Human endothelial cells express NOD2/CARD15 and increase IL-6 secretion in response to muramyl dipeptide

Microvascular Research 71 (2006) 103 – 107 www.elsevier.com/locate/ymvre

Human endothelial cells express NOD2/CARD15 and increase IL-6 secretion in response to muramyl dipeptide Michael P. Davey a,b,d,*, Tammy M. Martin c,d, Stephen R. Planck b,c,e, Jack Lee e, David Zamora c, James T. Rosenbaum b,c,e a Department of Veterans Affairs Medical Center, Portland, OR 97239, USA Department of Medicine, Oregon Health and Science University, Portland, OR 97239, USA c Casey Eye Institute, Oregon Health and Science University, Portland, OR 97239, USA Department of Molecular Microbiology and Immunology, Oregon Health and Science University, Portland, OR 97239, USA e Department of Cell and Developmental Biology, Oregon Health and Science University, Portland, OR 97239, USA b

d

Received 5 October 2005; revised 6 November 2005; accepted 29 November 2005 Available online 18 January 2006

Abstract Mutations in the human NOD2/CARD15 gene cause Blau syndrome, an autoinflammatory disorder involving the joints, skin and eyes. Insights into the mechanism of this association may be gained by a further understanding of where NOD2 is expressed. The objective of this study was to analyze ocular endothelial cells for NOD2 expression. Human ocular tissue was analyzed by immunohistology using anti-NOD2 antisera. RNA isolated from iris, choroid and endothelial cell lines was analyzed by reverse transcription-PCR and real-time quantitative PCR. Gene regulation was studied by treating endothelial cells with TNF-a and IFN-g. Functional responses were assessed by measuring IL-6 release from endothelial cells treated with muramyl dipeptide (MDP), synthetic lipopeptide (Pam3CSK4) and lipopolysaccharide (LPS). Immunohistological analysis revealed staining of endothelial cells in the uveal tract. NOD2 expression was detected in primary ocular endothelial cell cultures, and levels increased in response to inflammatory cytokines. Endothelial cells from choroid demonstrated enhanced release of IL-6 in response to MDP, and synergy was observed following treatment with MDP and either Pam3CSK4 or LPS. The observations that endothelial cells express NOD2, upregulate NOD2 in response to stimuli known to promote NOD2 expression and show synergistic cytokine responses to MDP and TLR ligands previously shown to be mediated by NOD2 are informative since they may be relevant to pathogenic mechanisms leading to the spectrum of inflammation seen in Blau syndrome. D 2005 Elsevier Inc. All rights reserved. Keywords: NOD2 gene; Endothelial cells; Gene expression; Muramyl dipeptide; Interleukin-6; Blau syndrome

Introduction Blau syndrome (OMIM #186580), or arthrocutaneouveal granulomatosis, is a rare autosomal dominant disorder characterized by inflammatory arthritis, dermatitis and uveitis (Blau, 1985; Jabs et al., 1985). Blau syndrome is caused by a mutation in NOD2/CARD15 (Miceli-Richard et al., 2001), a cytosolic protein influencing innate immune responses (Ogura et al., 2001a,b), pathogen survival (Hisamatsu et al., 2003) and * Corresponding author. Department of Veterans Affairs Medical Center (R&D), 3710 S.W. U.S. Veterans Hospital Road, Portland, OR 97239-2999, USA. Fax: +1 503 273 5351. E-mail address: [email protected] (M.P. Davey). 0026-2862/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.mvr.2005.11.010

intracellular signaling (Watanabe et al., 2004). A different set of mutations in NOD2 are found in Crohn’s disease, a disorder characterized by granulomatous inflammation in the lower gastrointestinal tract (Hugot et al., 2001; Ogura et al., 2001a,b). The mechanism by which mutated NOD2 causes chronic inflammation in multiple organ systems is unknown. Much has been learned regarding the role of NOD2 in the normal homeostasis of a cell (reviewed in Eckmann and Karin, 2005; Murray, 2005). Equally important in understanding the link between NOD2 and granulomatous inflammatory syndromes is determining the range of cell types that express NOD2. While initially it was reported that the expression of NOD2 was restricted to monocytes (Ogura et al., 2001a,b), subsequent studies found expression in granulocytes, dendritic

