Macrophage-Stimulating Protein Polymorphism Rs3197999 Is Associated with a Gain of Function: Implications for Inflammatory Bowel

Home , MST1

ORIGINAL ARTICLE Macrophage-stimulating protein polymorphism rs3197999 is associated with a gain of function: implications for inﬂammatory bowel disease

FHa¨user1, C Deyle1, D Berard1, C Neukirch1, C Glowacki1, JK Bickmann1, JJ Wenzel1,2, KJ Lackner1 and H Rossmann1

Crohn’s disease and ulcerative colitis, the two main types of inflammatory bowel disease (IBD), were reported to be associated with a variety of genetic polymorphisms. A subset of these polymorphisms was identified in both diseases and only three of them were found in primary sclerosing cholangitis (PSC). rs3197999 (Arg689Cys) located in the MST1 gene is one of the most convincingly replicated IBD/PSC-associated polymorphisms but its functional consequences have not been investigated, yet. We expressed both MST1 gene variants (Arg689 (MSPwt) and Cys689 (MSPmut)) in a eukaryotic cell system and compared their stimulatory effects on macrophage-like THP-1 cells. Except for the rate of apoptosis that remained unchanged, MSPmut significantly increased the stimulatory effect of MSP (macrophage-stimulating protein) on chemotaxis and proliferation by THP-1 cells, indicating a gain of function associated with the Arg689Cys exchange. A broad set of evidence reported previously suggests that pro-inflammatory changes in macrophage function have a major role in the initiation of the inflammatory process in IBD and PSC. Therefore, the gain of function observed with rs3197999 in MST1 might provide a cellular mechanism for the consistent association of this polymorphism with an increased risk for IBD and PSC.

Genes and Immunity (2012) 13, 321--327; doi:10.1038/gene.2011.88; published online 12 January 2012 Keywords: MST1; Crohn’s disease; ulcerative colitis; primary sclerosing cholangitis

INTRODUCTION macrophages in sinusoidal and perisinusoidal spaces.7 In all, Crohn’s disease (CD) and ulcerative colitis (UC), the two main types 75--90% of PSC patients have a history of or a co-existing IBD. For of inflammatory bowel disease (IBD), are both based on a that matter PSC is not dependent on an active intestinal disease chronically dysregulated mucosal homeostasis of the gut. The but may also occur in patients with preceding colectomy precise etiology remains unknown but environmental changes, indicating common pathogenetic factors in IBD and PSC. In genetic predisposition, gut microbial flora and mucosal immune parallel to IBD, PSC is thought to be set off by an activation of the activation contribute to the pathogenesis.1,2 GWAS (genome-wide innate immune system, possibly triggered by bacteria or patho- association studies) revealed B100 loci that are significantly gen-associated molecular patterns delivered to the liver by the associated with CD, UC or both.3,4 The diversity of the identified portal vein. A common genetic basis of IBD and PSC has been loci reflects the multifactorial pathogenesis of IBD but potentially hypothesized, but merely for three of the analyzed susceptibility affected physiological pathways cluster to innate and acquired genes significant associations with both IBD and PSC have been immunity and intestinal epithelial defense. demonstrated to date.8 The lamina propria of the gut epithelium contains the largest One of the loci reported to be associated with IBD and PSC by pool of macrophages in the body that are predominantly recruited several GWAS (odds ratio B1.22 in IBD and B1.4 in PSC),3,4,9 --15 is from the blood. On the one hand, these macrophages are exposed rs3197999 (c.2078C4T; minor allele frequency 0.297), a common to a vast amount of bacteria and antigenic stimuli, mostly benign polymorphism in the MST1 (macrophage stimulating 1) gene, commensals and food components that should not induce any resulting in the substitution of an arginine by a cysteine at inflammatory response. On the other hand, macrophages position 689 of MSP (macrophage-stimulating protein; synonym: represent the first line of defense against harmful pathogens; HGF (hepatocyte-growth factor) like protein). Arg689 is one out of the recognition of conserved structures of infectious agents five arginine residues suggested to be essential for the binding of triggers a rapid response. Moreover, macrophages, along with MSP to its receptor RON (recepteur d’origine nantais).16 MSP dendritic cells, presenting antigens to lymphocytes instigate the belongs to the plasminogen-related kringle domain family. It is a response of the more flexible, but slower adaptive immune 78-kDa serum protein consisting of two disulfide-linked polypep- system. In UC as well as in CD, excessive activation of innate tide chains and displays a 45% sequence homology to HGF.17 MSP immunity seems to initiate inflammation and to be a prerequisite is mainly expressed in the liver from where it is released into the for the activation of the adaptive immune system.2,5 circulating blood as a biologically inactive pro-protein.18 Pro-MSP In 2.4% of all patients, IBD is accompanied by primary sclerosing is activated according to the local requirements by trypsin-like cholangitis (PSC), a chronic inflammatory and cholestatic liver proteases such as kallikrein.19 Unlike most proteins of the disease,6 which is, inter alia, characterized by the accumulation of plasminogen-related family that are serine proteases, both MSP

