Molecular and Functional Characterization of Mouse S5D-SRCRB: A New Group B Member of the Scavenger Receptor Cysteine-Rich Superfamily

The innate immune system represents the first line of host defense for multicellular organisms to maintain homeostasis of the internal environment against foreign pathogens (microorganisms, chemicals), as well as altered self-components (1). To do this, the humoral (i.e., complement) and the cellular (e.g., mucocutaneous barriers, macrophages) arms of the innate immune system are equipped with a set of inborn (germ-line encoded) receptors named pattern recognition receptors (PRRs). These PRRs recognize a relatively small spectrum of highly conserved structures of protein, saccharide, lipid, and nucleic acid nature of endogenous and exogenous origin. Typically, PRRs recognize pathogen-associated molecular patterns (PAMPs), which are conserved components essential for pathogen survival and are not shared by the host. Examples of PAMPs include LPS from Gram-negative bacteria, lipotheicoic acid (LTA) and peptidoglycan (PGN) from Gram-positive bacteria, dsRNA from virus, lipoarabinomannan from mycobacteria, and mannan and β-glucan from fungi. Recognition of PAMPs by PRRs usually initiates an inflammatory cascade that involves recruitment of leukocytes to the site of infection, activation of antimicrobial effector mechanisms, and induction of adaptive immune responses that promote clearance of infection and long-term immune memory. In this way, the mammalian immune system eliminates invading pathogens with maximal efficacy and minimal damage to self (2). There are two main groups of PRRs: secreted molecules that circulate in blood and lymph and cell-surface receptors present on hemopoietic and epithelial cells. Each group is formed by several classes of PRRs, which have emerged and coevolved independently, giving rise to structurally and functionally different molecules, although with cooperative and complementary defensive and homeostatic functions. Examples of PRRs include proteins with leucine-rich repeats (i.e., TLRs), C-type lectin domains (i.e., dectin-1, DC-SIGN), and scavenger receptors (i.e., SR-AI/II, CD36, MARCO).

The scavenger receptor cysteine-rich superfamily (SRCR-SF) is an ancient and highly conserved group of membrane-bound and/or soluble proteins mostly reported in the animal kingdom, from low invertebrates to mammals (3–5), as well as in some aquatic plants [i.e., unicellular green alga (6)]. In mammals, SRCR-SF members can be expressed by hemopoietic and nonhemopoietic cells, at embryonic and adult developmental stages, depending on species and tissue type. These proteins were reported to play a role in the regulation of innate and adaptive immune responses, as well as in the development of the immune system (3–5). Members of the SRCR-SF are characterized by the presence of one or several repeats of a highly conserved cysteine-rich extracellular scavenger receptor cysteine-rich (SRCR) domain (∼100–110 aa in size), which was first reported on the mouse macrophage scavenger receptor type I (SR-AI) (7). Depending on the number of cysteine residues present in the SRCR domain and the number of exons coding for each domain, the SRCR-SF can be divided into two mutually exclusive groups: A and B. Group A contains SRCR domains encoded by two or more exons and including six cysteines forming three disulphide bonds, whereas group B domains are encoded by a single exon and contain eight cysteines forming four disulphide bonds (3–5).

The extracellular regions of SRCR-SF members may present as exclusively composed of SRCR domains repeated in tandem or as multidomain mosaic proteins in which the SRCR domains seem to be combined with other types of protein modules, such as epithelial growth factor, C1r/C1s Uegf Bmp1, zona pellucida, collagenous regions, fibronectin, and short consensus repeats. The presence of short Pro, Ser, and Thr (PST)-rich polypeptides interspaced with contiguous SRCR domains is also frequently observed among SRCR-SF members. Available three-dimensional structures obtained from crystallization experiments indicate that group A and B SRCR domains present a highly conserved and compact core folding (a curved six-stranded β sheet cradling an α helix) with variable outer loop regions, likely giving rise to functional diversity (8–13).

Despite the overall structure conservation of the SRCR domains across different species, no unifying biological function has been reported. They do not possess enzymatic activity; however, some SRCR domains have been involved in protein–protein interactions, the best studied examples being those of CD6 with CD166/ALCAM (4) and CD163 with the haptoglobin–hemoglobin complex (14). In recent years, a number of studies also supported the recognition of PAMPs by some, but not all, group A (i.e., MARCO) (15) and B (i.e., DMBT1/SAG/gp340, Spα, CD6, CD5, and CD163) (16–20) SRCR-SF members. Group B is composed of about a dozen members expressed in mammals by immune cells, such as macrophages (i.e., CD163/M130, CD163L1/M160, Spα/AIM) or lymphocytes (i.e., CD5, CD6, SCART, WC1), as well as by cells of the gastrointestinal, respiratory, and genitourinary tracts (i.e., DMBT1/SAG/gp340, S4D-SRCRB, 18-B) (5, 21). The present work reports on the characterization, at the molecular and functional levels, of S5D-SRCRB (soluble 5 domain SRCR type B), the mouse homolog of a recently cloned human soluble member belonging to group B of SRCR-SF (SSc5D), for which no biochemical and functional data are available (22). By generating a recombinant protein form and a set of specific mAbs, we showed that mouse S5D-SRCRB is a secreted glycoprotein with a restricted tissue-expression pattern (mainly cells of the genitourinary and digestive tracts), which is able to bind to endogenous extracellular matrix components (laminin, galectin-1), as well as to constitutive components of bacterial (Gram-negative or -positive) and fungal (saprophytic and pathogenic) surfaces. These data argue in favor of mouse S5D-SRCRB being a component of the humoral arm of the innate immune system likely involved in the defense and homeostasis of host epithelial surfaces.

PBS (Roche Diagnostics, Mannheim, Germany) contained 137 mM NaCl, 3 mM KCl, 8 mM Na₂HPO₄, 1.5 mM KH₂PO₄ (pH 7.4). TBS contained 50 mM Tris-HCl (pH 7.4), 140 mM NaCl. Radioimmunoprecipitation assay (RIPA) buffer contained 50 mM Tris (pH 8), 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS supplemented with complete protease inhibitors (Roche Diagnostics), and 1 mM Na₃VaO₄.