104

M.P. Davey et al. / Microvascular Research 71 (2006) 103 – 107

cells, intestinal epithelial cells, Paneth cells and osteoblasts (Gutierrez et al., 2002; Marriott et al., 2005; Ogura et al., 2003; Rosenstiel et al., 2003). Further advances in understanding expression patterns of NOD2 may provide insights into the role that NOD2 plays in promoting uveitis, arthritis and dermatitis. In the present study, we have used immunohistochemistry, reverse transcription-PCR (RT-PCR) and real-time quantitative PCR to study NOD2 expression in vascular endothelial cells isolated from different locations within the eye and then confirmed in endothelial cells from more diverse sources. The eye is the focus of this report because uveitis is a characteristic finding in Blau syndrome and can occasionally develop in Crohn’s disease. Our results show that endothelial cells express NOD2. Furthermore, treatment of endothelial cells with muramyl dipeptide (MDP), a bacteria breakdown product known to trigger NF-nB activation and cytokine responses in a NOD2-dependent manner (Girardin et al., 2003; Inohara et al., 2003), resulted in secretion of IL-6. Materials and methods Immunostaining Human eyes were obtained from the Oregon Eye Bank, Portland, OR, with approval of the Oregon Health and Science University Institutional Review Board. After dissection, tissues were fixed in IHC Zinc Fixative (BD Pharmingen, San Diego, CA) prior to embedding in paraffin. After deparaffinization, 5 Am thick tissue sections on slides were incubated with 0.1% (w/v) of pepsin in 0.01 N HCl at room temperature (RT) for 20 min for antigen retrieval. All reagents were diluted in Tris-buffered saline (TBS; 50 nM Tris, 0.15 M NaCl, pH 7.5) unless otherwise noted. Tissue sections were blocked with 2% goat serum/1% bovine serum albumin/0.4% Triton X-100 in TBS for 1 h at room temperature. Sections were incubated with primary antibody overnight at 4-C. Nod2 was detected with polyclonal rabbit-anti-human Nod2 antiserum (1:200, Cayman, Ann Arbor, MI), and vascular endothelium was identified by positive staining with rabbit-antihuman von Willebrand Factor antiserum (1:100, Dako, Carpinteria, CA). Control staining was performed with purified rabbit IgG (Vector Laboratories, Burlingame, CA). Following 2 washes with TBS and 1 wash with 1% Tween 20 (v/v) in TBS, sections were incubated with secondary antibody (goat-anti-rabbit IgG-biotin conjugate, Vector Laboratories) for 1 h at room temperature. After washing (as above), sections were incubated for 45 min at RT with avidin-biotinylated alkaline phosphatase using the Vectastain ABC kits following manufacturer’s instructions (Vector Laboratories). Sections were washed as indicated above, and Fast Red substrate (BioGenex, San Ramon, CA) was added to develop the reaction. Sections were counterstained with hematoxylin and preserved with Crystal/mount (Biomeda, Foster City, CA).

Cells and reagents Human donor eyes were obtained within 24 h of death from the Oregon Eye Bank (Portland, OR). Iris, choroid and retinal tissues were carefully dissected from the donor eyes. Primary cultures of vascular endothelial cells from these tissues were established by our published techniques (Silverman et al., 2001). Cells were stimulated with TNF-a (10 ng/ml) (R&D Systems, Minneapolis, MN) for up to 24 h. Human umbilical vein, umbilical artery, retinal, microvascular and aortic endothelial cells (Cascade Biologics, Portland, OR) were cultured in MCDB131 growth medium (Sigma, St. Louis, MO) supplemented with all components (except hydrocortisone) of the EGM-2 SingleQuots kit (Cambrex BioSciences, Walkersville, MD) according to the manufacturer’s instructions and containing 10% fetal bovine serum. Peripheral blood mononuclear cells (PBMC) were prepared by Ficoll-Paque centrifugation.

