Immunology 719-724

Human complement regulatory proteins expressed on mouse A9 cells containing a human chromosome 1
T. SEYA, M. OKADA, T. HARA, M. MATSUMOTO, S. MIYAGAWA & M. OSHIMURA* Department of Immunology, Center for Adult Diseases Osaka, Higashinari-ku, Japan and *School of Life Sciences, Tottori University Yonago, Tottori, Japan
Acceptedfor publication August 1991
SUMMARY The structural genes of human complement regulatory proteins are clustered on chromosome I at position q3.2. Human chromosome 1 was transferred into a mouse fibroblast cell line, A9 [designated as A9(neo- 1)], and the surface expression of its gene products participating in complement regulation, namely C3b/C4b receptor (CRI, CD35), decay-accelerating factor (DAF, CD55), membrane cofactor protein (MCP, CD46) and C3d/EB virus receptor (CR2, CD2 1), were assessed using respective monoclonal antibodies by flow cytometry. CR1 became positive within 7 days of culture. MCP appeared in a small population ofcells by Day 3 and, together with DAF, began to increase on Day 7. CR2 appeared on Day 14. The order of the expression was CR1 > DAF = MCP > CR2. On Day 42, however, all became negative except for MCP, which was markedly diminished. These human regulatory proteins were specifically associated with the presence of human chromosome 1, since none of them were expressed on human chromosome 12-transferred A9 cells [A9(neo- 12)]. Intact A9 and A9(neo- 12) cells activated human complement via the alternative pathway. The activation of this pathway was suppressed in the A9(neo- 1) cells that expressed CR 1, DAF and MCP. Slight protective activity was still observed in the 42-day cultured A9(neo- 1) cells expressing only trace MCP. These results suggest that human complement regulators, expressed via the transferred human chromosome 1, can protect heterologous cells from complement, overcoming their ability to activate the human alternative pathway.
INTRODUCTION Activation of the complement system leads to the formation of the C3 convertases, which in turn promote the deposition of C3 on foreign material (reviewed by refs I and 2). The C3 activation and deposition serve as the basis of many biological functions, including immune cytolysis, immune clearance, potentiation of phagocytosis, enhancement of natural killer activity, and modulation of the inflammatory response by the release of anaphylatoxins (reviewed by ref. 3). Cells targeted by C3 (deposited as
C3b/C3bi) are thereby damaged. Normal human cells are not damaged by autologous complement. Membrane regulatory proteins are considered to play an important role in protecting host tissues from autologous complement attack.4 The regulation of C3 convertase
involves one regulatory group, consisting of two plasma
Abbreviations: CR1, C3b/C4b receptor (CD35); DAF, decayaccelerating factor (CD55); FCS, foetal calf serum; MCP, membrane co-factor protein (CD46); PBS, phosphate-buffered saline; SDSPAGE, sodium dodecyl sulphate-polyacrylamide gel electrophoresis. Correspondence: Dr T. Seya, Dept. of Immunology, Center for Adult Diseases Osaka, Nakamichi, Higashinari-ku, Osaka 537, Japan.
proteins, C4b-binding protein (C4bp) and factor H, and four membrane proteins, C3b/C4b receptor (CR1, CD35), C3dg/EB virus receptor (CR2, CD21), decay-accelerating factor (DAF, CD55), and membrane co-factor protein (MCP, CD46) (reviewed by ref. 5). These proteins commonly possess tandemly arranged -60 amino acid repeats, namely short consensus repeats (SCR), and their structural genes are clustered within an 800 kb fragment on the long arm of chromosome 1, lq3.2.67 Most tissue of non-human origin, on the other hand, activates human complement regardless of sensitization with antibody (Ab). The activation is promoted by an activator of the alternative complement pathway."8 Severe complement activation is actually observed in the transplantation of xenografts which may express activator.9 If xenografts could be protected against complement by human regulatory proteins, their survival time would be prolonged. The present study investigated the role of human regulatory proteins expressed on mouse cells containing a human chromosome 1.

