A Novel Chicken Membrane-Associated Complement Regulatory Protein: Molecular Cloning and Functional Characterization1 Naokazu Inoue,*† Aya Fukui,*‡ Midori Nomura,* Misako Matsumoto,* Kumao Toyoshima,* and Tsukasa Seya2*‡ A cDNA encoding a membrane-associated complement (C) regulatory protein was identified here for the first time in an oviparous vertebrate, chicken. This protein, named Cremp, possessed five short consensus repeats (SCRs) and one SCR-like domain followed by a transmembrane domain and a cytoplasmic tail. SCR1/SCR2 of Cremp were 43.6% identical with SCR2/SCR3 of human decay-accelerating factor (CD55), and SCR3/SCR4 were 45.3% identical with those of human membrane cofactor protein (CD46). Cremp is likely to be an ancestral hybrid protein of human decay-accelerating factor and membrane cofactor protein rather than a homolog of rodent C receptor 1-related protein y, which structurally resembles human CR1 (CD35). Chinese hamster ovary cells transfected with Cremp were efficiently protected from chicken C but not from human or rabbit C in both classical and alternative pathways. Thus, chicken Cremp is a membrane C regulator for cell protection against homologous C. Cremp mRNA was seen as a doublet comprised of a faint band of 2.2 kb and a thick band of 3.0 kb on RNA blotting analysis. An Ab against chicken Cremp recognized a single band of 46.8 kDa on immunoblotting. mRNA and protein of Cremp were ubiquitously expressed in all chicken organs tested. Minute amounts of dimer were present in some tissues. Surface expression of Cremp was confirmed by flow cytometry and immunofluorescence analysis. These results suggested that even in nonmammals a C regulatory membrane protein with ubiquitous tissue distribution should be a prerequisite for protection of host cells from homologous C attack. The Journal of Immunology, 2001, 166: 424 – 431.
T
he complement (C)3 system plays an important role in the elimination of microorganisms in the innate immune system that developed earlier in evolution than the acquired immune system (1, 2). There are many soluble and cell-associated regulatory proteins for the activated form of C3 that acts as the main effector in the C system. In humans, C4b-binding protein (C4bp) and factor H belong to the soluble form C regulators, and C3b receptor or C receptor type 1 (CR1; CD35), C receptor type 2 or C3d/EBV receptor (CR2; CD21), decay-accelerating factor (DAF; CD55), and membrane cofactor protein (MCP; CD46) belong to the membrane-associated forms (3). These molecules consist of a tandem arrangement of a motif of ⬃60 aa, termed the short consensus repeat (SCR), and their genes are clustered in 1q32 (3,
*Department of Immunology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan; †Division of Environmental Pharmacology, Department of Pharmaceutical Sciences, Osaka University, Osaka, Japan; and ‡Department of Molecular Immunology, Nara Institute for Science and Technology, Ikoma, Japan Received for publication August 15, 2000. Accepted for publication October 10, 2000. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported in part by Grants-in-Aid from the Ministries of Education, Science, and Culture, and Health and Welfare of Japan, and the Organization for Pharmaceutical Safety and Research (OPSR). 2 Address correspondence and reprint requests to Dr. T. Seya, Department of Immunology, Osaka Medical Center for Cancer and Cardiovascular Diseases, Higashinariku, Osaka 537 Japan. E-mail address:
[email protected] 3 Abbreviations used in this paper: C, complement; CHO, Chinese hamster ovary; Cremp, C regulatory membrane protein; Crry, C receptor 1-related protein y; DAF, decay-accelerating factor; DT40, chicken B-lymphocyte cell line; hMCP, human MCP; MCP, membrane cofactor protein; RK13, rabbit kidney cell line; SCR, short consensus repeat; ORF, open reading frame; UT, untranslated region; IAA, iodoacetamide; GVB, gelatin veronal buffer.
Copyright © 2001 by The American Association of Immunologists
4). All of these proteins physically bind C3 fragments (3). It is currently accepted that the role of the soluble SCR C regulatory proteins is to prevent excess C activation in the fluid phase, whereas the role of DAF and MCP is to protect host cells on the same membrane from attack by autologous C (3). CR1 and CR2 are involved in immune complex clearance in addition to C regulation and enhancement of Ag-C3d-dependent activation of B cells, respectively (5– 8). In rodents, these sets of SCR proteins are conserved, although their distribution profiles and predicted roles are not always consistent with those of humans. In addition, a specific SCR protein, C receptor 1-related protein y (Crry), has been identified (9, 10) that like DAF and MCP protects the rodent cells from complement attack (11, 12). However, in nonmammalian lower vertebrates the membrane-regulatory system of C has not been identified. Soluble form C regulators were presumed to exist in lower vertebrates and some invertebrates (13–16), and the fluid phase C regulatory system has been hypothesized to have evolved simultaneously with C proteins. However, no membrane-associated forms of C regulators have yet been identified in lower vertebrates other than mammals. In this study, we cloned a cDNA encoding a novel transmembrane SCR protein in the chicken, which we named chicken C regulatory membrane protein (Cremp). Chicken Cremp was first identified in a chicken cDNA library as a fragment containing the consensus sequence shared with MCP of all species reported to date (17, 18). The predicted amino acid sequence of Cremp suggests a hybrid consisting of human DAF-like and MCP-like sequences with five complete and one incomplete SCR domains. This protein was shown to protect mammalian transfectants from chicken C-mediated cytolysis. Thus, this protein is the first nonmammalian cell-associated C regulator with the ability to protect cells from attack by homologous C. We propose the evolutional 0022-1767/01/$02.00
The Journal of Immunology
425
Table I. Percent homology among SCR domains of chicken Cremp, human DAF, human MCP, and mouse Crrya Cremp
SCR 1 SCR 2 SCR 3 SCR 4 SCR 5 SCR-like
hDAF
hMCP
mCrry
45.2% SCR 2 43.3% SCR 3 41.8% SCR 3 22.7% SCR 2 19.3% SCR 4
29.7% SCR 3 43.7% SCR 3 47.5% SCR 4
31.8% SCR 1 34.0% SCR 5 44.3% SCR 4 39.0% SCR 5
⬍10%
⬍10%
23.1% SCR 2
⬍10%
⬍10%
⬍10%
a Maximum homologous values are indicated. SCR domain ⬎40% homology are underlined.
importance of membrane-associated C regulatory proteins in parallel with C proteins.
