1,25-dihydroxyvitamin D3 Protects against Macrophage-Induced Activation of NFκB and MAPK Signalling and Chemokine Release in Human Adipocytes

Increased accumulation of macrophages in adipose tissue in obesity is linked to low-grade chronic inflammation, and associated with features of metabolic syndrome. Vitamin D3 may have immunoregulatory effects and reduce adipose tissue inflammation, although the molecular mechanisms remain to be established. This study investigated the effects of vitamin D3 on macrophage-elicited inflammatory responses in cultured human adipocytes, particularly the signalling pathways involved. Macrophage-conditioned (MC) medium (25% with adipocyte maintenance media) markedly inhibited protein expression of the nuclear factor-κB (NFκB) subunit inhibitor κBα (IκBα) (71%, P<0.001) and increased NFκB p65 (1.5-fold, P = 0.026) compared with controls. Treatment with 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) abolished macrophage-induced activation of NFκB signalling by increasing IκBα expression (2.7-fold, P = 0.005) and reducing NFκB p65 phosphorylation (68%; P<0.001). The mitogen-activated protein kinase (MAPK) signalling was activated by MC medium, which was also blunted by 1,25(OH)2D3 with a downregulation of phosphorylated p38 MAPK (32%, P = 0.005) and phosphorylated Erk1/2 (49%, P = 0.001). Furthermore, MC medium (12.5% or 25%) dose-dependently upregulated secretion of key proinflammatory chemokines/cytokines (22-368-fold; all P<0.001) and this was significantly decreased by 1,25(OH)2D3: IL-8 (61% and 31%, P<0.001), MCP-1 (37%, P<0.001 and 36%, P = 0.002), RANTES (78% and 62%, P<0.001) and IL-6 (29%, P<0.001 and 34%, P = 0.019). Monocyte migration-elicited by adipocytes treated with 1,25(OH)2D3 was also reduced (up to 25%, P<0.001). In conclusion, vitamin D3 could be anti-inflammatory in adipose tissue, decreasing macrophage-induced release of chemokines and cytokines by adipocytes and the chemotaxis of monocytes. Our data suggests these effects are mediated by inhibition of the NFκB and MAPK signalling pathways.


Introduction
Growing evidence suggests that vitamin D 3 has pleiotropic functions, beyond its well established roles in bone and mineral metabolism, particularly with regards to insulin secretion and action [1,2]. Vitamin D 3 deficiency may contribute to the pathogenesis of a number of disorders, including obesity and metabolic syndrome [3,4,5]. Epidemiological studies and clinical trials have shown that obese individuals tend to have low vitamin D 3 status [6,7,8]. Although the mechanisms are not clear, sequestration of vitamin D by adipose tissue, less exposure to sunlight and low intake of vitamin D in obese individuals may contribute [8,9,10]. 25-hydroxycholecalciferol (25(OH)D 3 ) is the major circulating form of vitamin D 3 , which is converted to the active form 1,25-dihydroxycholecalciferol (1,25(OH) 2 D 3 ). 1,25(OH) 2 D 3 acts as a ligand for the vitamin D receptor (VDR) that facilitates the transcription of target genes [11,12]. Interestingly, recent studies demonstrate the presence of VDR and vitamin D-metabolizing enzymes in human adipose tissue [13,14]. Therefore, human adipose tissue could be a direct target of vitamin D 3 , and deficiency may have pathological consequences in this tissue [15].
With adipose tissue expansion in obesity, there is a marked increase in the synthesis and release of proinflammatory factors (e.g. TNFa, IL-6, IL-8 and MCP-1), and this may contribute to the elevated circulating levels seen as well as to local tissue inflammation [16,17]. Adipose tissue inflammation, exacerbated by increased infiltration of macrophages and other immune cells, is a central pathological process of adipose tissue dysfunction in obesity [18,19]. Recent work from our group and others has demonstrated that macrophage-derived factors potently stimulate the release of proinflammatory chemokines/cytokines and a number of proteins involved in extracellular matrix remodelling from human preadipocytes and adipocytes; these factors are known to induce inflammation, fibrosis and insulin resistance in adipose tissue, which is associated with metabolic disorders [20,21,22,23]. Evidence has accumulated that vitamin D 3 exerts potent immunoregulatory effects, such as inhibiting the production of TNFa, IL-6 and IL-8 by peripheral blood mononuclear cells in humans [24,25,26]. The effects of vitamin D 3 may be through targeting the nuclear factor-kB (NFkB) and mitogen-activated protein kinase (MAPK) signalling pathways [27,28,29,30]. The emerging role of adipose tissue in adaptive immunity has raised the question whether vitamin D 3 could protect against adipose tissue inflammation.
Studies in murine 3T3-L1 adipocytes have produced inconsistent results, 1,25(OH) 2 D 3 being reported to increase or decrease gene expression of IL-6 and MCP-1 [31,32,33]. Information concerning vitamin D 3 action in human adipose tissue is scarce. Recent studies from our group and others have shown that 1,25(OH) 2 D 3 decreased cytokine-induced expression and release of MCP-1 by human preadipocytes and mature adipocytes [34,35]. However, the mechanisms and the extent to which vitamin D 3 modulates inflammation in human adipose tissue, especially in macrophage-adipocyte crosstalk, remains to be established. These studies were therefore conducted to investigate the effect of 1,25(OH) 2 D 3 on macrophage-induced inflammatory responses in human adipocytes. The molecular mechanisms particularly the NFkB and MAPK signalling pathways and the downstream effects of vitamin D 3 were also studied.

