Diacylglycerol kinase inhibitor R59022 attenuates conjugated linoleic acid-mediated inflammation in human adipocytes.

Diacylglycerol kinases (DGK) convert diacylglycerol to phosphatidic acid, which has been reported to stimulate calcium release from the endoplasmic reticulum. Based on our published data showing that trans-10, cis-12 conjugated linoleic acid (t10,c12 CLA)-mediated intracellular calcium accumulation is linked to inflammation and insulin resistance, we hypothesized that inhibiting DGKs with R59022 would prevent t10,c12 CLA-mediated inflammatory signaling and insulin resistance in human adipocytes. Consistent with our hypothesis, R59022 attenuated t10,c12 CLA-mediated i) increased gene expression and protein secretion of interleukin (IL)-8, IL-6, and monocyte chemoattractant protein-1 (MCP-1); ii) increased activation of extracellular signal-related kinase (ERK), cJun-NH2-terminal kinase (JNK), and cJun; iii) increased intracellular calcium levels; iv) suppressed mRNA or protein levels of peroxisome proliferator activated receptor γ, adiponectin, and insulin-dependent glucose transporter 4; and v) decreased fatty acid and glucose uptake and triglyceride content. DGKη was targeted for investigation based on our findings that i) DGKη was highly expressed in primary human adipocytes and time-dependently induced by t10,c12 CLA and that ii) t10,c12 CLA-induced DGKη expression was dose-dependently decreased with R59022. Small interfering RNA (siRNA) targeting DGKη decreased t10,c12 CLA-induced DGKη, IL-8, and MCP-1 gene expression, as well as activation of JNK and cJun. Taken together, these data suggest that DGKs mediate, in part, t10,c12 CLA-induced inflammatory signaling in primary human adipocytes.

kinase substrate phosphorylation domain, and four ankyrin repeats. Type V DGKs, including DGK , contain three C1 domains, a Gly/Pro-rich domain, and a PH-domain-like region (reviewed in Ref. 13 ). DGKs also display tissue-specifi c expression. DGKs are highly expressed in the brain, thymus, and muscle ( 13 ). However, DGK expression in adipose tissue or primary human adipocytes is poorly defi ned. Thus, DGKs are a complex family of kinases, and little is known regarding their potential function in adipose tissue.
Several lines of evidence support the involvement of DGKs in t10,c12 CLA-mediated infl ammation and insulin resistance. First, DGK-generated PA levels activate mTOR and S6 kinase (S6K) in HEK 293 cells ( 14 ); and we reported that t10,c12 CLA activated these two proteins in primary human adipocytes ( 15 ). Second, mTOR and S6K activation have been implicated in the development of insulin resistance, a side effect of CLA supplementation ( 16,17 ). Third, DGK-mediated PA production has also been shown to increase calcium release from the ER ( 18 ). This fi nding could provide a mechanism by which t10,c12 CLA increases intracellular calcium levels in newly differentiated primary human adipocytes ( 6 ). Moreover, DGK has been reported to regulate ERK activation, which we have found to be necessary, in part, for t10,c12 CLA-mediated insulin resistance ( 4,19 ). Indeed, Yasuda et al. ( 19 ) found that DGK facilitated the transport of c-Raf to the plasma membrane, upstream of MEK/ERK activation in response to epidermal growth factor treatment in HeLa cells. Therefore, it is tempting to speculate that t10,c12 CLA-mediated activation of MEK/ERK may involve similar signaling mechanisms. Additionally, DGK ␣ has been shown to regulate tumor necrosis factor (TNF) ␣ -mediated NF B activation ( 20 ), which we have reported to be activated by t10,c12 CLA treatment in adipocytes ( 5 ). Finally, a mixture of CLA isomers has been shown to increase expression of DGK and increase PA levels in cardiomyocytes ( 21 ). Taken together, there are several interesting fi ndings in the literature that suggest a potential role for DGKs in t10,c12 CLA-mediated signaling in primary human adipocytes.
Based on the close similarity between pathways activated by DGK and t10,c12 CLA, we hypothesized that DGKs play an important role in t10,c12 CLA-mediated infl ammation, insulin resistance, and delipidation. To test this hypothesis, we employed the chemical DGK inhibitor R59022 and siRNA targeting DGK . In this study, we demonstrated that DGKs, particularly DGK , may be involved in the regulation of t10,c12 CLA-mediated infl ammatory signaling, insulin resistance, and delipidation in primary human adipocytes.

