- Email: [email protected]
Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men
Steroids xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men Pascalin Roy a, Mélanie Nadeau a,b,c, Marion Valle a,d,e, Kerstin Bellmann a,d,e, André Marette a,d,e, André Tchernof a,b,c, Claudia Gagnon a,c,e,⇑ a
Department of Medicine, Laval University, 1050, de la Médecine avenue, Québec G1V 0A6, Canada Obesity and Metabolism Unit, Quebec Heart and Lung Institute Research Centre, 2725, Sainte-Foy Road, Québec G1V 4G5, Canada Endocrinology and Nephrology Unit, CHU de Québec Research Centre, 2705, Laurier Boulevard, Québec G1V 4G2, Canada d Cardiology Unit, Quebec Heart and Lung Institute Research Centre, 2725, Sainte-Foy Road, Québec G1V 4G5, Canada e Institute of Nutrition and Functional Foods, 2440, Hochelaga Boulevard, Québec G1V 0A6, Canada b c
a r t i c l e
i n f o
Article history: Received 9 April 2015 Received in revised form 4 August 2015 Accepted 22 August 2015 Available online xxxx Keywords: Vitamin D Inflammation Adipose tissue Human
a b s t r a c t Context: Both vitamin D deficiency and inflammation have been associated with insulin resistance and type 2 diabetes risk. In vitro vitamin D treatment of subcutaneous (SC) adipose tissue (AT) may reduce inflammation, but data are conflicting. Objectives: To evaluate the effects of vitamin D (25(OH)D3 and 1,25(OH)2D3) on the secretion of inflammatory cytokines (TNF-a and IL-6) in omental (OM) and SC human AT and to explore factors that could correlate with the individual response to vitamin D including age, smoking status, BMI, comorbidities, medication, HbA1c, apolipoprotein B, serum 25-hydroxyvitamin D and high sensitivity C-reactive protein. Patients: 7 men and 8 women with severe obesity undergoing bariatric surgery. Intervention: Fresh OM and SC AT explants sampled during surgery (n = 15) were incubated for 24 h in a control, 25(OH)D3 (150 nM) or 1,25(OH)2D3 (1 nM) medium. Lipopolysaccharide (LPS) (10 ng/ml) was added for another 24 h. Main outcome measure: Change in TNF-a and IL-6 levels in collected media after vitamin D treatment (ELISA). Results: Mean age and BMI of the patients were 46.4 ± 10.9 years and 48.8 ± 7.5 kg/m2, respectively. Eleven patients had type 2 diabetes. 25(OH)D3 and 1,25(OH)2D3 reduced the LPS-induced increases in cytokine levels in OM AT of women but not in men. No effect was observed in SC AT. Apart from gender, none of the factors analyzed correlated with vitamin D response. Conclusion: We showed that 25(OH)D3 and 1,25(OH)2D3 can lower cytokine release from OM but not SC AT explants and only in women. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction Approximately 20% of the Canadian adult population is vitamin D deficient, with serum 25-hydroxyvitamin D (25(OH)D) concentrations <50 nM [1]. This proportion increases to 50% among morbidly obese patients given the linear inverse relationship between Abbreviations: 1,25(OH)2D3, calcitriol; 25(OH)D3, calcifediol; AT, adipose tissue; BMI, body mass index; DMSO, dimethyl sulfoxide; HbA1C, glycated hemoglobin; hs-CRP, high sensitivity C-reactive protein; IL-6, interleukin-6; LPS, lipopolysaccharide; OM, omental; PTH, parathyroid hormone; TNF-a, tumor necrosis factor alpha; SC, subcutaneous; VDR, vitamin D receptor. ⇑ Corresponding author at: Department of Medicine, Laval University, 2705 Boul. Laurier, Québec, Qc G1V4G2, Canada. E-mail address: [email protected] (C. Gagnon).
serum 25(OH)D concentrations and body mass indices (BMI) [2]. Recently, epidemiological data suggested that vitamin D deficiency is an independent risk factor contributing to the development of type 2 diabetes through a potential deleterious effect on insulin sensitivity [3–7]. However, randomized placebo-controlled trials studying the effect of vitamin D3 supplementation on insulin sensitivity have generated conflicting results [8–12]. Several potential mechanisms by which vitamin D could affect insulin sensitivity in various cell types have been proposed [13–15]. However, they remain uncertain for the most part. Studies specifically evaluating the effect of vitamin D on adipose tissue inflammation are scarce and inconclusive. Sun and Zemel demonstrated a release of pro-inflammatory cytokines following calcitriol
http://dx.doi.org/10.1016/j.steroids.2015.08.014 0039-128X/Ó 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
2
P. Roy et al. / Steroids xxx (2015) xxx–xxx
(1,25(OH)2D3) treatment of cultured adipocytes and other similar studies published since concluded on an anti-inflammatory effect of 1,25(OH)2D3 [16–20]. Several other groups also documented an anti-inflammatory effect of 1,25(OH)2D3 on fresh subcutaneous human adipose tissue [21–23]. However, despite the critical role of abdominal, visceral adipose tissue in the pathophysiology of type 2 diabetes in humans, no study has evaluated the effect of vitamin D on fat tissue from intra-abdominal compartments such as the greater omentum. Furthermore, only the effect of 1,25(OH)2D3, the activated form of vitamin D, has been studied so far. Yet, calcifediol (25(OH)D3), its precursor, might be effective, as the converting enzyme 1-a-hydroxylase is present in adipose tissue [24]. Thus, the objectives of this study were, through ex vivo experiments: (1) to evaluate the effects of two forms of vitamin D (25 (OH)D3 and 1,25(OH)2D3) on the secretion of the two most studied inflammatory cytokines (TNF-a and IL-6) in omental (OM) and subcutaneous (SC) human whole adipose tissue; and (2) to explore clinical and biochemical factors that could explain inter-individual variability in the response to vitamin D.
1,25(OH)2D3 (1 nM in DMSO) or control M199 medium at 37 °C under a 5% CO2 atmosphere. M199 medium didn’t contain any 25 (OH)D3 or 1,25(OH)2D3 and only a small, non significant quantity of vitamin D2. Culture media were then replaced with the same fresh media, with or without lipopolysaccharide (LPS) (10 ng/ml), for another 24 h to stimulate cytokine release. Each medium was thus duplicated. Media were then collected and frozen at 20 °C while explants were weighed and frozen at 80 °C until cytokine measurements were performed.
2. Experimental
Statistical analyses were performed using JMP 4.0 software (SAS Institute, Cary, NC). Results were normalized for explant weight and, because of the high inter-individual variability in absolute cytokine concentrations, were expressed relative to the LPS medium. Moreover, variables that were not normally distributed were log- or boxcox-transformed before analysis. Student’s t-tests were used to compare the secretion of cytokines in each medium and to assess differences between men and women, as well as between vitamin D-deficient and vitamin D-sufficient participants. Pearson correlation coefficients were computed to evaluate whether clinical and biochemical factors were associated with vitamin D responses in terms of cytokine release by adipose tissue.
2.1. Participants Between July 17th 2013 and March 18th 2014, 15 patients of Europid background undergoing bariatric surgery (gastrectomy or biliopancreatic derivation) at the Quebec Heart and Lung Institute located in Quebec City, Canada, were consecutively recruited for this study. Participants were chosen without regard to any criteria, except for a sex ratio of 1:1. Tissue specimens were obtained from the Biobank of the Quebec Heart and Lung Institute according to institutionally-approved management modalities. All participants provided written, informed consent.
2.5. Cytokine measurements Tumor necrosis factor alpha (TNF-a) and interleukine-6 (IL-6) were measured in collected media with TNF-a HS Quantikine ELISA (R&D systems, Minneapolis, USA) and IL-6 Quantikine ELISA (R&D systems, Minneapolis, USA) kits, respectively, according to the manufacturer’s instructions. 2.6. Statistical analyses
3. Results 2.2. Assessment of clinical and biochemical factors Age, smoking status, anthropometric measurements (weight and height), comorbidities including type 2 diabetes, medication and most recent glycated hemoglobin (HbA1c) and apolipoprotein B levels were retrieved from medical files. Blood samples were collected the night before the surgery or on the morning of the surgery after a 12 h overnight fast and were frozen at 80 °C until measured in batch. Plasma 25(OH)D concentration was measured by radioimmunoassay (Roche Modular, Roche Diagnostics, Laval, QC, Canada), plasma high sensitivity C reactive protein (hs-CRP) was measured by turbidimetry (Roche Integra, Roche Diagnostics, Laval, QC, Canada) and plasma parathyroid hormone (PTH) was measured by chemiluminescence (Roche Modular, Roche Diagnostics, Laval, QC, Canada). 2.3. Adipose tissue sampling Fresh OM and SC fat tissue biopsies were sampled during surgery. Each sample was brought immediately to the lab in sterile containers. Between 600 mg and 800 mg were rinsed with M199 medium supplemented (30 mM HEPES and 30 mM NaHCO3 in M199 medium adjusted to pH 7.1, 50 lg/ml of gentamicin, 2.5 lg/ml of amphotericin B and 0.7 mM of L-glutamine), heated beforehand at 37 °C. 2.4. Explants processing Each sample was cut into pieces of 5–10 mg that were divided in 12 macroscopically equal portions in a 12-well plate. The plate was then incubated for 24 h in 25(OH)D3 (150 nM in DMSO),
3.1. Characteristics of the participants Characteristics of the study sample are shown in Table 1. Of the 15 participants included, there were 7 men and 8 women with a mean age of 46.4 ± 10.9 years. They were obese with a mean BMI Table 1 Characteristics of the study sample. Characteristic
All
Men
Women
P value*
n Age (years) BMI (kg/m2) Type 2 diabetes, n (%) 25(OH)D (nmol/L) Hs-CRP (mg/L) PTH (ng/L) HbA1c (%) Apo B (g/L) Current smoker, n (%)y Anti-inflammatory medication, n (%)à CVD, n (%) COPD, n (%)
15 46.4 ± 10.9 48.8 ± 7.5 11 (73) 60.5 ± 24.0 7.3 ± 5.3 47.4 ± 11.8 6.4 ± 1.3 1.1 ± 0.3 0 (0) 3 (20)
7 45.2 ± 13.0 51.2 ± 8.5 5 (71) 56.1 ± 22.0 3.6 ± 2.5 41.9 ± 6.7 6.3 ± 1.0 1.1 ± 0.4 0 (0) 1 (14)
8 47.4 ± 9.4 46.6 ± 6.2 6 (75) 64.4 ± 26.4 10.6 ± 4.9 52.3 ± 13.4 6.6 ± 1.6 1.0 ± 0.2 0 (0) 2 (25)
0.71 0.25 0.88 0.52 <0.01 0.09 0.83 0.76 – 0.60
0 (0) 1 (7)
0 (0) 0 (0)
0 (0) 1 (13)
– 0.33
BMI, body mass index; 25(OH)D, 25-hydroxyvitamin D; hs-CRP, highly sensitive C reactive protein; PTH, parathyroid hormone; Apo B, apolipoprotein B; CVD, cardiovascular disease; COPD, chronic obstructive pulmonary disease. Data are presented as mean ± SD or n (%). * Using an unpaired t-test or a chi-square test, as appropriate. y All patients had to quit smoking to be eligible for the surgery. à Includes non-steroidal anti-inflammatory drugs (NSAIDs) (2 patients) and leukotriene receptor inhibitors (1 patient) but not inhaled corticosteroids. No patient was using oral corticosteroids. Patients had to withdraw NSAIDs 1 week before the intervention and thus were not excluded.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
P. Roy et al. / Steroids xxx (2015) xxx–xxx
3
of 48.8 ± 7.5 kg/m2 and 11 had a diagnosis of type 2 diabetes. Mean plasma 25(OH)D concentration was 61 ± 24 nmol/L, ranging from 19 to 103 nmol/L, with 33.3% of participants having a plasma 25 (OH)D concentration <50 nmol/L and 13.3% having a plasma 25 (OH)D concentration <25 nmol/L. There were no statistically significant differences for any of the clinical and biochemical characteristics between men and women, except for hs-CRP concentrations, which, as expected, were higher in women (p = 0.005). 3.2. The effect of vitamin D on TNF-a release from omental and subcutaneous adipose tissue explants Fig. 1 shows cytokine release in various media from OM and SC adipose tissue explants after 24 h of incubation. Because of the high inter-individual variability in absolute cytokine concentrations (TNF-a baseline values ranged from 1.53 to 16.50 fg/ml/mg of OM explant and from 0.45 to 29.50 fg/ml/mg of SC explant; IL-6 baseline values ranged from 2.50 to 154.60 pg/ml/mg of OM explant and from 1.52 to 52.99 pg/ml/mg of SC explant), data are expressed as percentage relative to the medium containing LPS in Fig. 2. Thus, means were not driven only by explants secreting higher concentrations of cytokines and individual effect was better circumscribed. Figs. 1A and 2A show results for TNF-a. Baseline TNF-a concentrations were not statistically different between
Fig. 2. Vitamin D effects on LPS-enhanced cytokine secretion in omental and subcutaneous adipose tissue explants. TNF-a ((A) OM: n = 14; SC: n = 13) and IL-6 ((B) OM: n = 14; SC: n = 13) concentrations in media of explants preincubated for 24 h in control, 25(OH)D3 (150 nM) or 1,25(OH)2D3 (1 nM) medium in which LPS was added (10 ng/ml) for another 24 h. Data are expressed relative to the value in the LPS-only medium. Data are presented as mean ± SEM. ELISA was used to assess cytokine concentrations in media. *p 6 0.02 vs. LPS-only. **p 6 0.06 vs. LPS-only.
