Expression of the LXRα Protein in Human Atherosclerotic Lesions
Objective— Liver X–activated receptor α (LXRα) regulates multiple genes controlling cholesterol metabolism and transport. To clarify its role in atherogenesis, we established a monoclonal antibody recognizing native human LXRα protein and studied the expression pattern in human atherosclerotic lesions.
Methods and Results— A novel monoclonal antibody PPZ0412 was raised against the ligand-binding domain of LXRα, which can be used for immunostaining of human LXRα protein. LXRα protein was detected in the nucleus of macrophages in the liver, spleen, or lung and also in hepatocytes and adipocytes. In atherosclerotic lesions, the LXRα protein was detected in macrophages positive for scavenger receptor class A and/or CD68.
Conclusions— In the human body, the LXRα protein is highly expressed in macrophage lineage cells and foam cells in atherosclerotic lesions and is identified as a target for intervention in atherosclerotic disease.
Liver X–activated receptor α (LXRα; NR1H3) is a member of the nuclear receptor superfamily that forms a functional heterodimer with retinoid X receptors (RXRs).1,2 LXRα/RXRs heterodimers bind to DR-4–type sequence elements known as the LXR response element in their target genes. LXRα is activated by oxidized derivatives of cholesterol (oxysterols), such as 22(R)-hydroxycholesterol, 24(S)-hydroxycholesterol, or 24(S), 25-epoxycholesterol.3,4 In experimental animals, LXRα mRNA is abundantly expressed in tissues that participate in lipid metabolism, such as white adipose tissue, liver, intestine, and also in macrophages.5–7 In the liver, where LXRα mRNA is highly expressed, activation of LXR induces de novo fatty acid biosynthesis, which has led to the suggestion that LXRα is a sensor of the balance between cholesterol and fatty acid metabolism.8,9 In macrophages, LXRα induces its target genes, such as ABCA1, ABCG1, and apolipoprotein (apo) E, which are involved in cholesterol efflux.10–12
The identification of LXRα as a potential cholesterol sensor that governs cholesterol metabolism and transport opens up new possibilities for intervention in the treatment of atherosclerosis. One potential problem with an LXRα agonist is that it upregulates fatty acid synthesis, resulting in hypertriglyceridemia.6 On the other hand, recent studies have demonstrated that LXR may play an atheroprotective role. Systemic administration of an LXR agonist reduced atherosclerosis in LDLR−/− and apoE−/− mice.13 Further, bone marrow transplantation from LXRα/β−/− mice increased lesion formation in these same models.14
These results suggested that LXRα expressed in macrophages in atherosclerotic lesions may play a critical role in atherosclerotic disease. Given the complexity of the LXRα effects, it is of great interest to identify the LXRα protein expression pattern in the human atherosclerotic lesion. Such studies to date have been hampered by the lack of antibodies capable of detecting the native LXRα protein.
Previously, we reported the establishment of a monoclonal antibody against the N-terminal domain of human LXRα, and we also reported the induction of endogenous human LXRα protein K-8607 during differentiation from monocytes to macrophages.15 Unfortunately, the reported anti–N-terminal antibody did not possess sufficient specificity to detect endogenous LXRα protein on immunohistochemical analysis. Therefore, tissue distribution and the subcellular localization of the endogenous LXRα protein remain unelucidated.
To identify the endogenous LXRα protein in human tissues, we established a novel monoclonal antibody against the human LXRα ligand-binding domain (LBD). This monoclonal antibody can specifically recognize endogenous human LXRα on immunoblot and immunohistochemical analyses. Using this antibody, we studied the expression of the human LXRα protein both in normal human tissues and in atherosclerotic lesions.
Expression of the LXRαLBD and Monoclonal Antibody Generation
The LBD of the human LXRα (amino acids 164 to 447) was expressed as a glutathione S-transferase (GST) fusion protein using the expression vector pGEX4T-2 (Amersham Biosciences). Fusion proteins were induced in BL-21 (Stratagene) and purified using Glutathione Sepharose 4B (Amersham Biosciences). Recombinant GST-LXRαLBD was used for 3 cycles of immunization against female BALB/c mice. Three days after the final administration, mice were euthanized and lymphocytes from the spleen were isolated and fused with NS-1 myeloma cells, as previously described.16 The fused cells were cultured in HAT (0.1 mmol/L hypoxanthine, 0.1 mmol/L aminopterin, and 0.16 mmol/L thymidine) selection medium for 10 to 14 days at 37°C to select for the surviving fusion clones. Hybridomas were selected by ELISA against the purified recombinant GST-fused LXRαLBD using tissue culture supernatant. Selected hybridoma clones were purified by limited dilution. For mass production, 9 hybridoma clones were grown in mice ascites. Ascitic fluids were collected and purified using ammonium sulfate.