The human embryonic kidney cell line HEK 293-EBNA, which stably expresses the EBNA-1 Ag from EBV, was kindly provided by Takako Sasaki (Max-Planck-Institut für Biochemie, Martinsried, Germany). HEK 293 cells stably expressing TLR2 were a kind gift from Dr. Golenbock (University of Massachusetts Medical School, Worcester, MA). HEK 293-EBNA cells were grown in DMEM/F12 (Life Technologies Life Science, Grand Island, NY) supplemented with antibiotics (100 U/ml penicillin, 100 μg/ml streptomycin), 10% heat-inactivated FCS (Walkersville, MD), and 250 μg/ml Geneticin (Life Technologies).

Bioinformatic tools available in public databases (http://www.ncbi.nlm.nih.gov) were used to search for new members of group B of the SRCR-SF. The high-throughput genome sequences and the nonredundant database sections of GenBank (http://www.ncbi.nlm.nih.gov/Genbank) were screened using the TBLASTN algorithm with the protein sequences of CD5 and CD6 as templates (Acc. No. XO4391 and U34623, respectively). Mouse genomic clones with significant similarities (Evalue < 10⁻¹⁰) were selected and analyzed through the BLASTN algorithm against the mouse expressed sequences tagged (ESTs) database of GenBank. The ESTs or full-length clones of interest were obtained from distribution centers already cloned into pFLCI and were further sequenced in both directions with the ABI PRISM dRhodamine terminator Cycle Sequencing Ready Reaction kit (PE Applied Biosystems, Warrington, U.K.) using universal M13 primers (M13-20.Fw 5′-GTAAAACGACGGCCAGT-3′; M13.Rv 5′-AACAGCTATGACCATG-3′). The AK079906 or A430110N23 full-length cDNA sequences (gi: 26348280) were used for mouse chromosome mapping of s5d-srcrb (http://www.ncbi.nlm.nih.gov/mapview). Similarly, determination of exon/intron boundaries was done by alignment of AK079906 with the mouse genomic sequence AC0794876 (gi: 9964851) using the BLAST2 tool (http://www.ncbi.nlm.nih.gov/BLAST).

The full-length cDNA of s5d-srcrb tagged with a C-terminal hemagglutinin (HA) sequence was cloned into a modified version of the pCEP-Pu/AC7 mammalian expression vector (kindly provided by Takako Sasaki, Max-Planck-Institut für Biochemie). To this purpose, the AK079906 full-length cDNA sequence was used as a template for PCR amplification (Taq Expand High Fidelity Polymerase System; Roche Diagnostics) with the specific primers msd1.Fw (5′-CCCAAGCTTTCAGCAGCTGCTTCCCTCCCT-3′) and msd2.Rv (5′-GAAGATCTCACATCTCCCCTCAGAGGCCT-3′) containing HindIII and BglII restriction sites, respectively (underlined). The cycling conditions were one step of 95°C for 5 min; 35 cycles of 95°C for 45 sec, 63°C for 45 sec, and 72°C for 4 min 30 sec; and a final extension step of 72°C for 10 min. The resulting PCR product was purified (PCR purification kit, Qiagen, Hilden, Germany) and then digested with HindIII and BglI (MBI Fermentas, Glen Burnie, MD). The restricted product was gel purified (QIAquick Gel Extraction Kit, Qiagen) and cloned into an appropriately digested (HindIII and BamHI) pCep-Pu/AC7-HAtag vector using the T4 DNA Ligase (Takara, Shiga, Japan). The ligation products were transformed into DH5α Escherichia coli-competent cells, and the resulting clones were checked by sequencing in both directions with vector-specific primers and the Prism Ready Reaction Big Dye Deoxy Terminator sequencing kit (PE Applied Biosystems). HEK293-EBNA cells were transfected with the above-mentioned construct using Lipofectamine 2000 reagent (Invitrogen, Paisley, U.K.), according to the manufacturer’s protocol. Transfectants were selected 48 h later in DMEM/FCS/Geneticin plus 1 μg/ml Puromycin (Sigma, St. Louis, MO) and left to grow to confluence. To perform protein-expression analysis, confluent transfectants were cultured in serum-free DMEM/F12 medium (without selecting antibiotics), and supernatants were collected every 48–72 h; storage of supernatants was at −20°C after supplementation with 0.5 mM PMSF, 0.02% sodium azide, and 2.5 mM EDTA. Supernatant samples (1.5 ml) were precipitated with TCA and run over 6–8% SDS-PAGE under reducing conditions, followed by electrotransfer to nitrocellulose membranes (Bio-Rad Laboratories, Richmond, CA). For Western blot analysis, membranes were blocked for 1 h at room temperature with PBS containing 10% nonfat milk and then incubated with a mouse anti-HA mAb (cat. #sc-7392; Santa Cruz, Santa Cruz, CA). After extensive washing with PBS 0.01% Tween20, membranes were further incubated with a 1:1000 dilution (in blocking buffer) of HRP-conjugated sheep anti-mouse IgG (Amersham Biosciences, Fairfield, CT). Blots were visualized by chemiluminescence with Supersignal West Dura Extended Duration Solution (Pierce, Rockford, IL).