OM and SC fat depots in the control medium. The addition of 25 (OH)D3 or 1,25(OH)2D3 did not change TNF-a concentrations compared to control in either depot. LPS induced a significant increase in TNF-a concentrations in both depots (p < 0.0001) compared with control with a trend towards a higher relative response to LPS in OM fat explants (p = 0.06). Addition of 25(OH)D3 to LPS did not change TNF-a concentrations in either depot. However, in OM fat tissue explants, 1,25(OH)2D3 significantly lowered the LPS-induced increases in TNF-a concentrations by 22% (p = 0.02) while it did not have a statistically significant effect in SC fat tissue. The apparent rise in TNF-a concentrations by vitamin D3 in SC fat tissue was driven by one explant that excessively secreted cytokines in the vitamin D3 and LPS media compared with the other explants. Since the relative excess of cytokine release was not due to an absence of response to LPS in the LPS-only medium and that the OM version of this explant did not express such discrepancies, we did not register this explant as dysfunctional. Fig. 1. Vitamin D effects on basal cytokine secretion in omental and subcutaneous adipose tissue explants. TNF-a ((A) OM: n = 14; SC: n = 13) and IL-6 ((B) OM: n = 15; SC: n = 14) concentrations in media of explants preincubated for 24 h in control, 25 (OH)D3 (150 nM) or 1,25(OH)2D3 (1 nM) medium. Data are expressed in fg/ml/mg of explant (TNF-a) or pg/ml/mg of explant (IL-6). Data are presented as mean ± SEM. ELISA was used to assess cytokine concentrations in media.
3.3. The effect of vitamin D on IL-6 release from OM and SC adipose tissue explants Figs. 1B and 2B show results for IL-6. Baseline IL-6 concentrations did not differ significantly between the OM and SC depots
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
4
P. Roy et al. / Steroids xxx (2015) xxx–xxx
Fig. 3. Sex differences in 1,25(OH)2D3 effects on LPS-induced cytokine secretion in omental fat tissue. TNF-a (men: n = 6; women: n = 8) and IL-6 (men: n = 7; women: n = 7) concentrations in media of explants preincubated for 24 h in a 1,25(OH)2D3 (1 nM) medium in which LPS was added (10 ng/ml) for another 24 h. Data are expressed relative to the value in the LPS-only medium. Data are presented as mean ± SEM. ELISA was used to assess cytokine concentrations in media. *p 6 0.01 vs. men. **p 6 0.08 vs. men.
in the control medium. IL-6 concentrations were not altered by the addition of 25(OH)D3 or 1,25(OH)2D3 compared with control in either depot. LPS induced a significant increase in IL-6 concentrations in both depots (p < 0.0001) compared with control with a trend towards a higher relative response to LPS in OM fat explants (p = 0.06). The addition of 25(OH)D3 to LPS did not change IL-6 concentrations in either depot. However, there was a trend for 1,25 (OH)2D3 to lower the LPS-induced increase in IL-6 concentrations by 15% in OM fat tissue explants (p = 0.06) while it did not affect the LPS-induced increases in IL-6 concentrations in SC fat tissue. As mentioned above for TNF-a, the apparent rise in IL-6 concentrations following vitamin D supplementation in SC fat tissue was driven by the odd response of one particular explant. 3.4. The effect of sex on cytokine release by adipose tissue explants after vitamin D treatment Fig. 3 compares the effect of 1,25(OH)2D3 on LPS-induced TNF-a and IL-6 concentrations between men and women in OM adipose tissue. TNF-a concentrations were not affected by 1,25(OH)2D3 treatment in men. However, it decreased TNF-a concentrations significantly by 39% in women (p = 0.0008). The difference in vitamin D response between sexes was statistically significant (p = 0.01). Furthermore, 25(OH)D3 also significantly reduced the LPS-induced increase in TNF-a concentrations by 25% in women (p = 0.04, data not shown). This effect was not seen when considering men only or both sexes together. Moreover, the tendency of 1,25(OH)2D3 to lower the LPSinduced increase in IL-6 concentrations became statistically significant when considering women alone (28% reduction, p = 0.03). The difference in vitamin D response between men and women was, however, slightly less apparent (p = 0.08). Similar to what was seen with TNF-a, there was also a trend for 25(OH)D3 to lower the LPSinduced increase in IL-6 concentrations by 35% among women (p = 0.08, data not shown). To verify if these differences were due to individual explants responses, we compared absolute cytokine concentrations in the control medium and relative explants response to LPS of men and women. No statistically significant difference was demonstrated (data not shown). We performed the same analyses with the data from SC adipose tissue, but no sex difference stood out. 3.5. The effect of other clinical or biochemical factors on cytokine release by adipose tissue explants after vitamin D treatment Even if we noticed a significant variation in vitamin D response among the participants, none of the clinical (age, BMI)
or biochemical (HbA1c, apolipoprotein B, plasma 25(OH)D, hsCRP, PTH and baseline IL-6 and TNF-a concentrations) factors examined correlated consistently with vitamin D effects on cytokine release (data not shown). No difference in cytokine concentrations was noted when vitamin D-deficient and non-deficient participants (under and over 50 nmol/L) or even vitamin D-deficient and sufficient participants (under 50 nmol/L and over 75 nmol/L) were compared. The fold-increase in IL-6 concentrations in response to LPS correlated with 1,25(OH)2D3 effects on IL-6 concentrations in OM adipose tissue explants (r = 0.53, p = 0.05), but the same trend was not seen when considering women or men alone, for TNF-a concentrations or the SC depot. Hs-CRP concentrations also correlated with 1,25(OH)2D3 effects on IL-6 concentrations in OM adipose tissue explants (r = 0.56, p = 0.04), but the correlation disappeared when sexes were considered separately and was not present for TNF-a concentrations or when SC tissue was examined. Interestingly, age correlated inversely with the fold-increase in TNF-a concentrations in response to LPS in both depots (OM: r = 0.62, p = 0.02; SC: r = –0.79, p = 0.001). Considering women alone, age correlated inversely with the fold-increase in TNF-a and IL-6 concentrations in response to LPS in OM adipose tissue (TNF-a: r = 0.77, p = 0.03; IL-6: r = 0.77, p = 0.04), but not significantly in SC adipose tissue (TNF-a: r = 0.72, p = 0.07; IL-6: r = 0.09, p > 0.10). 4. Discussion In this ex vivo experiment in which OM and SC adipose tissue were sampled from obese men and women undergoing bariatric surgery, we evaluated the effects of two forms of vitamin D, 25 (OH)D3 and 1,25(OH)D3, on the secretion of inflammatory cytokines (TNF-a and IL-6) in both fat depots. We also explored some clinical and biochemical factors that could explain the interindividual variability of this anti-inflammatory vitamin D response. We found that 1,25(OH)D3 reduced the LPS-induced increases in TNF-a and, to a lesser extent, in IL-6 concentrations in OM, but not SC, adipose tissue. When considering men and women separately, it appeared that these effects were restricted to women. Similarly, 25(OH)D3 reduced the LPS-induced increase in TNF-a and, to a lesser extent, IL-6 concentrations in OM, but not SC, adipose tissue only in women. Other than sex, we failed to demonstrate the influence of factors including age, BMI, HbA1c, apolipoprotein B, plasma 25(OH)D, hs-CRP and PTH on the variability in the anti-inflammatory response to vitamin D in adipose tissue. To our knowledge, this is the first demonstration of an anti-inflammatory effect of 25(OH)D3 and 1,25(OH)2D3 on cytokine release in OM adipose tissue explants. It is also the first study reporting a sex difference in such anti-inflammatory response to vitamin D in adipose tissue. Although we found a statistically significant reduction of 22% in LPS-induced TNF-a secretion under 1,25(OH)2D3 treatment in OM adipose tissue of both men and women, and a 25% statistically significant decrease after 25(OH)D3 treatment in women only, the effect on IL-6 was not as strong (men and women: 15% reduction, p = 0.06). In fact, reduction of LPS-induced IL-6 secretion by 1,25 (OH)2D3 was only significant among women (28%, p = 0.03) and the effect of 25(OH)D3 was only close to be statistically significant among women (35%, p = 0.08). Several research groups showed a lowering of IL-6 expression or secretion after incubation of preadipocytes or mature human adipocytes with 1,25(OH)2D3 [18– 20,22]. However, TNF-a was not measured in these studies. Most importantly, they found that the steroid hormone acted by diminishing the phosphorylation of ERK 1/2, MAPK p38 and NFjB p65, allowing for less transcription of pro-inflammatory genes. These pathways are regulators of both IL-6 and TNF-a and thus both cytokines should theoretically respond similarly to vitamin D.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
P. Roy et al. / Steroids xxx (2015) xxx–xxx
Although effects on each cytokines were in the same direction in our study, the magnitude of the effect and their significance were different for each cytokine. Several cell types in adipose tissue secrete TNF-a and IL-6 in unequal proportions [25]. Although Wamberg et al. showed no difference in vitamin D receptor (VDR) mRNA expression between stroma-vascular cell fraction and isolated adipocytes [26], VDR gene expression does not seem to correlate with its protein level [27] and VDR concentration differences among the stroma-vascular cell fraction itself remains unknown. Thus, 1,25(OH)2D3 could have a different range of effect on these cells, perhaps explaining the greater effect of vitamin D on TNF-a than on IL-6 concentrations. Unfortunately, our study design did not allow us to assess the effects of 25(OH)D3 and 1,25(OH)D3 or specific VDR density among the different cell fractions in our model. Differences in the effect of 1,25(OH)2D3 on each cytokine could also be explained by the less sensitive assay used for IL-6 than TNF-a, allowing for wider variation among measurements. We report an anti-inflammatory effect of 1,25(OH)2D3 in OM but not SC adipose tissue. No study has included OM adipose tissue so far and those conducted on SC adipose tissue have demonstrated a favorable anti-inflammatory effect of 1,25 (OH)2D3 [16–23]. Noteworthy, as opposed to our study, the experiments conducted previously used cultured preadipocytes or adipocytes, not explants. As mentioned, explants contain all adipose tissue cells, macrophages, adipocytes and endothelial cells among others, and may provide a better reflection of the in vivo environment. Adipocytes secrete less than 10% of the cytokines released by adipose tissue [28]. Hence, when explants are used, an effect of 1,25(OH)2D3 on SC adipocytes could be concealed by the relatively higher concentrations of cytokines secreted by non-fat cells. Besides, macrophages are known to respond to vitamin D, even to its less potent 25(OH)D3 form [20,29], and a greater macrophage infiltration of OM vs. SC adipose tissue in obese humans has been reported [30]. A predominant effect of vitamin D on macrophages could explain the differences observed in terms of vitamin D response in OM vs. SC adipose tissue and also among studies conducted on isolated adipocytes. On the other hand, inflammatory markers in adipose tissue strongly correlate with macrophage infiltration [25]. Examination of baseline IL-6 and TNF-a concentrations and its relation with the effect of 1,25(OH)2D3 on cytokine release showed no constant correlation. Still, many metabolic differences could potentially explain the different vitamin D-induced anti-inflammatory response observed between SC and OM adipose tissues including the number of cells expressing the VDR [26]. Again, further studies are required to help explain such results. When we analyzed sexes separately, women were found to be responsible for the global lowering in LPS-induced cytokine release observed in OM adipose tissue while no effect was seen among men. Similar sex-specific anti-inflammatory effects of 25(OH)D3 were also seen in this fat depot. This sex difference in the antiinflammatory response to vitamin D is intriguing. Previous studies showed that estrogen increases 1,25(OH)2D3 activity and/or VDR expression in human osteoblasts and colonic mucosa of rats [31,32]. However, dihydrotestosterone was also noted to produce similar effects when compared with estradiol [33]. Little is known on such effects in adipose tissue, but we can speculate that both sex steroids affect the anti-inflammatory vitamin D response differently. Other sex differences in the structure of adipose tissue could influence the metabolic effects of vitamin D. Among others, premenopausal women tend to have smaller adipocytes in OM compared to SC adipose tissue, while men do not show differences in adipocyte size in these fat depots [34,35]. It is well known that cell size in adipose tissue is a critical determinant of its metabolic activity [36,37]. Further work is needed to fully understand the source of this difference in vitamin D response in OM adipose tissue between men and women. As for the effect of 25(OH)D3,
5
these results support previous work that showed the bioactivity of the 1-a-hydroxylase enzyme, converting 25(OH)D3 to 1,25 (OH)2D3, in adipose tissue and a potential effect of 25(OH)D3 on human adipocytes [24]. We have examined a large number of potential clinical and biochemical factors that could further explain the inter-individual variability in the anti-inflammatory response to vitamin D including age, BMI, HbA1c, apolipoprotein B, plasma 25(OH)D, hs-CRP and PTH and baseline IL-6 and TNF-a concentrations. We also used the fold-increase in LPS-induced cytokine production from both depots to analyze if explants quality intrinsically influenced results. None of these factors was significantly correlated with the effect of 1,25 (OH)2D3 on cytokine release. Other factors that we did not assess could also contribute to the inter-individual variability including genetic factors or adipose tissue variation in cell composition or fibrosis, which cannot be easily accounted for in this model. In recent studies, genetic variants in genes involved in vitamin D metabolism, insulin sensitivity and inflammation have been shown to modulate the individual response to vitamin D [38,39]. Besides, the number of cells with ability to secrete cytokines and respond to vitamin D could vary among explants and explain the variation in our experiment. While the fold-increase in cytokine concentrations in response to LPS varied extensively from one patient to another, this variation did not correlate with our results. It is thus less likely that this factor contributed significantly to the inter-individual variability that we observed. Besides, we found a negative correlation between fold-increase in cytokine concentrations in response to LPS and the participants’ age (TNF-a: OM: r = 0.62, p = 0.02, SC: r = 0.79, p = 0.001). Explants from older patients thus seem to respond less to LPS. This is an interesting fact to consider for further experiments in this model. We suggest that explants remain a better representation of the in vivo environment than isolated adipocytes or differentiated preadipocytes in adherent cultures. Our study has limitations. First of all, although advantageously comparable to previous studies on this topic, our sample of 8 women and 7 men remains relatively small, thus precluding us to conduct multivariate regression analyses and adjust results for potential confounders. However, our study still represents by far the largest sample to date among studies analyzing the effect of vitamin D on cytokine release by adipose tissue (highest n = 8) [16–23]. Also, cytokine stimulation with LPS may not be considered as representative of the low-grade inflammation found in obesity. Nevertheless, it has been recently shown that a pathogenic shift in the gut microbiota reduces gut integrity in diabetic patients, allowing for more LPS to reach the circulation (metabolic endotoxemia) [40,41]. Additionally, all patients from our study were severely obese. Our results are thus difficult to extrapolate to lean or less obese patients. Yet, an important strength of our study is the collection of OM human adipose tissue samples. Indeed, the anti-inflammatory effect of vitamin D in this fat depot had not been assessed so far and thus clearly provides novel information. Moreover, the study design allowed us to explore, for the first time, several important clinical and biochemical factors that could explain the inter-individual variability in the anti-inflammatory response to vitamin D. Finally, while other measurements like mRNA expression would have been possible, we believe that protein levels are a more direct reflection of cytokine secretion, since there are key regulatory steps between mRNA production and its transcription, as well as its secretion, that can impact on the actual production of the proteins. We thus felt that assessment of cytokine production at the protein level was the most relevant for the purpose of this study, which was not to explore genomic or intracellular effects of vitamin D. Nonetheless, we tried to perform additional experiments on our explants to assess the activation of the NFjB pathway. Although the tissue availability was very limited (the explants had been cut into 12 little pieces as outlined in
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
6
P. Roy et al. / Steroids xxx (2015) xxx–xxx
the Section 2), we were able to isolate protein and mRNA, which gave enough protein to do a single Western blot with explants from 3 participants. Determination of NFjB phosphorylation demonstrated that the NFjB pathway was not in an activated state after 24 h of LPS treatment as was revealed by NFjB-p65-phosphorylation (data not shown). Generally, in vitro, an inflammatory stimulus such as LPS will lead to a rapid degradation of IjB followed by the phosphorylation of NFjB and its translocation to the nucleus where it will exert its effect on certain genes to increase their transcription. The increase in NFjB phosphorylation is very rapid and happens within minutes. However, this study was designed to measure inflammatory mediators that accumulate in the supernatant after hours of stimulation with LPS and which are at the very end of the NFjB cascade. Thus, early events such as NFjB phosphorylation could not be evaluated at this time of treatment. Finally, further studies allowing for measurements of VDR and macrophage density in whole adipose tissue could certainly help clarify some unanswered issues raised in our study. To conclude, we showed that two forms of vitamin D, 25(OH)D3 and 1,25(OH)2D3, can lower cytokine release from OM but not SC adipose tissue explants only in women. If reproduced and confirmed in larger cohorts, these results could pave the way to a better targeting of patients in randomized placebo-controlled trials evaluating the effect of vitamin D on inflammation and insulin sensitivity. Conflicts of interest statement The authors declare no conflict of interest. Acknowledgments P.R. and M.V. carried out the experiments and analyzed data. M. N. and K.B. carried out and conceived the experiments. A.M., A.T. and C.G. conceived the experiments and analyzed data. All authors were involved in the writing of the paper and approved the final version. We would like to thank Diabète Québec – Canada for providing the funds to conduct this study. Part of the work was also funded by a Food & Health Programmatic Grant from the Canadian Institutes of Health Research – Canada to A.M. and C.G. We would also like to acknowledge the contribution of the members of Dr. Tchernof’s laboratory. We acknowledge the invaluable collaboration of the surgery team, bariatric surgeons and biobank staff of the IUCPQ. References [1] L.S. Greene-Finestone, C. Berger, M. de Groh, D.A. Hanley, N. Hidiroglou, K. Sarafin, et al., 25-Hydroxyvitamin D in Canadian adults: biological, environmental, and behavioral correlates, Osteoporos. Int. 22 (5) (2011) 1389–1399. [2] L.K. Johnson, D. Hofsø, E.T. Aasheim, T. Tanbo, K.B. Holven, L.F. Andersen, et al., Impact of gender on vitamin D deficiency in morbidly obese patients: a crosssectional study, Eur. J. Clin. Nutr. 66 (1) (2012) 83–90. [3] C. Gagnon, Z.X. Lu, D.J. Magliano, D.W. Dunstan, J.E. Shaw, P.Z. Zimmet, et al., Serum 25-hydroxyvitamin D, calcium intake, and risk of type 2 diabetes after 5 years: results from a national, population-based prospective study (the Australian Diabetes, Obesity and Lifestyle Study), Diabetes Care 34 (5) (2011) 1133–1138. [4] K.C. Chiu, A. Chu, V.L. Go, M.F. Saad, Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction, Am. J. Clin. Nutr. 79 (5) (2004) 820–825. [5] N.G. Forouhi, J. Luan, A. Cooper, B.J. Boucher, N.J. Wareham, Baseline serum 25hydroxy vitamin D is predictive of future glycemic status and insulin resistance: the Medical Research Council Ely Prospective Study 1990–2000, Diabetes 57 (10) (2008) 2619–2625. [6] P. Knekt, M. Laaksonen, C. Mattila, T. Härkänen, J. Marniemi, M. Heliövaara, et al., Serum vitamin D and subsequent occurrence of type 2 diabetes, Epidemiology 19 (5) (2008) 666–671. [7] A.G. Pittas, Q. Sun, J.E. Manson, B. Dawson-Hughes, F.B. Hu, Plasma 25hydroxyvitamin D concentration and risk of incident type 2 diabetes in women, Diabetes Care 33 (9) (2010) 2021–2023.