Materials
All cell cultureware was purchased from Fisher Scientifi c (Norcross, GA). Lightning chemiluminescence substrate was purchased from Perkin Elmer Life Science (Boston, MA). Immunoblotting buffers and precast gels were purchased from Invitrogen by Life Technologies (Carlsbad, CA). Polyclonal antibodies for antiglyceraldehyde-3-phosphate dehydrogenase (GAPDH), Beta Consuming a mixture of c9,t11 and t10,c12 CLA isomers or consuming t10,c12 CLA alone reduces body fat mass in rodents, particularly mice, and in some humans ( 3 ). However, the isomer-specifi c mechanism by which CLA reduces adiposity is unclear. Proposed antiobesity mechanisms of t10, c12 CLA include regulation of i ) energy metabolism, ii ) adipogenesis, iii ) lipid metabolism, iv ) infl ammation, and v ) adipocyte apoptosis (reviewed in Ref. 3 ). However, direct linkage of these potential mechanisms to body fat loss, especially infl ammation, is unclear. We have demonstrated that activation of extracellular signal-regulated kinase (ERK) ( 4 ) and nuclear factor kappa B (NF B) play a role in t10,c12 CLA-mediated delipidation and insulin resistance ( 5 ). We have also shown that t10,c12 CLA-mediated activation of ERK, cJun N-terminal kinase (JNK), NF B, and production of reactive oxygen species (ROS) was dependent on accumulation of intracellular calcium levels ( 6 ). Additionally, we demonstrated that TMB-8, an inhibitor of calcium release from the endoplasmic reticulum (ER), prevented t10,c12 CLA-mediated NF B binding to promoters of interleukin (IL)-8 and cyclooxygenase (COX)-2 ( 6 ). Moreover, activated NF B ( 7-9 ) and ERK ( 10-12 ) induce markers of infl ammation and antagonize peroxisome proliferator activated receptor (PPAR) ␥ activity, thereby causing insulin resistance. These data suggest that t10,c12 CLA mediates infl ammatory signaling that antagonizes adipogenic processes in adipocytes. However, the upstream signals responsible for t10,c12 CLA-mediated increases in intracellular calcium levels, infl ammatory signaling, insulin resistance, and reduced triglyceride (TG) content in human adipocytes are unknown.
Diacylglycerol kinases (DGK) are a family of kinases that phosphorylate diacylglycerol (DAG), resulting in the conversion of DAG into phosphatidic acid (PA). DAG and PA act as second messengers that activate an array of target proteins, resulting in signifi cant changes in cellular signaling (reviewed in Ref. 13 ). For example, DAG activates conventional protein kinase C (cPKC), Unc-13, and protein kinase D, whereas PA activates atypical PKC, phosphatidylinositol (PI)-4-phosphate 5-kinase, and mammalian target of rapamycin (mTOR), RasGAP, and Raf-1 kinase (reviewed in Ref. 13 ). Therefore, DGKs are critical in terminating DAG signaling and initiating PA signaling. In addition to this well-characterized function, DGKs act as scaffolding proteins and regulate subcellular signaling via endosomal and nuclear transport. To date, 10 different DGK isozymes have been identifi ed. Each of the DGKs has up to three PKC-like C1 domains and a catalytic region. DGKs are grouped into fi ve different types, based on their structural and functional features. For example, type 1 DGKs, which include DGK ␣ , ␤ , and ␥ , contain recoverin homology domains and EF-hand motifs that serve as calciumbinding domains (reviewed in Ref. 13 ). Thus, these DGKs are activated in part by calcium binding. Type II DGKs, including DGK ␦ , , and , contain pleckstrin homology domains, sterile ␣ motif domain, and a separated catalytic region. Type III DGKs, including DGK , contain no additional functional domains different than other DGK isoforms. Type IV DGKs, including DGK and , contain a nuclear localization signal, a myristolated alanine-rich C 14 C-oleic acid uptake Cultures of human primary adipocytes were supplemented with low-glucose DMEM on day 10. The following day, cultures were pretreated with 0.1, 1, or 10 M R59022 for 30 min and subsequently treated with BSA vehicle or 50 µM t10,c12 CLA for 48 h. Cultures were treated with 12.5 nM 14 C-oleic acid (0.2 µCi; specifi c activity = 40-60 mCi/mmol) for 120 min. The amount of 14 C-oleic acid oleic acid was measured via scintillation counting as described previously ( 23 ).