Macrophage-conditioned Medium
Human THP-1 myelomonocytic cell line was purchased from Health Protection Agency Culture Collections (Porton Down, Salisbury UK). THP-1 monocytes (1610 6 cells/ml) were cultured in a 150 cm 2 flask in Roswell Park Memorial Institute (RPMI-1640) medium (containing 10% FCS, 100 U/ml penicillin and 100 mg/ml streptomycin) at 37uC in a humidified atmosphere of O 2 :CO 2 (95:5%). For the preparation of macrophage-conditioned (MC) medium, THP-1 monocytes were differentiated into macrophages with 100 nM phorbol 12-myristate 13-acetate (PMA) (Sigma) for 48 h. The medium was replaced with PMAfree and FCS-free RPMI-1640 medium for 24 h; this medium was collected, filtered through a 0.22 mm filter and stored at 280uC for later use.

Cell Treatment
To examine the effect of vitamin D 3 on basal levels of NFkB and MAPK signalling, adipocytes (at day 11 post-differentiation) were treated with 1,25(OH) 2 D 3 (10 211 and 10 28 M) (ENZO Life Sciences, Plymouth Meeting, PA, USA) for 24 h, and another group of adipocytes received no treatment as controls. To further assess whether vitamin D 3 reduces macrophage-induced inflammatory response, adipocytes were pretreated with vitamin D 3 (10 211 and 10 28 M) for 48 h and then exposed to the MC medium (12.5% or 25% in adipocyte maintenance media), in the presence or absence of 1,25(OH) 2 D 3 (10 211 and 10 28 M) for a further 4 h, 6 h or 24 h. Separate groups of cells were treated with the RPMI medium (12.5% or 25% in adipocyte maintenance media) for the same period as controls. At the end of each experiment, cells and the culture media were collected and stored at 280uC until analysis.
To evaluate the effect vitamin D 3 on the migration of monocytes, adipocytes were treated with vitamin D 3 (10 211 and 10 28 M) or without (control) for 24 h; the culture media was then collected for performing the chemotaxis assay.