MATERIALS AND METHODS Cells and antibodies A human T-cell line, MT1, and mouse A9 cells containing a

T. Seya et al.

pg of human factor I and 10 pg of methylamine-treated fluorescent-labelled human C3, f-C3(MA), for 5 hr at 370. The reaction was then stopped by the addition of 10,lp of 10% SDS and 3 p1 of 2-mercaptoethanol. The samples were analysed by SDS-PAGE and the percentage ac-chain cleavage of the substrates was determined by spectrofluorometry.'8
Assessment of C3 deposition C3 fragments deposited on the cell-surface were analysed by flowcytometry. 15 Cells (1 x 106) were pretreated with about 25 pg of M 177, IAl 0 or non-immune mouse IgG and incubated with 50 p1 of Mg2+-EGTA human serum and 150 p1 of EGTAGVB2+ for 90 min at 37. As a control, EDTA-human serum was used instead of Mg2+-EGTA serum. The cells were washed twice in PBS containing 2% bovine serum albumin (PBS/BSA), then 3 pg of anti-C3c rabbit polyclonal antibody were added as the first antibody. After 45 min, the cells were again washed twice in PBS/BSA, and 5 pg of the second antibody (FITClabelled goat anti-mouse F(ab')2, Cappel) were added. The mixtures were allowed to stand for 45 min, washed twice, and analysed by flow cytometry.
human chromosome I or 12, i.e., A9(neo-1) and A9(neo-12),"0 were provided by the Japanese Cancer Research Resources Bank (JCRB, Tokyo, Japan). The preparation of these cells" and their properties'0" 2 have been described previously. Briefly, human fibroblast cells transfected with pSVZ neomycinresistant gene were hybridized with mouse A9 cells using polyethyleneglycol. The hybrid cells were treated with colcemid and cytochalasin B to obtain microcells containing separated human chromosomes. The microcells were fused again with A9 cells. The A9-microcell hybrids with a neomycin-resistant gene-tagged (neo-tagged) human chromosome were screened by chromosome analysis."' Mouse monoclonal antibodies against human CR1 (31R)'4 and MCP (M 177)'5 were prepared in our laboratory, and those against human DAF (including IA10)'6 were a gift from Dr Kinoshita (Osaka University, Osaka, Japan). A mouse monoclonal antibody against CR2 (OKB7)'7 was purchased from Ortho Diagnostic Systems Inc. (Raritan, RJ). Rabbit polyclonal antibody to human C3c (which recognizes C3b and C3bi but not C3dg) was from Behringwerke (Marburg, Germany). FITClabelled goat antibodies (directed specifically against mouse IgG or rabbit IgG) for flowcytometry were from Cappel (Westchester, PA).
Culture conditions A9(neo-1) and A9(neo-12) cells were maintained in RPMI-1640 containing 10% foetal calf serum (FCS) and 0-8 mg/ml of G418 for more than 6 weeks, and then stored. The cells for storage expressed little or virtually no regulatory proteins. When the cells were kept in a - 80 freezer, dimethylsulphoxide (10%) was added; this reagent was removed when culture of the cells was restarted. The experimental cultures were maintained in the above culture medium and kept in a 5% C02/95% air atmosphere at 37. No stimulating reagent was used during the culture.

Assessment of surface-expressed proteins Approximately 106 cells were incubated with 10-20 pg of the primary antibody, namely anti-MCP (M 177), anti-DAF (IA 10), or anti-CR 1 (31 R), together with 100 p1 of platelet-poor plasma for 40 min. Mouse non-immune IgG (10 pg; Cappel) was used as a control. The cells were washed in phosphate-buffered saline containing 2% foetal calf serum (PBS/FCS), and treated with 5 pg of the FITC-conjugated second antibody in 100 pl of PBS/ FCS and 100 p1 of platelet-poor plasma. After 30 min, the cells were washed twice with PBS and fixed with 0-5% paraformaldehyde.'5 The samples were analysed on a FACScan (BectonDickinson, Mountain View, CA) within I week. The protein A-rosette assay'4 was used to simultaneously assess the surface-expressed regulatory proteins in some cases. The sensitivity of this method was tested using anti-CR 1 (31 R) and human erythrocytes, having 100-950 copies/cell of CR1. This method was found to be sensitive enough to detect < 100 copies/cell of antigens.
Cytolysis assay Cells (1 x 107) were incubated with 200 pCi of 5'Cr (Amersham, Arlington Height, IL) in 400 p1 of RPMI for 60-120 min at 37".19 About 70,000 c.p.m. were incorporated into 106 cells. The labelled cells were pretreated with antibodies against MCP or non-immune IgG and incubated with 15-30% Mg2+-EGTA serum or EDTA serum for 90 min at 37. The radioactivity in the supernatants was measured in a y counter.

