Adipocytes As Immune Cells: Differential Expression of TWEAK

Adipocytes as Immune Cells: Differential Expression of TWEAK, BAFF, and APRIL and Their Receptors (Fn14, BAFF-R, TACI, and BCMA) at Different Stages of Normal This information is current as and Pathological Adipose Tissue of September 26, 2021. Development Vassilia-Ismini Alexaki, George Notas, Vassiliki Pelekanou, Marilena Kampa, Maria Valkanou, Panayiotis

Theodoropoulos, Efstathios N. Stathopoulos, Andreas Tsapis Downloaded from and Elias Castanas J Immunol published online 14 October 2009 http://www.jimmunol.org/content/early/2009/10/14/jimmuno l.0901186 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2009/10/13/jimmunol.090118 Material 6.DC1

Why The JI? Submit online. by guest on September 26, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published October 14, 2009, doi:10.4049/jimmunol.0901186 The Journal of Immunology

Adipocytes as Immune Cells: Differential Expression of TWEAK, BAFF, and APRIL and Their Receptors (Fn14, BAFF-R, TACI, and BCMA) at Different Stages of Normal and Pathological Adipose Tissue Development1

Vassilia-Ismini Alexaki,* George Notas,* Vassiliki Pelekanou,2* Marilena Kampa,* Maria Valkanou,* Panayiotis Theodoropoulos,† Efstathios N. Stathopoulos,‡ Andreas Tsapis,3§ and Elias Castanas3*

Adipose tissue represents a rich source of multipotent stem cells. Mesenchymal cells, isolated from this source, can differentiate

to other cell types in vitro and therefore can be used for a number of regenerative therapies. Our view of adipose tissue has recently Downloaded from changed, establishing adipocytes as new members of the immune system, as they produce a number of proinﬂammatory cytokines (such as IL-6 and TNF␣ and chemokines, in addition to adipokines (leptin, adiponectin, resistin) and molecules associated with the innate immune system. In this paper, we report the differential expression of TNF-superfamily members B cell activating factor of the TNF Family (BAFF), a proliferation inducing ligand (APRIL), and TNF-like weak inducer of apoptosis (TWEAK) in immature-appearing and mature adipocytes and in benign and malignant adipose tissue-derived tumors. These ligands act

through their cognitive receptors, BAFF receptor, transmembrane activator and calcium signal-modulating cyclophilic ligand http://www.jimmunol.org/ (TACI), B cell maturation Ag (BCMA), and fibroblast growth factor-inducible 14 (Fn14), which are also expressed in these cells. We further report the existence of functional BCMA, TACI, and Fn14 receptors and their ligands BAFF, APRIL, and TWEAK on adipose tissue-derived mesenchymal cells, their interaction modifying the rate of adipogenesis. Our data integrate BAFF, APRIL, and TWEAK and their receptors BCMA, TACI, and Fn14 as novel potential mediators of adipogenesis, in addition to their specific role in immunity, and define immature and mature adipocytes as source of immune mediators. The Journal of Immunology, 2009, 183: 5948–5956.