1Institute of Clinical Chemistry and Laboratory Medicine, University Medical Centre Mainz, Mainz, Germany and 2Institute of Medical Microbiology and Hygiene, University of Regensburg, Regensburg, Germany. Correspondence: Dr H Rossmann, Institute of Clinical Chemistry and Laboratory Medicine, University Medical Centre Mainz, Langenbeckstr. 1, 55131 Mainz, Germany. E-mail: [email protected] Received 22 September 2011; revised 18 November 2011; accepted 28 November 2011; published online 12 January 2012 Macrophage-stimulating protein polymorphism rs3197999 FHaüser et al 322 and HGF are enzymatically inactive,17,20,21 but regulate growth characteristic morphology. Furthermore, RON expression---which and motility. Binding of MSP to RON sets off multiple signal is a prerequisite for the study of MSP-dependent functional cascades affecting cell motility, proliferation, adhesion, apoptosis/ responses---was demonstrated by western blot and RT--PCR anoikis and cytokine secretion. RON is expressed in several tissues, analysis (Supplementary Figure 2). such as bone marrow,22 lung, colon and skin, and other cells of epithelial origin.23,24 In macrophages, RON expression is restricted to tissue macrophages25 --28 and monocytes adhering to the vessel MSP-induced chemotaxis and migration in macrophage-like THP-1 endothelium.22 RON is not expressed in peripheral blood cells monocytes, neutrophils or lymphocytes.25 Figure 2 shows MSP-induced migration of macrophage-like THP-1 To date, it is still unknown whether rs3197999 itself or a genetic cells from the upper Boyden chamber compartment, which had variation in linkage disequilibrium with rs3197999 is responsible for been supplemented with different concentrations of MSPwt and its consistently reported association with IBD.4 Potential functional MSPmut into the lower compartment containing EAMS (endotoxin- consequences of Arg689Cys in MSP that might explain this activated mouse serum) as a chemoattractant. Supernatants of at association have not been investigated, yet. Therefore, we expressed least three wt and three mut MSP expressing clones were diluted human wild-type (wt; Arg689) and mutant-type (mut; Cys689)MSPin with culture medium up to the desired MSP concentrations. Each a eukaryotic cell system and evaluated these recombinant proteins set of experiments included negative controls, that is, wells with respect to functional differences in cell culture. containing plain cell culture medium and wells containing supernatant of non-transfected CHO-K1 cells in the upper Boyden chamber compartment. The number of migrated cells was either RESULTS counted (Figure 2a) or quantified indirectly by measuring ATP Analysis of MSP clones concentrations after cell lysis (data not shown). Both methods Human wt and mut (c.2078C4T) MSP cDNAs were cloned and produced similar results. In the negative control wells, cell stably expressed in CHO-K1 (Chinese hamster ovarian) cells. Re- migration rates were low and no significant difference between sequencing of MSP amplicons from genomic DNA and cDNA of the cell culture medium and the supernatant of non-transfected the transfected CHO-K1 cells confirmed the expression of human CHO-K1 cells was observed. The mean value of the negative MSP thereby ruling out any unintended mutations of the control wells was defined as basal migration rate and subtracted expression constructs (Figure 1a). Western blot and RT--PCR from the MSP-stimulated cell migration rates. MSP increased the number of migrated cells in a concentration-dependent manner. (reverse transcriptase polymerase chain reaction) analysis ex- wt cluded the expression of hamster MSP in non-transfected CHO-K1 Comparison of the migration curves of cells incubated with MSP and MSPmut clearly demonstrates the increased stimulatory effect control cells (Supplementary Figure 1). Kallikrein digestion and mut western blotting of CHO-K1 cell lysates and culture supernatant of MSP . The bar graphs of Figures 2b and c are deduced from confirmed the correct and complete activation of MSP (Figure 1b). the data presented in Figure 2a: Figure 2b depicts the number of Neither by native nor by denaturing polyacylamide electrophor- migrated cells as percentage of the total number of cells subjected esis differences in the effectiveness of the kallikrein digest or the to the Boyden chamber experiment at the MSP concentration (0.5 ng ml -- 1 MSP) that lead to the maximum difference between resulting band patterns of the wt and mut MSP expressing CHO- wt mut K1 clones were observed. MSP and MSP : The migration rate differed 3.2-fold between MSPwt and MSPmut-stimulated cells. Since MSPwt approached saturation at much lower concentrations than MSPmut and since Differentiation of THP-1 cells into macrophage-like cells not all mut MSP clones actually reached saturation at the tested Stimulation of THP-1 monocytic leukemia cells by interferon (IFN)- concentrations, the actual difference may be even larger. The g and tumor necrosis factors a (TNF-a) induces a macrophage-like tested concentration range is much smaller for MSPmut than for phenotype.29 Successful differentiation was confirmed by semi- MSPwt, reflecting the fact that expression of recombinant MSP has quantitative RT--PCR: The expression of a macrophage marker, always been much lower in mut than in wt clones. Figure 2c GILT (IFN-g-inducible lysosomal thiol reductase), was quantified summarizes the maximal slope of the migration curves, which relative to the house keeping gene 18S rRNA. GILT expression differed 2.3-fold between MSPwt and MSPmut (Pp0.05). increased at least fivefold in differentiated THP-1 cells. Wright- Non-differentiated THP-1 cells migrated highly effectively Giemsa staining (HARLECO Wright-Giemsa Solution, Merck, through the Boyden chamber membrane but did not respond Darmstadt, Germany) of these cells revealed macrophage- to MSP (data not shown).