Analysis of the expression pattern of s5d-srcrb was performed in C57BL/6J mouse tissues by real-time retrotranscription quantitative PCR (RT-qPCR) analysis. Total RNA was extracted using the TRIzol reagent method, as recommended by the manufacturer (Invitrogen). cDNA synthesis was performed using the GeneAmp PCR kit (Roche). Briefly, 0.5–1 μg total RNA was mixed with 2.5 μM oligo-d(T)16 primer, 1 mM 2′-deoxynucleoside 5′-triphosphates, 1× PCR Buffer II, and 5 mM MgCl₂. After denaturing for 5 min at 65°C, 20 U RNase inhibitor and 50 U Moloney murine leukemia virus reverse transcriptase were added (total reaction volume, 20 μl). The reaction was incubated for 90 min at 42°C and stopped for 15 min at 99°C. cDNA integrity was checked by PCR using primers specific for the housekeeping 18S gene (18S.Fw: 5′-AAACGGCTACCACATCCAAG-3′ and 18S.Rv: 5′-CCTCCAATGGATCCTCGTTA-3′) and s5d-srcrb gene (mS5D.Fw: 5′-AAGACGTGGTGCTCACCTGC-3′ and mS5D.Rv: 5′-TCCCACGTCCAGGAAGACTC-3′) using the following cycling conditions: 2 min at 93°C, followed by 30 cycles of 93°C for 45 s, 56°C for 45 s, and 72°C for 5 min. For RT-qPCR, the same s5d-srcrb and 18S gene-specific primers were used in a reaction mixture of cDNA (2 μl; 1:100 diluted) and LightCycler FastStart DNA Master SYBR Green I mix (Roche Diagnostics). Reactions were carried out on an ABI 7900HTT FastReal Time PCR System (Applied Biosystems) under the following cycling conditions: 10 min at 95°C, 40 cycles of 95°C for 15 s, and 60°C for 1 min. Gene expression was normalized to 18S and expressed in arbitrary units.

For immunization purposes, each Sprague-Dawley rat was injected i.p. fortnightly with TCA precipitates from 12 ml FCS-free supernatant samples of recombinant HA-tagged mouse S5D-SRCRB (rmS5D-SRCRB-HA)-producing stable HEK 293-EBNA transfectants. In brief, FCS-free supernatant was precipitated in 14% TCA and 0.14% Triton X-100 for 10 min on ice and then pelleted at 14,000 rpm for 30 min at 4°C, washed with cold acetone, and resuspended with PBS. The first dose of Ag was emulsified with CFA (Sigma), and the next two immunizations were emulsified with IFA (Sigma). The spleen was removed at day 3 after i.v. injection of the Ag without adjuvant, and cells were fused with the non-Ig–secreting myeloma X63-Ag8.653 using polyethylene glycol 1500 (Sigma) (23). Hybridomas were selected by growth in RPMI 1640 culture medium (Cambrex Bioscience, Walkersville, MD) containing hypoxanthine, aminopterin, and thymidine (10) (Sigma) and supplemented with 12% FCS (Life Technologies, Invitrogen), 2 mM glutamine (Life Technologies, Invitrogen), 1 mM pyruvate (Sigma), 100 U/ml penicillin, and 100 μg/ml streptomycin. Hybridoma cell supernatants were analyzed by ELISA using plates coated with FCS-free rmS5D-SRCRB-HA supernatants and HRP-conjugated goat anti-rat IgG antiserum (Sigma). Anti–rmS5D-SRCRB-HA–reactive hybridomas were subcloned twice by limiting dilution after discarding their reactivity against the HA tag. The immunoreactivity of the mAb obtained was analyzed by different immunochemical techniques (immunoprecipitation, ELISA, and Western blot).

Tissue samples from C57BL/6J males were snap-frozen in liquid nitrogen and then weighed, minced, and mixed with three volumes of RIPA lysis buffer for homogenization with a Teflon-glass Dounce, followed by sonication with two short bursts (2 s) in a Barnson Sonifier 250 (Danbury, CT). Tissues were kept on ice for 20 s in between bursts. After clarification at 12,000 rpm for 20 min at 4°C, total protein content from supernatants was quantified by the Bradford method (BCA Protein assay kit; Pierce), as recommended by the manufacturer. For control purposes, 1 × 10⁷ HEK 293-EBNA transfectants expressing rmS5D-SRCRB-HA were lysed following the same procedure. Protein samples (100 μg) were resolved by 6–7.5% SDS-PAGE and then electrotransferred to a nitrocellulose membrane (BioRad, Durham, NC). Subsequently, Western blot analysis was performed using a 1:2 dilution of rat 1H11.A8.G2 hybridoma supernatant plus goat HRP-conjugated anti-rat IgG (Sigma), as previously described.

The rmS5D-SRCRB-HA was immunoprecipitated using serum-free rat anti–S5D-SRCRB 1H11.A8.G2 hybridoma supernatant previously adsorbed for 1 h at 4°C under orbital rotation with 50 μl a 50% (v/v) suspension (in PBS) of Protein G Sepharose 4 Fast Flow (GE Healthcare, Uppsala, Sweden). The beads were pulled down, washed three times with PBS, and incubated with serum-free rmS5D-SRCRB-HA supernatant for 1 h at 4°C under gentle rotation. The beads were washed again, as before, and treated with different glycosidase combinations from the Enzymatic Protein Deglycosylation Kit (Sigma), following the manufacturer’s instructions. The resulting products were resolved by SDS-PAGE and analyzed by Western blot, as described above.