lthough adipose tissue has long been regarded as an al- The TNF superfamily is a group of receptors and ligands con- by guest on September 26, 2021 most inert tissue, dedicated solely to energy storage and sisting of 19 ligands and 29 receptors. They orchestrate a wide A release, recent data dramatically changed our opinion. range of biological functions, from the regulation of activation and Indeed, recent insights into the metabolic and immunological func- cell death in the immune system to tissue homeostasis and cancer tions of preadipocytes and adipocytes revealed that they are potent cell modulation (3). Among the different members of this family, producers of proinflammatory cytokines (such as IL-6 and TNF-␣) an increased interest has been shown recently for the subset of B and chemokines, regulating monocyte/macrophage function (1) cell activating factor of the TNF Family (BAFF),4 a proliferation and molecules associated with the innate immune system, such as inducing ligand (APRIL), and TNF-like weak inducer of apoptosis the C1q/TNF-related superfamily. Finally, preadipocytes and adi- (TWEAK). APRIL (CD256, TNF-SF13) (4–6) and BAFF pocytes express a broad spectrum of functional Toll-like receptors (CD257, TNF-SF13B) (6–9) have been identified as trophic fac- and can convert into macrophage-like cells (reviewed in Ref. 2). tors in lymphocyte malignancies and immune-related disorders These data clearly establish adipose tissue as a new player in im- (10). APRIL and BAFF have also been identified in a number of mune reactions. immune-related and immune-independent tissues (spleen, liver, lung, heart, intestine, kidney, and thymus) (11). They both bind to † *Department of Experimental Endocrinology, Department of Biochemistry, and two TNF-R family members, transmembrane activator and cal- ‡Department of Pathology, University of Crete, School of Medicine, Heraklion, Greece; and §UMR976 INSERM, Hoˆpital St Louis, Paris, France cium signal-modulating cyclophilic ligand (TACI) (TNFRSF13B) Received for publication April 14, 2009. Accepted for publication September 3, 2009. and B cell maturation Ag (BCMA) (TNFRSF17) (12, reviewed in The costs of publication of this article were defrayed in part by the payment of page Ref. 13). APRIL binds also syndecan-1 (CD138), a heparan sulfate charges. This article must therefore be hereby marked advertisement in accordance proteoglycan present on the surface of many cell types (14, 15), with 18 U.S.C. Section 1734 solely to indicate this fact. allowing ligand polymerization and cross-linking (14–17). BAFF 1 This work was supported by grants from INSERM and the University of Crete Research Committee. 2 Current address: Laboratoire J.-C. Heuson de Cance´rologie Mammaire, Institut Ju- les Bordet-Centre des Tumeurs de l’ULB, Brussels, 1000, Belgium. 3 Address correspondence and reprint requests to Dr. Elias Castanas, University of 4 Abbreviations used in this paper: BAFF, B cell activating factor of the TNF family; Crete, School of Medicine, Laboratory of Experimental Endocrinology, P.O. Box APRIL, a proliferation inducing ligand; TWEAK, TNF-like weak inducer of apopto- 2208, Heraklion, 71003, Greece. E-mail address: [email protected] or Dr. An- sis; TACI, transmembrane activator and calcium signal-modulating cytophilic ligand; dreas Tsapis, UMR976 INSERM, Hoˆpital St Louis, 1, Av. Claude Vellefaux, Paris, BCMA, B cell maturation Ag; BAFF-R, BAFF receptor; Fn14, fibroblast growth 75010, France. E-mail address: [email protected] factor-inducible 14; LPL, lipoprotein lipase; ADMC, adipose tissue-derived mesenchymal cell; MSC, mesenchymal stem cell; qRT-PCR, quantitative RT-PCR; SF, Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 Superfamily.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901186 The Journal of Immunology 5949 specifically binds to another TNF-R family member, BAFF-recep- Immunohistochemistry tor (BAFF-R) (TNFRSF13C) (18). Binding to these receptors trig- After deparaffinization and hydration, slides were subjected to three 5-min gers diverse signaling pathways, including translocation of NF␬B cycles of citrate buffer (0.01M, pH 6.0) incubation in a 500-W microwave or activation of mitogen-activated kinases (19, reviewed in Ref. oven and treated with 3% hydrogen peroxide for 15 min, as appropriate. 20). They were incubated for 30 min with primary Abs to APRIL (hAprily-8 TWEAK/Fn14 represents another ligand/receptor couple, also mouse mAb, 1/100 dilution, ALX-804–149, Alexis Biochemicals), BAFF (ALX-804–131 mAb/Buffy-2 clone, dilution 1/100, Alexis Biochemicals), mediating pluripotent responses. First described as a TNF-like TWEAK (sc-12405, 1/100 dilution, Santa Cruz Biotechnology), BCMA weak inducer of apoptosis, TWEAK (TNF-SF12) has since (Vicky-1 rat mAb, 1/100 dilution, ALX-804–151, Alexis Biochemicals), emerged as a cytokine, regulating multiple cellular responses, in- TACI (IMG-249 rabbit polyclonal Ab, 1/150 dilution, Imgenex), BAFF-R cluding proinflammatory activity, angiogenesis, cellular prolifera- (goat polyclonal, AF1162, dilution 1/100, R&D Systems), or Fn14 (mouse monoclonal, sc-56250, dilution 1/50, Santa Cruz Biotechnology). The tion, differentiation, and migratory capacity, suggesting a potential UltraVision LP Detection System (TL-060-AL, Lab Vision) with Fast Red, role in inflammation and cancer (reviewed in Ref. 21). TWEAK is the HRP detection kit (BD Biotechnologies) or the K0689 LSAB kits highly expressed by inflammatory cells, expressed as a type II- (DakoCytomation) were used for immunodetection, as appropriate. Coun- transmembrane protein that can be cleaved to function as a soluble terstaining was performed using Mayer’s hematoxylin. Known positive and cytokine (22). It acts through its cognitive receptor, fibroblast negative controls (omission of the primary antibody) were used in every run (supplemental Fig. 1). growth factor-inducible 14 (Fn14) (TNFRSF12A), identified also on nonlymphoid cell types and triggers multiple, often conflicting, Isolation, culture, and differentiation of mesenchymal cells cellular responses ranging from proliferation to cell death. Lipoaspirates were obtained by liposuction of s.c. adipose tissue of abdo- TWEAK, and in particular Fn14, was relatively low in normal men or hips of healthy donors (n ϭ 6), aged 42 Ϯ 5 years old (mean Ϯ SD) Downloaded from tissues, but TWEAK undergoes dramatic up-regulation in tissue after informed consent. Human ADMC were isolated as previously de- injury and diseases, whereas Fn14 is expressed by many tissue- scribed (27). Briefly, lipoaspirates were washed with PBS, incubated with 1 mg/ml collagenase type I (Sigma-Aldrich) for 45 min at 37°C, and tissue resident progenitor cells (23, 24). More recently, the concept that was centrifuged at 1200 rpm for 10 min. Erythrocytes were removed by a inflammatory cells regulate tissue regeneration via their effects on 10-min incubation at room temperature in an erythrocyte lysis buffer (150 tissue progenitor cells has also been advanced (25). In this context, mM NH4Cl, 10 mM NaHCO3, 1 mM EDTA, pH 7.3) and isolated ADMC it is now evident that TWEAK is a multifunctional cytokine, sim- were cultured in low glucose (1g/L) DMEM, supplemented with 10% FBS http://www.jimmunol.org/ and penicillin/streptomycin at 37°C and 5% CO , in a humidified atmo- ilar in this regard to its sibling TNF-␣. 2 sphere (28). Two days after seeding, floating cells were removed and there- Until recently, the presence of BAFF and APRIL or their re- after medium was changed every 3 days. Cells were passaged by ceptors in adipose tissue was not evidenced. However, in a recent trypsinization after reaching 70–90% confluence. work, we have reported the expression of BAFF, but not APRIL, For adipogenic induction, human ADMC were cultured in adipogenic in mature breast adipocytes (26). In this study, we performed a medium consisting of basal medium supplemented with 10 ␮g/ml insulin, 1 ␮M dexamethasone, 0.5 mM methylxanthine, and 200 ␮M indometha- thorough investigation of BAFF, APRIL, and TWEAK expression cin. Medium was changed every 2 days and after 2 weeks of culture, in normal, mature, and immature-appearing adipose tissue cells adipogenesis was evaluated by Oil Red O staining. For the latter, cells were from different origins (s.c., visceral, and perirenal fat tissue). We fixed with 4% formaldehyde for 12 h, stained for 45 min with 2 volumes report the expression of these ligands, as well as their respective 0.35% Oil Red O (diluted in isopropanol)/1 volume PBS and thoroughly by guest on September 26, 2021 receptors (BAFF-R, BCMA, TACI, and Fn14) at different stages washed with PBS. Pictures were taken with an inverted Leica microscope. For quantification of staining, the dye was solubilized with isopropanol and of human adipose tissue maturation, establishing this tissue as a red color intensity (proportional to the lipid content of cells) was measured novel source of TNF members and strengthening the link between in a plate photometer at 600 nm. Results were normalized per mg of total adipose tissue and the immune system. This distribution is further protein. altered in nonmalignant and malignant adipose tissue-derived tu- Direct immunofluorescence mors (lipoma, hibernoma, liposarcoma). The existence of their cognitive receptors, however, suggests an additional role for these ADMC (passage-2) were characterized by flow cytometry using the fol- lowing Abs conjugated to fluorochromes: anti-CD2FITC, anti-CD3FITC, ligands as auto- or paracrine modulators of the adipose tissue. We anti-CD4FITC, anti-CD8FITC, anti-CD1SFITC, anti-CD19PE, anti-HLA-DR have therefore examined their role during adipogenesis, using PE, anti-CD13PE, anti-CD14FITC, anti-CD34PE, anti-CD45FITC, anti- adult adipose tissue-derived mesenchymal cells (ADMC). We re- CD54PE, anti-CD73PE, anti-CD138FITC, anti-CD90FITC (all from BD port that TWEAK and, to a lesser extent, BAFF and APRIL are Pharmingen); anti-CD31 (Thermo Fisher Scientific); anti-CD44FITC potent modulators of adipogenesis, a result with potential applica- (J173; Beckman-Coulter); anti-CD133PE (Miltenyi Biotec); and anti- CD105PE and anti-CD144FITC (Santa Cruz Biotechnology). In brief, cells tions in the control of adiposity in humans. were detached by gentle scrapping, washed in PBS/1%BSA, and incubated with Abs, according to the manufacturer’s instructions. Fluorochrome-con- Materials and Methods jugated IgG1 or IgG2a were used as negative controls. Staining was quantified by flow cytometry using a FACScan operator (BD Biosciences) and Tissue selection results were analyzed with the FACScan and Cell Quest softwares (BD Specimens of human skin or breast or renal peritumoral fat (n ϭ 20, 52, Biosciences). and 32, respectively) were obtained from the University of Crete, School of Indirect immunofluorescence Medicine, Department of Pathology (Crete, Greece) tissue bank, after permission of the University Hospital Research and Ethics Committee. The For identification of TNFR-Superfamily (TNFR-SF) members, primary demographic characteristics of the patients are presented in supplemental Abs against BAFF-R (goat polyclonal, AF1162, dilution 1/100; R&D Sys- Table I.5 Furthermore, one hibernoma, two cases of lipomas, and two li- tems); BCMA (Vicky-1 rat mAb, 1/100 dilution, ALX-804–151; Alexis posarcomas were also examined. Eight tissue slides (3 ␮m) were cut se- Biochemicals); TACI (IMG-249 rabbit polyclonal Ab, 1/150 dilution; Im- rially, one for H&E staining and seven for specific immunostaining for genex); Fn14 (mouse monoclonal, sc-56250, dilution 1/50), TNF-␣ (2C8, BAFF, APRIL, TWEAK, BAFF-R, BCMA, TACI, and Fn14. The slides 1/50), and TNFR1 (H-271, 1/50) (Santa Cruz Biotechnology); and TNFR2 were reviewed by two investigators independently and blindly and the final (PA1–21148, 1/50; Thermo Fisher Scientific) have been used. Briefly, cells conclusion was reached in consensus. In the few cases with discrepancy were harvested by gentle scraping, washed with PBS/10%FBS/0.2% NaN3, between the two observers, the immuno-stained slides were reviewed in a and incubated with primary Abs at 100 ␮g/ml for 30 min at 4°C. Cells were double viewing microscope, so that the discrepancy was settled. then washed with ice-cold PBS and incubated with the according second-