Endothelial cells from choroid were seeded into 12-well plates (1.5  105/ well in 1 ml) and cultured overnight to become adherent. Cells were then treated with TNF-a (10 ng/ml) and IFN-g (100 U/ml) (Genentech, South San Francisco, CA) and cultured an additional 24 h. Cells were gently washed and cultured in triplicate in fresh media or media containing MDP (10 Ag/ml; Bachem, Torrance, CA), Pam3CSK4 (500 ng/ml; InvivoGen, San Diego, CA), LPS (1 Ag/ml; Salmonella abortus equi, Sigma) or the indicated combinations. Supernatants were harvested 24 h later and analyzed for IL-6 by ELISA (Human IL-6 DuoSet ELISA, R&D Systems) at a 1:1000 dilution according to the manufacturer’s instructions.

PCR techniques For RT-PCR, total RNA was extracted (RNeasy kit, Qiagen, Valencia, CA) and converted to cDNA with random primers and MultiScribe MULV reverse transcriptase (Applied Biosystems, Foster City, CA). Primers for NOD2 (forward: GCACGTGGCCTGAATGTTGG and reverse: CCGCGGCAGTGATGTAGTTATTC) were designed to amplify a 385 bp product corresponding to nucleotide positions 2380 – 2764 of the mRNA (Ogura et al., 2001a,b). GAPDH primers amplifying a 209 bp product were used as a housekeeping gene control. RNA from peripheral blood mononuclear cells was used as a positive control, and the identity of the PCR product amplified by the NOD2 primers was verified by sequencing representative gel-purified PCR products (data not shown). A 25 Al PCR reaction contained 1 Al of cDNA diluted 1:10, 20 pmol of each primer, each dNTP (0.8 mM), 5 mM of MgCl2, 1 PCR Buffer (Promega, Madison, WI) and 1 U of Taq DNA polymerase (Promega). Thermal cycling was performed in a GeneAmp PCR System 9700 thermocycler (Applied Biosystems). After an initial preheating step at 94-C for 4 min, a touchdown procedure consisting of denaturation at 94-C for 15 s, annealing at 69- – 59-C for 1 min decreasing 1-C for every 2 cycles and extension at 72-C for 2 min was applied during the first 20 cycles. Subsequently, additional thermal cycling consisting of denaturation at 94-C for 15 s, annealing at 55-C for 1 min and extension at 72-C for 2 min was performed for 24 cycles. After the last cycle, an additional extension at 72-C for 7 min was carried out. Amplified fragments were electrophoresed through a 3% agarose gel and stained with ethidium bromide. Quantitative real-time PCR (qRT-PCR) was performed on cDNA in 96-well MicroAmp optical plates (Applied Biosystems) using the following primers: 5VCAGCCAGTATGAATGTGATGAAATC-3V (forward), 5V-GTGGCAAGATCAAGC AGCCT-3V (reverse) and 5V-FAM/ATCTTCACACCGTCCCAGA/ TAMRA-3V (probe). FAMi is a 5V reporter dye (6-carboxy-fluorescein), and TAMRAi is a 3V quencher dye (6-carboxy-N,N,NV,NV-tetramethyl-rhodamine). A 25 Al reaction consisted of 1 Al of 1:10 diluted cDNA, TaqMan Universal Master Mix (Applied Biosystems), 300 nM of each primers, 250 nM of the probe and GAPDH primer/probe mix (Applied Biosystems). Thermal cycling and detection were performed in an ABI Prism 7700 Sequence Detector (Applied Biosystems). An initial cycle of 50-C for 2 min and 95-C for 10 min was followed by 45 cycles of 95-C for 15 s and 60-C for 1 min. Data are expressed according to the Comparative Method (PE ABI PRISM 7700 User Bulletin #2) using untreated cells as the relative calibrator.

Statistics Sample differences with P < 0.05 determined by a single factor analysis of variance (ANOVA) were considered statistically significant.

Results Expression of NOD2 in human uveal tissue Human ocular tissue was analyzed with commercially available antisera against human NOD2. Positive staining was demonstrated in the vascular endothelium of the iris, ciliary body and choroid (Fig. 1A). Endothelial cells were identified by staining serial sections with anti-von Willebrand Factor antiserum (Fig. 1B). Note the staining with anti-NOD2

M.P. Davey et al. / Microvascular Research 71 (2006) 103 – 107

105

Fig. 2. Expression of NOD2 mRNA in human iris, choroid and retinal microvascular endothelial cells derived from the same donor. Cells were stimulated with TNF-a as described in the Methods. Total RNA was subjected to RT-PCR using NOD2 and GAPDH specific primers. Cells were unstimulated (0) or stimulated for 3 or 24 h (indicated at the bottom of each lane).