E. coli ATCC25922 and Staphylococcus aureus ATCC29213 were obtained from the American Type Culture Collection. Schizosaccharomyces pombe and Saccharomyces cerevisiae were kindly provided by the Cell Biology Unit of the University of Barcelona. The rest of the bacterial and fungal strains used in this study are clinical isolates characterized by the Department of Microbiology, Hospital Clinic of Barcelona, using standard biochemical procedures. Bacteria were grown overnight at 37°C in Luria-Bertani medium, and fungi were grown at 28°C in Sabouraud’s medium with aeration, harvested by centrifugation at 3500 rpm for 10 min, and resuspended in TBS to a final density of 10¹⁰ bacteria/ml or 10⁸ fungi/ml. Aspergillus fumigatus was grown at 30°C in Sabouraud’s for 1 wk, and the cells were resuspended in TBS to a final DO_530nm = 0.2, which corresponds to a final density of 10⁸ fungi/ml. Under these conditions, the cell suspension was a mixture of mainly conidiophores, conidia, and filaments. Quantification was performed by plating bacteria and fungi dilutions on agar. For microbial-binding assays, aliquots ∼5 × 10⁷ bacteria and ∼5 × 10⁶ fungi were incubated for 1 h at 4°C under gentle orbital rotation with different amounts of serum-free supernatant from HEK 293-EBNA transfectants expressing rmS5D-SRCRB-HA in binding buffer (TBS plus 1% BSA and 5 mM CaCl₂) to a final volume of 0.5 ml. In parallel, 100 μl testis and liver lysates was incubated with ∼5 × 10⁷ Gram-positive bacteria (S. aureus) in binding buffer to a final volume of 0.5 ml by gentle orbital rotation for 1 h at 4°C. The lysates were obtained from 95 mg tissue samples, minced, and mixed with three volumes of binding buffer for homogenization with a Teflon-glass Dounce, followed by sonication with two short bursts in a Branson Sonifier 250. Following incubation, microbial cells were pelleted and washed once with 0.5 ml binding buffer and twice with 0.5 ml TBS plus 5 mM CaCl₂ to remove nonspecifically bound proteins. The washed pellets were resuspended in reducing Laemmli’s sample buffer for 15 min at 100°C and separated on 6–7.5% SDS-PAGE gels, followed by electrotransfer to a nitrocellulose membrane (BioRad). Membranes were subjected to Western blot analysis, as described above. Aggregation assays were performed using fresh overnight cultures of E. coli, Acinetobacter baumannii, S. aureus, and Candida albicans harvested by centrifugation at 3500 rpm for 10 min and then resuspended in PBS to a final density of 10¹⁰ bacteria or 10⁸ fungi per ml. FITC (Sigma) was dissolved in PBS to a concentration of 10 mM and then added to the bacterial/fungal suspensions to a final concentration of 1 mM for 30 min at room temperature. Following removal of excess unbound FITC by extensive washing with PBS, 10 μl FITC-labeled bacterial/fungal suspension (5 × 10⁷ bacteria and 5 × 10⁶ fungi) in TBS containing 5 mM CaCl₂ was incubated overnight at room temperature with 20 μg/ml affinity-purified rmS5D-SRCRB-HA or BSA. Then, 10 μl the samples was examined by fluorescence microscopy.

The plasmid used for bacterial expression of GST-Galectin 1 (GST-Gal1) was kindly provided by Dr. Fu-Tong Liu (UC Davis, Davis, CA). Overnight cultures of transformed BL21 E. coli cells were diluted 1:10 in 500 ml Luria-Bertani broth supplemented with ampicillin (0.1 mg/ml) and grown for ∼4 h at 37°C until OD_600nm ∼0.8. Then, expression of GST-Gal1 was induced by the addition of isopropyl β-D-1-thiogalactopyranoside (0.1 mM) for an additional 4 h. The bacterial cell pellet was subjected to three cycles of freezing/thawing and was sonicated and solubilized with 10% Triton X-100 in the presence of protease inhibitors (Complete; Roche). The clarified supernatant was mixed with Glutathione Sepharose 4B beads (GE Healthcare) for 2 h at 4°C under gentle orbital rotation. The beads were then washed thoroughly with TBS 1% Triton X-100, resuspended in TBS, and kept at 4°C until used. GST-Gal1–Sepharose beads were incubated for 90 min at 4°C in blocking buffer (TBS plus 5% BSA). Then, 10 μl 50% (v/v) beads were incubated overnight at 4°C with 25 μl serum-free rmS5D-SRCRB-HA–containing supernatants in the presence or absence of increasing concentrations (3, 15, 30 mM) of competing sugars (lactose or sucrose). Beads were washed twice with TBS 0.01% Tween 20 and once with TBS alone to remove unbound protein. The proteins were eluted with Laemmli’s sample buffer and separated on SDS-PAGE under reducing conditions for further Western blot analysis.

Interaction of rmS5D-SRCRB-HA with PAMP was analyzed by coating 96-well microtiter plates (Nunc, Roskilde, Denmark) overnight at 4°C with 2 μg/well purified LPS from E. coli K12 (InvivoGen, San Diego, CA), PGN (cat. #77140; Sigma), LTA from S. aureus (cat. #2515; Sigma), Zymosan A (cat. #z4250; Sigma), glucan from bakers’ yeast (cat. #g5011; Sigma), β(1→3)-glucan (cat. #89862; Sigma), β-d-glucan (cat. #g6513; Sigma), or mannan (cat. #m7504; Sigma) in coating buffer (100 mM NaHCO₃ [pH 9.5]). Nonspecific binding to plastic wells was prevented by incubation for 1 h at room temperature in blocking solution (PBS plus 3% BSA). Serial 2-fold dilutions of serum-free rmS5D-SRCRB-HA supernatants were added to the wells and incubated for 2 h at room temperature. Bound protein was detected by a 1-h incubation at room temperature with neat serum-free 1H11.A8.G2 rat hybridoma supernatant, followed by a 30-min incubation with a 1:1000 dilution of goat anti-rat IgG/HRP antiserum (Sigma). Between each incubation step, excess unbound proteins were washed off three times with PBS 0.01% Tween 20. Color was developed by adding 3,3′,5,5′-tetramethylbenzidine (TMB) liquid substrate (Sigma), and the absorbance was measured at 450 nm. The assay was repeated twice with similar results.

Interaction of rmS5D-SRCRB-HA with extracellular matrix proteins was analyzed by coating ELISA microtiter plates overnight at 4°C with 0.5 μg/well purified fibronectin, laminin, collagen I, and collagen IV (supplied by Angeles García-Pardo, Centro de Investigaciones Biológicas-Centro Superior de Investigaciones Científicas, Madrid) in PBS (Roche). The assay conditions were the same as described above, with the exception that incubations were performed at 4°C, and the 4D11.A2.H4 hybridoma supernatant was used for detection purposes. The assay was repeated twice with similar results.