ary Abs in 3%BSA/PBS/0.2% NaN3 for 30 min at 4°C in the dark. Donkey anti-goat-Alexa 546 and goat anti-rabbit-FITC (Molecular Probes) were 5 The online version of this article contains supplemental material. used at 1:1000, and rat anti-mouse-PE (BD Pharmingen) was used at 1:50. 5950 TNF-SF MEMBERS IN ADIPOCYTES

FIGURE 1. Expression of APRIL, BAFF, TWEAK, and their receptors (BCMA, TACI, BAFF-R, and Fn14) in immature- and mature-appearing adipose tissue. APRIL, BAFF, TWEAK, BCMA, TACI, BAFF-R, and Fn14 staining is presented in selected cases of immature-appearing or mature adipose tissue. Arrowheads show vascular endothelial cells. Bar ϭ 50 ␮m. Downloaded from http://www.jimmunol.org/

Cells were again washed and resuspended in PBS/3%BSA/0.2% NaN3. For TNF superfamily BAFF, APRIL, TACI, BCMA, BAFF-R, TWEAK, and fluorescent staining against BCMA, we used a biotinylated goat anti-rat FN14 between adipocytes and MSCs. secondary Ab (used at 1/100; Sigma-Aldrich), and FITC-conjugated avidin (1/200; BD Biosciences). Staining was quantified by flow cytometry. Assay of secreted APRIL, BAFF, and TWEAK Detection of functional TNFR-SF members on cells Assays of secreted TNF-SF ligands in the culture medium were performed on supernatant medium after 24 h incubation with cells. The ELISA for

Cells were harvested in cold 1 mM EDTA in PBS and incubated for 20 min these ligands were provided by Bender MedSystems (catalogue numbers: by guest on September 26, 2021 in a buffer containing 10% human serum, 1 mM EDTA, 1% BSA, and human BAFF BMS2007INST, human APRIL BMS2008, human TWEAK

0.2% NaN3 in PBS. FLAG-BAFF, FLAG-APRIL, or FLAG-TWEAK BMS2006INST). The sensitivity of the assays was 130, 400, and 9.8 pg/ml were added to a concentration of 100 ng/ml. Incubation buffer was used as for BAFF, APRIL, and TWEAK, respectively, whereas the corresponding a mock control. Then monoclonal mouse (M2; Sigma-Aldrich), anti-FLAG intra-assay coefficient of variation were 6.7%, 7.3% and 7.9%, respec- (8 ␮g/ml), and FITC-conjugated goat anti-mouse IgG (1:50) were added tively. Due to the assay characteristics, production of BAFF, APRIL, and sequentially. Staining was quantified by flow cytometry using the afore- TWEAK content was assayed in 5ϫ concentrated supernatant medium. mentioned operator. Statistical analysis Quantitative RT-PCR (qRT-PCR) Statistical analysis was performed with the Statistical Package for Social Total RNA was isolated from cell cultures with the Miniprep RNA isola- Sciences, v 16 (SPSS) using the appropriate parametric tests. Statistical Ͻ tion kit (Macherey-Nagel) and from harvested (adipose) tissue with Trizol significance was set to p 0.05. Reagent (Invitrogen). After DNase treatment (Invitrogen), absence of DNA was verified by a control semiquantitative PCR using Platinum TaqDNA Results polymerase (Invitrogen) for GAPDH. RNA (1 ␮g) was subjected to Ther- Expression of BAFF, APRIL, TWEAK, and their receptors in moscript RT-PCR (Invitrogen), using primers as presented in supplemental adipose tissue sections Table II (VBC Biotech). Real-time PCR with SYBR Green was performed with DyNAmo SYBR Green quantitative PCR kit with ROX (Finnzymes) Using immunological staining, we have tested for the presence of according to the manufacturer’s instructions, using the ABI Prism 7000 BAFF, APRIL, TWEAK, and their receptors in adipose tissue sec- sequence detection system (Applied Biosystems). The reaction conditions tions (Fig. 1). APRIL was constantly expressed in visceral adipose for real-time PCR were 95°C for 3 min, followed by 40 cycles of 95°C for 15 s and then 60°C for 60 s. Changes were normalized according to 18S tissue (perinephric fat), in which adipocytes appeared immature RNA expression. Each set of primers was tested with at least three different with multilocular cytoplasm (herein described as immature-ap- RNA samples treated independently. ADMC mRNA expression was used pearing adipocytes). Staining was restricted in well described in- as control. tracellular dots in specific cytoplasmic areas in relation to the Analysis of published gene-array data mode of APRIL synthesis and secretion through secretory vesicles. In contrast, in peripheral adipocytes (s.c. fat, with mature-appear- Published gene array data that included expression analysis of mesenchy- ing adipocytes, herein described as mature cells), APRIL immu- mal stem cells (MSCs) that were driven to adipocytes in vitro were used. In the Gene Expression Omnibus database, we identified dataset GSE9451 noreactivity was absent. However, BAFF immunoreactivity was (K. Shimizu et al., Hiroshima University, Hiroshima, Japan) that included constantly present in immature-appearing and mature adipose tis- gene array data from MSCs that were in vitro differentiated to either fi- sue. BAFF-immunoreactivity was uniform in the cytoplasm, while broblast, osteoblast, chondrocyte, or adipocyte in triplicates. The control cell periphery was more heavily stained. In some cases, an addi- (MSC) and adipocyte gene-array data were extracted and loaded to Gene- spring GX V10.0.1 (Agilent Technologies) in order to calculate fold tional perinuclear staining was also observed. BAFF, but not changes in gene expression (adipocytes/MSCs), and to identify statistically APRIL, was also expressed in the vascular endothelium (Fig. 1, significant changes (unpaired t test) in the expression of members of the arrowheads). The Journal of Immunology 5951

Table I. Relative distribution of TNF-SF ligands and receptors in immature- and mature-appearing human adipocytes, adipose tissue-derived tumors, adipose tissue-derived mesenchymal cells, and differentiated adipocytesa

Normal Tissue Adipose Tissue-Derived Tumors Isolated Cells

Immature-Appearing Mature-Appearing Differentiated Adipocytes Adipocytes Endothelial Cells Lipoma Hibernoma Liposarcoma ADMC Adipocytes