cular cells or hMVEC), human retinal endothelial cells (hREC) and human aortic endothelial cells (hAEC) were cultured for 24 h in the presence of IFN-g and TNF-a. qRTPCR analysis was then performed to assess upregulation of NOD2 mRNA. Peripheral blood mononuclear cells (PBMC) were included as a positive control. As shown in Fig. 3, NOD2 was upregulated following treatment with IFN-g and TNF-a in all cultures. IL-6 response of endothelial cells treated with MDP Fig. 1. NOD2 is expressed in vascular endothelium of human choroid. Sections from human eyes were subjected to immunohistology. The neuroretina was pulled away from the bottom and the sclera was pulled from the top of the tissue as seen. Panel A: Immunostaining with anti-NOD2 antiserum. Panel B: The vascular endothelium is identified with anti-von Willebrand Factor antiserum. Panel C: Control immunostaining with purified rabbit IgG. All panels: Fast Red antibody detection, hematoxylin counterstain, original magnification 400; natural brown pigment is evident in retinal pigment epithelial cells and melanocytes.

antisera matches that seen with anti-von Willebrand Factor antiserum. Staining was not observed with control rabbit IgG (Fig. 1C). Detection of NOD2 mRNA by PCR To confirm the observations made using immunohistology, RT-PCR was performed. Low levels of NOD2 mRNA were detected in unstimulated ocular microvascular endothelial cell cultures from 3 of 5 donors (data not shown). Previous studies had shown that NOD2 mRNA can be upregulated by TNF-a, IFN-g or TNF-a and IFN-g in combination (Gutierrez et al., 2002; Hisamatsu et al., 2003; Rosenstiel et al., 2003). From one set of donor eyes, iris, choroid and retinal microvascular endothelial cells were isolated and stimulated in culture with TNF-a. At 0, 3 and 24 h time points, RNA was prepared and analyzed by RT-PCR. Upregulation of NOD2 expression in vascular endothelial cells from all 3 tissues was observed (Fig. 2). The upregulation of NOD2 expression was also seen in ocular endothelial cells from 4 other donor specimens (data not shown). Quantitative real-time PCR Expression of NOD2 in vascular cell cultures was further evaluated by qRT-PCR analysis. Endothelial cells derived from human umbilical vein (hUVEC), skin (human microvas-

Given that endothelial cells are known to express IL-6 and expression of this cytokine can be stimulated by MDP (Sanceau et al., 1990), we chose to study IL-6 secretion in response to MDP. Endothelial cells from choroid were stimulated with MDP alone or in combination with a TLR2 ligand (synthetic bacterial lipopeptide-Pam3SCK4, which cooperates with TLR1 to induce signaling) and a TLR4/ CD14 ligand (LPS). As shown in Fig. 4, stimulation of choroid cells with MDP results in a significant increase in IL-6 production (P = 0.002 compared to cells maintained in media alone). A synergistic response was noted when MDP was added in the presence of Pam3CSK4 or LPS (Fig. 4). These data are consistent with NOD2-mediated synergy of TLR

Fig. 3. NOD2 is upregulated by TNF-a and INF-g in endothelial cell cultures. Cells were cultured in the presence of media alone (open bars) or TNF-a (10 ng/ml) and IFN-g (100 U/ml) (solid bars) for 24 h. RNA was then isolated and analyzed by real-time quantitative PCR and data are expressed as fold induction, using 1 as the reference for cells grown in media alone. Error bars represent standard deviation of samples analyzed in quadruplicate. The assay was repeated three times and a representative example is shown. Endothelial cells were derived from human umbilical vein (hUVEC), skin (human microvascular cells or hMVEC), human retinal endothelial cells (hREC) and human aortic endothelial cells (hAEC). PBMC were included as a positive control.