RESULTS

Dynamics of human complement regulatory proteins expressed on mouse A9 cells Human complement regulators, CR1, DAF and MCP, and an alternative pathway activator, CR2,20 expressed on mouse A9(neo-1) cells were assessed. CR1, DAF and CR2 were all undetectable on A9(neo-1) cells by flow cytometry (Fig. 1) and protein A-assay (not shown) 3 days after the initial culture of thawed cells. Trace amounts of MCP were detected in a small population of cells on Day 3. On Day 7, CR1 was strongly expressed. MCP and DAF began appearing uniformly around Day 7. On Day 14, CR2s appeared and the other three formed symmetric, unimodal shifts, suggesting that their expression had reached a plateau (Fig. 1). The mean fluorescent shifts caused by the expression of CR1, CR2, DAF and MCP are plotted in Fig. 2. The Ab we used did not induce significant fluorescent shift on A9(neo- 12) cells at any time-points, suggestive of no cross-reactive antigen on A9 cells. Regarding A9(neo- 1), the order of expression of the regulators was CR1 > DAF = MCP > CR2. The regulators once expressed became undetectable by both flowcytometry and protein A-assay by Day 42, except for MCP, which was still weakly expressed (not shown).
Function of the expressed human complement regulatory proteins

Assay for factor I co-factor activity A fluid-phase assay system was used.'8 The buffer was 20 mM phosphate buffer/0-02% NP-40, pH 6-2. A9 cells (5 x 108) were solubilized with I ml of PBS/1% NP-40 containing EDTA (10
mM), iodoacetamide (25 mM) and phenylmethylsulphonyl floride (I mM), and the supernatants (100 pl) were incubated with
A9(neo-1) and A9(neo-12) cells cultured for 7 days were incubated with EGTA-Mg2+ human serum (Fig. 3). More than 20% of the serum concentrations were required for induction of
complement regulators expressed on mouse cells

Fluorescent intensity

Control

02,;,..

Fluorescent intensity Figure 1. Expression of human complement regulatory proteins on mouse A9(neo- 1) cells. The cells were thawed and cultured for the indicated periods. Surface-expressed DAF, MCP, CR1 (a) and CR2 (b) were evaluated by flow cytometry using mouse monoclonal antibodies and FITC-labelled goat anti-mouse IgG. Non-immune mouse IgG was used as control.