Four-micrometer sections from neutral-buffered formaldehyde-fixed paraffin-embedded tissue blocks were mounted on ChemMate Capillary Gap Slides (Dako, Glostrup, Denmark), dried at 60°C, deparaffinized, and hydrated. Prior to Ag retrieval, blocking of endogenous peroxidase was performed in 1.5% hydrogen peroxide in TBS buffer (pH 7.4) for 10 min. Ag retrieval was carried out using microwave heating in target retrieval solution (Dako) or pepsin. Three Tissue-Tek containers (Miles, Elkhart, IN), each with 24 slides in 250 ml buffer, were placed on the edge of a turntable inside the microwave oven. Slides were heated for 11 min at full power (900 W) and then for 15 min at 400 W. After heating, slides were left in buffer for 15 min. Incubation with 1:10 diluted supernatants (in Ab Diluent S2022; Dako) from anti–S5D-SRCRB hybridomas (1H11.A2.G8, 4D11.A2.H4, 8F4.F3.F9, 5B7.B8.A1, 8C11.D6.C9, 7G4.E1.H7, and 5E12.G3.B1) was carried out for 60 min at room temperature. No immunostaining was seen when the primary Ab was omitted or substituted with an isotype control. Immunostaining was automated using the PowerVision+ HRP detection system DPVB+500HRP (ImmunoVision Technologies, Brisbane, CA) on the TechMate 500 instrument (Dako). DAB+ (K3468; Dako) was used as substrate-chromogen system. Immunostaining was followed by brief nuclear counterstaining in Mayer’s hematoxylin. Finally, cover slips were mounted with AquaTex (Merck, Darmstadt, Germany).

HEK 293 cells and HEK 293 cells stably expressing TLR2 were plated on a 96-well plate at a density of 5 × 10⁴ cells/well for 24 h and then cultured in serum-free media for an additional 24 h. Cells were then pulsed with 20 μg/ml PGN for 24 h in the presence of supernatants from HEK 293-EBNA transfectants expressing rmS5D-SRCRB-HA or from untransfected cells as a control. Culture supernatant samples (25 μl) were collected and assayed for human IL-8 by ELISA (BD OptEIA, Human IL-8 ELISA Set; BD Biosciences, San Diego, CA), following the manufacturer’s instructions.

The screening of GenBank databases with the CD5 and CD6 protein sequences rendered two Mus musculus P1-derived artificial chromosome genomic clones (Acc. No. AC079486 and AC079542) predicted to contain a gene coding for a protein containing several group B SRCR domains. Also identified were 69 EST clones with significant similarities (Evalue < 10⁻¹⁰) to PAC genomic clones, 17 of which showed the highest score (Supplemental Table I) and were expressed in different tissues (muscle, brain, mammary gland, pancreas, liver, eye, gastrointestinal tract) and at different developmental stages (from embryo to adult). Two 4286 bp-long cDNA clones (Acc. No: AK079906 and A430110N23) obtained from thymus tissue of C57BL/6J mice at day 0 of neonatal development were cloned into the pFLCI vector to be fully resequenced. They were shown to contain an open reading frame of 4116 bp encoding a polypeptide chain of 1371 aa (Acc. No: BAC37780 and NP_766596), with a predicted M_r of 144.6 kDa (Supplemental Fig. 1). This putative new protein (Acc. No: EU850434, gi: 194354452) contained a short leader peptide and five SRCR domains (each was ∼100 aa in size) (http://www.ncbi.nlm.nih.gov/structure/cdd). The N-terminal signal peptide sequence was 16 aa long, and the most likely signal peptidase cleavage site was IQA-IE (Signal IP 3.0; http://www.cbs.dtu.dk). Four Asn-X-Ser/Thr motifs were found, three of which (NETA, NSTA, and NTTS) are predicted to be N-glycosylated (NetNGlyc 1.0; http://www.cbs.dtu.dk). All five SRCR domains fulfilled the characteristics of group B (encoded by a single exon and containing eight properly spaced cysteines) (5) and were interspersed by short (between SRCR1–SRCR2, and SRCR3–SRCR4), intermediate (between SRCR4–SRCR5), or long (C-terminal to SRCR5) PST-rich regions. Multiple O-glycosylation sites were predicted, which were mainly located within the PST-rich sequences (NetOGlyc 3.1; http://www.cbs.dtu.dk) (Fig. 1A, 1B). The PST-rich region C-terminal to SRCR5 was predicted to contain a 155-aa-long sequence (from aa 1050–1205) with homology to syndecan domains (http://www.ncbi.nlm.nih.gov/structure/cdd). No transmembrane region could be predicted (TMHMM; http://www.cbs.dtu.dk), as would be expected for a secreted protein. The mouse gene encoding this cDNA sequence was named s5d-srcrb, for soluble protein with 5 domains of the SRCR group B, and was mapped to chromosome 7 (http://www.ncbi.nlm.nih.gov/mapview). It spans 15 kb and encompasses 14 exons (Fig. 1A). Exons 1 and 2 encode the 5′-untranslated region and the signal peptide. Exons 3, 6, 7, 9, and 11 encode each of the five SRCR domains as expected for group B members. Exons 4, 5, 8, and 10 encode the short and intermediate PST-rich intervening regions. Exons 12–14 encode the long PST-rich C-terminal region (encompassing the putative syndecan domain), the stop codon, and the 3′-untranslated region with the poly-adenylation signal sequence (ACAAAA) at 14 bp upstream from the poly(A) tail.

Through the same bioinformatic tools used to identify s5d-srcrb, a homologous gene mapping to an orthologous region of human chromosome 19 was also identified. It corresponds to a recently reported human gene (SSc5D), for which no biochemical and functional data are available (22). Multiple alignments of the amino acid sequences of the SRCR domains from the human and mouse genes show a high degree (∼60%) of interdomain, as well as interspecies amino acid identity (Fig. 1C). This indicates that the five SRCR domains may have arisen from exon duplication.

The full-length cDNA sequence of mouse s5d-srcrb was fused in frame with a C-terminal HA tag and then cloned into the pCEP-Pu vector for further expression into HEK 293-EBNA cells. Serum-free supernatants from stable transfectants expressing this rmS5D-SRCRB-HA protein were precipitated using TCA, separated on SDS-PAGE under reducing conditions, and subjected to Western blot with an anti-HA mAb. This analysis rendered a single broad band of ∼200 kDa in size (Fig. 2A), which is far greater than the predicted M_r (144.6 kDa) for the s5d-srcrb gene product. The difference between the observed and expected M_r argues in favor of rmS5D-SRCRB-HA undergoing posttranslational modifications. This was further confirmed by simultaneous Western blot analysis of cell solubilizates and culture supernatants from HEK 293-EBNA transfectants expressing rmS5D-SRCRB-HA. As seen in Fig. 2C, the intracellular form of rmS5D-SRCRB-HA was shown to display a smaller M_r (∼150 kDa) than did the extracellular one (>200 kDa).