Western Blotting
Western blotting was performed as previously described [21]. Briefly, total cellular protein was obtained using lysis buffer (50 mM Tris-HCl, pH 6.7, 10% Glycerol, 4% SDS, 2% 2mercaptoethanol) with freshly added protease inhibitor cocktail and phosphatase inhibitor cocktail (both from Sigma). Protein concentrations were determined by the BCA method. Protein samples (40 mg/lane) were separated on 10% Tricine-SDS polyacrylamide slab gels (Mini Protean Tetra, Bio-Rad, Hemel Hempstead, UK) and transferred to nitrocellulose membranes (Hybond-C Extra, Amersham Bioscience, UK) by wet transfer (Trans Blot, Bio-Rad). The successful transfer of proteins to the membranes was assessed by Ponceau S staining.
For immunodetection, the membranes were blocked for 1 hour at room temperature in Tris-buffered saline (TBS) containing 5% BSA and 0.1% Tween-20. The membranes were then incubated with the primary antibody, including Ikba (New England Biolabs Ltd, Hitchin, Hertfordshire, UK), phosphorylated NFkB p65 (Sigma) and phosphorylated p38 MAPK and phosphorylated Erk1/2 (both from New England Biolabs Ltd, Hitchin, Hertfordshire, UK), at 1:1000 dilution at 4uC overnight. Subsequently, membranes were washed in PBS with 0.1% Tween-20 and then incubated with a HRP-conjugated secondary antibody (Bio-Rad, Hertfordshire, UK or Cell Signalling, Danvers, MA, US). Signals were detected by chemiluminescence using a SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, US). The intensity of signals was evaluated using the Molecular Imager ChemiDoc XRS+ System (Bio-Rad). The size of the protein bands was estimated with PageRuler protein markers (Fermentas, York, UK). The membranes were further probed with GAPDH (Abcam, Cambridge, UK) or total Akt (Cell Signalling) as a loading control. The results were normalised to the value of GAPDH or total Akt.

Real-time PCR
Total RNA was extracted from cells using Trizol (Invitrogen, Paisley, UK). For reverse transcription, 0.5 mg of total RNA was converted to first-strand cDNA in a volume of 10 ml reaction using an iScript first strand synthesis kit (Bio-Rad), which was then Data are means 6 SEM, normalised to GAPDH levels, n = 3 per group. { P,0.05, {{{ P,0.001 vs controls; **P,0.01 vs MC medium. The results were confirmed by three independent experiments. doi:10.1371/journal.pone.0061707.g001  diluted at 1:4. Real-time PCR was carried out in a final volume of 12.5 ml, containing 1 ml cDNA (equivalent to 10 ng of RNA), optimized concentrations of primers, TaqMan probe (FAM-TAMRA) and a master mix made from qPCR core kit (Eurogentec, Seraing, Belgium) using a Stratagene Mx3005P instrument. The sequences of primer and probe used for human IL-8, MCP-1, RANTES (regulated on activation, normal T cell expressed and secreted), IL-1b, IL-6 and b-actin were as described previously [22,36]. PCR reactions were performed in duplicate and the PCR amplification was initiated at 95uC for 10 min, followed by 40 cycles (95uC for 15 sec and 60uC for 1 min). Nontemplate controls were run in parallel. All Ct values were within the range of 20-33 cycles. The results were normalised to the house-keeping gene b-actin values and expressed as fold changes of Ct value relative to controls using the 2 2DDct formula.

Enzyme-Linked Immunosorbent Assay
Protein release of IL-8, MCP-1, RANTES and IL-6 by adipocytes, and by THP-1 macrophages were measured as protein concentrations in cell culture medium, using DuoSet ELISA Development kits (R&D Systems, Abingdon, UK).

Transmigration Assay
THP-1 monocytes at a density of 2610 6 cells/ml were suspended in RPMI-1640 and 100 ml of monocyte suspension was added to the upper chamber of QCM TM chemotaxis transwells (Fisher Scientific, Loughborough, UK) with a pore size of 5 mm. 150 ml of adipocyte culture medium, harvested from the cells treated with vitamin D 3 (10 28 M) or without (control) for 24 h, was added to the lower chamber of transwells. After incubation for 4 h at 37uC in a humidified atmosphere of 5% CO 2 and 95% air, the number of monocytes that migrated to the lower chamber of transwells was determined using the MTT assay with a cell density standard curve.

LDH Assay
Adipocyte viability following various treatments was assessed as the release of lactate dehydrogenase (LDH) into the cell culture medium, using a colourimetric cytotoxicity detection kit (Roche Diagnostics GmbH, Mannheim, Germany). LDH levels were measured by a spectrophotometer at 492 nm with a reference wavelength of 620 nm at room temperature.

Statistical Analysis
Results are presented as means 6 SEM. Comparison of means between two groups was analysed using Student's t-test. Comparison among more than two groups was performed by one-way ANOVA coupled with Bonferroni's t-test. Differences were considered as statistically significant at P,0.05.