106

M.P. Davey et al. / Microvascular Research 71 (2006) 103 – 107

Fig. 4. MDP enhances IL-6 production by endothelial cells and is synergistic with Pam3CSK4 or LPS. Endothelial cells were cultured in media alone or media supplemented with MDP, Pam3CSK4, LPS or the indicated combinations. IL-6 levels were detected using an ELISA. IL-6 levels detected for cells grown in media alone (56.7 T 1.4 ng/ml) were subtracted from each condition. Data represent the mean values of triplicate samples and error bars show standard deviation. P values were calculated by ANOVA. One of two experiments with comparable results is shown.

activation previously observed by others (Pauleau and Murray, 2003; Uehara et al., 2005) and suggest that endothelial cells cultured from choroid display NOD2-dependent cytokine responses.

inducer of cytokine secretion in mice and in in vitro culture systems with murine or human cells (reviewed in Takada et al., 2002). However, in combination with several different TLR ligands, MDP significantly enhances toxicity and cytokine release (Netea et al., 2005; Takada et al., 2002; van Heel et al., 2005a,b; Wolfert et al., 2002; Yang et al., 2001). Studies with NOD2-deficient mice or suppression of NOD2 mRNA by RNA interference showed that NOD2 was required for a synergistic response (Pauleau and Murray, 2003; Uehara et al., 2005). Our results show that endothelial cells, when treated with MDP and ligands for TLR2 (Pam3CSK4) and TLR4 (LPS), likewise show synergistic responses that previously were shown to be NOD2-dependent. Acknowledgments This work was supported by grants from the Department of Veterans Affairs (MPD), NIH R03EY015137 (TMM), RO1EY06484, T32EY007123 and P30EY010572 (JTR), Research to Prevent Blindness (TMM, SRP, JTR) and funds from the Rosenfeld Family Trust (JTR). We are grateful to Monica Hermann, Caroline Suing and Paul Matiaco for technical assistance.

Discussion References Using immunohistochemistry, RT-PCR, qRT-PCR and a functional assay, this study shows that NOD2 is expressed in endothelial cells and thus further expands the range of tissues found to express NOD2. Appreciating the range of expression of NOD2 is important in understanding the mechanism by which mutations in NOD2 cause Blau syndrome. Endothelial cells represent the portal of entry of lymphocytic and mononuclear cells into joints, skin and eyes, and it is possible that the Blau mutation alters a function associated with NOD2 within endothelial cells leading to inflammation. Studies by others, using reporter assays to measure NF-nB levels, have shown that mutations in NOD2 associated with Blau syndrome result in a higher ‘‘set point’’ of NF-nB levels (Chamaillard et al., 2003; Kanazawa et al., 2005; Tanabe et al., 2004). It is speculated that the elevated function of NF-nB may lead to the inflammation and granuloma formation characteristic of Blau syndrome. Other functions associated with NOD2, such as recognition of bacterial cell wall components, altering growth of intracellular pathogens and modulating other cytokine responses all remain to be further investigated with regard to possible roles in the pathogenesis of Blau syndrome. NOD2 expression can be induced by TNF-a and IFN-g in intestinal epithelial cells (Rosenstiel et al., 2003). Our analysis of endothelial cells from several different organs likewise showed upregulation in response to these cytokines, indicating that at sites where endothelial cells are involved in inflammatory responses, NOD2 levels would most likely be enhanced. Once expressed, NOD2 participates in the detection of intracellular MDP, leading to activation of NF-nB and induction of genes encoding inflammatory cytokines (Girardin et al., 2003; Inohara et al., 2003). MDP by itself is a very weak