APRIL ϩϩϩ Ϫ Ϫ ϩ ϩϩ ϩϩϩ/Ϫϩ ϩ BAFF ϩϩϩ ϩϩϩ ϩϩ ϩϩϩ ϩϩϩ ϩϩϩ/Ϫ ϩϩ ϩϩϩ BAFFR ϩϩϩ ϩϩ Ϫ ϩϩϩ ϩϩϩ ϩϩ/Ϫϩ ϩ BCMA ϩϩ Ϫ Ϫ Ϫ ϩϩ ϩ/Ϫϩ ϩϩ TACI ϩϩ Ϫ ϩ ϩ ϩϩ ϩϩ/Ϫϩϩ ϩ TWEAK ϩϩ/Ϫ Ϫ Ϫ Ϫ ϩϩ ϩϩ ϩϩ ϩϩϩ Fn14 ϩϩ ϩ ϩϩ ϩ ϩϩϩ ϩϩ/Ϫ ϩϩϩ Membrane ϩϩ Intracellular

a The number of crosses represents the intensity of staining. ϩ/Ϫ, heterogeneous staining with areas of positive (ϩ) and negative (Ϫ) staining.

The ﬁnding that BAFF and APRIL are expressed in normal resembling immature-appearing adipose tissue (30), and two cases adipose tissue suggests a possible local action. Therefore, we of malignant liposarcomas (Fig. 2). have assayed whether their receptors BCMA, TACI, and Lipomas showed staining for BAFF and BAFF-R; Fn14 staining

BAFF-R are also present. BCMA and TACI were equally present was also present, but at lower intensity than in normal mature Downloaded from in immature-appearing adipocytes, but not in mature adipose tissue. It is notable that vessels (Fig. 1, arrowheads) were negative for BCMA but moderately positive for TACI staining. In contrast, BAFF-R was constantly present in immature-appearing and mature adipocytes, but negative in endothelial cells. These data sug-

gest that in mature adipose tissue, BAFF might exert its action http://www.jimmunol.org/ through the equally expressed BAFF-R, while in immature-appearing adipose tissue both BAFF and APRIL, as well as their three receptors (BCMA, TACI, and BAFF-R) are expressed, suggesting more complex interactions. TWEAK was heterogeneously present in immature-appearing adipose tissue, in which negative and positive areas were found, while no TWEAK expression has been evidenced in mature adipose tissue. In contrast, Fn14 was always present as discrete membrane dots, both in immature-appearing and mature adipose tissue, by guest on September 26, 2021 although mature adipocytes exhibited an apparent lower staining of Fn14 compared with that of immature tissue. In addition, blood vessels (Fig. 1, arrowheads) were constantly negative for TWEAK but positive for Fn14 staining. A comparison of staining intensities of TNF-SF ligands and receptors is presented in Table I. The above data indicate that TNF-SF members APRIL, BAFF, and TWEAK are differentially expressed in the adipose tissue. TNF-␣ has been reported to be produced by adipocytes, preadipocytes, and inﬁltrating macrophages in the adipose tissue of (predominantly obese) humans, exerting auto- or paracrine effects (29). In this paper, we report that TNF-␣ and its receptors (TNFR1 and TNFR2) were also expressed differentially in adipose tissue (supplemental Fig. 2). TNF-␣ was preferentially expressed in immature-appearing adipocytes and endothelial cells. The same pattern was found for TNFR2. In contrast, TNFR1 present in mature-appearing adipocytes with a greater intensity. Endothelial cells were equally stained for TNFR1 and TNFR2.

Expression of BAFF, APRIL, TWEAK, and their receptors in adipose tissue-derived tumors The above data indicate a differential expression of BAFF, APRIL, TWEAK, and their receptors in nonneoplastic, immature-, and mature-appearing adipose tissue in a different pattern than TNF-␣ and its receptors. This result suggests a possible differential role of these ligands in adipogenesis. We have assayed whether the expression of these molecules varies in adipose tissue tumors, in which proliferation and differentiation of adipocytes display enor- FIGURE 2. Expression of APRIL, BAFF, TWEAK, and their receptors mous variability. We have examined three cases of lipomas (be- (BAFF-R, TACI, BCMA, and Fn14) in adipose tissue-derived benign and nign tumors consisting of nonmalignant normal-appearing adipo- malignant tumors. A selected case of lipoma (left column), hibernoma cytes), one case of hibernoma, a rare benign brown fat tumor (middle column), and liposarcoma (right column) is shown. Bar ϭ 50 ␮m. 5952 TNF-SF MEMBERS IN ADIPOCYTES

FIGURE 3. Expression of BAFF, APRIL, TWEAK, and their receptors (BCMA, TACI, BAFF-R, and Fn14) in ADMC and differentiated adipocytes. Bottom left, Phase-contrast mi- crophotograph of ADMC and the same culture after adipocyte differentiation, both in phase contrast and after Oil Red O staining. Bar ϭ 50 ␮m. Downloaded from http://www.jimmunol.org/ adipocytes, and TWEAK staining was absent. In this respect, ex- BAFF-R was maintained at low levels. In contrast, an intense pression of these TNF-SF members in lipoma resembles the ex- membrane staining for Fn14 was found in ADMC; this staining pression observed in mature-appearing adipose tissue. However, was decreased after adipocyte differentiation and was mainly an additional low APRIL and TACI staining was evidenced. Hi- found intracellularly (Table I). bernoma was also heavily stained for BAFF, BAFF-R, and Fn14, ADMC cells, either nondifferentiated or differentiated toward whereas APRIL, TACI, BCMA, and TWEAK staining was also adipocytes, do not express (Ͻ1%) CD2, CD3, CD4, CD8, CD15, evident in a pattern similar to that observed in immature-appearing CD19, HLA-DR, and CD45, indicative of the nonexistence of he- adipose tissue. It is interesting that this staining was restricted to mopoietic cells (supplemental Table IV). In this respect, our data by guest on September 26, 2021 immature-appearing adipocytes, whereas, in mature appearing adi- indicate that TNF-SF ligands and receptors are expressed by adipocytes, the latter molecules were absent. Finally, in liposarcomas, pocytes themselves and that ADMC do not differentiate to blood a heterogeneous APRIL, BAFF, BAFF-R, BCMA, TWEAK, and cells (lymphocytes, macrophages/monocytes, granulocytes) that Fn14 staining was observed, whereas TACI presented a homoge- may also secrete these molecules. These results are consistent with neous staining, suggesting a potential role of this receptor in the immunostaining data presented in Figs. 1–3. development of sarcoma tumors (Table I). The atypical neoplastic The above data in isolated cells, cultured on plastic, present adipocytes show staining patterns similar to that of hibernomas. some discrepancies from those presented in tissue sections. Indeed, in fully differentiated adipose tissue, neither APRIL, TWEAK, TNF-SF ligands and receptors immunocytochemistry in ADMC TACI, nor BCMA are expressed, although they continue to be and in in vitro differentiated adipocytes expressed in isolated differentiated adipocytes. This result provides Our results indicate that normal, immature-appearing, and mature evidence about a more complex regulation of the TNF-SF system adipose tissue and adipose tissue-derived tumors express differen- in the adipose tissue, suggesting that potential interactions between tially BAFF, APRIL, TWEAK and their receptors. To further ex- immature-appearing adipocytes and humoral or tissue factors plore this differential expression, we isolated ADMC and followed might regulate their expression. the expression of these molecules during adipogenesis. These cells The expression of the above TNF-SF members and receptors were found to present striking similarities with bone-marrow-de- was further verified by qRT-PCR on total RNA isolated from rived mesenchymal cells, presenting also an additional partial ADMC, undifferentiated or differentiated in vitro toward adipo- staining for endothelial markers (supplemental Table III). cytes (Fig. 4A). Adipocyte differentiation was verified both phe- We investigated the presence of TNF-SF ligands (BAFF, notypically (Oil Red O staining; Fig. 3) and with qRT-PCR of APRIL, TWEAK) and their receptors (BAFF-R, TACI, BCMA, lipoprotein lipase (LPL) and peroxisome proliferator activated re- Fn14) both in ADMC and in vitro differentiated adipocytes (Fig. ceptor ␥ (Fig. 4B). After 2 weeks of differentiation, results of 3). APRIL presented a weak staining in ADMC, found as discrete BAFF and BAFF-R transcripts were comparable with their immu- intracellular dots, present also in differentiated cells. ADMC and nostaining on ADMC and differentiated cells. APRIL, as well as differentiated adipocytes expressed both BAFF and TWEAK. Dif- TACI and BCMA, exhibited a significant increase of their expres- ferentiated cells were more heavily stained than their precursors. sion in differentiated cells as compared with that of ADMC, par- The cognate receptors of these TNF-SF members were also evi- alleling the increased staining found in differentiated adipocytes. It denced in ADMC and in vitro differentiated adipocytes. BAFF-R, is interesting that Fn14 expression decreased dramatically in dif- BCMA, and TACI, receptors for BAFF and APRIL, were present ferentiated cells, paralleling the decreased staining observed in dif- in ADMC. After adipogenesis, BCMA staining intensity was in- ferentiated cells, whereas TWEAK mRNA expression showed a creased, TACI staining was decreased, while the presence of slight increase in differentiated adipocytes. In addition, the changes The Journal of Immunology 5953