The rmS5D-SRCRB-HA present in serum-free supernatants from stable transfectants was used as immunogen to generate specific rat mAbs. After several cloning and subcloning steps, seven rat hybridomas were selected (Supplemental Table II). Their specificity was initially analyzed using ELISA plates precoated with serum-free rmS5D-SRCRB-HA–containing supernatants. None of the selected hybridomas was reactive to other HA-tagged group B SRCR-SF members’ protein, such as S4D-SRCRB (24). The same results were obtained following immunoprecipitation and Western blot analyses of rmS5D-SRCRB-HA supernatant samples, further indicating that all rat mAbs generated recognized rmS5D-SRCRB-HA under native (ELISA, immunoprecipitation) and denaturing (Western blot) conditions; this suggests that the mAbs are likely recognizing lineal epitopes.

Because the predicted amino acid sequence of mouse S5D-SRCRB contains multiple N-and O-linked glycosylation sites, the possibility that glycosylation could account for the above-mentioned posttranslational modifications was assessed. rmS5D-SRCRB-HA was immunoprecipitated and then subjected to incubation with different glycosidases, either alone or in combination. As illustrated in Fig. 2B, little or no change in M_r was observed when rmS5D-SRCRB-HA was incubated with galactosidase, which releases only β(1→4) terminal galactose residues. Partial sensitivity could be observed to N-acetylglucosaminidase (Fig. 2B) and PNGase F (data not shown), as deduced from the generation of a faint band ∼150 kDa, which is the predicted M_r for the unprocessed intracellular form of S5D-SRCRB. Because N-acetylglucosaminidase cleaves all terminal β-linked N-acetylglucosamine residues typical of N-linked glycosylations, and PNGase F cleaves all types of Asn-linked sugars, some degree of N-glycosylation can be inferred for S5D-SRCRB, which is in agreement with the presence of putative N-glycosylation sites. Combined digestion with neuraminidase and O-glycosidase induced a small reduction in the observed M_r of rmS5D-SRCRB-HA (Fig. 2B), which was similar to that observed for neuraminidase or O-glycosidase alone (data not shown). This indicates that S5D-SRCRB also undergoes some degree of O-glycosylation, as predicted from its relatively high content of PST-rich sequences. In parallel experiments, all of the glycosidases gave optimal digestion results when bovine fetuin was used as a control glycoprotein (data not shown). These results indicated that, although likely glycosylated, rmS5D-SRCRB-HA is also relatively resistant to glycosidase treatment, as has often been reported for some heavily glycosylated high M_r proteins (25, 26).

The expression pattern of S5D-SRCRB in normal mouse tissues was assessed by immunohistochemistry (IHC) and RT-qPCR assays. Of the seven rat hybridomas generated, two (4D11.A2.H4 and 1H11.A8.G2) were shown to be appropriate for IHC studies on paraffin-embedded tissues. The two mAbs gave identical staining results with different retrieval methods (pepsin or target retrieval solution). As illustrated in Fig. 3A, positive immunostaining was detected throughout the gastrointestinal and genitourinary tracts. Strong staining of serosal salivary gland and the exocrine part of pancreas, as well as of testis, was observed. In kidney, selective staining of a few tubular structures (likely corresponding to the distal part of the collecting tubules) was observed. Lung and heart gave negative staining results. A few scattered positive cells were detected in spleen sections.

The RT-qPCR analysis of s5d-srcrb expression in normal mouse tissues was in agreement with the above-mentioned IHC results. Following normalization of gene expression to ribosomal 18S, the greatest relative expression levels were observed in testis, kidney, and pancreas. Low or negative relative expression was detected in the rest of the analyzed tissues (Fig. 3B). Similar negative results were observed when bone marrow-derived monocytes (resting or LPS stimulated), and different cell lines of monocytic (Raw 264.7, B10R), lymphocytic (EL-4, X63), or epithelial (266.6) origin were analyzed (Supplemental Fig. 2). Taken together, these results illustrate the restricted cell- and tissue-expression pattern of s5d-srcrb.

The information obtained from tissue-expression assays was used to validate the specificity of the anti–S5D-SRCRB rat mAbs generated. Western blot analysis of RIPA solubilizates showed mAb-reactive bands in testis but not in heart tissue, used as negative control (Fig. 2C). Similar negative results were obtained in kidney and pancreas (data not shown), likely due to the low and restricted expression observed in kidney and the relatively high abundance of proteolytic enzymes present in exocrine pancreas. Interestingly, the higher M_r bands observed in testis had a similar M_r to those detected intra- and extracellularly in the transfectants expressing rmS5D-SRCRB-HA, used as positive control. A band of lower M_r (∼80 kDa) was also repeatedly observed in testis, likely representing an alternatively spliced isoform or a partially degraded form. Several attempts to immunoprecipitate S5D-SRCRB from testis and other positive tissues (pancreas, kidney) were unsuccessful. This indicated that the epitopes recognized by the anti–S5D-SRCRB mAbs are inaccessible in the endogenous protein under the solubilization conditions used (RIPA buffer) or, more likely, that the mAbs do not bind properly in the stringent solubilization buffer conditions used during RIPA-immunoprecipitation.

We also set out to study whether the expression of mouse S5D-SRCRB is regulated during embryo development. Preliminary in situ RNA-hybridization analysis carried out at day 9.5 postcoitum showed that S5D-SRCRB is selectively detected in placodes, embryonic ectodermal thickenings where organs or structures will develop (Supplemental Fig. 3). RT-qPCR experiments also showed a substantial increase in S5D-SRCRB RNA expression from days 9–14 postcoitum (Supplemental Fig. 4). These data suggested that the expression of S5D-SRCRB is developmentally regulated.