Figure 1. Generation and activation of recombinant MSPwt and MSPmut.(a) Sequencing of cDNA from wt (C, above) and mut (T, below) MSP expressing CHO-K1 clones. (b) Western blot analysis of equal volumes of cell supernatant of MSPwt and MSPmut expressed in CHO-K1; supernatants were digested with kallikrein before loading; blots of one wt and one mut clone are shown exemplarily; MSP expression was signiﬁcantly lower in the mut compared with the wt clones. The MSP antibody detects predominantly the MSP a-chain.

Genes and Immunity (2012) 321 --327 & 2012 Macmillan Publishers Limited Macrophage-stimulating protein polymorphism rs3197999 FHa¨user et al 323

Figure 2. Effect of different concentrations of MSPmut and MSPwt on THP-1 migration. (a) Cells were incubated with different concentrations of MSPmut and MSPwt for 24 h. The number of migrated cells in the lower Boyden chamber compartment was determined and is shown as percentage of total cells±s.e.m. (nX4). Media without MSP (plain cell culture medium and supernatant of non-transfected CHO-K1 cells) served as negative controls. Negative control values were deﬁned as basal migration rate and subtracted from the MSP-stimulated cell migration rates. (b) Effect of wt and mut MSP on THP-1 migration at a concentration of 0.5 ng ml À 1 (the concentration of the greatest difference observed between MSPmut and MSPwt). Calculation based on Figure 2a. (c) Maximum slopes of the migration curves (Figure 2a) of MSPwt and MSPmut.

Figure 3. Effect of different concentrations of MSPmut and MSPwt on THP-1 proliferation. (a) Cells were incubated with different concentrations of MSPmut and MSPwt for 24 h. BrdU incorporation in newly synthesized DNA was determined and is shown as relative luminescence units per second ±s.e.m. (nX4). Media without MSP (plain cell culture medium and supernatant of non-transfected CHO-K1 cells) served as negative controls. Media without MSP served as negative controls (as described in Figure 1a) and the resulting proliferation rates were subtracted from the MSP-stimulated proliferation rates. (b) Effect of wt and mut MSP at a concentration of 0.17, 0.37 and 0.54 ng ml À 1 on THP-1 proliferation (0.54 ng ml -- 1 : the concentration of the greatest difference observed between MSPmut and MSPwt). Calculation based on Figure 3a. (c) Maximum slopes of the proliferation curves (Figure 2a) of MSPwt and MSPmut.

In order to rule out cell proliferation as a major cause of the dent proliferation curves clearly demonstrate that MSPmut is more observed differences in migration induced by MSPwt and MSPmut, effective than MSPwt (Figure 3a): Figure 3b shows BrdU cells of both chambers were counted at the end point of the incorporation at the MSP concentrations inducing the most experiment and compared with the initial cell count: The total cell striking differences between MSPwt and MSPmut The maximum count remained largely constant in the course of the experiment difference was observed at a concentration of 0.5 ng ml À 1, which (cell count at the end point: 80--100% of the initial cell count) stimulated proliferation threefold more effective when MSPmut pointing to a marginal impact of proliferation in this experimental was used (Pp0.05). Apparently neither MSP from wt nor from mut setting most likely due to the presence of the chemoattractant clones reached saturation within the tested range of MSP EAMS. concentrations, suggesting that the actual difference may be even larger. Figure 3c summarizes the maximal slope of the proliferation curves, which differed eightfold between MSPwt and MSP-induced proliferation in macrophage-like THP-1 cells MSPmut (Pp0.05). The MSP-activated proliferation rate of macrophage-like THP-1 cells was determined by the assessment of 5-bromo-20-deoxyuridine (BrdU) incorporation into newly synthesized DNA. As speciﬁed for MSP-induced apoptosis in macrophage-like THP-1 cells the Boyden chamber experiments, cell culture medium and MSP-dependent apoptosis was determined by TUNEL (TdT supernatant of non-transfected CHO-K1 cells served as negative (terminal deoxynucleotidyl transferase)-mediated dUTP-biotin controls. Again the negative control values were subtracted from nick-end labeling) staining (Figure 4) and by the assessment of the MSP-stimulated proliferation rates. MSP concentration-depen- caspase 3 and 7 activity (Figure 5). Neither MSPmut nor MSPwt

& 2012 Macmillan Publishers Limited Genes and Immunity (2012) 321 --327 Macrophage-stimulating protein polymorphism rs3197999 FHa¨user et al 324

Figure 4. Effect of different concentrations of MSPwt (a-- d) and MSPmut (e-- h) on THP-1 apoptosis (detected by TUNEL staining). Apoptotic cells are labeled green. (a) 0.4 ng ml --1 MSPwt,(b) 2 ng ml À 1 MSPwt,(c) 5 ng ml À 1 MSPwt,(d)10ngmlÀ 1 MSPwt,(e) 0.4 ng ml À 1 MSPmut,(f) 2 ng ml À 1 MSPmut,(g) 5 ng ml À 1 MSPmut,(h)10ngmlÀ 1 MSPmut,(i) Supernatant of non-transfected CHO-K1 cells, (j, k) positive controls (j: incubation with DNAse I, k: incubation with staurosporine). No signiﬁcant difference was detected between MSPwt, MSPmut and the supernatant of non- transfected cells.