Immunoblotting
After experimental treatments, cultures were washed once with ice-cold HBSS. The cells were then solubilized by the direct addition of a lysis buffer containing PBS (pH 7.5), 1% Nonidet P-40, 0.1% SDS, 0.5% sodium deoxycholate, 2 mM Na 3 VO 4 , 20 mM ␤glycerophosphate, 10 mM NaF, and a protease inhibitor mixture (Calbiochem) including 500 M AEBSF, 1 g/ml aprotinin, 1 M E-64, 500 M EDTA, and 1 M leupeptin. Monolayers were scraped and transferred to prechilled microfuge tubes. Cell lysates were then sonicated three times for 5 s and stored on ice for an additional 20 min. Cell debris was pelleted by centrifugation at 14,000 rpm at 4°C, and the resulting supernatant was collected for analysis. The protein concentration of each sample was determined using a bicinchoninic acid assay (Pierce). Subsequently, 20 g of protein from each sample was prepared with Nu-Page LDS sample buffer (Life Technologies) for denaturing gel electrophoresis using 4-12% NuPage precasted gels. Proteins were transferred to polyvinylidene difl uoride (PVDF) membranes that were next blocked with 5% milk in TBST for 1 h and washed three times in TBST for 5 min. Blots were incubated overnight at 4°C with primary antibodies targeting DGK , P-ERK, P-JNK, P-cJun, and total cJun at a dilution of 1:1,000, and then subsequently incubated in the respective horseradish peroxidase-conjugated secondary antibody at a dilution of 1:5,000 at room temperature for 1 h. Primary and secondary antibodies targeting GAPDH were used at a 1:5,000 dilution. Primary and secondary antibodies targeting PPAR ␥ were used at dilutions 1:200 and 1:2,000, respectively. After washing, blots were treated with chemiluminescence reagent for 1 min, and fi lm was exposed using a SRX-101A Konica Minolta fi lm developer. To quantify treatment differences, densitometry was performed using a Kodak Image Station 440 . Data were normalized to ␤ -actin loading control.

RNA isolation and PCR
Total RNA was isolated from the cultures using Tri Reagent purchased from Molecular Research Center (Cincinnati, OH), according to manufacturer's protocol. For quantitative real-time PCR, 1.0 g total RNA was converted into fi rst-strand cDNA using high-capacity cDNA archive kit (Applied Biosystems, Foster City, CA). Real-time PCR was performed in an Applied Biosystems 7500 FAST real-time PCR system using Taqman gene expression assays. To account for possible variation in cDNA input or the presence of PCR inhibitors, the endogenous reference gene GAPDH was simultaneously quantifi ed for each sample, and these data were normalized accordingly. Due to RNA interference of GAPDH in Fig. 6 , TATA box binding protein (TBP) was used as the endogenous reference gene. The relative standard curve method using seven 2-fold dilutions ranging from 100 to 1.56 ng RNA was used to check primer effi ciency and linearity of each transcript according to Applied Biosystems's Guide to Performing Relative Quantifi cation of Gene Expression Using Real-Time Quantitative PCR.