Materials and Methods Tissues, cells, and sera Fresh chicken, human, and rabbit sera were obtained from each species by the standard method (19). All serum samples were immediately stored at ⫺80°C until use. Chinese hamster ovary (CHO) cells were obtained from American Type Culture Collection (ATCC, Manassas, VA). RK13 cells (derived from rabbit kidney) were obtained from RIKEN Cell Bank (Wako, Saitima, Japan). DT40 cells (derived from chicken B lymphocyte) were a gift from Dr. S. Takeda (Kyoto University, Kyoto, Japan). For RNA and protein blot analysis, total RNA and proteins were obtained from each chicken tissue (excel link, 20 wk) and were stored at ⫺80°C until use.
Screening a chicken cDNA library
FIGURE 1. Strategy for cDNA cloning of Cremp. The Cremp-specific primers described in Table I were used for Cremp cDNA cloning. The specific primers are shown by arrows. Nested PCR was performed to amplify various cDNA fragments (1–5) of Cremp using 1–5 Crempspecific primers and vector (pME-18s)-specific primers. Each cDNA fragment (1–5) overlapped with adjacent fragments. Finally, to confirm the complete Cremp sequence, 12 RT-PCR products were sequenced using total RNA that was extracted from chicken lung.
PBS three times, and suspended in 0.5 ml of PBS. Then, RK13 cell suspensions were mixed with 0.6 ml of Freund’s complete adjuvant (FCA) (Difco, Detroit, MI) and extensively agitated. The mixture was used to immunize rabbits. Immunization was performed four times at 7-day intervals, and boosted before drawing blood. The antisera were harvested by centrifugation. IgG was purified from the sera according to the standard method.
Tissue protein blotting analysis Various chicken tissues were solubilized with lysis buffer containing 1% (v/v) Nonidet P-40, 0.14 M NaCl, 0.01 M EDTA, 20 mM Tris-HCl (pH 7.4), 1 mg/ml iodoacetamide (IAA), and 1 mM PMSF using a Potter type homogenizer. The tissue suspension was centrifuged at 15,000 rpm for 30 min at 4°C, the pellet was removed, and the supernatant was subjected to SDS-PAGE followed by transblotting onto polyvinylidene difluoride (PVDF) membranes by the method described previously (20). Cremp was detected with anti-Cremp Ab, peroxidase-conjugated secondary Ab (Cappel, West Chester, PA), and chemiluminescence (ECL system, Amersham).
Chicken thymus cDNA library was a gift from Dr. R. Goizuka (Tokyo Scientific University, Tokyo, Japan). The cDNA fragment was ligated into the mammalian expression vector pME18s at the BstXI site. A partial Cremp cDNA fragment was obtained by nested PCR using a degenerate primer and a vector (pME 18s)-specific primer (Fig. 1). One degenerate primer was designed from the junctional sequence of SCR3 and SCR4 of human MCP (hMCP) as described previously for molecular cloning of mouse MCP (18). First PCR was performed as follows: denaturation at 94°C for 2 min, 1 cycle, denaturation at 94°C for 30 s, annealing at 50°C for 1 min, polymerization at 72°C for 90 s, 20 cycles. In the second PCR, 1 l of the first PCR product was added to the PCR mixture of the nested PCR primers, and sequential PCR was performed under the same conditions for 35 cycles. The PCR products were cloned into the vector pCR-2.1 (Invitrogen, San Diego, CA) and subjected to DNA sequencing using an ABI 377 sequencer. The DNA sequence of the full-length open reading frame (ORF) and complete 3⬘ untranslated region (UT) were determined by consecutive rounds of PCR using the primers described in Fig. 1. The complete Cremp sequence was confirmed from chicken lung mRNA (white leghorn) by sequencing twelve RT-PCR amplicons.
Transfected CHO cells (5⫻105) were incubated with 30 l of 20 g/ml anti-Cremp Ab or 30 l of 5 g/ml mAb against hMCP (M177), which recognizes SCR2 of hMCP, for 1 h at 4°C. After three washes, cells were treated with FITC-conjugated secondary Ab. The stained cells were analyzed using a FACSCalibur. Mean fluorescence intensity was evaluated on the attached computer software, Cell Quest.
Tissue RNA blotting analysis
Immunofluorescence analysis of transfected cells
Total RNAs (20 g) were extracted from various chicken tissues using TRIZOL Reagent (Life Technologies, Grand Island, NY) and separated in a 1.0% (w/v) agarose gel. RNAs were transferred onto Hybond-N⫹ membrane (Amersham, Arlington Height, IL), which was prehybridized for 30 min at 68°C and hybridized for 1 h at 68°C in Express Hybridization buffer (Clontech, Palo Alto, CA) with 32P-labeled full-length Cremp ORF as a probe. After washing, the membrane was exposed to x-ray film at ⫺80°C. The blot was rehybridized with a GAPDH probe as a control.