Figure 5. Effect of different concentrations of MSPmut and MSPwt on THP-1 apoptosis (detected by caspase 3 and 7 activity). Cells were incubated with different concentrations of MSPmut and MSPwt for 2 h (a) or with different concentrations of staurosporine for 4 h (b, positive control). Activity of caspase 3 and 7 was determined and is shown as relative luminescence units per second (r.l.u. s À 1) ±s.e.m. (nX4). Caspase 3/7 activity obtained in the wells without any stimuli was low and was subtracted from the MSP- or staurosporine-stimulated caspase 3/7 activity.

induced apoptosis. Supernatant of non-transfected CHO-K1 cells polymorphisms).3,4,9 --11,15,30 Some of these SNPs were useful in was used as negative control. Staurosporine-induced apoptosis is defining IBD-associated genes, which were then traced to several shown in Figure 5 and revealed the functionality of the caspase 3 pathways including microbe recognition and autophagy by and 7 activity assay. macrophages, interleukin (IL) signaling and lymphocyte activation, survival and growth.3 Only one gene, namely NOD2 (nucleotide- binding oligomerization domain protein 2), contributes substan- DISCUSSION tially to the risk of CD (20-fold risk in homozygous carriers31 --33). GWAS are undertaken in order to identify genes contributing to Most other SNPs reported to date increase disease risk to a much the genetic backgrounds of complex, multifactorial diseases and lesser extent (odds ratio o1.5) but might be valuable in to generate hypotheses concerning disease pathogenesis. IBD was elucidating IBD pathophysiology. For the bulk of identified SNPs, reported to be associated with a variety of SNPs (single-nucleotide however, their respective pathophysiological significance remains