1,25-dihydroxyvitamin D 3 Inhibits Macrophage-Induced Activation of NFkB
As the activation of NFkB signalling pathway has a key role in the signal transduction of proinflammatory chemokines/cytokines, we first assessed whether vitamin D 3 affects basal and MC medium-stimulated protein expression of NFkB subunits IkBa and NFkB p65 by human adipocytes. As shown in Fig. 1A-B, a low dose of 1,25(OH) 2 D 3 (10 211 M) had no effect on IkBa while a higher dose (10 28 M) of 1,25(OH) 2 D 3 significantly increased basal IkBa protein abundance (by 1.4-fold, P = 0.042). Exposure to MC medium led to a marked reduction in IkBa protein abundance in adipocytes (by 71%, P,0.001) compared with controls ( Fig. 1C-D). Although the lower dose of 1,25(OH) 2 D 3 (10 211 M) did not reverse this reduction, 1,25(OH) 2 D 3 at higher dose (10 28 M) abolished the inhibitory effect of the MC medium, leading to a 2.7-fold increase (P = 0.005) in IkBa levels compared with the MC group ( Figure 1C-D).

1,25-dihydroxyvitamin D 3 Decreases Monocyte Migration
As vitamin D 3 reduces the adipocyte production of the chemokines (i.e MCP-1, IL-8 and RANTES) which are known to have chemotactic effects, we then explored whether vitamin D 3 affects chemotactic ability of adipocytes. This was determined as THP-1 monocyte migration induced by adipocytes pretreated with 1,25(OH) 2 D 3 or without (control) for 24 h. As shown in Fig. 8A-B, The medium of adipocytes pretreated with1,25(OH) 2 D 3 (10 211 and 10 28 M) resulted in a significant decrease in monocyte migration (by 25% and 21%, both P,0.001) compared with controls; maintenance medium alone (without cells) served as a negative control had the least effect on monocyte migration.

MC Medium or 1,25-dihydroxyvitamin D 3 Has No Cytotoxic Effect
Cell viability was assessed as LDH release by adipocytes and there were no significant differences between control (mean6-SEM: 0.7560.025) and treatment groups (