CHO cells expressing Cremp (CHO/Cremp) were incubated with 100 l of 2 g/ml anti-Cremp Ab for 1 h at 37°C in PBS containing 1% (w/v) BSA. The cells were washed, incubated with a 1:100 dilution of FITC-conjugated anti-rabbit IgG (Cappel) for 30 min at 37°C in PBS containing 10% (w/v) Block Ace (Yukijirushi, Sapporo, Japan), washed, and mounted on glass slides in PBS containing 2.3% 1,4-diazabiccyclo-2-octane and 50% glycerol. The stained cells were visualized at ⫻40 magnification under a FLUOVIEW (Olympus, Tokyo, Japan). Images were captured using the attached computer software, FLUOVIEW.
Establishment of stable transfectants expressing Cremp The cloned Cremp cDNA was ligated into the mammalian expression vector pCXN-2 (21), and CHO cells were transfected with this vector using lipofectAMINE. Thereafter, transfected cells were selected in medium containing 0.6 mg/ml of G418 (Life Technologies). The mean fluorescence shifts (MFS) of CHO cells expressing Cremp were assessed by flow cytometry using anti-Cremp Ab.
Flow cytometry
Preparation of rabbit anti-Cremp polyclonal Ab To prepare a rabbit polyclonal Ab against Cremp (anti-Cremp Ab), RK13 cells (1 ⫻ 107) were transiently transfected with pME18s-Cremp-His tag using lipofectAMINE Reagent (Life Technologies). After 2 days, transfected RK13 cells were collected with 10 mM EDTA-PBS, washed with
Deglycosylation assay of Cremp The methods for analyses using deglycosidases were described previously (20). Briefly, each transfectant (5 ⫻ 106) was solubilized with solubilization buffer containing 1% Nonidet P-40, 50 mM Tris-maleate (pH 8.6), 10
426
CHICKEN DAF/MCP-LIKE PROTEIN
The Journal of Immunology
427
mM EDTA, 1 mg/ml IAA, 1 mM PMSF for O-glycosidase, or 1% Nonidet P-40, 20 mM Tris-maleate (pH 6.0), 10 mM EDTA, 1 mg/ml IAA, and 1 mM PMSF for N-glycosidase. Solubilized proteins were centrifuged at 15,000 rpm for 30 min at 4°C, the pellets were removed, and the supernatants were incubated with 100 U of neuraminidase (Sigma) for 1 h at 37°C. Then, the samples were treated with either 250 mU of N-glycosidase or 1 mU of O-glycosidase (Genzyme, Cambridge, MA) for 16 h at 37°C. The samples were subjected to SDS-PAGE followed by immunoblotting. Cremp protein was detected with anti-Cremp Ab. 51
Cr release cytotoxicity assay
For cytotoxicity assay, 1 ⫻ 107 transfected CHO cells were collected in 10 mM EDTA-PBS and incubated with 100 l (3.7 MBq) of Na251CrO4 for 60 min at 37°C in 1 ml of serum-free Ham’s F12 medium. After three washes with PBS, 2 ⫻ 104 51Cr-labeled cells in 50 l of Ham’s F12 supplemented with 10% FCS were placed in the wells of 96-well plates and incubated with 50 l of various concentrations of rabbit Ab against CHO cells (precipitated with 33% ammonium sulfate) in gelatin veronal buffer (GVB)2⫹ or 0.03 M EGTA-GVB for 30 min at 4°C. Cells were subsequently incubated with 100 l of various concentrations of chicken, human, or rabbit serum diluted in GVB2⫹ or 0.03 M EGTA-GVB for 60 min at 37°C with gentle shaking. GVB2⫹ and 0.03 M EGTA-GVB represent the conditions for activation of the classical and the alternative C pathway, respectively, which was true in chicken C according to the criteria of C4b deposition (data not shown). The plates were centrifuged at 1500 rpm for 5 min, and radioactivity in aliquots of 150 l of supernatants were measured with an auto gamma counter. The percentage of cytotoxicity was calculated as follows: (sample cpm ⫺ control cpm)/(MAX cpm ⫺ control cpm) ⫻ 100. Untreated CHO cells were used to measure spontaneous 51Cr release (control cpm), and cells treated with 5% Triton X-100 were used to measure maximum release (MAX cpm). The experiments were performed three times in triplicate. Because we used the different sources of C in each assay to measure percent inhibition of cytotoxicity, the data show relative C-regulatory potencies of each C-regulatory protein.
Results cDNA cloning of chicken Cremp The aim of this study was the molecular cloning of nonmammalian MCP homolog. The chicken was chosen because a cell-level gene disruption system is available using the chicken B cell line DT40 (22). A set of degenerate PCR primer was designed referring to conserved sequence (Fig. 2A, long arrow) based on a homology search for MCP of various species (17). Nested PCR was performed with the degenerate primer as the forward primer, the vector (pME18s)-specific primer as the reverse primer, and a chicken thymus cDNA library as the template. After many trials, the second PCR yielded a 280-bp cDNA fragment. This cDNA fragment was cloned, sequenced, and found to be similar to SCR4 of hMCP. As shown in Fig. 1, consecutive rounds of PCR were performed by the same method. Then, the nucleotide sequence of the ORF and 3⬘UT of the chicken SCR protein cDNA was determined by the rapid amplification of cDNA end (RACE) method. The presence of this message was confirmed with chicken lung mRNA by sequencing twelve independent RT-PCR amplicons. We named this novel SCR protein, C regulatory membrane protein, Cremp. The Cremp cDNA consisted of 1910 bp including an incomplete polyadenylation signal and poly(A) tail, and it encoded a predicted protein of 451 aa (Fig. 2A). Cremp was a type 1 membrane protein consisting of five SCR domains, one SCR-like domain, a transmembrane domain, and a cytoplasmic tail. Cremp has
FIGURE 3. Tissue blot of Cremp RNA and protein. A, RNA expression of Cremp in various chicken tissues. Total RNA (20 g) from various tissues was separated in a 1.0% (w/v) agarose gel and hybridized with 32 P-labeled Cremp or GAPDH probes. Top, A ⬃3.0-kb band was detected in all lanes and a faint ⬃2.2-kb band was detected in some lanes. B, Immunoblotting analysis of various chicken tissues. Solubilized proteins (100 g) from various tissues were separated by SDS-PAGE in a 10% polyacrylamide gels under nonreducing conditions, transblotted onto membranes, and detected with anti-Cremp Ab. A 46.8-kDa band (putative mature molecular mass) were observed in all tissues.