VIs. -.~

~~~CP2
A9(neo- 1) cells, in the 42-day cultured A9(neo- 1) cells which expressed no CRI or DAF but MCP, reflecting the protective activity of MCP (Fig. 4). Seven- and 14-day cultured A9(neo- 1) and A9(neo- 1 2) cells were solubilized and factor I co-factor activity (presumably due to the produced MCP) was assessed using human factor I and C3b." By this method, mouse factor I co-factors are not detectable, since mouse co-factors are virtually incompatible with the human system (manuscript in preparation). Co-factor activity was negative in the A9(neo-12) cells but positive in the A9(neo-1) cells (data not shown). Expression of the MCP epitopes on A9(neo-1) cells On human cells and cell lines, MCP expresses the three epitopes defined by M75, M 160 and M 177. Their degrees of expression were different: most were M 160 < M75 M 177, an example of which, MT I, is shown in Fig. 5b. The amounts of the expressed epitopes on A9(neo-1) cells were assessed by flow cytometry (Fig. 5a). The 7-day cultured A9(neo-1) cells displayed the M 160 > M75 M 177 profile, which is different from that of the lower expression of the M 160 epitope seen in most human cells (Fig. 5). This profile was maintained until Day 14.
-50 Days cultured Figure 2. Human regulatory proteins were expressed on A9(neo- 1) cells but not on A9(neo- 12) cells. A9(neo- 1) (a) and A9(neo- 12) cells (b) were cultured under the same conditions and their surface-expressed human CR 1, DAF MCP and CR2 were assessed by flow cytometry. The same antibody lots were used to detect these factors.
the C3-deposition (Fig. 3b). C3 was deposited on A9(neo-12) cells, although no cytolysis occurred. On the other hand, less C3 was deposited on A9(neo- 1) cells that expressed CR 1, DAF and MCP. No lysis was observed, either. The clear biomodal shift in flowcytometry (Fig. 3a) was reproducible, for an unknown

reason.

mouse mouse
DISCUSSION CR2- and DAF-like molecules have been identified

no case

did EDTA human

induce C3-

deposition or cytolysis. The protective activity against C3-deposition of the 14-day cultured A9(neo-1) cells was slightly higher than that of the 7day cultured A9(neo-1) cells, an unlikely result if the expressed CR2 potentiated the C3-deposition overcoming the DAF/MCP on the same cell membrane (data not shown). The protective activity still remained, though far less than in the 7-day cultured
monoclonal Ab to human regulatory proteins, no crossreactive antigen was detected on A9(neo- 1 2) cells. Furthermore, with regard to M 177 and 1A IO, no mouse cross-reactive antigen has been identified on 20 mouse cell lines so far examined (not shown). These antibodies therefore can be applied to the assessment of human complement regulators expressed on mouse cells. The purpose of this study was to clarify whether the human C3-step regulatory proteins expressed by gene transfection on cells of other species can block human C3-deposition. On
cells." -21 Although a mouse molecule, p65, cross-reacts with anti-human CR I polyclonal antibod y,21 in this study using

Seya et al.

A9-1 A9-12

cc: 200)

100 10

Serum conc. (%)

Figure 4. Protective activity against human C3-deposition remaining on the 42-day cultured mouse A9(neo- 1) cells. A9(neo- 1) cells (designated as A9- 1 ) and A9(neo- 1 2) cells (designated as A9- 12) were incubated with the indicated concentrations of Mg" EGTA human serum then the deposited C3 was measured by flowcytometry. as in Fig. 3.
10l Fluorescent intensity

2 A9-1

10 Serum
Figure 3. C3-deposition induced via the human alternative pathway on 7-day cultured A9(neo-1) and A9(neo-12) cells. The 7-day cultured A9(neo-1) and A9(neo-12) cells were incubated with Mg2'-EGTA human serum and deposited human C3 was assessed with a rabbit polyclonal antibody to human C3. EDTA human serum was used to estimate the background bound C3. Less than 5% of the cells were killed during the treatment (not shown). (a) A9(neo-1) cells; similar results were obtained with 14-day cultured cells (not shown). (b) panel. A9 (neo-12) cells; two peaks indicating different efficiencies of C3 deposition were observed reproducibly and similar results were obtained with intact A9 cells. (c) C3-deposition was estimated under the various serum concentrations as indicated.

IFluorescent intensity

Figure 5. Varying expression rates of MCP epitopes between A9(neo- 1) cells and MTl. The 7-day cultured A9(neo- 1) cells (a) and MTI (b) were incubated with the monoclonal antibodies, M75, M160 and M177, which recognize different MCP epitopes. Binding was detected with the FITC-labelled goat anti-mouse IgG and analysed by flowcytometry. Non-immune mouse IgG was used as the control.