Discussion
In the present study, we used human THP-1 monocytes and human primary adipocytes as in vitro models to illustrate the inhibitory effects of 1,25(OH) 2 D 3 on macrophage-induced inflammatory responses in adipocytes. We first examined whether 1,25(OH) 2 D 3 prevents the activation of NFkB, which controls the transcription of proinflammatory cytokines in many cell types, including preadipocytes and adipocytes [21,37,38,39]. NFkB activation is initiated by the degradation of IkBa protein, which allows the translocation of NFkB subunits into the nucleus thereby regulating downstream transcriptional programmes [38,40]. In the present study we demonstrate that 1,25(OH) 2 D 3 has a strong inhibitory effect on NFkB signalling in human adipocytes, as 1,25(OH) 2 D 3 (10 28 M) increased basal IkBa levels and reversed inhibition of IkBa by the MC medium. Consistent with our data, recent studies have observed that in murine 3T3-L1 adipocytes, human preadipocytes and adipocytes, 1,25(OH) 2 D 3 also increased protein abundance of IkBa [33,34,41]. Thus, 1,25(OH) 2 D 3 could enhance the stability of IkBa to inhibit NFkB activation in adipocytes. In addition, we show that 1,25(OH) 2 D 3 reduced basal and completely attenuated MC medium-induced phosphorylation of NFkB p65 in human adipocytes. NFkB p65 has been shown to be essential in the production of proinflammatory cytokines in human preadipocytes as NFkB p65 knockdown markedly reduced the release of IL-6 and IL-8 [20]. Recently,1,25(OH) 2 D 3 (10 27 M) was shown to block NFkB p65 translocation to the nucleus in hMSC-derived adipocytes [41]. Taken together, these results The signal transduction of inflammatory mediators may also involve the activation of the MAPK signalling. MAPK of the serine/threonine family, such as p38 MAPK, the extracellular signal-regulated kinases (Erk1/2) and the c-jun N-terminal kinase (JNK), contribute to the inflammatory response in various cell types [29,42] although responses in human adipose tissue are largely unknown. We recently found that MAPK signalling is required in macrophage-induced increases in MMP1 and MMP3 production by human preadipocytes [21]. The release of MCP-1 from explants of human visceral adipose tissue was reduced by inhibitors of the p38 MAPK and NFkB pathways [43]. In the present study, we provide clear evidence that in human adipocytes, MC medium strongly induces phosphorylation of the p38 MAPK and Erk1/2 kinases. Of interest, the activation of MAPK signalling in omental fat in obese subjects has been suggested, as protein expression of phosphorylated p38 MAPK was increased by over 2-fold in obese women compared with lean controls and further, the expression level was positively correlated with clinical parameters such as plasma triglycerides and HOMA-IR (homeostasis model assessment for insulin resistance) [44]. We show in the present study that 1,25(OH) 2 D 3 effectively decreased macrophage-induced phosphorylation of p38 MAPK in a dose-dependent manner. In addition to inhibiting p38 MAPK, 1,25(OH) 2 D 3 also reduced basal, and totally abolished phosphorylation of Erk1/2 elicited by MC medium. Therefore, 1,25(OH) 2 D 3 could act as a potent negative regulator of the MAPK signalling pathway in adipocytes, thereby blocking the transcriptional induction of proinflammatory factors. The molecular mechanisms by which vitamin D 3 exerts effects on the signalling pathways remain to be established. Vitamin D 3 acts by binding to its nuclear receptor VDR which then forms heterodimer with retinoid X receptors, and binds to vitamin D response elements (VDREs) located in promoter regions, thereby regulating the transcription of many target genes [45,46]. In a recent study, vitamin D 3 increased VDR binding to a putative VDRE in MKP-1 promoter and upregulated MKP-1 expression, which led to the inhibition of LPS-induced p38 MAPK phosphorylation and cytokine production in human blood monocytes [29]. VDR could be important in mediating the effects of 1,25(OH) 2 D 3 on signalling pathways in human adipocytes and further studies are warranted.
Since 1,25(OH) 2 D 3 inhibits the NFkB and MAPK pathways, we subsequently examined the downstream effects of 1,25(OH) 2 D 3 , particularly the production of the proinflammatory factors by adipocytes upon macrophage stimulation. We show that exposure of adipocytes to MC medium induced a striking increase in gene expression (14-to 169-fold) and protein release (22-to 368-fold) of the major chemokines/ cytokines, including IL-8, MCP-1, RANTES, IL-1b and IL-6 ( Fig.7). Our data suggests that macrophages are strong inducers of a proinflammatory state in adipocytes, which may form a positive autocrine/paracrine feedback circuit and also provide signals for recruiting macrophages and other immune cells. Additionally, chemokines (i.e. IL-8, MCP-1 and RANTES) are known to be produced at high levels by macrophages [22], which would cause further monocyte/macrophage accumulation in adipose tissue.