that the observed BCMA membrane expression is rare, because this receptor is usually identiﬁed intracellularly (12). APRIL has been reported to bind also to syndecan-1 (a membrane heparan sulfate proteoglycan) (31, 32) through its N-terminal domain, inducing its oligomerization, TACI- and/or BCMA- mediated activation, and resulting in the generation of migration or survival signals (14, 15). However, we have not detected the presence of syndecan-1 (CD138) on adipocyte-derived mesenchymal cells (not shown). Furthermore, FLAG-tagged APRIL binding on ADMC was not decreased in presence of a 250-fold molar concentration of heparin, suggesting that membrane proteoglycans might not be implicated in the binding of APRIL and therefore in its potential action in ADMC (supplemental Fig. 3). In conclusion, BAFF, APRIL, and TWEAK could exert their effects through functional TACI, BCMA, and Fn14 receptors expressed on ADMC. We further investigated whether the addition of human recombinant BAFF, APRIL, and TWEAK can modify adipogenesis.

FIGURE 4. A, Expression of BAFF, APRIL, TWEAK, and their recep- However, as these ligands have, in other systems, notable effects Downloaded from tors (BCMA, TACI, BAFF-R, and Fn14) mRNA in ADMC and the same on cell proliferation and survival, we have tested whether this Ϯ cells after 1 or 2 wk of adipocyte differentiation. Mean SEM of ﬁve might be the case in ADMC differentiated toward adipocytes. As different cases. B, Expression of LPL and PRAP␥ mRNA in ADMC un- presented in supplemental Fig. 4, neither ligand (100 ng/ml) had an differentiated and differentiated for 1 and 2 wk toward adipocytes. C, Anal- ysis of the NCBI GSE9451 gene-array dataset for the expression of effect on cell survival after a short or a long incubation time period TNF-SF ligands and receptors, as in A. (4 and 7 days, respectively). Conversely, at the same concentra-

tions, we observed an increased differentiation of ADMC toward http://www.jimmunol.org/ adipocytes in the presence of BAFF, whereas addition of APRIL in the expression of these TNF-SF ligands and their receptors fol- and especially of TWEAK significantly decreased the adipogenic low in a parallel way the gradual differentiation of ADMC toward differentiation. Addition of APRIL resulted in a dose-dependent adipocytes, which is evidenced by a gradual increase of adipogenic decrease of adipogenesis after 14 days, whereas the inhibitory ef- markers (LPL, PPAR␥; Fig. 4B) expression, with the exception of fect of TWEAK was evident as early as day 5 (not shown) and Fn14 expression, which shows a dramatic decrease, suggesting a maximal at concentrations as low as 25 ng/ml (Fig. 5C). This result potential inhibitory role of the tandem TWEAK/Fn14 in adipogen- further suggests the involvement of BAFF as a trophic factor of the esis (Fig. 4A). Our results are partially in accordance with the mature adipose tissue with a proadipogenic potential and provides results of the GSE9451 gene-array dataset, which includes samples a hint about the absence of the antiadipogenic APRIL, BCMA, and by guest on September 26, 2021 from MSCs that were in vitro driven to adipogenesis (Fig. 4C). TACI system in mature adipose tissue. However, the absence on Analysis of these data showed a significant reduction of Fn14 cell membranes of mature adipocytes of Fn14 suggests an addi- (4.61-fold reduction, p Ͻ 0.0001) as compared with mesenchymal tional role of TWEAK/Fn14. progenitor cells, whereas changes of the other transcripts were As presented above, Fn14 is constantly present in endothelial minor. The differences observed might be due to the different or- cells. Therefore, the system TWEAK/Fn14 could play a role in the igin of mesenchymal cells used in the GSE9451 gene-array dataset angiogenesis of the adipose tissue. To investigate a possible role of study (bone marrow-derived cells) as compared with ADMC used TWEAK/Fn14 in angiogenesis and as ADMC contain endothelial in our study. precursor cells (33, 34), we have investigated the expression of the Traces of secreted APRIL were detected in ADMC supernatants endothelial marker ICAM-1 (CD54). ICAM-1 is an intercellular with ELISA before or after adipogenic differentiation. However, adhesion factor, constantly present in leukocytes and endothelial we were unable to detect BAFF in ADMC supernatant medium by cells (35) that is greatly enhanced upon cytokine stimulation (IL-1 ELISA. TWEAK was detected in ADMC supernatants at 5.08 Ϯ and TNF-␣). Incubation of ADMC with TWEAK results in an 0.90 ng/pg protein, while it was decreased to 2.87 Ϯ 0.23 and enhanced expression of CD54 (Fig. 5D), a result not observed in 3.14 Ϯ 0.70 ng/pg protein after 1 and 2 weeks of adipogenesis, the case of BAFF or APRIL (not shown). respectively. Discussion Effect of BAFF, APRIL, and TWEAK on adipogenesis After their initial characterization, MSCs were found within the As presented above, isolated ADMC express BAFF, APRIL, and connective tissue of many organs (36), although their proliferation TWEAK, as well as BCMA, TACI, and Fn14, whereas BAFF-R is and differentiation characteristics vary. Similarities in surface expressed at low quantities. The existence of both ligands and marker profiles, morphology, and differentiation potentials assayed receptors on the same cell population is indicative of a possible in vitro led to suggestions that each of these multipotent cell types autocrine/paracrine action and suggests a possible role of these arise from a common adult progenitor cell (37) and then may adopt factors in adipogenesis. Indeed, we found that FLAG-tagged re- tissue-specific attributes according to its particular niche (38). Ad- combinant BAFF, APRIL, and TWEAK can bind on ADMC, ev- ipose tissue represents a rich source of such cells. Indeed, mesen- idencing the existence of functional receptors on the cell surface chymal cells, isolated from this easy source (28), can differentiate (Fig. 5A). The presence of BCMA, TACI, and Fn14 molecules on to other cell types in vitro (27) and therefore have been used for a ADMC surface was also confirmed by flow cytometry (Fig. 5B), number of regenerative therapies currently under investigation. although we have not observed the presence of BAFF-R. These However, recent data indicate that adult mesenchymal cells can results suggest that APRIL, BAFF, and TWEAK possibly act on also be transformed in cancer-associated fibroblasts, a necessary adipocytes precursors through these three receptors. It is notable stroma element promoting the growth of epithelial solid tumors 5954 TNF-SF MEMBERS IN ADIPOCYTES Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 5. A, Binding of FLAG-tagged APRIL, BAFF, and TWEAK on ADMC assayed by ﬂow cytometry. Black hollow curves represent the omission of the corresponding ligand, while grey curves represent binding of FLAG-tagged APRIL, BAFF, and TWEAK. B, Binding of anti-BAFF-R, anti-BCMA, anti-TACI, and anti-Fn14 Abs on ADMC by ﬂow cytometry. Hollow curves show isotype matching controls (replacement of Abs by corresponding pre-immune Igs), while grey curves show binding of anti-BAFF-R, anti-TACI, anti-BCMA, and anti-Fn14. C, Effect of BAFF, APRIL, and TWEAK on adipocyte differentiation of ADMC. Mean Ϯ SEM of two different cases, assayed in triplicate. D, Effect of APRIL, BAFF, and TWEAK on the ICAM-1 (CD54) expression in ADMC.