Culturing of human primary adipocytes
Abdominal white adipose tissue was obtained with consent from the Institutional Review Boards at the University of North Carolina at Greensboro and the Moses Cone Memorial Hospital during elective abdominoplasty of nondiabetic Caucasian and African American females between the ages of 20 and 50 years with a body mass index р 32.0. These selection criteria allowed for reduced variation in gender, age, and obesity status. Tissue was digested using collagenase; stromal vascular cells were isolated as previously described ( 4 ). Stromal vascular cells were differentiated with adipocyte media (AM-1) containing 1 M rosiglitazone and 250 M 1-methyl-3-isobutylxanthine for three days, which yielded cultures containing ‫ف‬ 30-50% adipocytes. On days 7-14, cells were pretreated with 0.1, 1, 3, 10, or 30 M R59022 dissolved in DMSO for 30 min and subsequently treated with 50-150 M t10,c12 CLA or BSA (BSA) vehicle control for 5 min to 48 h depending on the experimental outcome measured. All cultures were normalized to contain the same amount of BSA and DMSO vehicles. Each independent experiment was repeated at least twice using a mixture of cells from three subjects, unless otherwise indicated.

R59022 attenuates t10,c12 CLA-mediated activation of MAPK and cJun and [Ca 2+ ] i levels
We have previously shown that t10,c12 CLA-induced infl ammation is dependent on MAPK activation ( 3 ) and [Ca 2+ ] i accumulation ( 6 ). Therefore, we examined the role of DGKs in t10,c12 CLA-mediated activation of MAPK (i.e., ERK and JNK) and activator protein (AP)-1 (i.e., cJun), due to their role in upregulating infl ammatory gene expression. Indeed, 30 µM R59022 attenuated t10,c12 CLA-mediated ERK, JNK, and c-Jun phosphorylation ( Fig. 4A ). These data suggest that DGKs are involved in CLAmediated MAPK and c-Jun phosphorylation. Due to the involvement of [Ca 2+ ] i in t10,c12 CLA-mediated infl ammatory signaling, the role of DGKs in elevating [Ca 2+ ] i levels by t10,c12 CLA was determined. Cultures were pretreated with increasing doses of R59022 for 10 min and subsequently treated with t10,c12 CLA, after which [Ca 2+ ] i levels were measured using the fl uorescent calcium indicator Fluo3-AM. As expected, t10,c12 CLA increased [Ca 2+ ] i levels within 1 min, which were decreased by R59022 ( Fig. 4B ). To better understand how R59022 decreases [Ca 2+ ] accumulation, [Ca 2+ ] i levels were measured in cultures treated with following the manufacturer's protocol. Briefl y, media was collected from cultures that were pretreated with 30 M R59022 for 30 min, and subsequently treated with 50 M t10,c12 CLA or BSA for 24 h. This time point was based on previous time course studies showing that the maximum level of cytokine secretion occurred after 24 h of t10,c12 CLA treatment ( 5 ). The media was centrifuged at 13,200 g for 10 min at 4°C to clear the samples of cellular debris. Samples and standards were run in duplicate. Based on the manufacturer's report Bio-Plex Pro Human Cytokine, Chemokine, and Growth Factor Assays -Bulletin 5828, the intra-assay percentage coeffi cient of variation for IL-6, IL-8, and MCP-1 are 7, 9, and 9%, respectively. The interassay percentage CVs are 11, 4, and 7%, respectively. [Ca 2+ ] i levels were determined using the calcium sensitive fl uorescent dye Fluo-3 AM. Briefl y, cells were preloaded with 5 µM Fluo-3 AM and an anionic detergent, 10% Pluronic F-127, at 37°C for 30 min in the dark. Next, cells were washed with HBSS containing CaCl 2 and probenecid, which prevents Fluo-3 AM leakage from cells. Cells were pretreated with R59022 or DMSO vehicle control for 10 min. Subsequently, baseline fl uorescence was measured using a synergy multidetection microplate reader (BioTek Inc., Winooski, VT) for 1 min at 10 s intervals. Cells were then treated with 5 µM thapsigargin (positive control), 1 µg/ml ionomycin (positive control), or 150 µM t10,c12 CLA, and fl uorescence was monitored at 20 s intervals for 7 min. Excitation wavelength was 485 nm, and fl uorescence was collected at 528 nm. Changes in the ratio of calcium-dependent fl uorescence to prestimulus background fl uorescence (F/F 0 ) were plotted over time. For simplicity, single representative experiments are shown.

