CD59a Is the Primary Regulator of Membrane Attack Complex Assembly in the Mouse

This information is current as Sivasankar Baalasubramanian, Claire L. Harris, Rossen M. of September 27, 2021. Donev, Masashi Mizuno, Nader Omidvar, Wen-Chao Song and B. Paul Morgan J Immunol 2004; 173:3684-3692; ; doi: 10.4049/jimmunol.173.6.3684

http://www.jimmunol.org/content/173/6/3684 Downloaded from

References This article cites 27 articles, 12 of which you can access for free at: http://www.jimmunol.org/content/173/6/3684.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 27, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2004 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CD59a Is the Primary Regulator of Membrane Attack Complex Assembly in the Mouse1

Sivasankar Baalasubramanian,* Claire L. Harris,* Rossen M. Donev,* Masashi Mizuno,* Nader Omidvar,* Wen-Chao Song,† and B. Paul Morgan2*

Gene-deleted mice have provided a potent tool in efforts to understand the roles of complement and complement-regulating in vivo. In particular, mice deficient in the membrane regulators 1-related / y, decay- accelerating factor, or CD59 have demonstrated homeostatic relevance and backcrossing between the strains has revealed coop- erativity in regulation. In mouse, encoding decay-accelerating factor and CD59 have been duplicated and show differential expression in tissues, complicating interpretation and extrapolation of findings to man. The first described form of CD59, CD59a, is broadly distributed and of the cd59a gene causes a mild hemolytic with increased susceptibility in comple- ment-mediated disease models. The distribution of the second form, CD59b, was originally described as testis specific, but later Downloaded from by some as widespread. Deletion of the cd59b gene caused a severe hemolytic and thrombotic phenotype. To apply data from these mouse models to man it is essential to know the relative distribution and functional roles of these two forms of CD59. We have generated new specific reagents and used them in sensitive quantitative analyses to comprehensively characterize expression of mRNA and protein and functional roles of CD59a and CD59b in wild-type (wt) and CD59a-negative mice. cd59b mRNA was detected only in testis and, at very low levels, in bone marrow. CD59b protein was present on mature spermatozoa and precursors and, in trace amounts, erythrocytes. Erythrocyte CD59b did not inhibit complement lysis except when CD59a was absent or http://www.jimmunol.org/ blocked. These data confirm that CD59a is the primary regulator of complement membrane attack in mouse. The Journal of Immunology, 2004, 173: 3684–3692.

ubiquitously expressed GPI-anchored , thrombosis (6–11). In PNH, other cell types express normal levels CD59 is the sole membrane regulator of the membrane of CD59 and are not compromised. However, it has been sug- A attack complex (MAC)3 of C (1–4). The critical role of gested that glycation-induced inactivation of human CD59 on en- CD59 in homeostasis is revealed in the acquired hematological dothelium may contribute to diabetic vascular damage (12).