(39, 40). In this respect, a detailed analysis of the differentiation way, regulating the differentiation potential of these cells. We also potential and the involved mechanisms are of primary importance. show that this expression is modulated along the differentiation of The results of the present study integrate this effort by reporting ADMC to mature adipocytes. BAFF, APRIL, and TWEAK and that the subset of the TNF-SF ligands APRIL, BAFF, and TWEAK their receptors were, for a long time, regarded as elements of im- and their receptors BCMA, TACI, BAFF-R, and Fn14 are involved mune system, regulating the differentiation and fate of B lympho- in the differentiation process of ADMC toward adipocytes and cytes (reviewed in Ref. 44). However, in recent years these mol- (possibly) endothelial cells. In this respect, they may provide an ecules have been identified in a number of normal and pathological additional spectrum of agents for use in therapeutic interventions. tissues (11, 21, 45–47). In this paper, we report, for the first time, TNF was the first proinflammatory mediator shown to be se- that adipose tissue-resident adult mesenchymal cells express and creted by adipocytes (Ref. 41 and supplemental Fig. 2). Addition- synthesize these ligands. In addition, the same population of mes- ally, recent data identified a whole new family of C1q-TNF mem- enchymal cells expresses their cognitive receptors BAFF-R, TACI, bers, synthesized and secreted by the adipose tissue (42, 43), BCMA, and Fn14 on the surface of the cells. An interesting result including adiponectin, C1q, and CORS-26 (collagenous repeat is the rare localization of BCMA on the cell membrane of ADMC. containing sequence of 26 kDa protein, also known as CTRP-3). They are highly expressed and secreted by adipose tissue (re- Indeed, usually, staining for BCMA is found intracellularly, lo- viewed in Ref. 2) and linked to a number of processes, such as cated in the Golgi apparatus (12). In addition, Fn14 (the specific inflammation, apoptosis, autoimmunity, host defense, hibernation, TWEAK receptor) is predominantly present at the membrane level organogenesis, cell differentiation, metabolism, and insulin signal- in ADMC, whereas it translocates intracellularly in in vitro ma- ing. In this paper, we report that other members of the TNF-SF turing adipocytes. These receptors are functional as they can bind (BAFF, APRIL, and TWEAK) are also synthesized and secreted in vitro their corresponding ligands. from adipocytes and ADMC, which can act in an auto/paracrine The Journal of Immunology 5955

The expression of ligands and receptors on ADMC indicate a References possible role in their differentiation, expanding their potential tar- 1. Tsatsanis, C., V. Zacharioudaki, A. Androulidaki, E. Dermitzaki, gets beyond the immune system. We report that BAFF enhances I. Charalampopoulos, V. Minas, A. Gravanis, and A. N. Margioris. 2006. Pe- ripheral factors in the metabolic syndrome: the pivotal role of adiponectin. Ann. adipogenesis, while APRIL and especially TWEAK inhibit it. NY Acad. Sci. 1083: 185–195. APRIL, in contrast to many other TNF-related ligands, is a factor 2. Schaffler, A., J. Scholmerich, and B. Salzberger. 2007. Adipose tissue as an promoting proliferation (4, 48) of cells in tissue culture. Addition immunological organ: Toll-like receptors, C1q/TNFs and CTRPs. Trends Immu- nol. 28: 393–399. of recombinant APRIL or transfection of tumor lines with APRIL 3. Bodmer, J. L., P. Schneider, and J. Tschopp. 2002. The molecular architecture of provided a significant growth advantage to lymphoid and nonlym- the TNF superfamily. Trends Biochem. Sci. 27: 19–26. phoid cells (49). Our data show that APRIL decreases adipocyte 4. Hahne, M., T. Kataoka, M. Schroter, K. Hofmann, M. Irmler, J. L. Bodmer, P. Schneider, T. Bornand, N. Holler, L. E. French, et al. 1998. APRIL, a new differentiation, whereas no effect of APRIL on ADMC prolifera- ligand of the tumor necrosis factor family, stimulates tumor cell growth. J. Exp. tion was evident (not shown), suggesting other roles of this ligand Med. 188: 1185–1190. in ADMC that have yet to be determined. It is interesting to note 5. Kelly, K., E. Manos, G. Jensen, L. Nadauld, and D. A. Jones. 2000. APRIL/ TRDL-1, a tumor necrosis factor-like ligand, stimulates cell death. Cancer Res. that in vitro differentiated adipocytes continue to express all 60: 1021–1027. ligands and receptors, suggesting that the final differentiation 6. Shu, H. B., W. H. Hu, and H. Johnson. 1999. TALL-1 is a novel member of the process is the result of a fine interplay between elements of this TNF family that is down-regulated by mitogens. J. Leukocyte Biol. 65: 680–683. 7. Gross, J. A., J. Johnston, S. Mudri, R. Enselman, S. R. Dillon, K. Madden, system. In addition, Fn14 internalizes in differentiated adipocytes. W. Xu, J. Parrish-Novak, D. Foster, C. Lofton-Day, et al. 2000. TACI and BCMA However, TWEAK has been recognized in many tissue-resident are receptors for a TNF homologue implicated in B-cell autoimmune disease. progenitor cells (23, 24) and a potential role in tissue regeneration Nature 404: 995–999. 8. Schneider, P., F. MacKay, V. Steiner, K. Hofmann, J. L. Bodmer, N. Holler, Downloaded from has been proposed (25). In addition, TWEAK enhances endothelial C. Ambrose, P. Lawton, S. Bixler, H. Acha-Orbea, et al. 1999. BAFF, a novel cell maturation of ADMC, a result supporting the recent finding of ligand of the tumor necrosis factor family, stimulates B cell growth. J. Exp. Med. the perivascular residence of ADMC (50) and providing a biolog- 189: 1747–1756. 9. Moore, P. A., O. Belvedere, A. Orr, K. Pieri, D. W. LaFleur, P. Feng, D. Soppet, ical support of the reported role of the system (TWEAK/Fn14) in M. Charters, R. Gentz, D. Parmelee, et al. 1999. BLyS: member of the tumor the pathogenesis of atherogenesis (51). necrosis factor family and B lymphocyte stimulator. Science 285: 260–263. The role and regulation of this TNF-SF system appears to be 10. Ng, L. G., C. R. Mackay, and F. Mackay. 2005. The BAFF/APRIL system: life beyond B lymphocytes. Mol. Immunol. 42: 763–772. more complicated at the organism level. Indeed, staining of tissue 11. Guan, Z. B., W. B. Dan, Y. Shui, J. L. Ye, and S. Q. Zhang. 2007. cDNA cloning, http://www.jimmunol.org/ sections revealed some discrepancies between tissue expression expression and bioactivity of porcine BAFF. Dev. Comp. Immunol. 31: and in vitro differentiated adipocytes. In fully differentiated adi- 1211–1219. 12. Gras, M. P., Y. Laabi, G. Linares-Cruz, M. O. Blondel, J. P. Rigaut, J. C. Brouet, pose tissue, neither APRIL, TWEAK, TACI, nor BCMA are ex- G. Leca, R. Haguenauer-Tsapis, and A. Tsapis. 1995. BCMAp: an integral mem- pressed, whereas they are expressed in fully isolated differentiated brane protein in the Golgi apparatus of human mature B lymphocytes. Int. Im- adipocytes. These results suggest that there are potential interac- munol. 7: 1093–1106. 13. Dillon, S. R., J. A. Gross, S. M. Ansell, and A. J. Novak. 2006. An APRIL to tions between adipocytes and humoral or tissue factors that might remember: novel TNF ligands as therapeutic targets. Nat. Rev. Drug Discov. 5: regulate the expression of these ligands and receptors. These data 235–246. further indicate that dedifferentiation of cells results in the acqui- 14. Hendriks, J., L. Planelles, J. de Jong-Odding, G. Hardenberg, S. T. Pals,