siRNA-mediated knockdown of DGK
Transfection of human primary adipocytes with DGK siRNA was conducted on day 7 of differentiation in 35 mm cell culture plates. Cells were seeded and differentiated as previously described. Cultures were supplemented with either 50 nM DGK siRNA, GAPDH siRNA, or nontargeting siRNA complexed with Dharmafect reagent 1 (2 µl/ml) from Dharmacon for 72 h. Transfection reagent and undelivered siRNA were removed 24 h posttransfection by removing the media and replacing with complete adipocyte media (AM-1, Zen-Bio Inc., Research Triangle Park, NC).

Statistical analyses
Data are expressed as the means ± SE. Data were analyzed using one-way ANOVA followed by Student t -tests for each pair for multiple comparisons. Alternatively, data were analyzed using two-way ANOVA with interaction followed by Tukey's honestly signifi cant difference test . Differences were considered signifi cant if P < 0.05. All analyses were performed using JMP , version 9 (SAS Institute, Cary, NC).

DGK inhibitor R59022 attenuates t10,c12 CLA-mediated suppression of glucose, FA uptake, and TG content
It has been previously reported that DGKs increase [Ca 2+ ] i levels via PA-mediated secretion from the ER ( 18 ). Because we previously found that t10,c12 CLA-mediated infl ammatory signaling was dependent on increased [Ca 2+ ] i ( 6 ), we hypothesized that DGKs played an important role in t10,c12 CLA-mediated infl ammation, insulin resistance, and delipidation. To determine the extent to which t10,c12 CLA-mediated delipidation involves DGK to the plasma membrane by t10, c12 CLA, we targeted DGK as the candidate DGK isoform responsible for t10,c12 CLA-induced infl ammation and insulin resistance. To identify a specifi c role of DGK in t10,c12 CLA-mediated infl ammatory signaling, we employed RNA interference to selectively deplete DGK before treatment with t10,c12 CLA. DGK siRNA modestly decreased DGK protein levels without affecting GAPDH or ␤ -actin protein levels ( Fig. 6A ). Successful knockdown was supported by using GAPDH siRNA as a positive control, which decreased GAPDH protein levels without affecting ␤ -actin levels. Densitometry revealed a 20 and 24% knockdown of DGK and GAPDH protein levels, respectively ( Fig. 6A ). To confi rm these results, mRNA levels of DGK and GAPDH positive control were measured. Treatment with DGK siRNA resulted in a 50% knockdown of DGK mRNA levels compared with the siRNA control ( Fig. 6B ). GAPDH siRNA selectively decreased GAPDH mRNA levels by 70% ( Fig. 6B ). Next, cultures were pretreated with 50 nM DGK siRNA for 72 h and subsequently treated with t10,c12 CLA ionomycin, which causes calcium infl ux from outside the cell, and thapsigargin, which inhibits calcium-ATPases on the ER causing calcium release from the ER, in the presence or absence of R59022. Interestingly, R59022 completely blocked ionomycin-mediated calcium and partially attenuated thapsigargin-mediated calcium accumulation ( Fig. 4C ). These data suggest that DGKs may be involved in t10,c12 CLA-mediated increase in [Ca 2+ ] i from both intra-and extracellular stores.