[8] M.B. Davidson, P. Duran, M.L. Lee, T.C. Friedman, High-dose vitamin D supplementation in people with prediabetes and hypovitaminosis D, Diabetes Care 36 (2) (2013) 260–266. [9] A.M. Belenchia, A.K. Tosh, L.S. Hillman, C.A. Peterson, Correcting vitamin D insufficiency improves insulin sensitivity in obese adolescents: a randomized controlled trial, Am. J. Clin. Nutr. 97 (4) (2013) 774–781. [10] S. Shab-Bidar, T.R. Neyestani, A. Djazayery, M.R. Eshraghian, A. Houshiarrad, A. Kalayi, et al., Improvement of vitamin D status resulted in amelioration of biomarkers of systemic inflammation in the subjects with type 2 diabetes, Diabetes Metab. Res. Rev. 28 (5) (2012) 424–430. [11] S. Soheilykhah, M. Mojibian, M.J. Moghadam, A. Shojaoddiny-Ardekani, The effect of different doses of vitamin D supplementation on insulin resistance during pregnancy, Gynecol. Endocrinol. 29 (4) (2013) 396–399. [12] O.H. Ryu, S. Lee, J. Yu, M.G. Choi, H.J. Yoo, F. Mantero, A prospective randomized controlled trial of the effects of vitamin D supplementation on long-term glycemic control in type 2 diabetes mellitus of Korea, Endocr. J. 61 (2) (2014) 167–176. [13] B. Garcia-Bailo, A. El-Sohemy, P.S. Haddad, P. Arora, F. Benzaied, M. Karmali, et al., Vitamins D, C, and E in the prevention of type 2 diabetes mellitus: modulation of inflammation and oxidative stress, Biologics 5 (2011) 7–19. [14] C. Ding, D. Gao, J. Wilding, P. Trayhurn, C. Bing, Vitamin D signalling in adipose tissue, Br. J. Nutr. 108 (11) (2012) 1915–1923. [15] T. Mezza, G. Muscogiuri, G.P. Sorice, A. Prioletta, E. Salomone, A. Pontecorvi, et al., Vitamin D deficiency: a new risk factor for type 2 diabetes?, Ann Nutr. Metab. 61 (4) (2012) 337–348. [16] X. Sun, M.B. Zemel, Calcium and 1,25-dihydroxyvitamin D3 regulation of adipokine expression, Obesity (Silver Spring) 15 (2) (2007) 340–348. [17] X. Sun, M.B. Zemel, Calcitriol and calcium regulate cytokine production and adipocyte-macrophage cross-talk, J. Nutr. Biochem. 19 (6) (2008) 392–399. [18] J. Marcotorchino, E. Gouranton, B. Romier, F. Tourniaire, J. Astier, C. Malezet, et al., Vitamin D reduces the inflammatory response and restores glucose uptake in adipocytes, Mol. Nutr. Food Res. 56 (12) (2012) 1771–1782. [19] D. Gao, P. Trayhurn, C. Bing, 1,25-Dihydroxyvitamin D3 inhibits the cytokineinduced secretion of MCP-1 and reduces monocyte recruitment by human preadipocytes, Int. J. Obes. (Lond.) 37 (3) (2013) 357–365. [20] C. Ding, J.P. Wilding, C. Bing, 1,25-dihydroxyvitamin D3 protects against macrophage-induced activation of NFjB and MAPK signalling and chemokine release in human adipocytes, PLoS One 8 (4) (2013) e61707. [21] S. Lorente-Cebrián, A. Eriksson, T. Dunlop, N. Mejhert, I. Dahlman, G. Aström, et al., Differential effects of 1a,25-dihydroxycholecalciferol on MCP-1 and adiponectin production in human white adipocytes, Eur. J. Nutr. 51 (3) (2012) 335–342. [22] S.J. Mutt, T. Karhu, S. Lehtonen, P. Lehenkari, C. Carlberg, J. Saarnio, et al., Inhibition of cytokine secretion from adipocytes by 1,25-dihydroxyvitamin D3 via the NF-jB pathway, FASEB J. 26 (11) (2012) 4400–4407. [23] L. Wamberg, K.B. Cullberg, L. Rejnmark, B. Richelsen, S.B. Pedersen, Investigations of the anti-inflammatory effects of vitamin D in adipose tissue: results from an in vitro study and a randomized controlled trial, Horm. Metab. Res. 45 (6) (2013) 456–462. [24] H. Nimitphong, M.F. Holick, S.K. Fried, M.J. Lee, 25-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 promote the differentiation of human subcutaneous preadipocytes, PLoS One 7 (12) (2012) e52171. [25] S.P. Weisberg, D. McCann, M. Desai, M. Rosenbaum, R.L. Leibel, A.W. Ferrante, Obesity is associated with macrophage accumulation in adipose tissue, J. Clin. Invest. 112 (12) (2003) 1796–1808. [26] L. Wamberg, T. Christiansen, S.K. Paulsen, S. Fisker, P. Rask, L. Rejnmark, et al., Expression of vitamin D-metabolizing enzymes in human adipose tissue – the effect of obesity and diet-induced weight loss, Int. J. Obes. (Lond.) 37 (5) (2013) 651–657. [27] R.J. Wiese, A. Uhland-Smith, T.K. Ross, J.M. Prahl, H.F. DeLuca, Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization, J. Biol. Chem. 267 (28) (1992) 20082–20086. [28] J.N. Fain, A.K. Madan, M.L. Hiler, P. Cheema, S.W. Bahouth, Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans, Endocrinology 145 (5) (2004) 2273–2282. [29] A.E. Riek, J. Oh, I. Darwech, C.E. Moynihan, R.R. Bruchas, C. Bernal-Mizrachi, 25 (OH) vitamin D suppresses macrophage adhesion and migration by downregulation of ER stress and scavenger receptor A1 in type 2 diabetes, J. Steroid Biochem. Mol. Biol. (2013). [30] J. Aron-Wisnewsky, J. Tordjman, C. Poitou, F. Darakhshan, D. Hugol, A. Basdevant, et al., Human adipose tissue macrophages: m1 and m2 cell surface markers in subcutaneous and omental depots and after weight loss, J. Clin. Endocrinol. Metab. 94 (11) (2009) 4619–4623. [31] L.G. Rao, J.N. Wylie, M.S. Kung Sutherland, T.M. Murray, 17 beta-oestradiol enhances the stimulatory effect of 1,25-dihydroxyvitamin D3 on alkaline phosphatase activity in human osteosarcoma SaOS-2 cells in a differentiationdependent manner, J. Endocrinol. 148 (1) (1996) 181–187. [32] B. Schwartz, P. Smirnoff, S. Shany, Y. Liel, Estrogen controls expression and bioresponse of 1,25-dihydroxyvitamin D receptors in the rat colon, Mol. Cell. Biochem. 203 (1–2) (2000) 87–93. [33] M.T. Escaleira, S. Sonohara, M.M. Brentani, Sex steroids induced up-regulation of 1,25-(OH)2 vitamin D3 receptors in T 47D breast cancer cells, J. Steroid Biochem. Mol. Biol. 45 (4) (1993) 257–263.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
P. Roy et al. / Steroids xxx (2015) xxx–xxx [34] S.K. Fried, J.G. Kral, Sex differences in regional distribution of fat cell size and lipoprotein lipase activity in morbidly obese patients, Int. J. Obes. 11 (2) (1987) 129–140. [35] A. Boivin, G. Brochu, S. Marceau, P. Marceau, F.S. Hould, A. Tchernof, Regional differences in adipose tissue metabolism in obese men, Metabolism 56 (4) (2007) 533–540. [36] P. Björntorp, A. Gustafson, B. Persson, Adipose tissue fat cell size and number in relation to metabolism in endogenous hypertriglyceridemia, Acta Med. Scand. 190 (5) (1971) 363–367. [37] M. Lundgren, M. Svensson, S. Lindmark, F. Renström, T. Ruge, J.W. Eriksson, Fat cell enlargement is an independent marker of insulin resistance and ‘hyperleptinaemia’, Diabetologia 50 (3) (2007) 625–633. [38] T.R. Neyestani, A. Djazayery, S. Shab-Bidar, M.R. Eshraghian, A. Kalayi, N. Shariátzadeh, et al., Vitamin D receptor fok-I polymorphism modulates
7
diabetic host response to vitamin D intake: need for a nutrigenetic approach, Diabetes Care 36 (3) (2013) 550–556. [39] J.S. Zheng, L.D. Parnell, C.E. Smith, Y.C. Lee, A. Jamal-Allial, Y. Ma, et al., Circulating 25-hydroxyvitamin D, IRS1 variant rs2943641, and insulin resistance: replication of a gene-nutrient interaction in 4 populations of different ancestries, Clin. Chem. 60 (1) (2014) 186–196. [40] A.M. Caricilli, M.J. Saad, The role of gut microbiota on insulin resistance, Nutrients 5 (3) (2013) 829–851. [41] I. Moreno-Indias, F. Cardona, F.J. Tinahones, M.I. Queipo-Ortuño, Impact of the gut microbiota on the development of obesity and type 2 diabetes mellitus, Front. Microbiol. 5 (2014) 190.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
Contents lists available at ScienceDirect
Steroids journal homepage: www.elsevier.com/locate/steroids
Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men Pascalin Roy a, Mélanie Nadeau a,b,c, Marion Valle a,d,e, Kerstin Bellmann a,d,e, André Marette a,d,e, André Tchernof a,b,c, Claudia Gagnon a,c,e,⇑ a
Department of Medicine, Laval University, 1050, de la Médecine avenue, Québec G1V 0A6, Canada Obesity and Metabolism Unit, Quebec Heart and Lung Institute Research Centre, 2725, Sainte-Foy Road, Québec G1V 4G5, Canada Endocrinology and Nephrology Unit, CHU de Québec Research Centre, 2705, Laurier Boulevard, Québec G1V 4G2, Canada d Cardiology Unit, Quebec Heart and Lung Institute Research Centre, 2725, Sainte-Foy Road, Québec G1V 4G5, Canada e Institute of Nutrition and Functional Foods, 2440, Hochelaga Boulevard, Québec G1V 0A6, Canada b c
a r t i c l e
i n f o
Article history: Received 9 April 2015 Received in revised form 4 August 2015 Accepted 22 August 2015 Available online xxxx Keywords: Vitamin D Inflammation Adipose tissue Human
a b s t r a c t Context: Both vitamin D deficiency and inflammation have been associated with insulin resistance and type 2 diabetes risk. In vitro vitamin D treatment of subcutaneous (SC) adipose tissue (AT) may reduce inflammation, but data are conflicting. Objectives: To evaluate the effects of vitamin D (25(OH)D3 and 1,25(OH)2D3) on the secretion of inflammatory cytokines (TNF-a and IL-6) in omental (OM) and SC human AT and to explore factors that could correlate with the individual response to vitamin D including age, smoking status, BMI, comorbidities, medication, HbA1c, apolipoprotein B, serum 25-hydroxyvitamin D and high sensitivity C-reactive protein. Patients: 7 men and 8 women with severe obesity undergoing bariatric surgery. Intervention: Fresh OM and SC AT explants sampled during surgery (n = 15) were incubated for 24 h in a control, 25(OH)D3 (150 nM) or 1,25(OH)2D3 (1 nM) medium. Lipopolysaccharide (LPS) (10 ng/ml) was added for another 24 h. Main outcome measure: Change in TNF-a and IL-6 levels in collected media after vitamin D treatment (ELISA). Results: Mean age and BMI of the patients were 46.4 ± 10.9 years and 48.8 ± 7.5 kg/m2, respectively. Eleven patients had type 2 diabetes. 25(OH)D3 and 1,25(OH)2D3 reduced the LPS-induced increases in cytokine levels in OM AT of women but not in men. No effect was observed in SC AT. Apart from gender, none of the factors analyzed correlated with vitamin D response. Conclusion: We showed that 25(OH)D3 and 1,25(OH)2D3 can lower cytokine release from OM but not SC AT explants and only in women. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction Approximately 20% of the Canadian adult population is vitamin D deficient, with serum 25-hydroxyvitamin D (25(OH)D) concentrations <50 nM [1]. This proportion increases to 50% among morbidly obese patients given the linear inverse relationship between Abbreviations: 1,25(OH)2D3, calcitriol; 25(OH)D3, calcifediol; AT, adipose tissue; BMI, body mass index; DMSO, dimethyl sulfoxide; HbA1C, glycated hemoglobin; hs-CRP, high sensitivity C-reactive protein; IL-6, interleukin-6; LPS, lipopolysaccharide; OM, omental; PTH, parathyroid hormone; TNF-a, tumor necrosis factor alpha; SC, subcutaneous; VDR, vitamin D receptor. ⇑ Corresponding author at: Department of Medicine, Laval University, 2705 Boul. Laurier, Québec, Qc G1V4G2, Canada. E-mail address: [email protected] (C. Gagnon).