foreign material, C3 deposition depends upon membrane factors that act an alternative pathway activator (C3acceptor)'24 or modulate accessibility of factor H to membranebound C3b.25 Furthermore, C3-accepting molecules may influence the stability of the C3b deposited upon themselves.'Whether or not the human regulatory protein expressed on heterologous cells with potent abilities to activate human complement can still maintain protective function against human C3 has not been examined except for DAF.26 We found that transferred human chromosome 1 induced the expression of human CRI, DAF, MCP and CR2 on the surface of mouse A9 cells. Although A9(neo- 12) and A9 cells are capable of activating the human alternative pathway, promias
nent protective activity against human C3 attack was induced in parallel with the expression of DAF and MCP. Furthermore. MCP extracted from A9(neo-1) cells possessed full co-factor activity. The results imply that human regulatory proteins can be produced as functionally active forms from transferred human chromosome 1I and serve as effective regulators for human C3 activation even on the cells of differcnt species. That is, the human regulatory proteins can interact as inhibitors with human C3b on putative mouse activators. Human lymphocyte CR2 is known to act as an alternative pathway activator on which homologous C3 is deposited.2" This CR2 activity can not be suppressed by the DAF/MCP on the
Human complement regulators expressed on mouse cells
same cell membrane.27 Why the expressed CR2 was not an effective activator on A9 cells in unknown. It may have a limited function on human lymphocytes, or a qualitative (including maintaining a quaternary structure with other membrane proteins) or a quantitative alteration may abolish the alternative pathway-activating ability. Inability of CR2 to induce homologous C3 depositions may be convenient for protection of the chromosome 1-transfected cells. MCP was the last regulator to disappear from A9(neo- 1) cells. Thus, we examined the expression of each epitope recognized by the three different mAb. The three antibodies all recognized the primary sequences of MCP but not the tertiary structure or sugar. Interestingly, the M 160 epitope is expressed at higher levels than the other two on A9 cells, in contrast to most human cell lines and blood cells. Although the altered expression of the MCP epitopes on A9(neo-1) cells probably reflects a structural difference of the MCP expressed on this cell line, the findings that the MCP extracted from A9(neo-1) cells has full co-factor activity and that the M177 and M75 epitopes associated with MCP co-factor function are sufficiently expressed may account for the complement-protective activity in 42-day cultured A9-1(neo-1) cells. Clearer results will be obtained by the transfection of MCP cDNA into cells which do not express any regulatory factors for human complement. Expression of human chromosome 1 gene products in other species is particularly important in xenotransplantation since, within 30 mins, the graft undergoes hyperacute rejection which is evoked by activation of the host alternative complement pathway.9 Although the killing effectors are yet unidentified in this case, introduction of human chromosome 1 may facilitate xenograft survival by protecting the graft cells from human complement-mediated rejection. However, difficulties are that > 7 days of culture are required for the expression of the major protective factors, DAF and MCP, and they diminished within 42 days, for reasons as yet unknown. Gene transfection therapy could be applicable to a temporal xenotransplant such that short-term circumvention of host C3 attack would provide sufficient graft survival, which would facilitate the following permanent homo-or allo-transplantation. Note added in proof The ability of human MCP to protect host heterologous cells from human complement-mediated cell damage was tested using CHO cells transferred with MCP cDNa by Lublin & Coyne,28 whose report was published soon after the submission of this paper.

ACKNOWLEDGMENTS

We are grateful to JRCB for providing the cell lines, and to Dr Akedo for valuable discussions. Thanks are also due to Ms Ito for secretarial assistance. This work was supported in part by grants from the Cell Science Research Foundation, the Kowa Research Foundation, Pharmaceutical Research Foundation, and the Inoue Science Research Foundation.

REFERENCES

1. FEARON D.T. (1979) Activation of the alternative complement pathway. CRC Crit. Rev. Immunol. 1, 1.