t10,c12 CLA increases DGK and DGK ␦ expression, which is blocked by R59022
To determine the DGK isoform responsible for these effects, we analyzed basal gene expression of several DGK isoforms, including DGK ␣ , ␦ , ␥ , , and . These isoforms were chosen based on microarray analyses of all DGK isoforms (data not shown). DGK ␣ was the most highly expressed isoform ( Fig. 5A ). DGK ␦ and DGK were expressed at similar levels, whereas DGK and DGK ␥ were the least abundant isoforms ( Fig. 5A ). To assess the effect of t10,c12 CLA on DGK expression, cultures were treated with 50 M t10,c12 CLA from 3 to 24 h. DGK was induced by t10,c12 CLA treatment at 6, 12, and 24 h, and DGK ␦ was modestly induced by t10,c12 CLA treatment after 12-24 h. DGK ␥ was induced after 24 h of treatment ( Fig. 5B ). There was no effect of t10,c12 CLA on DGK ␣ or DGK expression at any time point (data not shown). Treatment with R59022 for 18 h decreased t10,c12 CLA-induced DGK and DGK ␦ , but not DGK ␥ mRNA levels ( Fig. 5C ).

Depletion of DGK by RNA interference attenuates t10, c12 CLA-induced infl ammatory gene expression and P-cJun activation
Preliminary experiments investigating DGK activity via translocation to the plasma membrane were conducted to evaluate the isoform likely responsible for t10,c12 CLA-mediated infl ammation and insulin resistance. Of each isoform examined, including DGK ␣ , ␤ , ␦ , ␥ , , and , only results examining DGK suggested translocation to the plasma membrane (data not shown). Based on t10,c 12 CLA-mediated induction of DGK and suppression by R59022 and preliminary data suggesting DGK translocation