disorder, paroxysmal nocturnal hemoglobinuria (PNH). Due to an Analyses of CD59 analogues from other species have not only by guest on September 27, 2021 acquired somatic mutation of the PIG-A gene in bone marrow helped our understanding of the evolution and structure-function hemopoietic precursor cells, a variable proportion of cells in relationships of CD59, but also have enabled the use of appropriate PNH patients lack GPI-anchored proteins, including the membrane animal models to study the physiological relevance of CD59. To- C regulators CD59 and decay-accelerating factor (DAF). As a con- ward this latter goal we earlier identified and characterized mouse sequence, erythrocytes and from these patients are highly CD59, which has a wide expression pattern, very similar to that of sensitive to autologous C-mediated lysis and activation, resulting human CD59 (13, 14), and went on to engineer a mouse lacking in and thrombosis (5). The key role of CD59 is the gene (15). Although erythrocytes from these mice were C emphasized in studies of isolated deficiencies of the two regula- sensitive in vitro, analyses in vivo revealed only a low degree of tors. Whereas isolated deficiency of DAF in several families was intravascular hemolysis and increased reticulocyte count, a mild not associated with PNH-like symptoms, the single described case phenotype in comparison with that seen in PNH. This observation of isolated deficiency of CD59 presented with hemolysis and in our CD59Ϫ/Ϫ mouse was supported by the phenotype observed in a mouse model of PNH in which a proportion of the hemopoi- etic cells are mutated in the PIG-A gene and lack GPI-anchored proteins (16, 17). Here too, the mice displayed only mildly ele- *Complement Biology Group, Department of Medical Biochemistry and Immunol- vated reticulocyte counts and were not anemic. Murine erythro- ogy, University of Wales College of Medicine, Cardiff, United Kingdom; and †Center for Experimental Therapeutics and Department of Pharmacology, University of Penn- cytes differ from human erythrocytes in that they express the pow- sylvania School of Medicine, Philadelphia, PA 19104 erful transmembrane C regulator -related Received for publication April 6, 2004. Accepted for publication June 2, 2004. gene/protein y (18). Further, mouse erythrocytes play no role in The costs of publication of this article were defrayed in part by the payment of page clearance (19). Together, these differences likely charges. This article must therefore be hereby marked advertisement in accordance explain the mild hemolytic phenotype both in the CD59Ϫ/Ϫ and with 18 U.S.C. Section 1734 solely to indicate this fact. PNH model mice. 1 This work was supported by the Wellcome Trust, through the award of International Travelling Research Fellowship 068280 to S.B., and Programme Grant 068590 fund- An additional level of complexity was introduced to the saga of ing to B.P.M. mouse erythrocyte C resistance by the demonstration that, in con- 2 Address correspondence and reprint requests to Dr. B. Paul Morgan, Department of trast with all other species studied so far, the cd59 gene in mouse Medical Biochemistry and Immunology, University of Wales College of Medicine, is duplicated (20). This second gene, termed cd59b, encoded a Cardiff CF14 4XN, U.K. E-mail address: [email protected] putative protein product CD59b that was 63% identical with the 3 Abbreviations used in this paper: MAC, membrane attack complex; CFD, comple- ment fixation diluent; CVF, cobra venom factor; DAF, decay-accelerating factor; m, originally described form, now termed CD59a. Analysis of mRNA murine; PNH, paroxysmal nocturnal hemoglobinuria; wt, wild type. expression indicated that, in contrast to the wide distribution of

Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00 The Journal of Immunology 3685