2. Ross G.D. & MEDOF M.E. (1985) Membrane complement receptors specific for bound fragments of C3. Adv. Immunol. 37, 217. 3. ARNAOUT M.A. & COLTEN H.R. (1984) Complement C3 receptors: structure and function. Molec. Immunol. 21, 1191. 4. ATKINSON J.P. & FARRIES T.C. (1987) Separation of self from nonself in the complement system. Immunol. Today, 8, 212. 5. REID K.B.M., BENTLEY D.R., CAMPBELL R.D., CHUNG L.P., SIM R.B., KRISTENSEN T. & TACK B.F. (1986) Complement system proteins which interact with C3b or C4b. Immunol. Today, 7, 230. 6. HOLERS V.M., LUBLIN D.M., COLE J.L., SEYA T. & ATKINSON J.P. (1985) Human C3b- and C4b-binding proteins: a new gene family. Immunol. Today, 6, 188. 7. HOURCADE D., HOLERS V.M. & ATKINSON J.P. (1989) The regulators of complement activation (RCA) gene cluster. Adv. Immunol. 45, 381. 8. PANGBURN M.K. & MULLER-EBERHARD H.J. (1984) The alternative pathway of complement. Springer Semin. Immunopathol. 7, 163. 9. MIYAGAWA S., HIROSE H., SHIRAKURA R., NAKA Y., NAKATA S., KAWASHIMA Y., SEYA T., MATSUMOTO M., UENAKA A. & KITAMURA H. (1988) The mechanism of discordant xenograft rejection. Transplantation, 46, 825. 10. Kol M., MORITA H., SHIMIZU M. & OSHIMURA M. (1989) construction of mouse A9 clones containing a single human chromosome (X/autosome translocation) via micro-cell fusions. Jpn. J. Cancer Res. 80, 122. 11. OSHIMURA, M., KUGOH H., Kol M., SHIMIZU M., YAMADA H., SATOH H. & BARRETT J.C. (1990a) Transfer of a normal human chromosome 11 suppresses tumorigenicity of some but not all tumor cell lines. J. Cell Biochem. 42, 1. 12. OSHIMURA M., KUGOH H., SHIMIZU M., YAMADA, H., HASHIBA H., HORIKI I. & SASAKI, M. (1990b) Multiple chromosomes carrying tumor suppressor activity via microcell-mediated chromosome transfer, for various tumor cell lines. In: Genetic Basis for Carcinogenesis: Tumor Suppressor Genes and Oncogenes, p. 247. Japan Scientific Society Press, Tokyo/Taylor & Francis, Ltd. London. 13. Kol M., SHIMIZU M., MORITA H., YAMADA H. & OSHIMURA M. (1989) Construction of mouse A9 clones containing a single human chromosome tagged with neomycin-resistance gene via microcell fusion. Jpn. J. Cancer Res. 80, 413. 14. SEYA T., HARA T., UENAKA A., NAKAYAMA E. & AKEDO H. (1990a) Application of protein A-rosette assay for screening of antibodies to complement regulatory proteins. Comp. Inflamm. 7, 78. 15. SEYA T., HARA T., MATSUMOTO M. & AKEDO H. (1990b) Quantitative analysis of membrane cofactor protein (MCP) of complement: high expression of MCP on human leukemia cell lines, which is down-regulated during cell differentiation. J. Immunol. 145, 238. 16. KINOSHITA T., MEDOF M.E., SILBER R. & NUSSENZWEIG V. (1985) Distribution of decay-accelerating factor in the peripheral blood of normal individuals and patients with paroxysmal nocturnal hemoglobinuria. J. exp. Med. 162, 75. 17. NEMEROW G. R., WOLFERT R., McNAUGHTON M E. & COOPER N. R. (1985) Identification and characterization of the EBV receptor on human B lymphocytes and its relationship to the C3d receptor (CR2). J. Virol. 55, 347. 18. SEYA T., HOLERS V.M. & ATKINSON J.P. (1985) Purification and functional analysis of the polymorphic variants of the C3b/C4b receptor (CR 1) and comparisons to H, C4b-binding protein (C4bp) and decay-accelerating factor (DAF). J. Immunol. 135, 2661. 19. SEYA T., HARA T., MATSUMOTO M., SUGITA Y. & AKEDO H. (1990c) Complement-mediated tumor cell damage induced by antibodies against membrane cofactor protein (MCP, CD46). J. exp. Med. 172, 1673. 20. MOLD C., NEMEROW G.R., BRADT B.M. & COOPER N.R. (1988) CR2 is a complement activator and the covalent binding site for C3 during alternative pathway activation by Raji cells. J. Immunol. 140, 1923. 21. KAMEYOSHI Y., MATSUSHITA M. & OKADA H. (1989) Murine

modulation by H of C5 cleavage by the cell-bound alternative pathway CS convertase of human complement. J. Immunol. 130, 2821. 26. MEDOF M.E., KINOSHITA T. & NUSSENZWEIG V. (1984) Inhibition of complement activation on surface of cells after incorporation of decay-accelerating factor (DAF) into their membranes. J. exp. Med. 160, 1558. 27. CAUDWELL V., PORTEU F., CALENDER A., PANGBURN M.K. & HALBWACHS-MECARELLI L. (1991) Complement alternative pathway activation and control on membranes of human lymphoid B cell lines. Eur. J. Immunol. 20, 2643. 28. LUBLIN D.M. & COYNE, K.E. (1991) Phospholipid-anchored and transmembrane versions of either decay-accelerating factor or membrane cofactor protein show equal efficiency in protection from complement-mediated cell damage. J. Exp. Med. 174, 35.
membrane inhibitor of complement which accelerates decay human C3 convertase. Immunology, 68, 439. KINOSHITA T., LAVOIE S. & NUSSENZWEIG V. (1985) Regulatory proteins for the activated third and fourth components of complement (C3b and C4b) in mice. II. Identification and properties of complement receptor type I (CR1). J. Immunol. 134, 2564. WONG W.W. & FEARON D.T. (1984) p65: a C3b-binding protein on murine cells that shares antigenic determinants with human C3b receptor (CR1) and is distinct from murine C3b receptor. J. Immunol. 134, 4048. MATSUMOTO M., SUGITA, Y. & SEYA T. (1991) Alternative complement pathway-mediated myeloid cell cytotoxicity: repertoire of membrane factors participating in regulation of C3-deposition and cytolysis. Eur. J. Immunol. 21, 1787. FISCHER A. & KAZATCHIKINE M.D. (1983) Surface-dependent