Given that some members of the SRCR-SF act as receptors for PAMPs, the microbial-binding properties of rmS5D-SRCRB-HA were assessed in vitro. Bacterial or fungal cell suspensions were incubated with serum-free rmS5D-SRCRB-HA–containing supernatants, and the presence of cell-bound protein was tested by Western blot. As shown in Fig. 4A (left panel), rmS5D-SRCRB-HA could be detected in cell pellets from most bacterial strains assayed, either Gram-negative (E. coli, Salmonella typhimurium, Yersinia enterocolitica, Shigella flexneri) or Gram-positive (S. aureus, Staphylococcus epidermidis). Relatively low binding signal was observed for A. baumannii. In parallel assays, rmS5D-SRCRB-HA also bound to all of the fungal strains analyzed, either saprophytic (S. cerevisiae, S. pombe) or pathogenic (C. albicans, Cryptococcus neoformans, A. fumigatus) (Fig. 4A, right panel). In all cases tested, the binding of rmS5D-SRCRB-HA was dose dependent (Fig. 4B). In full agreement with its pathogen-binding properties, rmS5D-SRCRB-HA was also able to induce bacterial and fungal aggregation, an effective mechanism to avoid dissemination of pathogenic agents and, thus, to control infection. As shown in Fig. 5, addition of rmS5D-SRCRB-HA, but not BSA, to microbial suspensions induced aggregation of Gram-positive (S. aureus) and Gram-negative (E. coli, A. baumannii) bacteria, as well as fungi (C. albicans).

The nature of the cell wall component(s) responsible for the bacterial- and fungal-binding ability of mouse S5D-SRCRB was studied by ELISA. Plastic plates were coated with different PAMPs (PGN, LPS, LTA, zymosan, glucan, mannan) and then incubated with serum-free rmS5D-SRCRB-HA supernatants. The presence of bound mS5D-SRCRB-HA was developed using anti–S5D-SRCRB rat mAb plus HRP-conjugated anti-rat IgG antiserum. As shown in Fig. 6, significant dose-dependent binding of rmS5D-SRCRB-HA could be detected to PGN, LPS, zymosan, and linear β-glucan but not to LTA, mannan, or branched β-glucans.

Next, we also studied whether the endogenous mouse S5D-SRCRB protein form(s) expressed in testis also retain(s) the microbial-binding properties of the recombinant protein (rmS5D-SRCRB-HA). S. aureus was chosen for these studies because our results showed that it gave the best binding. Thus, bacterial cell suspensions were incubated with liver and testis tissue solubilizates and then pelleted, extensively washed, and resolved by SDS-PAGE under reducing conditions for further Western blot analysis. As shown in Fig. 4C, anti–S5D-SRCRB–reactive bands were recovered in S. aureus cell pellets adsorbed onto testis but not liver homogenates. The observed bands corresponded to the greatest (fully processed) and smallest (alternatively spliced or proteolitically processed) M_r forms of mouse S5D-SRCRB (Fig. 2C). No evidence of reactivity with the intermediate M_r (intracellular, incompletely processed) form was observed.

Our results showed that rmS5D-SRCRB-HA bound to bacteria and was able to induce aggregation. We next studied whether rmS5D-SRCRB-HA has intrinsic bactericidal activity. As shown in Supplemental Fig. 5, our preliminary results suggested that rmS5D-SRCRB-HA does not display measurable bactericidal activity at a concentration of 1 μg/ml. Similar results were obtained with rmS5D-SRCRB-HA concentrations ≤20 μg/ml (data not shown).

We then sought to determine whether the interaction of S5D-SRCRB with PAMP could affect a biological outcome, such as cytokine release. HEK 293 cells stably transfected with TLR2 release IL-8 when exposed to PGN, a known PAMP and TLR2 ligand. We then exposed these cells, as well as untransfected HEK 293 cells, to PGN in the presence or absence of rmS5D-SRCRB-HA–containing supernatant. As shown in Fig. 7, release of IL-8 induced by PGN was dose-dependently inhibited by rmS5D-SRCRB-HA; inhibition was significant (p < 0.05) when 100 μl supernatant was used. Supernatant from untransfected HEK 293 cells did not affect IL-8 release by HEK 293 TLR2 cells (data not shown).

Archetypical members of the SRCR-SF (e.g., DMBT1/gp340/SAG) were reported to interact with exogenous pathogens, as well as with endogenous host components (27). Therefore, it was tested whether this could also be the case for S5D-SRCRB. First, we explored its binding to different purified extracellular matrix proteins by ELISA. As illustrated by the results presented in Fig. 8A, rmS5D-SRCRB-HA showed clear dose-dependent binding to laminin. Low (fibronectin) or negative (collagen I and IV) binding was observed for other extracellular matrix components. Another set of experiments demonstrated the sugar-dependent interaction of S5D-SRCRB with galectin-1, a broadly expressed homodimeric mammalian lectin secreted by epithelial cells (28). As shown in Fig. 8B, specific binding of rmS5D-SRCRB-HA to GST-Gal1 Sepharose beads was competed in a dose-dependent (3–30 mM) manner by lactose but not by another irrelevant sugar (sucrose). Taken together, the results indicated that, once secreted, S5D-SRCRB could be immobilized to relevant constituents of the extracellular milieu (e.g., laminin and galectin-1, and perhaps other unidentified proteins as well), thus putatively contributing to basic homeostatic epithelial cell functions.