serum 25(OH)D concentrations and body mass indices (BMI) [2]. Recently, epidemiological data suggested that vitamin D deficiency is an independent risk factor contributing to the development of type 2 diabetes through a potential deleterious effect on insulin sensitivity [3–7]. However, randomized placebo-controlled trials studying the effect of vitamin D3 supplementation on insulin sensitivity have generated conflicting results [8–12]. Several potential mechanisms by which vitamin D could affect insulin sensitivity in various cell types have been proposed [13–15]. However, they remain uncertain for the most part. Studies specifically evaluating the effect of vitamin D on adipose tissue inflammation are scarce and inconclusive. Sun and Zemel demonstrated a release of pro-inflammatory cytokines following calcitriol
http://dx.doi.org/10.1016/j.steroids.2015.08.014 0039-128X/Ó 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
2
P. Roy et al. / Steroids xxx (2015) xxx–xxx
(1,25(OH)2D3) treatment of cultured adipocytes and other similar studies published since concluded on an anti-inflammatory effect of 1,25(OH)2D3 [16–20]. Several other groups also documented an anti-inflammatory effect of 1,25(OH)2D3 on fresh subcutaneous human adipose tissue [21–23]. However, despite the critical role of abdominal, visceral adipose tissue in the pathophysiology of type 2 diabetes in humans, no study has evaluated the effect of vitamin D on fat tissue from intra-abdominal compartments such as the greater omentum. Furthermore, only the effect of 1,25(OH)2D3, the activated form of vitamin D, has been studied so far. Yet, calcifediol (25(OH)D3), its precursor, might be effective, as the converting enzyme 1-a-hydroxylase is present in adipose tissue [24]. Thus, the objectives of this study were, through ex vivo experiments: (1) to evaluate the effects of two forms of vitamin D (25 (OH)D3 and 1,25(OH)2D3) on the secretion of the two most studied inflammatory cytokines (TNF-a and IL-6) in omental (OM) and subcutaneous (SC) human whole adipose tissue; and (2) to explore clinical and biochemical factors that could explain inter-individual variability in the response to vitamin D.
1,25(OH)2D3 (1 nM in DMSO) or control M199 medium at 37 °C under a 5% CO2 atmosphere. M199 medium didn’t contain any 25 (OH)D3 or 1,25(OH)2D3 and only a small, non significant quantity of vitamin D2. Culture media were then replaced with the same fresh media, with or without lipopolysaccharide (LPS) (10 ng/ml), for another 24 h to stimulate cytokine release. Each medium was thus duplicated. Media were then collected and frozen at 20 °C while explants were weighed and frozen at 80 °C until cytokine measurements were performed.
2. Experimental
Statistical analyses were performed using JMP 4.0 software (SAS Institute, Cary, NC). Results were normalized for explant weight and, because of the high inter-individual variability in absolute cytokine concentrations, were expressed relative to the LPS medium. Moreover, variables that were not normally distributed were log- or boxcox-transformed before analysis. Student’s t-tests were used to compare the secretion of cytokines in each medium and to assess differences between men and women, as well as between vitamin D-deficient and vitamin D-sufficient participants. Pearson correlation coefficients were computed to evaluate whether clinical and biochemical factors were associated with vitamin D responses in terms of cytokine release by adipose tissue.
2.1. Participants Between July 17th 2013 and March 18th 2014, 15 patients of Europid background undergoing bariatric surgery (gastrectomy or biliopancreatic derivation) at the Quebec Heart and Lung Institute located in Quebec City, Canada, were consecutively recruited for this study. Participants were chosen without regard to any criteria, except for a sex ratio of 1:1. Tissue specimens were obtained from the Biobank of the Quebec Heart and Lung Institute according to institutionally-approved management modalities. All participants provided written, informed consent.
2.5. Cytokine measurements Tumor necrosis factor alpha (TNF-a) and interleukine-6 (IL-6) were measured in collected media with TNF-a HS Quantikine ELISA (R&D systems, Minneapolis, USA) and IL-6 Quantikine ELISA (R&D systems, Minneapolis, USA) kits, respectively, according to the manufacturer’s instructions. 2.6. Statistical analyses
3. Results 2.2. Assessment of clinical and biochemical factors Age, smoking status, anthropometric measurements (weight and height), comorbidities including type 2 diabetes, medication and most recent glycated hemoglobin (HbA1c) and apolipoprotein B levels were retrieved from medical files. Blood samples were collected the night before the surgery or on the morning of the surgery after a 12 h overnight fast and were frozen at 80 °C until measured in batch. Plasma 25(OH)D concentration was measured by radioimmunoassay (Roche Modular, Roche Diagnostics, Laval, QC, Canada), plasma high sensitivity C reactive protein (hs-CRP) was measured by turbidimetry (Roche Integra, Roche Diagnostics, Laval, QC, Canada) and plasma parathyroid hormone (PTH) was measured by chemiluminescence (Roche Modular, Roche Diagnostics, Laval, QC, Canada). 2.3. Adipose tissue sampling Fresh OM and SC fat tissue biopsies were sampled during surgery. Each sample was brought immediately to the lab in sterile containers. Between 600 mg and 800 mg were rinsed with M199 medium supplemented (30 mM HEPES and 30 mM NaHCO3 in M199 medium adjusted to pH 7.1, 50 lg/ml of gentamicin, 2.5 lg/ml of amphotericin B and 0.7 mM of L-glutamine), heated beforehand at 37 °C. 2.4. Explants processing Each sample was cut into pieces of 5–10 mg that were divided in 12 macroscopically equal portions in a 12-well plate. The plate was then incubated for 24 h in 25(OH)D3 (150 nM in DMSO),
3.1. Characteristics of the participants Characteristics of the study sample are shown in Table 1. Of the 15 participants included, there were 7 men and 8 women with a mean age of 46.4 ± 10.9 years. They were obese with a mean BMI Table 1 Characteristics of the study sample. Characteristic
All
Men
Women
P value*
n Age (years) BMI (kg/m2) Type 2 diabetes, n (%) 25(OH)D (nmol/L) Hs-CRP (mg/L) PTH (ng/L) HbA1c (%) Apo B (g/L) Current smoker, n (%)y Anti-inflammatory medication, n (%)à CVD, n (%) COPD, n (%)
15 46.4 ± 10.9 48.8 ± 7.5 11 (73) 60.5 ± 24.0 7.3 ± 5.3 47.4 ± 11.8 6.4 ± 1.3 1.1 ± 0.3 0 (0) 3 (20)
7 45.2 ± 13.0 51.2 ± 8.5 5 (71) 56.1 ± 22.0 3.6 ± 2.5 41.9 ± 6.7 6.3 ± 1.0 1.1 ± 0.4 0 (0) 1 (14)
8 47.4 ± 9.4 46.6 ± 6.2 6 (75) 64.4 ± 26.4 10.6 ± 4.9 52.3 ± 13.4 6.6 ± 1.6 1.0 ± 0.2 0 (0) 2 (25)
0.71 0.25 0.88 0.52 <0.01 0.09 0.83 0.76 – 0.60
0 (0) 1 (7)
0 (0) 0 (0)
0 (0) 1 (13)
– 0.33
BMI, body mass index; 25(OH)D, 25-hydroxyvitamin D; hs-CRP, highly sensitive C reactive protein; PTH, parathyroid hormone; Apo B, apolipoprotein B; CVD, cardiovascular disease; COPD, chronic obstructive pulmonary disease. Data are presented as mean ± SD or n (%). * Using an unpaired t-test or a chi-square test, as appropriate. y All patients had to quit smoking to be eligible for the surgery. à Includes non-steroidal anti-inflammatory drugs (NSAIDs) (2 patients) and leukotriene receptor inhibitors (1 patient) but not inhaled corticosteroids. No patient was using oral corticosteroids. Patients had to withdraw NSAIDs 1 week before the intervention and thus were not excluded.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
P. Roy et al. / Steroids xxx (2015) xxx–xxx
3
of 48.8 ± 7.5 kg/m2 and 11 had a diagnosis of type 2 diabetes. Mean plasma 25(OH)D concentration was 61 ± 24 nmol/L, ranging from 19 to 103 nmol/L, with 33.3% of participants having a plasma 25 (OH)D concentration <50 nmol/L and 13.3% having a plasma 25 (OH)D concentration <25 nmol/L. There were no statistically significant differences for any of the clinical and biochemical characteristics between men and women, except for hs-CRP concentrations, which, as expected, were higher in women (p = 0.005). 3.2. The effect of vitamin D on TNF-a release from omental and subcutaneous adipose tissue explants Fig. 1 shows cytokine release in various media from OM and SC adipose tissue explants after 24 h of incubation. Because of the high inter-individual variability in absolute cytokine concentrations (TNF-a baseline values ranged from 1.53 to 16.50 fg/ml/mg of OM explant and from 0.45 to 29.50 fg/ml/mg of SC explant; IL-6 baseline values ranged from 2.50 to 154.60 pg/ml/mg of OM explant and from 1.52 to 52.99 pg/ml/mg of SC explant), data are expressed as percentage relative to the medium containing LPS in Fig. 2. Thus, means were not driven only by explants secreting higher concentrations of cytokines and individual effect was better circumscribed. Figs. 1A and 2A show results for TNF-a. Baseline TNF-a concentrations were not statistically different between
Fig. 2. Vitamin D effects on LPS-enhanced cytokine secretion in omental and subcutaneous adipose tissue explants. TNF-a ((A) OM: n = 14; SC: n = 13) and IL-6 ((B) OM: n = 14; SC: n = 13) concentrations in media of explants preincubated for 24 h in control, 25(OH)D3 (150 nM) or 1,25(OH)2D3 (1 nM) medium in which LPS was added (10 ng/ml) for another 24 h. Data are expressed relative to the value in the LPS-only medium. Data are presented as mean ± SEM. ELISA was used to assess cytokine concentrations in media. *p 6 0.02 vs. LPS-only. **p 6 0.06 vs. LPS-only.