Cell Culture and Oxidized Low-Density Lipoprotein Preparation
Human primary monocytes/macrophages were obtained as previously described5 and maintained in RPMI 1640 medium supplemented with 10% FBS.
Human low-density lipoprotein (LDL) was isolated from the plasma of healthy volunteers by the method of Goldstein et al.17,18 After dialysis with EDTA-free PBS, 1.7 mg/mL LDL were oxidized with 100 μmol/L CuCl2 at 37°C for 18 hours.
Transient Transfection and Immunoblot Analysis
COS-7 cells were cultured in DMEM containing 10% FBS. Cells were plated in a 100-mm dish at 2.0×106 cells per dish for 16 hours before transfection. Transfections were performed with Effectene Transfection Reagent (QIAGEN) using 2 μg of the pcDNA3-hLXRα expression vector.
Nuclear extracts were obtained as previously described.19 Aliquots of each sample were resolved by SDS-PAGE (10%) and transferred to polyvinylidene difluoride membranes (ProBlott, Applied Biosystems). After blocking the membranes with BlockAce (Dainippon Pharmaceutical Co Ltd) for 3 hours at room temperature, immunoblotting was performed with an anti-LXRα antibody PPZ0412 (1 μg/mL) as the primary antibody. Peroxidase-conjugated anti-mouse IgG antibody (Sigma, St Louis, Mo) was used as the secondary antibody, and SuperSignal West Dura Extended Duration Substrate (Pierce) was used as the substrate for chemiluminescent detection. As a control, we used a nuclear extract of transfected COS-7 cells transfected with pcDNA3-LXRα.
Immunoprecipitation study was performed as follows. COS-7 cells transiently transfected with the FLAG-tag fusion human LXRα expression vector were used as source. Cells were scraped in immunoprecipitation buffer (50 mmol/L Tris-HCl, pH 7.5, 10 mmol/L EDTA, 150 mmol/L NaCl, 10% glycerol, and 0.1% NP-40) supplemented with protease inhibitors. The scraped cells were lysed by freezing and thawing 3×. The anti-FLAG mouse monoclonal antibody (Sigma; 20 μg/mL), control IgG (Sigma; 20 μg/mL), or PPZ0412 (20 μg/mL) and protein G Sepharose (Amersham) were added to the lysate and mixed by rotating the tubes at 4°C. The antibody-protein G Sepharose conjugate was collected by centrifugation. After washing twice, the conjugates were resolved by SDS-PAGE. To detect the FLAG-tagged fusion human LXRα, anti-FLAG antibody conjugated with horseradish peroxidase (Sigma) was used.
Immunohistochemical analysis was performed as described previously.20 Human tissues were fixed for 1 day at room temperature in 10% formalin. The samples were sequentially dehydrated with an alcohol series and embedded in paraffin. Antigen retrieval was performed by heating the sections in an autoclave at 121°C for 15 minutes. During heating, the sections were immersed in 0.1 N citrate buffer solution (pH 6.5). The paraffin sections (4 μm thick) were then treated with normal horse serum to minimize nonspecific staining. These tissues were incubated with a monoclonal antibody against LXRα (PPZ0412) dissolved in 1% BSA/PBS at a final concentration of 10 μg/mL for 2 hours at 25°C. After several washes with PBS, the sections were stained with a secondary antibody (Simple Stain MAX-PO, Nichirei, Tokyo, Japan) for 1 hour. To prevent endogenous peroxidase reactions, the samples were pretreated with 0.3% H2O2 in cold methanol for 30 minutes. Finally, 0.1 mg/mL of 3, 3′-diaminobenzidine tetrahydrochloride was applied to sections for 10 minutes. The sections were counterstained with hematoxylin. Cultured cells were fixed in 4% paraformaldehyde and immunostained by the same method as above.