one putative N-glycosylation site on SCR3. According to homology search analysis, this molecule was likely to differ from a simple MCP homolog as well as other SCR proteins (Table I). At the amino acid level, SCR1/SCR2 (combinational homology of two SCRs) of Cremp was 43.6% identical with SCR2/SCR3 of human DAF, and SCR3/SCR4 of Cremp was 45.4% identical with SCR3/ SCR4 of hMCP (Fig. 2B). These two sequential SCRs of Cremp were less similar to any two sequential SCRs of the other known SCR proteins than the above. For example, in homology search of each SCR of the SCR proteins, mouse Crry SCR1 and SCR5 were 31.8 and 34.0% homologous to SCR1 and SCR2 of Cremp, respectively. The only exception was SCR3 of Cremp, which was 44.3% identical with that of mouse Crry, showing a high degree of similarity. In total SCR pattern, Cremp was more similar to parts of human DAF and hMCP than Crry (Table I). We hypothesized that Cremp is a DAF/MCP hybrid molecule from which DAF and MCP, but not Crry, evolved.
FIGURE 2. Complete Cremp nucleotide and amino acid sequences. A, Cremp cDNA sequence and its deduced amino acid sequence. The deduced amino acid sequence is shown under the nucleotide sequence. The stop codon is indicated by an asterisk. The degenerate primer used for nested PCR is indicated by the long arrow, and a putative N-glycoside link motif is underlined. This nucleotide sequence contained both the polyadenylation signal (broken line) and poly(A) sequence (double underlined). The nucleotide sequence has been submitted to the EMBL Data Library/GenBank/DDBJ databases with the accession number AB035592. B, The Cremp amino acid sequence was compared with those of other SCR proteins using homology search software GENETYX-MAC 9.0. The similar and identical amino acid residues are shown by a dot and an asterisk, respectively. The boxes indicate possible consensus Cys, Trp frames of the SCR domain.
428
CHICKEN DAF/MCP-LIKE PROTEIN
FIGURE 4. Establishment of CHO cells stably expressing Cremp and localization of Cremp protein. 1, Flow cytometric analysis. DT40 cells (A) derived from chicken B-lymphocytes, and CHO/Cremp (B) established from a transfectant stably expressing Cremp were treated with anti-Cremp Ab. CHO cells transfected with hMCP (CHO/hMCP), STC, CYT2 type, (C) were treated with M177 Ab. Cremp protein and hMCP were expressed on the cell surface (filled area). Rabbit and mouse IgG were used as controls for DT40, CHO/Cremp and CHO/hMCP (solid line). Untransfected CHO cells were treated with anti-Cremp Ab (B) or M177 Ab (C), which showed no expression (broken line). 2, Immunostaining of Cremp protein. CHO/Cremp and untransfected CHO cells were treated with anti-Cremp Ab, and the cells were treated with FITC-conjugated anti-rabbit IgG, then mounted on glass slides and observed under a confocal microscope. Only the margins of Ab-treated CHO/Cremp cells were FITC positive.
Cremp RNA and protein expression
Localization of Cremp
We first analyzed the tissue distribution of Cremp mRNA by RNA blotting analysis. RNA blotting followed by hybridization to the full-length ORF of Cremp as a probe (1356 bp), which detects Cremp mRNA, revealed a doublet consisting of a minor 2.2-kb band and a major 3.0-kb band. The major 3.0-kb band was expressed in all tissues, whereas the minor 2.2-kb band showed restricted expression in the lung, kidney, bursa, testis, and ovary (Fig. 3A). The cDNA isolated here was 1.9 kb in length with an incomplete polyadenylation signal and poly(A) tail. The 1.9-kb cDNA may have corresponded to the minor 2.2-kb message, as in the case of the first identification of hMCP cDNA (23). If this is the case, alternative usage of polyadenylation signals, either incomplete or complete, may result in two messages with a short or long 3⬘UT. Additional experiments to clarify this point are currently in progress in our laboratory. To further determine the tissue distribution and relative levels of Cremp protein, we performed immunoblotting analysis (Fig. 3B). Rabbit anti-Cremp Ab was prepared as described in Materials and Methods. In this analysis, each lane reflected the amount of Cremp released from 50 g of tissue extracts with lysis buffer. All tissues tested showed fairly high levels of Cremp. Cremp protein had a molecular mass of 46.8 kDa, representing a putative mature molecule with no glycosylation. Minute amounts of a 92-kDa band were also observed in the spleen, lung, kidney, and ovary. This high molecular mass protein appeared to be a disulfide-linked dimer of Cremp because the band disappeared after reduction. Thus, both Cremp 3.0-kb mRNA and 46.8-kDa protein were ubiquitously expressed in parallel.