Genes and Immunity (2012) 321 --327 & 2012 Macmillan Publishers Limited Macrophage-stimulating protein polymorphism rs3197999 FHaüser et al 325 unclear and in many cases, the IBD-associated genes have not phages of the inflammatory phenotype induces T-helper cell been identified unequivocally, yet. rs3197999 in the MST1 gene, differentiation (Th1, Th17) and/or stabilization of the differ- one of the most convincingly replicated IBD-associated poly- entiation status,41 which is essential for the activation of the morphisms4,9 --12,15,30 causes the exchange of an amino acid adaptive immune response in IBD. conserved across species and predicted to alter protein structure 3. Furthermore, monoclonal antibodies targeting the p40 subunit and/or function (prediction tools: sorting tolerant from intolerant of IL-12 and IL-23 were effective in treating patients with CD in (SIFT) algorithm (http://sift.jcvi.org), Polymorphism Phenotyping preliminary studies.42 v2.1.0 (http://genetics.bwh.harvard.edu/pph2/index.shtml). We expressed the MST1 gene in a eukaryotic cell system and we were We conclude that MST1 is the most likely IBD/PSC-associated able to demonstrate a significantly enhanced activity of the mut gene defined by the rs3197999 association and rs3197999 itself is MSP protein (Cys689): MSPmut-induced migration and proliferation the genetic variant causing disease association. The gain of in macrophage-like THP-1 cells more effectively than MSPwt function caused by rs3197999 has at least two effects, which would (Figures 2 and 3). A broad set of evidence, detailed below, be expected to promote local inflammation: (1) increased suggests that this spectrum of macrophage functions play a major attraction of macrophages to chemotactic agents and (2) increased role in IBD and also PSC pathogenesis, strongly arguing for MST1 macrophage proliferation. as the disease-associated gene within region 3p21 and for rs3197999 as the association-causing genetic variant. Macrophages are essential in the rapid response to bacteria and MATERIALS AND METHODS other pathogens. They initiate the recruitment of cells of the Reagents and antibodies adaptive immunity. In the gut epithelium, a large amount of Recombinant human IFN-g and TNF-a, mouse serum and all cell culture resident macrophages meets with a mass of luminal bacteria, reagents were obtained from Invitrogen (Carlsbad, CA, USA). LPS from which does not induce inflammation in healthy individuals. Escherichia coli serotype 0111:B4, LPS from Samonella typhi, and Mucosal macrophages originate from CD14 þ blood monocytes staurosporine were purchased from Sigma (St Louis, MO, USA). Human expressing surface molecules facilitating the adhesion of mono- recombinant kallikrein was obtained from Enzyme Research Laboratories cytes especially to sites of infection and inflammation but also to Ltd. (Swansea, UK). Anti-human MSP antibody was purchased from R&D unaffected mucosa.34 --36 In the environment of a non-inflamed Systems (Minneapolis, MN, USA), anti-human RON antibody from Santa mucosa, recruited macrophage precursor cells lose their ability to Cruz Biotechnology (Santa Cruz, CA, USA). Anti-goat IgG and anti-rabbit IgG migrate towards chemoattractants, to proliferate, to secrete antibodies conjugated with HRP were obtained from Jackson (Suffolk, UK). cytokines and to respond by respiratory burst. Instead, they intensify their bactericidal activity and their phagocytic activity of Cell culture inert material. Thus, resident intestinal macrophages scavenge cell debris and harmless microorganisms without inducing any CHO-K1 cells were obtained from the ECACC (Salsbury, UK). Cells were inflammatory response. However, macrophages of an inflamma- grown in F12 medium supplemented with 10% fetal bovine serum (FBS) tory phenotype, as observed in infectious diseases and IBD for and 2 mM glutamine. instance, conserve the properties of the precursor cells in many HepG2 cells and THP-1 (acute monocytic leukemia) cells were purchased respects. In UC as well as in CD, excessive activation of the innate from the DSMZ (Braunschweig, Germany) and the latter were maintained immunity initiates the inflammatory process, which is a pre- in RPMI 1640 medium supplemented with 10% FBS, 1 mM sodium pyruvate requisite of the activation of the adaptive immune system, the and 2 mM glutamine. HepG2 cells were cultured in RPMI 1640 medium latter being the main cause of tissue damage in IBD.2,5 A similar supplemented with 10% FBS. The cell lines were cultured in a 37 1C mechanism is postulated in PSC except that Kupffer cells are not humidified atmosphere containing 5% CO2. For all subsequent experiments, THP-1 cells were pre-stimulated by IFN-g (400 U ml À 1) and TNF-a directly exposed to luminal gut microbes but to bacterial and À 1 pathogenic material from the portal vein. rs3197999 is one of the (10 ng ml ) in RPMI 1640 medium containing 10% FBS, 2 mM glutamine few SNPs associated with three different disease entities, UC, CD und 1 mM sodium pyruvate for 24 h. The assays were performed in RPMI and PSC, which although distinct from each other share a 1640 medium containing 1% FBS. common macrophage-dependent initiation mechanism. Whereas RON is not expressed in monocytes of the circulating blood, its Preparation of EAMS expression could be demonstrated in macrophage precursor cells À 1 34,35,37 Mouse serum was incubated with S. typhi LPS (10 mg ml ) for 90 min at adhering to the vessel wall. At this early stage of 1 mut 37 C. The endotoxin serum mixture was heat inactivated for 30 min at macrophage recruitment, the presence of MSP may impair 56 1C and centrifuged at 1000 g and 4 1C for 10 min. The maximal the differentiation into resident intestinal macrophages but chemotactic activity was obtained by a 100-fold dilution of EAMS in cell instead promote an inflammatory phenotype. This hypothesis is culture medium. supported by the fact that MSPmut enhances a certain spectrum of macrophage functions, which is characteristic of the inflammatory phenotype of macrophages. Cloning of MSP cDNA and in vitro mutagenesis A hyperinflammatory immune response of mononuclear cells Total RNA extracted from HepG2 cells was used as a template for synthesis similar to that induced by MSPmut is known to cause enterocolitis of oligo(dT)-primed cDNA, using Superscript-III (Invitrogen). Full-length as illustrated by the following examples: MSP was amplified by a proofreading polymerase (forward primer 50- CCAGGAATTCTGATGGGGTGGCTCCCACT-30; reverse primer 50-CTGGGCCT 1. The rare, homozygous IL-10 receptor defect prevents the IL-10- AACCCAGTCTCA-30). mediated inhibition of pro-inflammatory cytokine release and The PCR product was cloned into the pcDNA3.1/V5-His vector results in severe, early-onset IBD.38 (Invitrogen) and transformed into prokaryotic TOP10 cells. In vitro 2. Polymorphisms located in genes of the IL-23 pathway (for mutagenesis was performed using the QuikChange II Site-Directed example, IL-23R; JAK2, STAT3) were shown to be associated with Mutagenesis Kit (Stratagene, La Jolla, CA, USA) according to the both, UC and CD, with the minor allele being protective against manufacturer’s instructions. The following primers were used to generate IBD.3,39 In resident peritoneal macrophages stimulated by LPS the c.2078C4T/p.Arg689Cys MSP mut: forward primer 50-ATGCGC (lipopolysaccharide) and IFN-g, MSP was reported to suppress AAGGTCCTGCTGGCCAGCTGTCTTCACGC-30; reverse primer 50-GCGTGAAG IL-12/IL-23 p40 and TNF-a release independent from IL-10.40 ACAGCTGGCCAGCAGGACCTTGCGCAT-30. The sequence of wt and mut Deficient down-regulation of IL-12/IL-23 release by macro- MSP insert was checked by conventional sequencing.