In separate sets of experiments, sections of the human aorta were double-stained with the anti-CD68 or anti–scavenger receptor class A (SR-A) antibodies. In brief, the sections were immunostained using the first primary monoclonal antibody and diaminobenzidine as described above. After a wash with glycine/HCl buffer for 1 hour, the sections were incubated with a second primary antibody and the Vectastain ABC-PO substrate kit (Vector).21
Generation of Monoclonal Antibody Against LXRαLBD
Figure 1A indicates the expression of GST-LXRαLBD in Escherichia coli (E coli). The expression of this fusion protein was induced in E coli by isopropyl β-d-thiogalactoside (IPTG; lanes 1 and 2). The induced GST fusion protein was purified using Glutathione Sepharose 4B (lane 3). The monoclonal antibody PPZ0412 was raised against this purified GST fusion protein.
Figure 1B indicates the result of immunoblot study with PPZ0412 using protein obtained from COS-7 cells transfected with human LXRα or LXRβ expression vectors. The antibody bound specifically to proteins expressed in COS-7 cells transfected with the LXRα expression vector. PPZ0412 recognized a protein of apparent molecular weight (Mr) of 47 kDa. Additional proteins with an apparent Mr of 40 kDa were also detected only in cells transfected with the LXRα expression vector. This monoclonal antibody did not bind to the LXRβ protein, which has an amino acid sequence highly similar to LXRα.
Figure IA (available online at http://atvb.ahajournals.org) indicates the result of immunoblot study with PPZ0412 using protein obtained from COS-7 cells transfected with human LXRα or FLAG-tagged human LXRα expression vectors. PPZ0412 recognized a protein of apparent Mr of 47 kDa or 50 kDa. Additional proteins were detected in cells transfected with the human LXRα expression vector or FLAG-tagged human LXRα expression vector. Figure IB indicates the result of immunoblot study with anti-FLAG antibody using same protein. The protein of apparent Mr of 50 kDa was detected using anti-FLAG antibody. There were no additional bands.
Detection of Native Human LXRα Protein in Human Monocyte-Derived Macrophages by the Monoclonal Antibody PPZ0412
Figure 2A indicates the results of immunoblotting using whole-cell extracts obtained from human monocytes and macrophages. The monoclonal antibody PPZ0412 bound to the 47-kDa protein in human monocyte-derived macrophages. The apparent molecular weight of this protein is equal to that of the human LXRα protein expressed in COS-7 cells. An additional minor band of ≈40 kDa was also detected. As can be seen in Figure 2B, we could not detect significant effect of oxidation on the LXRα protein amount in macrophages.
Immunoprecipitation of the Human LXRα Protein
Figure 3 indicates the result of immunoprecipitation studies using the LXRα protein tagged for the FLAG epitope at the N-terminal domain expressed in COS-7 cells. This FLAG-tagged protein was recognized by anti-FLAG antibody and was immunoprecipitated. PPZ0412 also bound to this protein, and the tagged LXRα protein was precipitated efficiently. Control IgG was unable to bind to this protein. This indicates the anti-FLAG antibody and PPZ0412 are able to specifically recognize the FLAG-tag fusion human LXRα.
Immunohistochemical Study of the Human Liver
We initially studied the localization of the endogenous LXRα protein in the human liver, because earlier studies by Northern blotting and RT-PCR have reported relatively abundant expression of LXRα mRNA here. PPZ0412 recognizes the protein in the nucleus of Kupffer cells and hepatocytes. As can be seen in Figure 4C, the staining in the nucleus of Kupffer cells (arrow heads) is more prominent than that in hepatocytes. Control IgG did not exhibit significant binding. This result is consistent with the reported expression pattern of LXRα mRNA in the liver.
Expression of the LXRα Protein in Human Organs
Figure 5 depicts LXRα protein expression in human lung, spleen, thymus, and adipose tissue. In the lung, the LXRα protein was detected in the nucleus of alveolar macrophages. In the spleen and thymus, it was also detected in macrophage-like cells. These LXRα-positive cells were also positive for CD68, a marker for macrophage lineage cells (data not shown). However, the expression of LXRα was not limited to cells of macrophage lineage. In adipose tissue, LXRα was positive in the nucleus of adipocytes, which were negative for CD68.