OM and SC fat depots in the control medium. The addition of 25 (OH)D3 or 1,25(OH)2D3 did not change TNF-a concentrations compared to control in either depot. LPS induced a significant increase in TNF-a concentrations in both depots (p < 0.0001) compared with control with a trend towards a higher relative response to LPS in OM fat explants (p = 0.06). Addition of 25(OH)D3 to LPS did not change TNF-a concentrations in either depot. However, in OM fat tissue explants, 1,25(OH)2D3 significantly lowered the LPS-induced increases in TNF-a concentrations by 22% (p = 0.02) while it did not have a statistically significant effect in SC fat tissue. The apparent rise in TNF-a concentrations by vitamin D3 in SC fat tissue was driven by one explant that excessively secreted cytokines in the vitamin D3 and LPS media compared with the other explants. Since the relative excess of cytokine release was not due to an absence of response to LPS in the LPS-only medium and that the OM version of this explant did not express such discrepancies, we did not register this explant as dysfunctional. Fig. 1. Vitamin D effects on basal cytokine secretion in omental and subcutaneous adipose tissue explants. TNF-a ((A) OM: n = 14; SC: n = 13) and IL-6 ((B) OM: n = 15; SC: n = 14) concentrations in media of explants preincubated for 24 h in control, 25 (OH)D3 (150 nM) or 1,25(OH)2D3 (1 nM) medium. Data are expressed in fg/ml/mg of explant (TNF-a) or pg/ml/mg of explant (IL-6). Data are presented as mean ± SEM. ELISA was used to assess cytokine concentrations in media.
3.3. The effect of vitamin D on IL-6 release from OM and SC adipose tissue explants Figs. 1B and 2B show results for IL-6. Baseline IL-6 concentrations did not differ significantly between the OM and SC depots
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
4
P. Roy et al. / Steroids xxx (2015) xxx–xxx
Fig. 3. Sex differences in 1,25(OH)2D3 effects on LPS-induced cytokine secretion in omental fat tissue. TNF-a (men: n = 6; women: n = 8) and IL-6 (men: n = 7; women: n = 7) concentrations in media of explants preincubated for 24 h in a 1,25(OH)2D3 (1 nM) medium in which LPS was added (10 ng/ml) for another 24 h. Data are expressed relative to the value in the LPS-only medium. Data are presented as mean ± SEM. ELISA was used to assess cytokine concentrations in media. *p 6 0.01 vs. men. **p 6 0.08 vs. men.
in the control medium. IL-6 concentrations were not altered by the addition of 25(OH)D3 or 1,25(OH)2D3 compared with control in either depot. LPS induced a significant increase in IL-6 concentrations in both depots (p < 0.0001) compared with control with a trend towards a higher relative response to LPS in OM fat explants (p = 0.06). The addition of 25(OH)D3 to LPS did not change IL-6 concentrations in either depot. However, there was a trend for 1,25 (OH)2D3 to lower the LPS-induced increase in IL-6 concentrations by 15% in OM fat tissue explants (p = 0.06) while it did not affect the LPS-induced increases in IL-6 concentrations in SC fat tissue. As mentioned above for TNF-a, the apparent rise in IL-6 concentrations following vitamin D supplementation in SC fat tissue was driven by the odd response of one particular explant. 3.4. The effect of sex on cytokine release by adipose tissue explants after vitamin D treatment Fig. 3 compares the effect of 1,25(OH)2D3 on LPS-induced TNF-a and IL-6 concentrations between men and women in OM adipose tissue. TNF-a concentrations were not affected by 1,25(OH)2D3 treatment in men. However, it decreased TNF-a concentrations significantly by 39% in women (p = 0.0008). The difference in vitamin D response between sexes was statistically significant (p = 0.01). Furthermore, 25(OH)D3 also significantly reduced the LPS-induced increase in TNF-a concentrations by 25% in women (p = 0.04, data not shown). This effect was not seen when considering men only or both sexes together. Moreover, the tendency of 1,25(OH)2D3 to lower the LPSinduced increase in IL-6 concentrations became statistically significant when considering women alone (28% reduction, p = 0.03). The difference in vitamin D response between men and women was, however, slightly less apparent (p = 0.08). Similar to what was seen with TNF-a, there was also a trend for 25(OH)D3 to lower the LPSinduced increase in IL-6 concentrations by 35% among women (p = 0.08, data not shown). To verify if these differences were due to individual explants responses, we compared absolute cytokine concentrations in the control medium and relative explants response to LPS of men and women. No statistically significant difference was demonstrated (data not shown). We performed the same analyses with the data from SC adipose tissue, but no sex difference stood out. 3.5. The effect of other clinical or biochemical factors on cytokine release by adipose tissue explants after vitamin D treatment Even if we noticed a significant variation in vitamin D response among the participants, none of the clinical (age, BMI)
or biochemical (HbA1c, apolipoprotein B, plasma 25(OH)D, hsCRP, PTH and baseline IL-6 and TNF-a concentrations) factors examined correlated consistently with vitamin D effects on cytokine release (data not shown). No difference in cytokine concentrations was noted when vitamin D-deficient and non-deficient participants (under and over 50 nmol/L) or even vitamin D-deficient and sufficient participants (under 50 nmol/L and over 75 nmol/L) were compared. The fold-increase in IL-6 concentrations in response to LPS correlated with 1,25(OH)2D3 effects on IL-6 concentrations in OM adipose tissue explants (r = 0.53, p = 0.05), but the same trend was not seen when considering women or men alone, for TNF-a concentrations or the SC depot. Hs-CRP concentrations also correlated with 1,25(OH)2D3 effects on IL-6 concentrations in OM adipose tissue explants (r = 0.56, p = 0.04), but the correlation disappeared when sexes were considered separately and was not present for TNF-a concentrations or when SC tissue was examined. Interestingly, age correlated inversely with the fold-increase in TNF-a concentrations in response to LPS in both depots (OM: r = 0.62, p = 0.02; SC: r = –0.79, p = 0.001). Considering women alone, age correlated inversely with the fold-increase in TNF-a and IL-6 concentrations in response to LPS in OM adipose tissue (TNF-a: r = 0.77, p = 0.03; IL-6: r = 0.77, p = 0.04), but not significantly in SC adipose tissue (TNF-a: r = 0.72, p = 0.07; IL-6: r = 0.09, p > 0.10). 4. Discussion In this ex vivo experiment in which OM and SC adipose tissue were sampled from obese men and women undergoing bariatric surgery, we evaluated the effects of two forms of vitamin D, 25 (OH)D3 and 1,25(OH)D3, on the secretion of inflammatory cytokines (TNF-a and IL-6) in both fat depots. We also explored some clinical and biochemical factors that could explain the interindividual variability of this anti-inflammatory vitamin D response. We found that 1,25(OH)D3 reduced the LPS-induced increases in TNF-a and, to a lesser extent, in IL-6 concentrations in OM, but not SC, adipose tissue. When considering men and women separately, it appeared that these effects were restricted to women. Similarly, 25(OH)D3 reduced the LPS-induced increase in TNF-a and, to a lesser extent, IL-6 concentrations in OM, but not SC, adipose tissue only in women. Other than sex, we failed to demonstrate the influence of factors including age, BMI, HbA1c, apolipoprotein B, plasma 25(OH)D, hs-CRP and PTH on the variability in the anti-inflammatory response to vitamin D in adipose tissue. To our knowledge, this is the first demonstration of an anti-inflammatory effect of 25(OH)D3 and 1,25(OH)2D3 on cytokine release in OM adipose tissue explants. It is also the first study reporting a sex difference in such anti-inflammatory response to vitamin D in adipose tissue. Although we found a statistically significant reduction of 22% in LPS-induced TNF-a secretion under 1,25(OH)2D3 treatment in OM adipose tissue of both men and women, and a 25% statistically significant decrease after 25(OH)D3 treatment in women only, the effect on IL-6 was not as strong (men and women: 15% reduction, p = 0.06). In fact, reduction of LPS-induced IL-6 secretion by 1,25 (OH)2D3 was only significant among women (28%, p = 0.03) and the effect of 25(OH)D3 was only close to be statistically significant among women (35%, p = 0.08). Several research groups showed a lowering of IL-6 expression or secretion after incubation of preadipocytes or mature human adipocytes with 1,25(OH)2D3 [18– 20,22]. However, TNF-a was not measured in these studies. Most importantly, they found that the steroid hormone acted by diminishing the phosphorylation of ERK 1/2, MAPK p38 and NFjB p65, allowing for less transcription of pro-inflammatory genes. These pathways are regulators of both IL-6 and TNF-a and thus both cytokines should theoretically respond similarly to vitamin D.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
P. Roy et al. / Steroids xxx (2015) xxx–xxx
Although effects on each cytokines were in the same direction in our study, the magnitude of the effect and their significance were different for each cytokine. Several cell types in adipose tissue secrete TNF-a and IL-6 in unequal proportions [25]. Although Wamberg et al. showed no difference in vitamin D receptor (VDR) mRNA expression between stroma-vascular cell fraction and isolated adipocytes [26], VDR gene expression does not seem to correlate with its protein level [27] and VDR concentration differences among the stroma-vascular cell fraction itself remains unknown. Thus, 1,25(OH)2D3 could have a different range of effect on these cells, perhaps explaining the greater effect of vitamin D on TNF-a than on IL-6 concentrations. Unfortunately, our study design did not allow us to assess the effects of 25(OH)D3 and 1,25(OH)D3 or specific VDR density among the different cell fractions in our model. Differences in the effect of 1,25(OH)2D3 on each cytokine could also be explained by the less sensitive assay used for IL-6 than TNF-a, allowing for wider variation among measurements. We report an anti-inflammatory effect of 1,25(OH)2D3 in OM but not SC adipose tissue. No study has included OM adipose tissue so far and those conducted on SC adipose tissue have demonstrated a favorable anti-inflammatory effect of 1,25 (OH)2D3 [16–23]. Noteworthy, as opposed to our study, the experiments conducted previously used cultured preadipocytes or adipocytes, not explants. As mentioned, explants contain all adipose tissue cells, macrophages, adipocytes and endothelial cells among others, and may provide a better reflection of the in vivo environment. Adipocytes secrete less than 10% of the cytokines released by adipose tissue [28]. Hence, when explants are used, an effect of 1,25(OH)2D3 on SC adipocytes could be concealed by the relatively higher concentrations of cytokines secreted by non-fat cells. Besides, macrophages are known to respond to vitamin D, even to its less potent 25(OH)D3 form [20,29], and a greater macrophage infiltration of OM vs. SC adipose tissue in obese humans has been reported [30]. A predominant effect of vitamin D on macrophages could explain the differences observed in terms of vitamin D response in OM vs. SC adipose tissue and also among studies conducted on isolated adipocytes. On the other hand, inflammatory markers in adipose tissue strongly correlate with macrophage infiltration [25]. Examination of baseline IL-6 and TNF-a concentrations and its relation with the effect of 1,25(OH)2D3 on cytokine release showed no constant correlation. Still, many metabolic differences could potentially explain the different vitamin D-induced anti-inflammatory response observed between SC and OM adipose tissues including the number of cells expressing the VDR [26]. Again, further studies are required to help explain such results. When we analyzed sexes separately, women were found to be responsible for the global lowering in LPS-induced cytokine release observed in OM adipose tissue while no effect was seen among men. Similar sex-specific anti-inflammatory effects of 25(OH)D3 were also seen in this fat depot. This sex difference in the antiinflammatory response to vitamin D is intriguing. Previous studies showed that estrogen increases 1,25(OH)2D3 activity and/or VDR expression in human osteoblasts and colonic mucosa of rats [31,32]. However, dihydrotestosterone was also noted to produce similar effects when compared with estradiol [33]. Little is known on such effects in adipose tissue, but we can speculate that both sex steroids affect the anti-inflammatory vitamin D response differently. Other sex differences in the structure of adipose tissue could influence the metabolic effects of vitamin D. Among others, premenopausal women tend to have smaller adipocytes in OM compared to SC adipose tissue, while men do not show differences in adipocyte size in these fat depots [34,35]. It is well known that cell size in adipose tissue is a critical determinant of its metabolic activity [36,37]. Further work is needed to fully understand the source of this difference in vitamin D response in OM adipose tissue between men and women. As for the effect of 25(OH)D3,
5
these results support previous work that showed the bioactivity of the 1-a-hydroxylase enzyme, converting 25(OH)D3 to 1,25 (OH)2D3, in adipose tissue and a potential effect of 25(OH)D3 on human adipocytes [24]. We have examined a large number of potential clinical and biochemical factors that could further explain the inter-individual variability in the anti-inflammatory response to vitamin D including age, BMI, HbA1c, apolipoprotein B, plasma 25(OH)D, hs-CRP and PTH and baseline IL-6 and TNF-a concentrations. We also used the fold-increase in LPS-induced cytokine production from both depots to analyze if explants quality intrinsically influenced results. None of these factors was significantly correlated with the effect of 1,25 (OH)2D3 on cytokine release. Other factors that we did not assess could also contribute to the inter-individual variability including genetic factors or adipose tissue variation in cell composition or fibrosis, which cannot be easily accounted for in this model. In recent studies, genetic variants in genes involved in vitamin D metabolism, insulin sensitivity and inflammation have been shown to modulate the individual response to vitamin D [38,39]. Besides, the number of cells with ability to secrete cytokines and respond to vitamin D could vary among explants and explain the variation in our experiment. While the fold-increase in cytokine concentrations in response to LPS varied extensively from one patient to another, this variation did not correlate with our results. It is thus less likely that this factor contributed significantly to the inter-individual variability that we observed. Besides, we found a negative correlation between fold-increase in cytokine concentrations in response to LPS and the participants’ age (TNF-a: OM: r = 0.62, p = 0.02, SC: r = 0.79, p = 0.001). Explants from older patients thus seem to respond less to LPS. This is an interesting fact to consider for further experiments in this model. We suggest that explants remain a better representation of the in vivo environment than isolated adipocytes or differentiated preadipocytes in adherent cultures. Our study has limitations. First of all, although advantageously comparable to previous studies on this topic, our sample of 8 women and 7 men remains relatively small, thus precluding us to conduct multivariate regression analyses and adjust results for potential confounders. However, our study still represents by far the largest sample to date among studies analyzing the effect of vitamin D on cytokine release by adipose tissue (highest n = 8) [16–23]. Also, cytokine stimulation with LPS may not be considered as representative of the low-grade inflammation found in obesity. Nevertheless, it has been recently shown that a pathogenic shift in the gut microbiota reduces gut integrity in diabetic patients, allowing for more LPS to reach the circulation (metabolic endotoxemia) [40,41]. Additionally, all patients from our study were severely obese. Our results are thus difficult to extrapolate to lean or less obese patients. Yet, an important strength of our study is the collection of OM human adipose tissue samples. Indeed, the anti-inflammatory effect of vitamin D in this fat depot had not been assessed so far and thus clearly provides novel information. Moreover, the study design allowed us to explore, for the first time, several important clinical and biochemical factors that could explain the inter-individual variability in the anti-inflammatory response to vitamin D. Finally, while other measurements like mRNA expression would have been possible, we believe that protein levels are a more direct reflection of cytokine secretion, since there are key regulatory steps between mRNA production and its transcription, as well as its secretion, that can impact on the actual production of the proteins. We thus felt that assessment of cytokine production at the protein level was the most relevant for the purpose of this study, which was not to explore genomic or intracellular effects of vitamin D. Nonetheless, we tried to perform additional experiments on our explants to assess the activation of the NFjB pathway. Although the tissue availability was very limited (the explants had been cut into 12 little pieces as outlined in
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
6
P. Roy et al. / Steroids xxx (2015) xxx–xxx
the Section 2), we were able to isolate protein and mRNA, which gave enough protein to do a single Western blot with explants from 3 participants. Determination of NFjB phosphorylation demonstrated that the NFjB pathway was not in an activated state after 24 h of LPS treatment as was revealed by NFjB-p65-phosphorylation (data not shown). Generally, in vitro, an inflammatory stimulus such as LPS will lead to a rapid degradation of IjB followed by the phosphorylation of NFjB and its translocation to the nucleus where it will exert its effect on certain genes to increase their transcription. The increase in NFjB phosphorylation is very rapid and happens within minutes. However, this study was designed to measure inflammatory mediators that accumulate in the supernatant after hours of stimulation with LPS and which are at the very end of the NFjB cascade. Thus, early events such as NFjB phosphorylation could not be evaluated at this time of treatment. Finally, further studies allowing for measurements of VDR and macrophage density in whole adipose tissue could certainly help clarify some unanswered issues raised in our study. To conclude, we showed that two forms of vitamin D, 25(OH)D3 and 1,25(OH)2D3, can lower cytokine release from OM but not SC adipose tissue explants only in women. If reproduced and confirmed in larger cohorts, these results could pave the way to a better targeting of patients in randomized placebo-controlled trials evaluating the effect of vitamin D on inflammation and insulin sensitivity. Conflicts of interest statement The authors declare no conflict of interest. Acknowledgments P.R. and M.V. carried out the experiments and analyzed data. M. N. and K.B. carried out and conceived the experiments. A.M., A.T. and C.G. conceived the experiments and analyzed data. All authors were involved in the writing of the paper and approved the final version. We would like to thank Diabète Québec – Canada for providing the funds to conduct this study. Part of the work was also funded by a Food & Health Programmatic Grant from the Canadian Institutes of Health Research – Canada to A.M. and C.G. We would also like to acknowledge the contribution of the members of Dr. Tchernof’s laboratory. We acknowledge the invaluable collaboration of the surgery team, bariatric surgeons and biobank staff of the IUCPQ. References [1] L.S. Greene-Finestone, C. Berger, M. de Groh, D.A. Hanley, N. Hidiroglou, K. Sarafin, et al., 25-Hydroxyvitamin D in Canadian adults: biological, environmental, and behavioral correlates, Osteoporos. Int. 22 (5) (2011) 1389–1399. [2] L.K. Johnson, D. Hofsø, E.T. Aasheim, T. Tanbo, K.B. Holven, L.F. Andersen, et al., Impact of gender on vitamin D deficiency in morbidly obese patients: a crosssectional study, Eur. J. Clin. Nutr. 66 (1) (2012) 83–90. [3] C. Gagnon, Z.X. Lu, D.J. Magliano, D.W. Dunstan, J.E. Shaw, P.Z. Zimmet, et al., Serum 25-hydroxyvitamin D, calcium intake, and risk of type 2 diabetes after 5 years: results from a national, population-based prospective study (the Australian Diabetes, Obesity and Lifestyle Study), Diabetes Care 34 (5) (2011) 1133–1138. [4] K.C. Chiu, A. Chu, V.L. Go, M.F. Saad, Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction, Am. J. Clin. Nutr. 79 (5) (2004) 820–825. [5] N.G. Forouhi, J. Luan, A. Cooper, B.J. Boucher, N.J. Wareham, Baseline serum 25hydroxy vitamin D is predictive of future glycemic status and insulin resistance: the Medical Research Council Ely Prospective Study 1990–2000, Diabetes 57 (10) (2008) 2619–2625. [6] P. Knekt, M. Laaksonen, C. Mattila, T. Härkänen, J. Marniemi, M. Heliövaara, et al., Serum vitamin D and subsequent occurrence of type 2 diabetes, Epidemiology 19 (5) (2008) 666–671. [7] A.G. Pittas, Q. Sun, J.E. Manson, B. Dawson-Hughes, F.B. Hu, Plasma 25hydroxyvitamin D concentration and risk of incident type 2 diabetes in women, Diabetes Care 33 (9) (2010) 2021–2023.