The hydrophobicity plot suggested that Cremp protein is a type 1 membrane protein with a signal peptide (Fig. 2A). To determine whether Cremp was localized on the cell surface, we performed flow cytometry and immunostaining. On flow cytometry using anti-
FIGURE 5. Deglycosylation analysis of CHO/Cremp and DT40. Solubilized CHO/Cremp and DT40 were treated with N- or O-glycosidase (A). CHO/hMCP was used as a positive control (B). Untransfected cells were treated under the same conditions in the absence of N- or O-glycosidase. The samples were subjected to SDS-PAGE followed by immunoblotting, and detected with anti-Cremp Ab or M177 Ab.
The Journal of Immunology
429
Cremp Ab, DT40 and CHO cells expressing Cremp (CHO/Cremp) were found to express Cremp on the cell surface (Fig. 41). Even in unstimulated DT40 cells, Cremp was highly and constitutively expressed on the cell surface. Moreover, CHO/Cremp cells were also specifically detected by immunostaining with anti-Cremp Ab around the cell margins (Fig. 42). Therefore, Cremp protein is largely localized on the cell surface. Glycosylation status of Cremp The presence of an N-linked glycosylation site was predicted from the amino acid sequence of Cremp (Fig. 2A). To determine whether Cremp has N-linked and O-linked sugars, we performed deglycosylation analysis of Cremp-expressing cells, CHO/Cremp and DT40, using N- and O-glycosidases. In both CHO/Cremp and DT40 cells, the equivalent mobility of Cremp remained unchanged before and after glycosidase treatment on SDS-PAGE/immunoblotting (Fig 5A). Therefore, Cremp on both CHO and DT40 cells is likely to be mostly enzyme resistant. The Cremp protein was thus suggested to be an unglycosylated or glycosidase-resistant protein. C protection assay of Cremp To determine whether Cremp has the ability to protect host cells from attack by C, we measured C protection activity of CHO/ Cremp by cytotoxicity assay using 51Cr-labeled CHO cells and chicken serum (C source). We used CHO/hMCP as a positive control and untransfected CHO cells as a negative control. The levels of expression of Cremp and hMCP on CHO cells were 330.0 and 102.3, respectively, expressed as mean fluorescence intensity measured by flow cytometry. CHO cells were sensitized with an optimal concentration of the anti-CHO Ab based on our primary tests for C protection assay. CHO/Cremp completely inhibited chicken C attack in both of the classical and the alternative pathways (Fig. 6A). In contrast, CHO/hMCP and untransfected CHO induced cytolysis in a dose-dependent manner. In contrast, human serum lysed CHO/Cremp as well as CHO control but not CHO/hMCP in both pathways (Fig. 6B). Rabbit serum induced cytolysis in all cells irrespective of Cremp or MCP expression in either pathway (Fig. 6C). Therefore, Cremp selectively acted on chicken C to protect host cells from homologous C attack. The action of Cremp on C was thus indicated to be species specific.
Discussion Here, we identified a chicken C regulatory protein that acts on the membrane as a safeguard against homologous C similarly to DAF and MCP. This protein, Cremp, was a type 1 membrane protein with five SCRs, one SCR-like domain, a transmembrane region, and a cytoplasmic tail. SCR1/SCR2 of Cremp was 43.6% identical with SCR2/SCR3 of human DAF, whereas SCR3/SCR4 of Cremp was 45.4% identical with those of hMCP. It has been demonstrated that SCR2– 4 in both DAF and MCP play a major role in C regulation (24, 25). The scaffold structure of these domains may allow Cremp to encompass the C regulatory functions of DAF and MCP. This structural feature of Cremp does not resemble that of Crry in rodents, which consists of five to six SCRs homologous to human CR1 (CD35), although functionally Crry possesses DAF- and MCP-like activities (11). This structural or functional hybrid of DAF/MCP may be a primitive phenotype of membrane C regulators for host protection from C. The Cremp cDNA isolated here was 1.9 kb in length with an incomplete polyadenylation signal, AATTAAA, followed by a poly(A) tail. Its message consisted of a major species of 3.0 kb and a minor species of 2.2 kb as determined by RNA blotting analysis. The 2.2-kb band was expressed relatively thick in the lung, kidney,
FIGURE 6. CHO/Cremp selectively protected host cells from C attack in chicken serum. A, CHO/Cremp (E), CHO/hMCP (F), and control CHO cells (䡺) labeled with 51Cr were sensitized with 65 g/ml (classical pathway; CP) or 130 g/ml (alternative pathway; AP) of anti-CHO Ab and incubated with various concentrations of GVB2⫹ (CP) or 0.03 M EGTAGVB (AP)-chicken serum. The percent cytotoxicity was calculated as described in Materials and Methods. B, The same test performed with human serum. To sensitize CHO cells, the cells were incubated with 130 g/ml (CP) or 650 g/ml (AP) of anti-CHO Ab. C, The same test performed with rabbit serum. The cells were sensitized with 65 g/ml (both) and incubated with 10% GVB2⫹ or 20% EGTA-GVB-rabbit serum. Then, CHO/Cremp, CHO/hMCP, and control CHO were represented by open, gray, and black columns, respectively. The experiments were performed three times in triplicate, and similar results were observed. The error bars represent SD.
bursa, testis, and ovary. The most likely interpretation of these observations is that the 3.0-kb message is ubiquitous and the 2.2 kb is tissue specific, and we may have cloned the latter cDNA as shown in Fig. 2. Two similar messages were obtained with human and mouse MCP; ubiquitous 4.0 kb and testis-specific 1.5-kb species produced by alternative usage of incomplete and complete polyadenylation signals. The chicken C cascade has not been well delineated. Of the C components, C3 has been identified at the molecular level (26, 27). IgY has the ability to activate chicken C via the classical pathway, and in the absence of Ca2⫹ and IgY chicken C can be activated by the addition of foreign materials such as zymosan, suggestive of the presence of the alternative pathway (28, 29). Thus, we examined Cremp-dependent C inhibitory activity in the chicken classical and alternative pathways. The C-mediated CHO cell lysis system was used because this system worked well to test species specificity of C regulatory proteins in previous studies (19).