Epithelial cells form physical barriers that are covered by mucosal surfaces aimed at preventing the entry of invading pathogens. Moreover, spontaneously or following the sensing of microbial components, epithelial cells can also synthesize effector molecules that trigger or increase the defensive immune responses of the host (i.e., mucosal inflammation and other related innate and adaptive immune responses) (29–33). The present work reports the identification and further molecular and functional characterization of a new mouse receptor (S5D-SRCRB) mainly expressed by some epithelial cell types and likely belonging to the humoral arm of the innate immune system. This receptor possesses the characteristics of group B members of the SRCR-SF (3, 5), and it should be considered the mouse homolog of SSc5D (22). A relevant finding of the current study is the demonstration that mouse S5D-SRCRB behaves as a PRR. Thus, S5D-SRCRB should be added to the short, although still-growing, list of members of the SRCR-SF showing pathogen-binding properties. This ability has been unequivocally mapped to the SRCR itself in only a few members, including MARCO, CD163, DMBT1/SAG/gp340, CD5, CD6, and Spα. In some instances, the amino acid sequence motifs involved in SRCR-mediated pathogen recognition have been precisely identified (RxR, VEVLxxxxW) (15, 16, 20), but an exact match to these motifs is not observed among any of the SRCR domains of S5D-SRCRB. However, by no means does this exclude the SRCR domains of mouse S5D-SRCRB as being responsible for pathogen recognition, as exemplified by Spα, CD6, and CD5 (17–19). The SRCR domains are among the few protein modules from which evolution has settled a myriad of structurally related but functionally different proteins. Key residues that stabilize the core structure of SRCR domains are well conserved, whereas amino acid positions at externally oriented connecting loops are keen to change, thus generating protein versatility. This versatility may have given rise to a certain degree of functional diversity among the members of the SRCR-SF that would justify the lack of a unifying function for all SRCR-SF members; this is exemplified even when exploring the pathogen-binding properties of SRCR-SF members. However, mutational investigations to exclude the possibility that the non-SRCR domains of S5D-SRCRB could account for its pathogen-binding properties remain to be performed.

Intriguingly, mouse S5D-SRCRB presents several structural and functional analogies with the human secreted protein DMBT1, also known as gp340 or salivary agglutinin (SAG), and which also corresponds to rat ebnerin, mouse Crp-ductin, or rabbit hensin (5). DMBT1/gp340/SAG is an archetypal group B SRCR-SF protein involved in infection, inflammation, and cancer (34). From a structural point of view, it is a mosaic-type glycoprotein composed of 14 SRCR domains interspersed by PST-rich sequences and possessing two C1r/C1s Uegf Bmp1 and one zona pellucida domains at its C-terminal region. Functionally, DMBT1/gp340/SAG represents an innate defense factor interacting with a broad spectrum of pathogens (bacteria, fungi, viruses), as well as mucosal defense proteins (galectins, IgA, surfactant A and D proteins, MUC5B, TFF2). In contrast, it drives epithelial and stem cell differentiation as an extracellular matrix protein, whereas inactivation of its gene is associated with tumor formation (34). Mouse S5D-SRCRB should also be considered a mosaic protein because it is composed of five SRCR domains interspersed by PST-rich sequences and possessing a C-terminal syndecan-like domain. Syndecans are heparan sulfate proteoglycans that are known to mediate interactions with extracellular matrix proteins and heparin-binding growth factors. Preliminary data support the putative interaction of rmS5D-SRCRB-HA with some extracellular matrix proteins, such as laminin and, to a lesser extent, fibronectin, but not collagen I or IV. Moreover, we also provide evidence on the carbohydrate-dependent interaction of mS5D-SRCRB-HA with galectin-1, a host soluble lectin that functions as a damage-associated molecular pattern and a receptor for PAMPs (35). The mucosal-defense properties of mouse S5D-SRCRB against living organisms are supported by its broad binding and aggregating microbial spectrum that includes saprophytic and pathogenic bacteria and fungi; potential interactions with viruses remain to be studied. Our preliminary data indicated that this defensive role is played in the absence of detectable intrinsic bactericidal activity, which is not unusual for innate-immunity proteins; SAG is a well-known example of a nonbactericidal protein with defensive properties. By binding to microbial components, mouse S5D-SRCR seems to downregulate subsequent PAMP-induced cytokine release, as deduced from our experiments showing dose-dependent inhibition of IL-8 release following PGN stimulation of HEK 293 TLR2 transfectants. We consider that this could be important for preserving the integrity and the function of epithelia from excessive or prolonged inflammation caused by PAMPs in those tissues where S5D-SRCRB is expressed. This is best exemplified by the expression of mouse S5D-SRCRB in seminiferous testicular tubules, where infection and inflammation may cause male infertility, thus compromising murine reproductive capability (36). The expression of mouse S5D-SRCRB in these tissues also argues in favor of its putative role in cell-differentiation processes. Interestingly, preliminary data from our group indicate that expression of mouse S5D-SRCRB is regulated during embryo development. More precisely, in situ RNA hybridization results at day 9.5 postcoitum showed that S5D-SRCRB is selectively detected in placodes, embryonic ectodermal thickenings where organs or structures will develop (37). Moreover, RT-qPCR data indicate that there is an important relative increase in its RNA expression from days 9–14 postcoitum, the meaning of which remains to be explored further. Finally, the relative preferential expression of mouse S5D-SRCRB observed in adult testis also opens the possibility that it may behave as a nonchromosome X-encoded cancer/testis Ag (38). Therefore, a systematic investigation of mouse S5D-SRCRB expression in human and mouse tumors should be performed to exclude this possibility.

The very limited information available on human SSc5D highlights the need for future studies to explore whether the functional and structural characteristics reported in this article for mouse S5D-SRCRB also apply to its human homolog. At present, marked differences exist regarding their tissue-expression pattern. Although the highest expression level of SSc5D is reported in placenta, spleen, colon, and lung, mouse S5D-SRCRB is mainly expressed in testis, kidney, and the serosal region of salivary and pancreas glands. To further clarify this point, parallel IHC analyses of mouse and human tissues performed with the 1H11.A8.G2 mAb, which shows human–mouse species cross-reactivity (U. Holmskov, unpublished observations), are urgently needed. The demonstration of conserved binding capabilities to a broad spectrum of pathogens, as well as to endogenous proteins, by the human protein homolog also requires further investigation.

In summary, to our knowledge, the molecular and functional characteristics of a new mouse group B SRCR-SF member, S5D-SRCRB, have been reported in this article for the first time. Apart from its likely involvement in protection against pathogenic or saprophytic microorganisms of the restricted epithelial surfaces where mouse S5D-SRCRB is expressed, other functions related to epithelial cell differentiation and homeostasis should be taken into consideration in future studies.

We thank the Department of Microbiology, Hospital Clinic of Barcelona, and the Cell Biology Unit, Faculty of Medicine, University of Barcelona, for providing bacterial and fungal specimens. We also thank R. Fenutría, B. Suárez, M. Antón, C. Astasio, J. Milà, and M. Bayo for technical and administrative support.

The authors have no financial conflicts of interest.