& 2012 Macmillan Publishers Limited Genes and Immunity (2012) 321 --327 Macrophage-stimulating protein polymorphism rs3197999 FHaüser et al 326 Expression of human MSP in CHO-K1 cells Analysis of apoptosis CHO-K1 cells, a well-established system for the expression of human MSP-dependent programmed cell death was analyzed by determining MSP,43,44 were transfected with the eukaryotic expression vector caspase 3 and 7 activity and by quantifying the degree of DNA pcDNA3.1/V5-His, containing wt or mut MSP cDNA using FuGENE HD fragmentation: Transfection Reagent (Roche, Mannheim, Germany) and transfected cells A total of 25 000 THP-1 cells in 50 ml medium containing CHO-K1 were selected in 1 mg ml --1 G418 (Roth, Karlsruhe, Germany). Individual supernatant with MSPwt/MSPmut or supernatant of non-transfected cells clones were obtained by limiting dilution. Expression of human MSP mRNA were incubated for 1, 2 and 4 h in a 96-well microplate. Caspase 3 and 7 was checked by conventional PCR using the following primers: forward activities were determined using Caspase-Glo 3/7 assay according to the primer (sequence located in expression vector): 50-ATAGGGAGACC manufacturer’s instructions (Promega Corp.). As control, apoptosis was CAAGCTGGCTAGT-30; reverse primer (sequence located in MSP transcript): induced by different concentrations of staurosporine. 50-GGCCGCACGTCGTCTGTA-30. The primer design ensured the amplifica- Alternatively TUNEL staining was performed. THP-1 cells were pre- tion of human recombinant MSP. treated for 24 h with IFN-g (400 U ml À 1) and TNF-a (10 ng ml À 1) in eight- The supernatant of transfected and non-transfected CHO-K1 cells well Lab-Tek chamber slides (Nunc, Roskilde, Denmark) and incubated with (grown in RPMI 1640 supplemented with 1% FBS, without G418) was different concentrations of kallikrein-activated and non-activated MSPwt/ harvested. MSP concentrations were quantified by ELISA according to the MSPmut or supernatant of non-transfected cells for 24 h. DNA fragmenta- manufacturer’s instructions (R&D Systems). Pro-MSP was activated by tion was examined in situ by the TUNEL method using the In situ Cell Death kallikrein digestion for 30 min at 37 1C with 30 mg kallikrein ml À 1. Detection Kit (Roche) according to the manufacturer’s instructions. As a negative control, cells were incubated with all assay reagents omitting the labeling enzyme. As a positive control, cells were pre-treated with Western blot analysis 600 U ml À 1 DNAse I. The number of apoptotic cells per visual field was Equal amounts of pro-MSP and kallikrein-digested MSP in CHO-K1 lysates determined by counting the stained cells under a fluorescence microscope and supernatants were resolved by 10% SDS--PAGE (sodium dodecyl (Leica DM6000 B; Leica Microsystems, Wetzlar, Germany & Volocity 4.0; sulfate polyacrylamide gel electrophoresis) and examined by standard Perkin-Elmer, Waltham, MA, USA) western blotting procedures. Anti-human MSP antibody was used at a 1:1500 dilution, anti-goat IgG was diluted 1:10 000. RON expression was analyzed in THP-1 cells before and after induction of differentiation by Statistics SDS--PAGE and western blotting. Anti-human RON antibody was used at a MedCalc 4 (Mariakerke, Belgium) was used to perform all statistical 1:500 dilution, anti-rabbit IgG was diluted 1:10 000. analyses. Comparisons between the two groups were performed by paired two-tailed or one-tailed t-test. P-values p0.05 were considered significant.

RON and GILT-PCR RON and GILT mRNA was detected by real-time PCR in relationship to 18S CONFLICT OF INTEREST rRNA using the LightCycler 2.0 instrument (Roche) and the QuantiTect The authors declare no conflict of interest. SYBR Green PCR Kit at standard conditions (PCR primers: GILT forward: 50-GCTGTCGCCAGACACTATCA-30, GILT reverse: 50-AGCTGGGTCTGATCTTC- 0 45 0 0 0 CAA-3 ; RON forward: 5 -AGCCCACGCTCAGTGTCTAT-3 , RON reverse: 5 - ACKNOWLEDGEMENTS GGGCACTAGGATCATCTGTCA-30, 18S rRNA forward: 50-GATACCGCAGCTAG- This article includes experimental work performed by Friederike Haüser in fulfilment GAATAATG-30, 18S rRNA reverse: 50-GCGCAATACGAATGCC-30). Specificity of the requirements for her doctoral thesis. This work was supported by the research of the respective PCR reactions was assured by final melting curve analysis. and education funds of the University Clinic of Mainz (MAIFOR program, Julia K Bickmann). Staining of THP-1 cells The morphological properties of differentiated THP-1 cells were deter- REFERENCES mined by Wright-Giemsa staining of cytospin preparations. 1 Sartor RB. Microbial influences in inflammatory bowel diseases. Gastroenterology 2008; 134: 577 --594. Macrophage chemotaxis assay 2 Xavier RJ, Podolsky DK. Unravelling the pathogenesis of inflammatory bowel disease. Nature 2007; 448: 427 --434. A total of 50 000 THP-1 cells in 100 ml were added to the upper chamber of 3 Cho JH, Brant SR. Recent insights into the genetics of inflammatory bowel disease. a transwell filter insert (24-well plates, 5 mm pore size; Corning B.V. Life Gastroenterology 2011; 140: 1704 --1712.e2. Sciences, Amsterdam, The Netherlands) containing CHO-K1 supernatant 4 Franke A, McGovern DPB, Barrett JC, Wang K, Radford-Smith GL, Ahmad T et al. with different concentrations of MSPwt/MSPmut. The medium in the bottom Genome-wide meta-analysis increases to 71 the number of confirmed Crohn’s well contained 10 mlml--1 EAMS as a chemoattractant. After 24 h, the disease susceptibility loci. Nat Genet 2010; 42: 1118 --1125. medium of the lower chamber was collected and the number of migrated 5 Smith PD, Ochsenbauer-Jambor C, Smythies LE. Intestinal macrophages: unique cells per well was determined by cell counting in a hemocytometer or by effector cells of the innate immune system. Immunol Rev 2005; 206: 149 --159. 6 Bernstein CN, Blanchard JF, Rawsthorne P, Yu N. The prevalence of extraintestinal measuring ATP concentrations (Kinase-Glo Luminescent Kinase Assay; diseases in inflammatory bowel disease: a population-based study. Am J Promega Corp., Madison, WI, USA). Gastroenterol 2001; 96: 1116 --1122. 7 Cameron RG, Blendis LM, Neuman MG. Accumulation of macrophages in primary sclerosing cholangitis. Clin Biochem 2001; 34: 195 --201. Macrophage proliferation assay 8 Bowlus C. Cutting edge issues in primary sclerosing cholangitis. Clin Rev Allergy The cell proliferation rate was determined using a cell proliferation ELISA Immunol 2011; 41: 139 --150. (Roche) via BrdU incorporation into newly synthesized DNA according to 9 Anderson CA, Massey DCO, Barrett JC, Prescott NJ, Tremelling M, Fisher SA et al. the manufacturer’s protocol. Investigation of Crohn’s disease risk loci in ulcerative colitis further defines their BrdU was added to a total of 50 000 THP-1 cells in 100 ml medium molecular relationship. Gastroenterology 2009; 136: 523 --529.e3. 10 Barrett JC, Hansoul S, Nicolae DL, Cho JH, Duerr RH, Rioux JD et al. Genome-wide containing CHO-K1 supernatant with MSPwt/MSPmut. Cells were incubated association defines more than 30 distinct susceptibility loci for Crohn’s disease. for 24 h in a 96-well microplate. Subsequently, the plates were centrifuged Nat Genet 2008; 40: 955 --962. at 450 g for 10 min and the medium was removed. The cells were dried at 11 Fisher SA, Tremelling M, Anderson CA, Gwilliam R, Bumpstead S, Prescott NJ et al. 60 1C for 1 h and then treated as described in the manufacturer’s Genetic determinants of ulcerative colitis include the ECM1 locus and five loci instructions. implicated in Crohn’s disease. Nat Genet 2008; 40: 710 --712.