430 Cremp protects host cells from homologous C attack by inhibition of the chicken classical and alternative pathways. However, chicken Cremp hardly protects cells from heterologous C in both pathways. Cremp at least exerts host cell-protective activity against homologous C. This species specificity resembles those of DAF, MCP, and Crry. Therefore, Cremp is first identified as an SCR membrane protein of nonmammalian origin with C regulatory activity. The SCR protein-self protection theory in the C system should be adaptable to lower vertebrates including oviparous animals. In mammals, the fetus grows in the placenta, which originates from the embryo and expresses paternal allo-Ags. One interpretation of the necessity of high levels of C regulatory proteins in organs including the placenta was that the placenta and fetal organs would be targets for allo-Ab generated by maternal lymphocytes (12). This hypothesis will not be the case in nonmammals because an oviparous animal has no direct communication between maternal and antenatal immunological factors. Yet, the chicken possesses a ubiquitously expressed membrane C regulatory protein. We favor the interpretation of our results as indicating that Cremp-like proteins must be a prerequisite for survival of lower vertebrates with sufficient C function, rather than protection of the fetus from maternal C attack via the placenta. Cremp may represent an ancestral form of membrane C regulators. It is essentially a hybrid consisting of DAF-like and MCPlike elements participating in their functions, and like rodent Crry it contains five to six SCR domains. Most membrane SCR proteins protect the cells from C attack by two modes of action, decayaccelerating and factor I-cofactor activities. The C3 convertases, bimolecular proteases for activation of C3, are reversibly dissociated by decay-accelerating activity of SCR proteins, and are irreversibly inactivated by serine protease factor I and SCR proteinbearing cofactor activity. Although the functional profile of Cremp has not been defined, it effectively protects host CHO cells from chicken C similarly to DAF/MCP. A hybrid-like molecule of factor B/C2 was revealed in the lower vertebrate, chicken, bony fish, and lamprey (30 –32). These results, together with the recent finding that chicken MHC is 10-fold more compact than that of mammals (33–35), suggest that C and its regulatory genes were more compact in the ancestral vertebrates and evolved into a complex system through gene duplication and unequal crossing over, as proposed by phylogenists. We speculate that human DAF and MCP also evolved from a single ancestor molecule by gene recombination. The membrane SCR proteins have both or either DAF-like or MCP-like functions to regulate C3 activation on the membrane. In contrast, the SCR protein CR1 is present in B lymphocytes and phagocytes to serve as a receptor for C3 fragments deposited on opsonized cells, thereby allowing opsonized cells to be effectors for efficient Ag presentation or for phagocytosis. The SCR protein CR2 is expressed on B cells in humans and mice, and forms a complex with CD19 and TAPA-1 to signal the cells (6 – 8). The other SCR proteins Crry and MCP can serve as costimulators for proliferation of T cells and lead to secretion of IL-4 (36, 37). In addition, most C regulatory proteins are expressed at high levels in the testicular germ cells in mammals (18, 38), and their function may be related to fertilization (39, 40). It is still unknown whether Cremp is expressed predominantly in chicken testicular germ cells and plays some roles in fertilization as speculated in mammals. Also, it is possible that Cremp is a multifunctional protein, because almost all human SCR proteins found to date function as virus and/or bacterial receptors as well as C regulators (41, 42). Therefore, although Cremp is structurally similar to DAF and MCP,
CHICKEN DAF/MCP-LIKE PROTEIN further infection-related function should be tested to verify its physiological roles. Chicken immune-related genes appear to be simple and compact compared with those of mammals. A number of functions related to immunological phenotypes may be covered with a minimal number of molecules, and Cremp is a likely example of such a molecule. Cremp is expressed in the chicken B cell line DT40, which is a useful tool for gene disruption at the cell level (22). We will be able to analyze the functions of Cremp in future studies by gene targeting. The gene cluster of the human SCR proteins is located in a single region, 1q32 (3, 4). However, in mice, gene translocation has allowed the cluster to separate to two distinct regions (43– 45). In addition, the gene regulatory profile of DAF and MCP appears to be unique compared with that of humans (46, 47). It will be of interest in future studies to identify the SCR gene cluster and its constituents in chicken chromosomes, as it may represent an ancestral SCR gene cluster. Genomic analysis will be required to settle these issues.
Acknowledgments We thank Drs. J. Takeda (Osaka University, Osaka, Japan), R. Goizuka (Tokyo Scientific University, Tokyo, Japan) and T. Takeda (Kyoto University, Kyoto, Japan), S. Tsuji (Osaka Medical Center, Osaka, Japan), and Y. Nishizawa (Osaka University) for providing reagents and thoughtful discussions. Thanks are also due to Drs. N. A. Begum and K. Hazeki, and K. Shida, M. Taniguchi, and S. Kikkawa for technical assistance.