Immunohistochemical Study of Human Atherosclerotic Lesions
To clarify the distribution of the LXRα protein in human atherosclerotic lesions, we examined the lesioned aorta of human subjects. As can be seen in Figure 6A and 6D, in human plaque lesions the LXRα protein was mainly detected in the nucleus of mononuclear cells and foam cells. LXRα-positive cells were not detected in normal aorta (Figure 6C and 6F), suggesting that these LXRα-positive mononuclear cells were infiltrating during lesion formation. As can be seen in Figure 6B and 6E, in advanced lesions the number of LXRα protein–positive cells was decreased because of the decrease of cellularity.
To identify the cell type of LXRα protein–positive cell, we also stained the lesion with anti–SR-A and anti-CD68 antibodies. Figure 6G indicates that the LXRα and SR-A proteins were detected in the same cells. The SR-A protein (purple) was mainly detected in association with the cell membrane, and the LXRα protein (brown) was detected mainly in the nucleus. Figure 6H indicates the results of immunostaining for another macrophage lineage marker CD68 (dark purple) and the LXRα protein (brown). CD68 was detected mainly in the cytoplasm, and the LXRα protein was detected in the nucleus. These results indicate that the LXRα protein is expressed in macrophage lineage cells in various stages of atherosclerotic lesions.
In this study, we established a novel anti-LXRα monoclonal antibody PPZ0412. The apparent molecular weight of the protein detected by PPZ0412 (47 kDa) is in good agreement with the expected molecular weight deduced from its reported cDNA sequence (447 amino acids). This antibody can be used for immunoblotting, immunoprecipitation, and also immunohistochemistry for endogenous or overexpressed human LXRα protein in various cells.
PPZ0412 did not cross-react with the LXRβ protein, although the amino acid sequence of the LXRβ protein is closely similar to that of LXRα. One of the problems we found during these experiments was the recognition of the lower-molecular-weight protein by PPZ0412. This lower-molecular-weight protein (* in Figure 1B, * in Figure IA, and * in Figure 2A, lane 3) was detected only in cells expressing the LXRα protein, and it was originally considered to be a degradation product. The result of immunoblot study using FLAG-tagged human LXRα (Figure I) supports the hypothesis, because an anti-FLAG antibody did not recognize the lower-molecular-weight protein. The FLAG epitope was fused to the LXRα protein at N-terminal domain, and if degradation products were generated by N-terminal truncation, the FLAG antibody could not recognize the low-molecular-weight degradation product. Recently, the presence of a splicing variant of LXRα mRNA was also reported. It is an open question whether this protein was a degraded LXRα protein or a splicing variant or characterized by nonspecific binding. Further analysis will be needed to characterize this additional immunoreactive protein. It was found that PPZ0412 is able to specifically recognize the endogenous human LXRα protein.
The expression pattern of the LXRα protein in human tissue is consistent with the previously reported profile of mRNA expression in both human tissue and experimental animals.5–7 Using PPZ0412, LXRα protein was detected in the Kupffer cells of the liver, alveolar macrophages in the lung, and in other macrophages resident in the thymus and spleen. LXRα-positive cells were also positive for CD68, a well-studied marker for macrophage lineage cells. This result clearly indicates that major cell types expressing the LXRα protein in the human body are macrophage lineage cells. Previously, Kohro et al reported that LXRα mRNA is the most highly induced transcriptional regulator during differentiation from human primary culture monocytes to macrophages with M-colony–stimulating factor or granulocyte/macrophage colony–stimulating factor.5 LXRα mRNA in human macrophage lineage cells was far higher than that found in the liver or other organs. These results provide evidence that macrophage lineage cells in various human organs are positive for the LXRα protein under normal physiological condition. Previously, induction of LXRα mRNA by addition of oxidized LDL was reported in THP-1 cells or mouse macrophages,12 and we have reported that LXRα gene is activated during macrophage differentiation without further stimulation in human monocyte-derived macrophages.5,15 To examine the effect of oxidized LDL, we investigated LXRα protein expression in macrophages. As depicted in Figure 2B, we could not detect the significant increase of LXRα protein by addition of oxidized LDL.
In addition to its detected presence in macrophage lineage cells, LXRα protein was also detected in the hepatocytes and adipocytes. Both of these cell types are actively involved in the metabolism, transport, and storage of lipids. The intensity of immunostaining was weaker in both hepatocytes and adipocytes than in macrophage lineage cells. Recently, Seo et al reported that treatment with LXR agonist enhanced adipocyte differentiation from primary human stromal vascular cells obtained from subcutaneous adipose tissue. Treatment of these cells with a synthetic LXR agonist resulted in markedly enhanced adipocyte differentiation.22 LXRα plays a role in the execution of adipocyte differentiation by regulation of both lipogenesis and adipocyte-specific gene expression.