A key finding from the present study is the demonstration that 1,25(OH) 2 D 3 powerfully inhibits MC medium-induced expression and release of the chemokines (IL-8, MCP-1 and RANTES) by human adipocytes. IL-8, a member of the CXC chemokine family, has significant chemotactic activity towards neutrophils [47]. In mice fed with a high-fat diet, there is a transient increase in neutrophil infiltration in intra-abdominal fat and IL-8 stimulates neutrophils adhered to 3T3-L1 adipocytes [48]. Other cell types including macrophages also respond to IL-8 as lack of IL-8 receptor CXCR2 protects from adipose macrophage recruitment and insulin resistance in diet-induced obese mice [49]. In severely obese subjects, circulating levels of IL-8 are increased [50]. IL-8 mRNA levels are upregulated in breast adipose tissue of obese women and this is in parallel with increased macrophage infiltration [51].We found that 1,25(OH) 2 D 3 inhibited macrophage-induced IL-8 gene expression (by 53%) and release (up to 61%) from human adipocytes, suggesting vitamin D 3 suppresses IL-8 production in adipose tissue. MCP-1 (or CCL2) and RANTES (or CCL5) belong to CC chemokines which induce the migration of monocytes and other cell types [52]. MCP-1 and its receptor CCR2 are considered to be pivotal for macrophage infiltration in adipose tissue in obesity [53,54]. In MCP-1 or CCR2 knockout mice, there is a decrease in macrophage infiltration in adipose tissue [54,55] whereas overexpressing CCL2 enhances macrophage accumulation and insulin resistance [56]. The present study demonstrates an inhibitory effect of 1,25(OH) 2 D 3 on MCP-1 expression and release by human adipocytes stimulated with MC medium. This is consistent with recent studies by our group and others that 1,25(OH) 2 D 3 decreased MCP-1 secretion under stimulated (by TNFa, IL-1b and MC medium) conditions, in human preadipocytes and adipocytes [34,35]. In addition to MCP-1, recent evidence suggests that RANTES is another key player in the inflammation of adipose tissue in obesity [57].
Serum levels of RANTES and its gene expression in adipose tissue are increased in obese subjects [58]. Furthermore, RANTES promotes monocyte transmigration and macrophage survival in human adipose tissue [58]. In contrast, blocking RANTES with a neutralising antibody reduced T-cell chemotaxis induced by media conditioned by adipose tissue of obese mice [59], and deletion of RANTES receptor CCR5 in mice protected against macrophage recruitment and M2-to M1-type adipose tissue macrophage (ATM) polarization [52]. However, whether vitamin D 3 modulates RANTES production in human adipose tissue is not known. The present study reveals that 1,25(OH) 2 D 3 strongly reduced the expression (by 66%) and release (up to 78%) of RANTES from human adipocytes upon macrophage stimulation. Moreover, we show that 1,25(OH) 2 D 3 also inhibits adipocyte production of the major cytokines IL-1b and IL-6, both of which are critically involved in obesity associated inflammation and insulin resistance [60]. Although IL-1b and IL-6 do not possess chemotactic properties, indirect effects on monocyte recruitment for example via upregulation of chemokines cannot be excluded. A recent study from our group has reported that IL-1b provoked a large increase in MCP-1 release from human preadipocytes [34].
The vitamin D 3 doses used in our study are based from physiological (i.e. 10 211 and 10 210 M) levels and pharmacological (i.e. 10 29 and 10 28 M) levels, which have been similarly employed in several published studies [29,61,62]. It should be mentioned that since adipocytes and macrophages are able to convert 25(OH)D 3 to 1,25(OH) 2 D 3 [14,63], vitamin D 3 concentrations in adipose tissue might be higher than circulating levels. Currently, data on the exact 1,25(OH) 2 D 3 levels in human adipose tissue are scarce. In a small study of morbidly obese subjects (n = 17), 1,25(OH) 2 D 3 concentrations determined by liquid chromatography-MS (LC/MS) were considerably higher (.10-fold) in subcutaneous fat than in serum [64]. Further studies are needed to reveal the levels of vitamin D 3 in adipose tissue of lean and obese subjects.
Collectively, the results from the current study suggest that vitamin D 3 is able to counteract the stimulatory effect of macrophages on the production of chemoattractants, such as IL-8, MCP-1 and RANTES, by adipocytes. As a result, this may disrupt the vicious cycle of perpetuating immune cell infiltration into adipose tissue. Consistent with this notion, we demonstrate that 1,25(OH) 2 D 3 decreased the chemotactic ability of adipocytes since conditioned medium of adipocytes treated with 1,25(OH) 2 D 3 (10 211 and 10 28 M) reduced monocyte migration (Fig.8). It is, therefore, probable that vitamin D 3 acts favourably in adipose tissue to limit monocyte recruitment and its associated inflammation (Fig. 9).
In summary, we have shown that 1,25(OH) 2 D 3 reduces macrophage-induced inflammatory responses in human adipocytes. 1,25(OH) 2 D 3 strongly inhibits the activation of the NFkB and MAPK signalling pathways, which may prevent gene transcription of proinflammatory factors. Consistently, 1,25(OH) 2 D 3 significantly decreases macrophage-elicited expression and release of the major proinflammatory chemokines/ cytokines by human adipocytes. In addition, 1,25(OH) 2 D 3 is able to reduce the chemotactic activity of adipocytes towards monocytes, probably as the result of lowered chemoattractant production. Overall these results suggest that vitamin D 3 has an important role in adipocyte biology through its anti-inflammatory properties; this might be particularly beneficial when adipose tissue becomes inflamed in obesity.