References 1. Ji, X., K. Azumi, M. Sasaki, and M. Nonaka. 1997. Ancient origin of the complement lectin pathway revealed by molecular cloning of mannan binding protein-associated serine protease from a urochordate, the Japanese ascidian, Halocynthia roretzi. Proc. Natl. Acad. Sci. USA 94:6340. 2. Sunyer, J. O., I. K. Zarkadis, and J. D. Lambris. 1998. Complement diversity: a mechanism for generating immune diversity? Immunol. Today 19:519. 3. Liszewski, M. K., T. W. Post, and J. P. Atkinson. 1991. Membrane cofactor protein (MCP or CD46): newest member of the regulators of complement activation gene cluster. Annu. Rev. Immunol. 9:431. 4. Carroll, M. C., E. M. Alicot, P. J. Katzman, L. B. Klickstein, J. A. Smith, and D. T. Fearon. 1988. Organization of the genes encoding complement receptors type1 and 2, decay-accelerating factor, and C4-binding protein in the RCA locus on human chromosome 1. J. Exp. Med. 167:1271. 5. Fearon, D. T. 1984. Cellular receptors for fragments of the third component of complement. Immunol. Today 5:105. 6. Hebell, T., J. M. Ahearn, and D. T. Fearon. 1991. Suppression of the immune response by a soluble complement receptor of B lymphocytes. Science 254:102. 7. Fearon, D. T., and R. H. Carter. 1995. The CD19/CR2/TAPA-1 complex of B lymphocytes: linking natural to acquired immunity. Annu. Rev. Immunol. 13:127. 8. Molina, H., V. M. Holers, B. Li, Y.-F. Fung, S. Mariathasan, J. Goellner, J. Strauss-Schoenberger, R. W. Karr, and D. D. Chaplin. 1996. Markedly impaired humoral immune response in mice deficient in complement receptors 1 and 2. Proc. Natl. Acad. Sci. USA 93:3357. 9. Paul, M. S., M. Aegerter, S. E. O’Brien, C. B. Kurtz, and J. H. Weis. 1989. The murine complement receptor gene family: analysis of mCRY gene products and their homology to human CR1. J. Immunol. 142:582. 10. Sakurada, C., H. Seno, N. Dohi, H. Takizawa, M. Nonaka, N. Okada, and H. Okada. 1994. Molecular cloning of the rat complement regulatory protein, 5I2 antigen. Biochem. Biophys. Res. Commun. 198:819. 11. Kim, Y. U., T. Kinoshita, H. Molina, D. Hourcade, T. Seya, L. M. Wagner, and V. M. Holers. 1995. Mouse complement regulatory protein Crry/p65 uses the specific mechanisms of both human decay-accelerating factor and membrane cofactor protein. J. Exp. Med. 181:151. 12. Xu, C., D. Mao, V. M. Holers, B. Palanca, A. M. Cheng, and H. Molina. 2000. A critical role for murine complement regulator crry in fetomaternal tolerance. Science 287:498. 13. Kaidoh, T., and I. Gigli. 1987. Phylogeny of C4b-C3b cleaving activity: similar fragmentation patterns of human C4b and C3b produced by lower animals. J. Immunol. 139:194. 14. Kaidoh, T., and I. Gigli. 1989. Phylogeny of regulatory proteins of the complement system: isolation and characterization of a C4b/C3b inhibitor and a cofactor from sand bass plasma. J. Immunol. 142:1605. 15. Dahmen, A., T. Kaidoh, P. F. Zipfel, and I. Gigli. 1994. Cloning and characterization of a cDNA representing a putative complement-regulatory plasma protein from barred sand bass (Parablax neblifer). Biochem. J. 301:391. 16. Kemper, C., P. F. Zipfel, and I. Gigli. 1998. The complement cofactor protein (SBP1) from the barred sand bass (Paralabrax nebulifer) mediates overlapping regulatory activities of both human C4b binding protein and factor H. J. Biol. Chem. 273:19398.
The Journal of Immunology 17. Seya, T., Y. Murakami, M. Nomura, N. A. Begum, M. Matsumoto, and S. Nagasawa. 1998. CD46 (membrane cofactor protein (MCP) of complement, measles virus receptor): structural and functional divergence among species. Int. J. Mol. Med. 1:809. 18. Tsujimura, A., K. Shida, M. Kitamura, M. Nomura, J. Takeda, H. Tanaka, M. Matsumoto, K. Matsumiya, A. Okuyama, Y. Nishimune, et al. 1998. Molecular cloning of a murine homologue of membrane cofactor protein (CD46): preferential expression in testicular germ cells. Biochem. J. 330:163. 19. Kojima, A., K. Iwata, T. Seya, M. Matsumoto, H. Ariga, J. P. Atkinson, and S. Nagasawa. 1993. Membrane cofactor protein (CD46) protects cells predominantly from alternative complement pathway-mediated C3 fragment deposition and cytolysis. J. Immunol. 151:1519. 20. Hara, T., Y. Suzuki, T. Nakazawa, H. Nishimura, S. Nagasawa, M. Nishiguchi, M. Matsumoto, M. Hatanaka, M. Kitamura, and T. Seya. 1998. Post-translational modification and intracellular localization of a splice product of CD46 cloned from human testis: role of the intracellular domains in O-glycosylation. Immunology 93:546. 21. Niwa, H., K. Yamamura, and J. Miyazaki. 1991. Efficient selection for highexpression transfectants with a novel eukaryotic vector. Gene 108:193. 22. Bezzubova, O., A. Silbergleit, Y. Yamaguchi-Iwai, S. Takeda, and J. M. Buerstedde. 1997. Reduced X-ray resistance and homologous recombination frequencies in a RAD54⫺/⫺ mutant of the chicken DT40 cell line. Cell 89:185. 23. Lublin, D. M., M. K. Liszewski, T. W. Post, M. A. Arce, M. M. Le Beau, M. B. Rebentisch, L. S. Lemons, T. Seya, and J. P. Atkinson. 