The result of immunohistochemical study indicated that LXRα is expressed in macrophages present in atherosclerotic lesions. Cells in the lesion expressing LXRα were also positive for SR-A, indicating that they are active for the uptake of modified lipoprotein. The LXR signaling pathway in atherosclerosis has an established role in atherosclerosis.13,14,23 Joseph et al have shown that treatment with a synthetic LXR agonist GW3965 can reduce atherosclerotic lesion development in 2 mouse models (ie, LDLR−/− and apoE−/− mice).13 Terasaka et al have reported the effectiveness of another synthetic agonist T-0901317.23 Ligands for RXR, an LXR-heterodimer partner, were also efficacious in reducing atherosclerosis.24 Furthermore, bone marrow transplantation from LXRα/β−/− mice increases lesion formation in these same models.14 The abundant expression of LXRα protein in infiltrating macrophages supports the hypothesis that LXRα agonists have a beneficial effect against development of atherosclerosis in the arterial wall. If LXRα proteins were mainly located in the foam cells of atherosclerotic lesions, the activation of LXRα might activate the expression of ABC transporters and help eliminate accumulated cholesterol from the foam cells. Recently, Joseph et al and Fowler et al reported the reciprocal regulation of inflammation and lipid metabolism by LXR.25,26 These studies reported that LXR agonists can inhibit macrophage inflammatory gene expression. The presence of LXRα protein in infiltrating macrophages indicates that they are primed to respond LXRα agonists. Highly expressed LXRα proteins in various human tissues can respond to LXRα agonists, and may suppress the progression of inflammatory reactions under a variety of conditions. Further studies will be necessary to assess the effectiveness of treatment specifically targeted to LXRα activation. The monoclonal anti-human LXRα antibody described in this study will be a powerful tool to help analyze the precise expression, localization, and function of LXRα in human physiology and pathology and will greatly facilitate progress toward realizing the therapeutic potential suggested by the ongoing work in this field.
This work is supported by a grant from joint research & development projects with academic institutes and private companies. The authors acknowledge C. Nagao, A. Kikuchi, and A. Izumi for their excellent technical assistance and Dr Kevin Boru of Pacific Edit for review of the manuscript. We thank Dr David J. Mangelsdorf of the Howard Hughes Medical Institute at the University of Texas Southwestern Medical Center for providing the human LXRα expression vector for positive control.
- Received August 11, 2004.
- Accepted November 24, 2004.
Apfel R, Benbrook D, Lernhardt E, Ortiz MA, Salbert G, Pfahl M. A novel orphan receptor specific for a subset of thyroid hormone-responsive elements and its interaction with the retinoid/thyroid hormone receptor subfamily. Mol Cell Biol. 1994; 14: 7025–7035.
Willy PJ, Umesono K, Ong ES, Evans RM, Heyman RA, Mangelsdorf DJ. LXR, a nuclear receptor that defines a distinct retinoid response pathway. Genes Dev. 1995; 9: 1033–1045.
Kohro T, Nakajima T, Wada Y, Sugiyama A, Ishii M, Tsutsumi S, Aburatani H, Imoto I, Inazawa J, Hamakubo T, Kodama T, Emi M. J Atheroscler Thromb. Genomic structure and mapping of human orphan receptor LXR alpha: upregulation of LXRa mRNA during monocyte to macrophage differentiation. J Atheroscler Thromb. 2000; 7: 145–151.
Lu TT, Repa JJ, Mangelsdorf DJ. Orphan nuclear receptors as eLiXiRs and FiXeRs of sterol metabolism. J Biol Chem. 2001; 276: 37735–37738.
Annicotte JS, Schoonjans K, Auwerx J. Expression of the liver X receptor alpha and beta in embryonic and adult mice. Anat Rec A Discov Mol Cell Evol Biol. 2004; 277A: 312–316.