Blau, E.B., 1985. Familial granulomatous arthritis, iritis, and rash. J. Pediatr. 107, 689 – 693. Chamaillard, M., et al., 2003. Gene – environment interaction modulated by allelic heterogeneity in inflammatory diseases. Proc. Natl. Acad. Sci. U. S. A. 100, 3455 – 3460. Eckmann, L., Karin, M., 2005. NOD2 and Crohn’s disease: loss or gain of function? Immunity 22, 661 – 667. Girardin, S.E., et al., 2003. Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J. Biol. Chem. 278, 8869 – 8872. Gutierrez, O., et al., 2002. Induction of Nod2 in myelomonocytic and intestinal epithelial cells via nuclear factor-kappa B activation. J. Biol. Chem. 277, 41701 – 41705. Hisamatsu, T., et al., 2003. CARD15/NOD2 functions as an antibacterial factor in human intestinal epithelial cells. Gastroenterology 124, 993 – 1000. Hugot, J.P., et al., 2001. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease. Nature 411, 599 – 603. Inohara, N., et al., 2003. Host recognition of bacterial muramyl dipeptide mediated through NOD2. Implications for Crohn’s disease. J. Biol. Chem. 278, 5509 – 5512. Jabs, D.A., et al., 1985. Familial granulomatous synovitis, uveitis, and cranial neuropathies. Am. J. Med. 78, 801 – 804. Kanazawa, N., et al., 2005. Early-onset sarcoidosis and CARD15 mutations with constitutive nuclear factor-kappaB activation: common genetic etiology with Blau syndrome. Blood 105, 1195 – 1197. Marriott, I., et al., 2005. Induction of Nod1 and Nod2 intracellular pattern recognition receptors in murine osteoblasts following bacterial challenge. Infect. Immun. 73, 2967 – 2973. Miceli-Richard, C., et al., 2001. CARD15 mutations in Blau syndrome. Nat. Genet. 29, 19 – 20. Murray, P.J., 2005. NOD proteins: an intracellular pathogen-recognition system or signal transduction modifiers? Curr. Opin. Immunol. 17, 1 – 7. Netea, M.G., et al., 2005. Nucleotide-binding oligomerization domain-2 modulates specific TLR pathways for the induction of cytokine release. J. Immunol. 174, 6518 – 6523. Ogura, Y., et al., 2001. A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease. Nature 411, 603 – 606.

M.P. Davey et al. / Microvascular Research 71 (2006) 103 – 107 Ogura, Y., et al., 2001. Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes and activates NF-kappaB. J. Biol. Chem. 276, 4812 – 4818. Ogura, Y., et al., 2003. Expression of NOD2 in Paneth cells: a possible link to Crohn’s ileitis. Gut 52, 1591 – 1597. Pauleau, A.L., Murray, P.J., 2003. Role of nod2 in the response of macrophages to toll-like receptor agonists. Mol. Cell. Biol. 23, 7531 – 7539. Rosenstiel, P., et al., 2003. TNF-alpha and IFN-gamma regulate the expression of the NOD2 (CARD15) gene in human intestinal epithelial cells. Gastroenterology 124, 1001 – 1009. Sanceau, J., et al., 1990. Secretion of interleukin-6 (IL-6) by human monocytes stimulated by muramyl dipeptide and tumour necrosis factor alpha. Immunology 69, 52 – 56. Silverman, M.D., et al., 2001. Cell adhesion molecule expression in cultured human iris endothelial cells. Invest. Ophthalmol. Visual Sci. 42, 2861 – 2866. Takada, H., et al., 2002. Enhancement of endotoxin activity by muramyldipeptide. J. Endotoxin. Res. 8, 337 – 342. Tanabe, T., et al., 2004. Regulatory regions and critical residues of NOD2 involved in muramyl dipeptide recognition. EMBO J. 23, 1587 – 1597.

107

Uehara, A., et al., 2005. Muramyldipeptide and diaminopimelic acid-containing desmuramylpeptides in combination with chemically synthesized Toll-like receptor agonists synergistically induced production of interleukin-8 in a NOD2- and NOD1-dependent manner, respectively, in human monocytic cells in culture. Cell. Microbiol. 7, 53 – 61. van Heel, D.A., et al., 2005a. Muramyl dipeptide and toll-like receptor sensitivity in NOD2-associated Crohn’s disease. Lancet 365, 1794 – 1796. van Heel, D.A., et al., 2005b. Synergy between TLR9 and NOD2 innate immune responses is lost in genetic Crohn’s disease. Gut 54, 1553 – 1557. Watanabe, T., et al., 2004. NOD2 is a negative regulator of Toll-like receptor 2mediated T helper type 1 responses. Nat. Immunol. 5, 800 – 808. Wolfert, M.A., et al., 2002. The origin of the synergistic effect of muramyl dipeptide with endotoxin and peptidoglycan. J. Biol. Chem. 277, 39179 – 39186. Yang, S., et al., 2001. Synergistic effect of muramyldipeptide with lipopolysaccharide or lipoteichoic acid to induce inflammatory cytokines in human monocytic cells in culture. Infect. Immun. 69, 2045 – 2053.