Genes and Immunity (2012) 321 --327 & 2012 Macmillan Publishers Limited Macrophage-stimulating protein polymorphism rs3197999 FHaüser et al 327 12 Goyette P, Lefebvre C, Ng A, Brant SR, Cho JH, Duerr RH et al. Gene-centric 29 Chen Q, DeFrances MC, Zarnegar R. Induction of met proto-oncogene association mapping of chromosome 3p implicates MST1 in IBD pathogenesis. (hepatocyte growth factor receptor) expression during human monocyte- Mucosal Immunol 2008; 1: 131 --138. macrophage differentiation. Cell Growth Differ 1996; 7: 821 --832. 13 Karlsen TH, Franke A, Melum E, Kaser A, Hov JR, Balschun T et al. Genome-wide 30 Anderson CA, Boucher G, Lees CW, Franke A, D’Amato M, Taylor KD et al. Meta- association analysis in primary sclerosing cholangitis. Gastroenterology 2010; 138: analysis identifies 29 additional ulcerative colitis risk loci, increasing the number 1102 --1111. of confirmed associations to 47. Nat Genet 2011; 43: 246 --252. 14 Melum E, Franke A, Schramm C, Weismuller TJ, Gotthardt DN, Offner FA et al. 31 Hugot J, Chamaillard M, Zouali H, Lesage S, Cezard J, Belaiche J et al. Association Genome-wide association analysis in primary sclerosing cholangitis identifies two of NOD2 leucine-rich repeat variants with susceptibility to Crohn’s disease. Nature non-HLA susceptibility loci. Nat Genet 2011; 43:17--19. 2001; 411: 599 --603. 15 Waterman M, Xu W, Stempak JM, Milgrom R, Bernstein CN, Griffiths AM et al. 32 Abraham C, Cho JH. Functional consequences of NOD2 (CARD15) mutations. Distinct and overlapping genetic loci in Crohn’s disease and ulcerative colitis: Inflamm Bowel Dis 2006; 12: 641 --650. correlations with pathogenesis. Inflamm Bowel Dis 2011; 17: 1936 --1942. 33 Brant SR, Wang M, Rawsthorne P, Sargent M, Datta LW, Nouvet F et al. 16 Miller M, Leonard EJ. Mode of receptor binding and activation by plasminogen- A population-based case-control study of CARD15 and other risk factors related growth factors. FEBS Lett 1998; 429:1--3. in Crohn’s disease and ulcerative colitis. Am J Gastroenterol 2007; 102: 17 Yoshimura T, Yuhki N, Wang M, Skeel A, Leonard EJ. Cloning, sequencing, and 313 --323. expression of human macrophage stimulating protein (MSP, MST1) confirms MSP 34 Grimm MC, Pullman WE, Bennett GM, Sullivan PJ, Pavli P, Doe WF. Direct evidence as a member of the family of kringle proteins and locates the MSP gene on of monocyte recruitment to inflammatory bowel disease mucosa. J Gastroenterol chromosome 3. J Biol Chem 1993; 268: 15461 --15468. Hepatol 1995; 10: 387 --395. 18 Wang MH, Zhou YQ, Chen YQ. Macrophage-stimulating protein and RON receptor 35 Rugtveit J, Brandtzaeg P, Halstensen TS, Fausa O, Scott H. Increased macrophage tyrosine kinase: potential regulators of macrophage inflammatory activities. Scand subset in inflammatory bowel disease: apparent recruitment from peripheral J Immunol 2002; 56: 545 --553. blood monocytes. Gut 1994; 35: 669 --674. 19 Wang M, Yoshimura T, Skeel A, Leonard EJ. Proteolytic conversion of single chain 36 Smythies LE, Maheshwari A, Clements R, Eckhoff D, Novak L, Vu HL et al. precursor macrophage-stimulating protein to a biologically active heterodimer by Mucosal IL-8 and TGF-beta recruit blood monocytes: evidence for cross-talk contact enzymes of the coagulation cascade. J Biol Chem 1994; 269: 3436 --3440. between the lamina propria stroma and myeloid cells. J Leukoc Biol 2006; 80: 20 Donate LE, Gherardi E, Srinivasan N, Sowdhamini R, Aparicio S, Blundell TL. 492 --499. Molecular evolution and domain structure of plasminogen-related growth factors 37 Smith PD, Smythies LE, Shen R, Greenwell-Wild T, Gliozzi M, Wahl SM. Intestinal (HGF/SF