1988. Molecular cloning and chromosomal localization of human membrane cofactor protein (MCP): evidence for inclusion in the multigene family of complement-regulatory proteins. J. Exp. Med. 168:181. 24. Iwata, K., T. Seya, Y. Yanagi, J. M. Pesando, P. M. Johnson, M. Okabe, S. Ueda, H. Ariga, and S. Nagasawa. 1995. Diversity of sites for measles virus binding and for inactivation of complement C3b and C4b on membrane cofactor protein CD46. J. Biol. Chem. 270:15148. 25. Brodbeck, W. G., D. Liu, J. Sperry, C. Mold, and M. E. Medof. 1996. Localization of classical and alternative pathway regulatory activity within the decayaccelerating factor. J. Immunol. 156:2528. 26. Laursen, I., and C. Koch. 1989. Purification of chicken C3 and a structural and functional characterization. Scand. J. Immunol. 30:529. 27. Mavroidis, M., J. O. Sunyer, and J. D. Lambris. 1995. Isolation, primary structure, and evolution of the third component of chicken complement and evidence for a new member of the ␣ 2-macroglobulin family. J. Immunol. 154:2164. 28. Koch, C., J. Josephsen, E. M. Nicolaisen, and M. Simonsen. 1982. Complement mediated lysis in chickens. Dev. Comp. Immunol. 6:141. 29. Ohta, H., Y. Yoshikawa, C. Kai, K. Yamanouchi, and H. Okada. 1984. Lysis of horse red blood cells mediated by antibody-independent activation of the alternative pathway of chicken complement. Immunology 52:437. 30. Kjalke, M., K. G. Welinder, and C. Koch. 1993. Structural analysis of chicken factor B-like protease and comparison with mammalian complement proteins factor B and C2. J. Immunol. 151:4147. 31. Nonaka, M., M. Takahashi, and M. Sasaki. 1994. Molecular cloning of a lamprey homologue of the mammalian MHC class III gene, complement factor B. J. Immunol. 152:2263.
431 32. Nakao, M., Y. Fushitani, K. Fujiki, M. Nonaka, and T. Yano. 1998. Two diverged complement factor B/C2-like cDNA sequences from a teleost, the common carp (Cyprinus carpio). J. Immunol. 161:4811. 33. The MHC sequencing consortium. 1999. Complete sequence and gene map of a human major histocompatibility complex. Nature 401:921. 34. Kaufman, J., S. Milne, T. W. Gobel, B. A. Walker, J. P. Jacob, C. Auffray, R. Zoorob, and S. Beck. 1999. The chicken B locus is a minimal essential major histocompatibility complex. Nature 401:923. 35. Kaufman, J. 1999. Co-evolving genes in MHC haplotypes: the “rule” for nonmammalian vertebrates? Immunogenetics 50:228. 36. Fernandez-Centeno, E., G. de Ojeda, J. M. Rojo, and P. J. Portoles. 2000. Crry/ p65, a membrane complement regulatory protein, has costimulatory properties on mouse T cells. J. Immunol. 164:4533. 37. Astier, A., M. C. Trescol-Biemont, O. Azocar, B. Lamouille, and C. J. Rabourdin-Combe. 2000. CD46, a new costimulatory molecule for T cells, that induces p120CBL and LAT phosphorylation. J. Immunol. 164:6091. 38. Simpson, K. L., and C. H. Holmes. 1994. Differential expression of complement regulatory proteins decay-accelerating factor (CD55), membrane cofactor protein (CD46) and CD59 during human spermatogenesis. Immunology 81:452. 39. Okabe, M., M. Nagira, Y. Kawai, S. Matzno, T. Mimura, and T. Mayumi. 1990. A human sperm antigen possibly involved in binding and/or fusion with zona-free hamster eggs. Fertil. Steril. 54:1121. 40. Anderson, D. J., A. F. Abbott, and R. M. Jack. 1993. The role of complement component C3b and its receptors in sperm-oocyte interaction. Proc. Natl. Acad. Sci. USA 90:10051. 41. Seya, T. 1995. Human regulator of complement activation gene family proteins and their relationship to microbial infection. Microbiol. Immunol. 39:s295. 42. Lindahl, G., U. Sjobring, and E. Johnsson. 2000. Human complement regulators: a major target for pathogenic microorganisms. Curr. Opin. Immunol. 12:44. 43. Kingsmore, S. F., D. P. Vik, C. B. Kurtz, P. Leroy, B. F. Tack, J. H. Weis, and M. F. Seldin. 1989. Genetic organization of complement receptor-related genes in the mouse. J. Exp. Med. 169:1479. 44. Holers, V. M., T. Kinoshita, and H. Molina. 1992. The evolution of mouse and human complement C3-binding proteins: divergence of form but conservation of function. Immunol. Today 13:231. 45. Nomura, M., A. Tsujimura, K. Shida, M. Matsumoto, Y. Matsuda, K. Toyoshima, and T. Seya. 1999. Membrane and secretory forms of mouse membrane cofactor protein (CD46) generated from a single gene through alternative splicing. Immunogenetics 50:245. 46. Spicer, A. P., M. F. Seldin, and S. J. Gendler. 1995. Molecular cloning and chromosomal localization of the mouse decay-accelerating factor genes: duplicated genes encode glycosylphosphatidylinositol-anchored and transmembrane forms. J. Immunol. 155:3079. 47. Nomura, M., A. Tsujimura, N. A. Begum, M. Matsumoto, K. Toyoshima, and T. Seya. 2000. Mechanism of gene regulation of membrane cofactor protein (CD46) in mice: identification and characterization of a silencer regulatory element in the 3⬘-flanking region of the murine CD46 gene. Biochem. J. 351:353.