Repa JJ, Liang G, Ou J, Bashmakov Y, Lobaccaro JM, Shimomura I, Shan B, Brown MS, Goldstein JL, Mangelsdorf DJ. Regulation of mouse sterol regulatory element-binding protein-1c gene (SREBP-1c) by oxysterol receptors, LXRalpha and LXRbeta. Genes Dev. 2000; 14: 2819–2830.
Repa JJ, Turley SD, Lobaccaro JA, Medina J, Li L, Lustig K, Shan B, Heyman RA, Dietschy JM, Mangelsdorf DJ. Regulation of absorption and ABC1-mediated efflux of cholesterol by RXR heterodimers. Science. 2000; 289: 1524–1529.
Laffitte BA, Repa JJ, Joseph SB, Wilpitz DC, Kast HR, Mangelsdorf DJ, Tontonoz P. LXRs control lipid-inducible expression of the apolipoprotein E gene in macrophages and adipocytes. Proc Natl Acad Sci U S A. 2001; 98: 507–512.
Joseph SB, McKilligin E, Pei L, Watson MA, Collins AR, Laffitte BA, Chen M, Noh G, Goodman J, Hagger GN, Tran J, Tippin TK, Wang X, Lusis AJ, Hsueh WA, Law RE, Collins JL, Willson TM, Tontonoz P. Synthetic LXR ligand inhibits the development of atherosclerosis in mice. Proc Natl Acad Sci U S A. 2002; 99: 7604–7609.
Tangirala RK, Bischoff ED, Joseph SB, Wagner BL, Walczak R, Laffitte BA, Daige CL, Thomas D, Heyman RA, Mangelsdorf DJ, Wang X, Lusis AJ, Tontonoz P, Schulman IG. Identification of macrophage liver X receptors as inhibitors of atherosclerosis. Proc Natl Acad Sci U S A. 2002; 99: 11896–11901.
Watanabe Y, Tanaka T, Uchiyama Y, Takeno T, Izumi A, Yamashita H, Kumakura J, Iwanari H, Shu-Ying J, Naito M, Mangelsdorf DJ, Hamakubo T, Kodama T. Establishment of a monoclonal antibody for human LXRalpha: Detection of LXRalpha protein expression in human macrophages. Nucl Recept. 2003; 1: 1.
Watanabe A, Hippo Y, Taniguchi H, Iwanari H, Yashiro M, Hirakawa K, Kodama T, Aburatani H. An opposing view on WWOX protein function as a tumor suppressor. Cancer Res. 2003; 63: 8629–8633.
Takabe W, Kanai Y, Chairoungdua A, Shibata N, Toi S, Kobayashi M, Kodama T, Noguchi N. Lysophosphatidylcholine enhances cytokine production of endothelial cells via induction of L-type amino acid transporter 1 and cell surface antigen 4F2. Arterioscler Thromb Vasc Biol. 2004; 24: 1640–1645.
Jiang S, Tanaka T, Iwanari H, Hotta H, Yamashita H, Kumakura J, Watanabe Y, Uchiyama Y, Aburatani H, Hamakubo T, Kodama T, Naito M. Expression and localization of P1 promoter-driven hepatocyte nuclear factor-4alpha (HNF4alpha) isoforms in human and rats. Nucl Recept. 2003; 1: 5.
Seo JB, Moon HM, Kim WS, Lee YS, Jeong HW, Yoo EJ, Ham J, Kang H, Park MG, Steffensen KR, Stulnig TM, Gustafsson JA, Park SD, Kim JB. Activated liver X receptors stimulate adipocyte differentiation through induction of peroxisome proliferator-activated receptor gamma expression. Mol Cell Biol. 2004; 24: 3430–3444.
Claudel T, Leibowitz MD, Fievet C, Tailleux A, Wagner B, Repa JJ, Torpier G, Lobaccaro JM, Paterniti JR, Mangelsdorf DJ, Heyman RA, Auwerx J. Reduction of atherosclerosis in apolipoprotein E knockout mice by activation of the retinoid X receptor. Proc Natl Acad Sci U S A. 2001; 98: 2610–2615.
Fowler AJ, Sheu MY, Schmuth M, Kao J, Fluhr JW, Rhein L, Collins JL, Willson TM, Mangelsdorf DJ, Elias PM, Feingold KR. Liver X receptor activators display anti-inflammatory activity in irritant and allergic contact dermatitis models: liver-X-receptor-specific inhibition of inflammation and primary cytokine production. J Invest Dermatol. 2003; 120: 246–255.