and HGF1/MSP). Prot Sci 1994; 3: 2378 --2394. macrophages and response to microbial encroachment. Mucosal Immunol 2011; 21 Han S, Stuart LA, Degen SJ. Characterization of the DNF15S2 locus on human 4:31--42. chromosome 3: identification of a gene coding for four kringle domains with 38 Glocker E, Kotlarz D, Boztug K, Gertz EM, SchA˜ ffer AA, Noyan F et al. Inflammatory homology to hepatocyte growth factor. Biochemistry 1991; 30: 9768 --9780. bowel disease and mutations affecting the interleukin-10 receptor. N Engl J Med 22 Gaudino G, Follenzi A, Naldini L, Collesi C, Santoro M, Gallo KA et al. Ron is a 2009; 361: 2033 --2045. heterodimeric tyrosine kinase receptor activated by the HGF homologue MSP. 39 van Limbergen J, Wilson DC, Satsangi J. The genetics of Crohn’s disease. Annu Rev EMBO J 1994; 13: 3524 --3532. Genom Human Genet 2009; 10: 89 --116. 23 Gaudino G, Avantaggiato V, Follenzi A, Acampora D, Simeone A, Comoglio PM. 40 Morrison AC, Wilson CB, Ray M, Correll PH. Macrophage-stimulating protein, The proto-oncogene RON is involved in development of epithelial, bone and the ligand for the stem cell-derived tyrosine kinase/RON receptor tyrosine neuro-endocrine tissues. Oncogene 1995; 11: 2627 --2637. kinase, inhibits IL-12 production by primary peritoneal macrophages 24 Quantin B, Schuhbaur B, Gesnel M, Dolle´ P, Breathnach R. Restricted expression of stimulated with IFN-gamma and lipopolysaccharide. J Immunol 2004; 172: the RON gene encoding the macrophage stimulating protein receptor during 1825 --1832. mouse development. Am J Anat 1995; 204: 383 --390. 41 Bettelli E, Korn T, Oukka M, Kuchroo VK. Induction and effector functions of TH17 25 Iwama A, Wang MH, Yamaguchi N, Ohno N, Okano K, Sudo T et al. Terminal cells. Nature 2008; 453: 1051 --1057. differentiation of murine resident peritoneal macrophages is characterized by 42 Mannon PJ, Fuss IJ, Mayer L, Elson CO, Sandborn WJ, Present D et al. Anti expression of the STK protein tyrosine kinase, a receptor for macrophage- interleukin-12 antibody for active Crohn’s disease. N Engl J Med 2004; 351: stimulating protein. Blood 1995; 86: 3394 --3403. 2069 --2079. 26 Kurihara N, Iwama A, Tatsumi J, Ikeda K, Suda T. Macrophage-stimulating protein 43 Danilkovitch A, Miller M, Leonard EJ. Interaction of macrophage-stimulating activates STK receptor tyrosine kinase on osteoclasts and facilitates bone protein with its receptor. J Biol Chem 1999; 274: 29937 --29943. resorption by osteoclast-like cells. Blood 1996; 87: 3704 --3710. 44 Yoshikawa W, Hara H, Takehara T, Shimonishi M, Sakai H, Shimizu N et al. 27 Liu Q, Fruit K, Ward J, Correll PH. Negative regulation of macrophage activation in Characterization of free alpha- and beta-chains of recombinant macrophage- response to IFN-g and lipopolysaccharide by the STK/RON receptor tyrosine stimulating protein. Arch Biochem Biophys 1999; 363: 356 --360. kinase. J Immunol 1999; 163: 6606 --6613. 45 Verhoeckx KCM, Bijlsma S, de Groene EM, Witkamp RF, van der Greef J, 28 Nanney LB, Skeel A, Luan J, Polis S, Richmond A, Wang MH et al. Proteolytic Rodenburg RJT. A combination of proteomics, principal component analysis cleavage and activation of pro-macrophage-stimulating protein and upregulation and transcriptomics is a powerful tool for the identification of biomarkers of its receptor in tissue injury. J Invest Dermatol 1998; 111: 573 --581. for macrophage maturation in the U937 cell line. Proteomics 2004; 4: 1014 --1028.

Supplementary Information accompanies the paper on Genes and Immunity website (http://www.nature.com/gene)

& 2012 Macmillan Publishers Limited Genes and Immunity (2012) 321 --327