M. Hahne, M. Spaargaren, and J. P. Medema. 2005. Heparan sulfate proteoglycan by guest on September 26, 2021 sition of a stem cell-like phenotype and that the TNF-SF members binding promotes APRIL-induced tumor cell proliferation. Cell Death Differ. 12: APRIL and TWEAK could possibly be used for the therapeutic 637–648. manipulation of these tumors, promoting the possible reprogram- 15. Ingold, K., A. Zumsteg, A. Tardivel, B. Huard, Q. G. Steiner, T. G. Cachero, F. Qiang, L. Gorelik, S. L. Kalled, H. Acha-Orbea, et al. 2005. Identification of ming of tumor cells. proteoglycans as the APRIL-specific binding partners. J. Exp. Med. 201: Adipose tissue was long regarded as a silent and passive organ, 1375–1383. storing excess energy as triglycerides and releasing energy as fatty 16. Sakurai, D., H. Hase, Y. Kanno, H. Kojima, K. Okumura, and T. Kobata. 2007. TACI regulates IgA production by APRIL in collaboration with HSPG. Blood acids. However, adipose tissue is now recognized as an endocrine 109: 2961–2967. and immune organ, secreting a wide variety of hormones, cyto- 17. Kimberley, F. C., L. van Bostelen, K. Cameron, G. Hardenberg, J. A. Marquart, kines, chemokines, and growth factors that influences metabolism, M. Hahne, and J. P. Medema. 2009. The proteoglycan (heparan sulfate proteoglycan) binding domain of APRIL serves as a platform for ligand multimerization vascular and endothelial function, appetite and satiety, immunity, and cross-linking. FASEB J. 23: 1584–1595. fertility, inflammation, tumor growth, and many other processes 18. Thompson, J. S., S. A. Bixler, F. Qian, K. Vora, M. L. Scott, T. G. Cachero, (for recent reviews, see Refs. 2, 52). In this respect, a thorough C. Hession, P. Schneider, I. D. Sizing, C. Mullen, et al. 2001. BAFF-R, a newly identified TNF receptor that specifically interacts with BAFF. Science 293: investigation of this tissue/organ might provide new insights for a 2108–2111. wide spectrum of diseases, from the metabolic syndrome to spe- 19. Hatzoglou, A., J. Roussel, M. F. Bourgeade, E. Rogier, C. Madry, J. Inoue, O. Devergne, and A. Tsapis. 2000. TNF receptor family member BCMA (B cell cific immunological and inflammatory conditions. Results pre- maturation) associates with TNF receptor-associated factor (TRAF) 1, TRAF2, sented in this paper integrate BAFF, APRIL, and TWEAK as and TRAF3 and activates NF-␬B, elk-1, c-Jun N-terminal kinase, and p38 mi- novel potential mediators of additional immune actions, taking into togen-activated protein kinase. J. Immunol. 165: 1322–1330. 20. Mackay, F., and C. Ambrose. 2003. The TNF family members BAFF and APRIL: consideration their direct implications in the development and mat- the growing complexity. Cytokine Growth Factor Rev. 14: 311–324. uration of the immune system and autoimmunity (44), and provide 21. Burkly, L. C., J. S. Michaelson, K. Hahm, A. Jakubowski, and T. S. Zheng. 2007. further evidence of the immune role of the adipose tissue (2). TWEAKing tissue remodeling by a multifunctional cytokine: role of TWEAK/ Moreover, as shown, these ligands and cognitive receptors might Fn14 pathway in health and disease. Cytokine 40: 1–16. 22. Chicheportiche, Y., P. R. Bourdon, H. Xu, Y. M. Hsu, H. Scott, C. Hession, be factors influencing the genetic programming of mesenchymal I. Garcia, and J. L. Browning. 1997. TWEAK, a new secreted ligand in the tumor cells toward adipogenesis, providing further elements for the ma- necrosis factor family that weakly induces apoptosis. J. Biol. Chem. 272: 32401–32410. nipulation and/or treatment of adipose tissue-derived benign and 23. Girgenrath, M., S. Weng, C. A. Kostek, B. Browning, M. Wang, S. A. Brown, malignant tumors. J. A. Winkles, J. S. Michaelson, N. Allaire, P. Schneider, et al. 2006. TWEAK, via its receptor Fn14, is a novel regulator of mesenchymal progenitor cells and skeletal muscle regeneration. EMBO J. 25: 5826–5839. Acknowledgments 24. Jakubowski, A., C. Ambrose, M. Parr, J. M. Lincecum, M. Z. Wang, T. S. Zheng, We thank Maria Klinaki for excellent technical assistance. B. Browning, J. S. Michaelson, M. Baetscher, B. Wang, D. M. Bissell, and L. C. Burkly. 2005. TWEAK induces liver progenitor cell proliferation. J. Clin. Invest. 115: 2330–2340. Disclosures 25. Duffield, J. 2003. The inflammatory macrophage: a story of Jekyll and Hyde. The authors have no financial conflict of interest. Clin. Sci. (Lond.) 104: 27–38. 5956 TNF-SF MEMBERS IN ADIPOCYTES