[8] M.B. Davidson, P. Duran, M.L. Lee, T.C. Friedman, High-dose vitamin D supplementation in people with prediabetes and hypovitaminosis D, Diabetes Care 36 (2) (2013) 260–266. [9] A.M. Belenchia, A.K. Tosh, L.S. Hillman, C.A. Peterson, Correcting vitamin D insufficiency improves insulin sensitivity in obese adolescents: a randomized controlled trial, Am. J. Clin. Nutr. 97 (4) (2013) 774–781. [10] S. Shab-Bidar, T.R. Neyestani, A. Djazayery, M.R. Eshraghian, A. Houshiarrad, A. Kalayi, et al., Improvement of vitamin D status resulted in amelioration of biomarkers of systemic inflammation in the subjects with type 2 diabetes, Diabetes Metab. Res. Rev. 28 (5) (2012) 424–430. [11] S. Soheilykhah, M. Mojibian, M.J. Moghadam, A. Shojaoddiny-Ardekani, The effect of different doses of vitamin D supplementation on insulin resistance during pregnancy, Gynecol. Endocrinol. 29 (4) (2013) 396–399. [12] O.H. Ryu, S. Lee, J. Yu, M.G. Choi, H.J. Yoo, F. Mantero, A prospective randomized controlled trial of the effects of vitamin D supplementation on long-term glycemic control in type 2 diabetes mellitus of Korea, Endocr. J. 61 (2) (2014) 167–176. [13] B. Garcia-Bailo, A. El-Sohemy, P.S. Haddad, P. Arora, F. Benzaied, M. Karmali, et al., Vitamins D, C, and E in the prevention of type 2 diabetes mellitus: modulation of inflammation and oxidative stress, Biologics 5 (2011) 7–19. [14] C. Ding, D. Gao, J. Wilding, P. Trayhurn, C. Bing, Vitamin D signalling in adipose tissue, Br. J. Nutr. 108 (11) (2012) 1915–1923. [15] T. Mezza, G. Muscogiuri, G.P. Sorice, A. Prioletta, E. Salomone, A. Pontecorvi, et al., Vitamin D deficiency: a new risk factor for type 2 diabetes?, Ann Nutr. Metab. 61 (4) (2012) 337–348. [16] X. Sun, M.B. Zemel, Calcium and 1,25-dihydroxyvitamin D3 regulation of adipokine expression, Obesity (Silver Spring) 15 (2) (2007) 340–348. [17] X. Sun, M.B. Zemel, Calcitriol and calcium regulate cytokine production and adipocyte-macrophage cross-talk, J. Nutr. Biochem. 19 (6) (2008) 392–399. [18] J. Marcotorchino, E. Gouranton, B. Romier, F. Tourniaire, J. Astier, C. Malezet, et al., Vitamin D reduces the inflammatory response and restores glucose uptake in adipocytes, Mol. Nutr. Food Res. 56 (12) (2012) 1771–1782. [19] D. Gao, P. Trayhurn, C. Bing, 1,25-Dihydroxyvitamin D3 inhibits the cytokineinduced secretion of MCP-1 and reduces monocyte recruitment by human preadipocytes, Int. J. Obes. (Lond.) 37 (3) (2013) 357–365. [20] C. Ding, J.P. Wilding, C. Bing, 1,25-dihydroxyvitamin D3 protects against macrophage-induced activation of NFjB and MAPK signalling and chemokine release in human adipocytes, PLoS One 8 (4) (2013) e61707. [21] S. Lorente-Cebrián, A. Eriksson, T. Dunlop, N. Mejhert, I. Dahlman, G. Aström, et al., Differential effects of 1a,25-dihydroxycholecalciferol on MCP-1 and adiponectin production in human white adipocytes, Eur. J. Nutr. 51 (3) (2012) 335–342. [22] S.J. Mutt, T. Karhu, S. Lehtonen, P. Lehenkari, C. Carlberg, J. Saarnio, et al., Inhibition of cytokine secretion from adipocytes by 1,25-dihydroxyvitamin D3 via the NF-jB pathway, FASEB J. 26 (11) (2012) 4400–4407. [23] L. Wamberg, K.B. Cullberg, L. Rejnmark, B. Richelsen, S.B. Pedersen, Investigations of the anti-inflammatory effects of vitamin D in adipose tissue: results from an in vitro study and a randomized controlled trial, Horm. Metab. Res. 45 (6) (2013) 456–462. [24] H. Nimitphong, M.F. Holick, S.K. Fried, M.J. Lee, 25-hydroxyvitamin D3 and 1,25-dihydroxyvitamin D3 promote the differentiation of human subcutaneous preadipocytes, PLoS One 7 (12) (2012) e52171. [25] S.P. Weisberg, D. McCann, M. Desai, M. Rosenbaum, R.L. Leibel, A.W. Ferrante, Obesity is associated with macrophage accumulation in adipose tissue, J. Clin. Invest. 112 (12) (2003) 1796–1808. [26] L. Wamberg, T. Christiansen, S.K. Paulsen, S. Fisker, P. Rask, L. Rejnmark, et al., Expression of vitamin D-metabolizing enzymes in human adipose tissue – the effect of obesity and diet-induced weight loss, Int. J. Obes. (Lond.) 37 (5) (2013) 651–657. [27] R.J. Wiese, A. Uhland-Smith, T.K. Ross, J.M. Prahl, H.F. DeLuca, Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization, J. Biol. Chem. 267 (28) (1992) 20082–20086. [28] J.N. Fain, A.K. Madan, M.L. Hiler, P. Cheema, S.W. Bahouth, Comparison of the release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from visceral and subcutaneous abdominal adipose tissues of obese humans, Endocrinology 145 (5) (2004) 2273–2282. [29] A.E. Riek, J. Oh, I. Darwech, C.E. Moynihan, R.R. Bruchas, C. Bernal-Mizrachi, 25 (OH) vitamin D suppresses macrophage adhesion and migration by downregulation of ER stress and scavenger receptor A1 in type 2 diabetes, J. Steroid Biochem. Mol. Biol. (2013). [30] J. Aron-Wisnewsky, J. Tordjman, C. Poitou, F. Darakhshan, D. Hugol, A. Basdevant, et al., Human adipose tissue macrophages: m1 and m2 cell surface markers in subcutaneous and omental depots and after weight loss, J. Clin. Endocrinol. Metab. 94 (11) (2009) 4619–4623. [31] L.G. Rao, J.N. Wylie, M.S. Kung Sutherland, T.M. Murray, 17 beta-oestradiol enhances the stimulatory effect of 1,25-dihydroxyvitamin D3 on alkaline phosphatase activity in human osteosarcoma SaOS-2 cells in a differentiationdependent manner, J. Endocrinol. 148 (1) (1996) 181–187. [32] B. Schwartz, P. Smirnoff, S. Shany, Y. Liel, Estrogen controls expression and bioresponse of 1,25-dihydroxyvitamin D receptors in the rat colon, Mol. Cell. Biochem. 203 (1–2) (2000) 87–93. [33] M.T. Escaleira, S. Sonohara, M.M. Brentani, Sex steroids induced up-regulation of 1,25-(OH)2 vitamin D3 receptors in T 47D breast cancer cells, J. Steroid Biochem. Mol. Biol. 45 (4) (1993) 257–263.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014
P. Roy et al. / Steroids xxx (2015) xxx–xxx [34] S.K. Fried, J.G. Kral, Sex differences in regional distribution of fat cell size and lipoprotein lipase activity in morbidly obese patients, Int. J. Obes. 11 (2) (1987) 129–140. [35] A. Boivin, G. Brochu, S. Marceau, P. Marceau, F.S. Hould, A. Tchernof, Regional differences in adipose tissue metabolism in obese men, Metabolism 56 (4) (2007) 533–540. [36] P. Björntorp, A. Gustafson, B. Persson, Adipose tissue fat cell size and number in relation to metabolism in endogenous hypertriglyceridemia, Acta Med. Scand. 190 (5) (1971) 363–367. [37] M. Lundgren, M. Svensson, S. Lindmark, F. Renström, T. Ruge, J.W. Eriksson, Fat cell enlargement is an independent marker of insulin resistance and ‘hyperleptinaemia’, Diabetologia 50 (3) (2007) 625–633. [38] T.R. Neyestani, A. Djazayery, S. Shab-Bidar, M.R. Eshraghian, A. Kalayi, N. Shariátzadeh, et al., Vitamin D receptor fok-I polymorphism modulates
7
diabetic host response to vitamin D intake: need for a nutrigenetic approach, Diabetes Care 36 (3) (2013) 550–556. [39] J.S. Zheng, L.D. Parnell, C.E. Smith, Y.C. Lee, A. Jamal-Allial, Y. Ma, et al., Circulating 25-hydroxyvitamin D, IRS1 variant rs2943641, and insulin resistance: replication of a gene-nutrient interaction in 4 populations of different ancestries, Clin. Chem. 60 (1) (2014) 186–196. [40] A.M. Caricilli, M.J. Saad, The role of gut microbiota on insulin resistance, Nutrients 5 (3) (2013) 829–851. [41] I. Moreno-Indias, F. Cardona, F.J. Tinahones, M.I. Queipo-Ortuño, Impact of the gut microbiota on the development of obesity and type 2 diabetes mellitus, Front. Microbiol. 5 (2014) 190.
Please cite this article in press as: P. Roy et al., Vitamin D reduces LPS-induced cytokine release in omental adipose tissue of women but not men, Steroids (2015), http://dx.doi.org/10.1016/j.steroids.2015.08.014