P < 0.0001) but also between IL-18R and IL-18 at the levels of gene (r ¼ 0.66 P ¼ 0.0003) and protein (r ¼ 0.91 P < 0.0001) expression. It may thus imply that the IL-18R/IL-18 axis changes that are induced in obesity/T2D are mutually concordant at the transcriptional and translational levels. However, the question remains that which immunometabolic factor(s) play a role in the induction or upregulation of IL-18R and IL-18 in the adipose tissue in obesity and/or T2D. Interestingly, Esposito et al. demonstrated that hyperglycemia and oxidative stress were able to elevate the circulating levels of proinflammatory cytokines including TNF-a, IL-6, and IL-18 [37]. Regarding previous studies on IL-18R and IL-18 expression, proinflammatory cytokines such as TNF-a was shown to upregulate the expression of IL-18R [38] whereas IL-18 was reported to promote the release of TNF-a by macrophages [39]. Adipose tissue infiltrating monocytes/ macrophages can play a role in the adipose tissue inflammation in morbid obesity and, in this regard, our data show that the increased adipose tissue IL-18R/IL-18 gene expression in obesity/T2D was associated positively with monocyte/macrophage markers such as CD11c, CD86, CD68, and CD163. The expression of IL-18 mRNA also correlated positively with the gene expression of TNF-a and CCL-5/RANTES which may have pathobiological significance in metabolic inflammation. TNF-a is a well-known proinflammatory cytokine that is secreted mainly by activated monocytes/macrophages and is a potent regulator/mediator of inflammation, an acute phase reactant and endogenous pyrogen [40]. Due to its role in metabolic inflammation and systemic insulin resistance, TNF-a is regarded as a key component of the obesity-diabetes link [41]. CCL-5/RANTES has been shown to cause inflammation by promoting macrophage recruitment and survival in human adipose tissue [42]. Monocytes/macrophages are also known to produce CCL-2/MCP-1 which is one of the key chemokines that regulate monocyte/ macrophage migration and infiltration into sites of inflammation [43]. In our study, however, the correlation between IL-18 and CCL-2/MCP-1 did not reach statistical significance. A hyperglycemic proinflammatory milieu that is consistently found in obesity/T2D can be expected to induce and promote the IL-18R/IL-18 expression in the adipose tissue which may have pathobiological implications in metabolic disease. In corroboration with this line of argument, Troseid et al. suggested that hyperglycemia and inflammation had a mutually potentiating effect in cardiovascular risk prediction and they also showed that IL-18 was a strong independent predictor of cardiovascular events in metabolic syndrome patients, especially in those with elevated fasting glucose levels [31].
Overall, our results are limited by the small number of participants that could be recruited at the initial phase of this study. Besides, the present findings highlight correlation of the altered adipose tissue expression of IL-18R/IL-18 with tissue inflammation and insulin resistance in obesity while the systemic changes in relation to inflammatory biomarkers such as TNF-a, IL-1b, IFN-g, IL-18, CCL-2, CCL-5, and C-reactive protein still remain unclear. Therefore, further studies will be required to validate these preliminary findings as well as to identify key factor(s) that can lead to upregulation of IL-18R/IL-18 expression in the adipose tissue in metabolic disease.
In conclusion, our data show that IL-18R/IL-18 mRNA and protein expression was elevated (although mainly on immune cells) in the adipose tissue in obesity. This upregulated IL-18R/IL-18 gene expression in the adipose tissue was found to be associated with tissue inflammation and insulin resistance. The bold text shows significant markers.

SUPPORTING INFORMATION
Additional supporting information may be found in the online version of this article at the publisher's web-site. Figure S1. Insulin resistance profiles of subjects with type-2 diabetes are associated with BMI. Figure S2. Concordance between mRNA/protein expression of IL-18R and IL-18 as well as between IL-18/IL-18R at gene and protein expression levels in the adipose tissue.