26. Pelekanou, V., M. Kampa, M. Kafousi, K. Darivianaki, E. Sanidas, D. D. Tsiftsis, 40. Mishra, P. J., P. J. Mishra, R. Humeniuk, D. J. Medina, G. Alexe, J. P. Mesirov, E. N. Stathopoulos, A. Tsapis, and E. Castanas. 2008. Expression of TNF-super- S. Ganesan, J. W. Glod, and D. Banerjee. 2008. Carcinoma-associated fibroblast- family members BAFF and APRIL in breast cancer: immunohistochemical study like differentiation of human mesenchymal stem cells. Cancer Res. 68: in 52 invasive ductal breast carcinomas. BMC Cancer 8: 76. 4331–4339. 27. Bunnell, B. A., M. Flaat, C. Gagliardi, B. Patel, and C. Ripoll. 2008. Adipose- 41. Hotamisligil, G. S., N. S. Shargill, and B. M. Spiegelman. 1993. Adipose ex- derived stem cells: isolation, expansion and differentiation. Methods 45: 115–120. pression of tumor necrosis factor-␣: direct role in obesity-linked insulin resis- 28. Hauner, H., G. Entenmann, M. Wabitsch, D. Gaillard, G. Ailhaud, R. Negrel, and tance. Science 259: 87–91. E. F. Pfeiffer. 1989. Promoting effect of glucocorticoids on the differentiation of 42. Kishore, U., M. S. Kojouharova, and K. B. Reid. 2002. Recent progress in the human adipocyte precursor cells cultured in a chemically defined medium. understanding of the structure-function relationships of the globular head regions J. Clin. Invest. 84: 1663–1670. of C1q. Immunobiology 205: 355–364. 29. Cawthorn, W. P., and J. K. Sethi. 2008. TNF-␣ and adipocyte biology. FEBS Lett. 43. Kishore, U., and K. B. Reid. 2000. C1q: structure, function, and receptors. Im- 582: 117–131. munopharmacology 49: 159–170. 30. Furlong, M. A., J. C. Fanburg-Smith, and M. Miettinen. 2001. The morphologic 44. Mackay, F., P. A. Silveira, and R. Brink. 2007. B cells and the BAFF/APRIL spectrum of hibernoma: a clinicopathologic study of 170 cases. Am. J. Surg. axis: fast-forward on autoimmunity and signaling. Curr. Opin. Immunol. 19: Pathol. 25: 809–814. 327–336. 31. Burjanadze, M., E. Roosnek, J. Passweg, and B. Huard. 2009. [APRIL, the spring 45. Pelekanou, V., M. Kampa, and E. Castanas. 2007. Potentiation of erythropoietin of plasmocytes.] Med. Sci. (Paris) 25: 5–8. (EPO) action by membrane steroid receptor agonists. Breast Cancer 16 (Suppl. 32. Rafei, M., J. Hsieh, S. Fortier, M. Li, S. Yuan, E. Birman, K. Forner, 1): s14–s15. M. N. Boivin, K. Doody, M. Tremblay, B. Annabi, and J. Galipeau. 2008. Mes- 46. Moreaux, J., J. L. Veyrune, J. De Vos, and B. Klein. 2009. APRIL is overex- enchymal stromal cell-derived CCL2 suppresses plasma cell immunoglobulin pressed in cancer: link with tumor progression. BMC Cancer 9: 83. production via STAT3 inactivation and PAX5 induction. Blood 112: 4991–4998. 47. Roosnek, E., M. Burjanadze, P. Y. Dietrich, T. Matthes, J. Passweg, and 33. Daher, S. R., B. H. Johnstone, D. G. Phinney, and K. L. March. 2008. Adipose B. Huard. 2009. Tumors that look for their springtime in APRIL. Crit. Rev. stromal/stem cells: basic and translational advances: the IFATS collection. Stem Oncol. Hematol. In press. Cells 26: 2664–2665. 48. Deshayes, F., G. Lapree, A. Portier, Y. Richard, P. Pencalet, D. Mahieu-Caputo, 34. Jeong, J. 2008. Adipose stem cells as a clinically available and effective source P. Horellou, and A. Tsapis. 2004. Abnormal production of the TNF-homologue of adult stem cell therapy. Int. J. Stem Cells 1: 43–48. APRIL increases the proliferation of human malignant glioblastoma cell lines via Downloaded from 35. Tonnesen, M. G. 1989. Neutrophil-endothelial cell interactions: mechanisms of a specific receptor. Oncogene 23: 3005–3012. neutrophil adherence to vascular endothelium. J. Invest. Dermatol. 93: 53S–58S. 49. Chiu, A., W. Xu, B. He, S. R. Dillon, J. A. Gross, E. Sievers, X. Qiao, P. Santini, 36. Young, H. E., M. L. Mancini, R. P. Wright, J. C. Smith, A. C. Black, Jr., E. Hyjek, J. W. Lee, et al. 2007. Hodgkin lymphoma cells express TACI and C. R. Reagan, and P. A. Lucas. 1995. Mesenchymal stem cells reside within the BCMA receptors and generate survival and proliferation signals in response to connective tissues of many organs. Dev. Dyn. 202: 137–144. BAFF and APRIL. Blood 109: 729–739. 37. Case, J., T. L. Horvath, J. C. Howell, M. C. Yoder, K. L. March, and E. F. Srour. 50. Tang, W., D. Zeve, J. M. Suh, D. Bosnakovski, M. Kyba, R. E. Hammer, 2005. Clonal multilineage differentiation of murine common pluripotent stem M. D. Tallquist, and J. M. Graff. 2008. White fat progenitor cells reside in the

cells isolated from skeletal muscle and adipose stromal cells. Ann. NY Acad. Sci. adipose vasculature. Science 322: 583–586. http://www.jimmunol.org/ 1044: 183–200. 51. Blanco-Colio, L. M., J. L. Martin-Ventura, B. Munoz-Garcia, J. A. Moreno, 38. Case, J., T. L. Horvath, C. B. Ballas, K. L. March, and E. F. Srour. 2008. In vitro O. Meilhac, A. Ortiz, and J. Egido. 2007. TWEAK and Fn14: new players in the clonal analysis of murine pluripotent stem cells isolated from skeletal muscle and pathogenesis of atherosclerosis. Front Biosci. 12: 3648–3655. adipose stromal cells. Exp. Hematol. 36: 224–234. 52. Schaffler, A., U. Muller-Ladner, J. Scholmerich, and C. Buchler. 2006. Role of 39. Mishra, P. J., P. J. Mishra, J. W. Glod, and D. Banerjee. 2009. Mesenchymal stem adipose tissue as an inflammatory organ in human diseases. Endocr. Rev. 27: cells: flip side of the coin. Cancer Res. 69: 1255–1258. 449–467. by guest on September 26, 2021