DISCUSSION
Consistent with our hypothesis, the DGK inhibitor R59022 attenuated t10,c12-mediated suppression of TG levels, radiolabeled oleic acid uptake, insulin-stimulated glucose uptake, and PPAR ␥ protein levels and target gene expression. Additionally, R59022 attenuated t10,c12 CLAinduced infl ammatory gene and protein secretion, MAPK and cJun phosphorylation, and [Ca 2+ ] i accumulation. In further support of our hypothesis, gene silencing of DGK with siRNA also attenuated t10,c12 CLA-mediated cJun and JNK phosphorylation and induction of infl ammatory genes. Taken together, these data suggest that DGKs play role in t10,c12 CLA-mediated induction of infl ammation and insulin resistance in cultures of human adipocytes.
As with most chemical inhibitors, the effi cacy and specifi city of R59022 has been questioned ( 22 ). Because we were unsure which DGK isoform might be involved in CLA-mediated infl ammatory signaling in adipocytes, we decided to use R59022 rather than the more specifi c type for 18 h. Treatment with DGK siRNA modestly decreased CLA-mediated cJun and JNK phosphorylation ( Fig. 6C ).
To determine whether DGK specifi cally played a role in t10,c12 CLA-mediated infl ammatory signaling, infl ammatory gene expression was measured. Notably, siRNA targeting DGK attenuated t10,c12 CLA-induced IL-8, MCP-1, and DGK expression ( Fig. 6D ), as well as calmodulin kinase (CaMK)II ␤ expression (data not shown). These data suggest that DGK is involved in the upregulation of infl ammatory genes and may also be self-regulating at the transcriptional level.
Lastly, to determine whether DGK contributes to t10,c12 CLA-mediated insulin resistance, insulin-stimulated glucose uptake was measured in cultures treated with DGK siRNA. Treatment with siRNA targeting DGK did not attenuate t10,c12 CLA-mediated suppression of insulin-stimulated glucose uptake (data not shown). Collectively, these data suggest that DGKs, particularly DGK , play a role in CLAmediated infl ammation, but the specifi c role of DGK in t10,c12 CLA-mediated insulin resistance is unclear.  and provide another reason why the type II DGKs were targeted instead of other DGK isoforms. Based on the fi nding that DGK ␣ displayed the highest level of expression, it is possible that DGK ␣ may have important functions in adipocytes, but perhaps not specifi cally related to infl ammatory signaling induced by t10,c12 CLA. Gene silencing I DGK inhibitor R59949. In preliminary experiments using R59949, we found it had little effect on reducing infl ammatory gene expression and did not signifi cantly attenuate t10,c12 CLA-mediated suppression of insulin-stimulated glucose uptake (data not shown). These fi ndings suggest that type I DGKs may not play a role in t10,c12 CLA signaling  expression (data not shown), thus providing another mechanism by which DGKs mediate t10,c12 CLA-induced infl ammation. Collectively, these fi ndings support mechanisms by which t10,c12 CLA may trigger intracellular calcium accumulation and subsequent activation of infl ammatory pathways.
DGK-generated PA has been shown to activate other kinases, such as mTOR and S6K. Both of these proteins have been implicated in regulating insulin resistance ( 16,17 ). For example, it was shown in 3T3-L1 adipocytes that expression of a dominant negative mutant of S6K blunted the suppression of insulin-stimulated glucose uptake mediated by TNF ␣ . The insulin-resistant effects of S6K were found to be due to phosphorylation of IRS-1 on Ser-265/270 ( 16 ). It was also found that insulin-resistant ob/ ob mice ( 16 ) or mice fed a high-fat, high-sucrose diet ( 17 ) had increased levels of mTOR and S6K activation compared with lean controls ( 16,17 ). Interestingly, we have previously shown that t10,c12 CLA robustly increases the phosphorylation of mTOR and S6K in primary human adipocytes ( 15 ). Furthermore, it was shown that inhibitors of GPCRs (pertussis toxin), mTOR (rapamycin), phosphatidylinositol 3-kinase (LY-294002), and protein kinase C (calphostin C) blocked the phosphorylation of mTOR and S6K ( 14 ). The activation of these proteins could provide a mechanism by which DGKs mediate t10,c12 CLA-induced insulin resistance in human adipocytes. However, future studies are needed to test this hypothesis.
Based on the data reported here and reports in the literature, we propose a working model ( Fig. 7 ) in which t10,c12 CLA activates a GPCR linked to PLC, which generates IP3 and DAG. DGKs convert DAG into PA, and together with IP3, stimulate calcium secretion from the ER. Elevated calcium levels activate calcium-sensitive kinases, such as CaMKII, which promote ROS production and MAPK activation, leading to the induction of infl ammatory genes by NF B and AP-1. Secreted infl ammatory proteins, such as IL-6, IL-8, and MCP-1, exacerbate infl ammatory signaling in adipocytes in an autocrine and paracrine fashion. These inflammatory signals, including activation of NF B or AP-1, antagonize PPAR ␥ , leading to decreased glucose and FA uptake, resulting in delipidation and insulin resistance in adipocytes. Lastly, t10,c12 CLA-mediated induction of DGK was inhibited with studies are needed to confi rm a role of candidate DGK isoforms in t10,c12 CLA-mediated infl ammation. Based on data presented here, we speculate that the type II DGK DGK may be involved in CLA-mediated infl ammatory signaling.
A common mechanism by which FAs increase calcium levels is through activation of G-protein coupled receptors (GPCR), such as GPR40. GPR40, coupled to the G-protein subunit G ␣ q/11 , activates phospholipase C (PLC), which hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) into DAG and inositol phosphates (IP). IPs activate receptors on the ER, triggering calcium mobilization (reviewed in Ref. 24 ). Intriguingly, Schmidt and colleagues recently reported that CLA increases [Ca 2+ ] i and stimulates insulin release from INS-1E pancreatic cells via activation of the cell surface receptor free fatty acid receptor 1 (FFA1)/GPR40 ( 25 ). An alternative mechanism has been proposed by Camina and colleagues, whereby PC-specifi c PLC, activated by a pertussis toxin-sensitive Gprotein, generates choline and DAG, which is converted into PA by DGKs. Subsequently, PA triggers calcium mobilization from IP3-independent calcium pools ( 18,(26)(27)(28).