cd59a, expression of cd59b was restricted to testis (20). CD59 is nylenediamine and H2O2). The absorbance was read in a Bio-Rad plate not alone among C proteins in being genetically duplicated in the reader at 490 nm. Wells positive for anti-CD59b mAbs (mAb) were sub- mouse. The genes encoding DAF, C1r, and C1s are also dupli- cloned by limiting dilution to monoclonality. The mAb were isotyped using the IsoStrip mouse mAb isotyping (Boehringer Mannheim, Sussex, cated, with expression of one form being confined to the repro- U.K.). The specificity of mAb to CD59b was confirmed by flow cytometry ductive organs (21, 22). Our early studies examining the expres- and Western blotting of EL4, a murine lymphoma cell line because it does sion at the protein level of CD59b in wt and CD59aϪ/Ϫ mice were not express CD59a or CD59b, transfected with CD59a, CD59b, or with in accord with the mRNA findings and confirmed that CD59b ex- empty vector. The function-blocking ability of anti-CD59b mAb was as- sessed by testing their capacity to enhance C lysis of Ab-sensitized CD59b- pression was restricted to the testis (23). However, later studies expressing EL4 cells essentially as described (23). from Qin, et al. (24) described a widespread distribution of CD59b both at the mRNA level and, using isoform-specific polyclonal Immunofluorescent localization of CD59a and CD59b antisera, at the protein level. These authors stated that CD59b was Tissue sections of brain, lungs, heart, , , kidney, and testis from widely and abundantly expressed in mouse tissues in a distribution wt and CD59aϪ/Ϫ C57/BL6 mouse were prepared as described previously pattern that mirrored that of CD59a. Critically, they went on to (23). Briefly, 7-␮m sections were cut from snap-frozen tissues, placed onto Ϫ Ϫ delete the cd59b gene and reported that CD59b / mice displayed Snowcoat X-tra glass slides (Surgipath, Peterborough, U.K.), and fixed a severe PNH-like phenotype with spontaneous hemolytic anemia with acetone for 5 min. Fixed tissue sections were probed with mAb with a marked reticulocytosis and abnormal erythrocyte morphol- against mouse CD59b directly labeled with N-hydroxysuccinimido-FITC Ϫ/Ϫ as instructed by the manufacturer (Pierce, Cheshire, U.K.). To examine the ogy, together with activation (25). Male CD59b mice, distribution of CD59a, tissue sections were pretreated with Avidin-Biotin though fertile immediately after puberty, then developed a pro- blocking kit (Vector Laboratories, Peterborough, U.K.), then probed with gressive decline into early infertility. a biotinylated mAb against CD59a (mCD59.1(23)), followed by washing Downloaded from There are compelling reasons to investigate further the relative and incubating with FITC-labeled streptavidin (Sigma-Aldrich, Dorset, U.K.). After further washing, the tissue sections were placed in mounting roles of CD59a and CD59b in the protection of tissues from C Ϫ/Ϫ medium (Vector Laboratories) and examined using a fluorescence micro- attack. Crucially, the phenotype described in the CD59b mice scope (Leica Microsystems, Knowlhill, U.K.) with image analysis system does not fit with data from other sources that have indirectly ad- (Openlab, Coventry U.K.). dressed the roles of CD59a and CD59b. Affected erythrocytes To isolate motile spermatozoa, two cauda epididymii from an adult wt C57/BL6 mouse were minced in 1 ml of DMEM and incubated at room from the PIG-A-mutated PNH mice described above would lack http://www.jimmunol.org/ temperature for 15 min to allow sedimentation of large cellular aggregates. all GPI-anchored molecules, including both CD59a and CD59b, The supernatant was transferred to a fresh tube and cells washed twice with yet the in vitro and in vivo properties of these cells resembles 500 ␮l of DMEM by centrifugation (300 ϫ g) for 5 min at room temper- closely those described for the CD59aϪ/Ϫ erythrocytes; indeed, the ature. Cell pellets were resuspended in 100 ␮l of DMEM, smeared on glass calculated erythrocyte half-life for affected cells in the PNH mice slides, and immediately air dried. The smeared slides were fixed in acetone Ϫ Ϫ at room temperature for 2 min. The expression of CD59a and CD59b on is almost identical with that in the CD59a / mice (7.3 days vs 8.7 Ϫ/Ϫ sperm cells was examined using mAb to CD59a and CD59b as described days) (15, 16). The CD59a mouse has been used to test the role for immunohistochemistry above. of C activation and MAC formation in a large and diverse group of model diseases, including experimental nephritis, experimental de- Real-time PCR analysis for expression of cd59a and myelination, and experimental arthritis (26–29). In each of these cd59b mRNA by guest on September 27, 2021 models, deficiency of CD59a, either alone or combined with de- Total RNA was isolated from different mouse tissues (brain, heart, kidney, ficiency of DAF, markedly enhances disease and pathology, dem- liver, spleen, lungs, testis, and bone marrow) obtained from both wt and onstrating that in the absence of CD59a there is a failure to reg- CD59aϪ/Ϫ C57/BL6 mice using the RNeasy kit (Qiagen, Sussex, U.K.). ulate MAC in diverse organs and tissues. Aliquots of these RNAs, 1 ␮g each, were reverse-transcribed using random All of our data to date indicate that CD59a is the sole regulator hexamers and multiscribe reverse transcriptase according to the manufac- turer’s instructions (Applied Biosystems, Warrington, U.K.). A common of MAC assembly in the large majority of tissues and that deletion primer pair for cd59a (GenBank accession number, NM_007652) and of CD59a, while causing only a mild spontaneous phenotype, cd59b (GenBank accession number, NM_181858) (GCCGGAATG markedly enhances susceptibility to MAC-mediated injury in nu- CAAGTGTATCA, forward; and GTCCCCAGCAATGGTGTCTT, re- merous models. The work of others flatly contradicts these clear verse) was designed using Primer Express software (Applied Biosystems) within regions of identity in the two sequences. This primer pair surrounds findings. In an effort to bring this controversy to a conclusion, we DNA sequences that differ between cd59a and cd59b. Within this region, have generated new reagents and applied highly sensitive and we designed TaqMan probes specific for each of the mRNAs to be exam- quantitative methods to ascertain the relative functional activities ined (cd59a,5Ј-FAM-CATGGTGAGATCATTATGGACCAATTA and distribution patterns at mRNA and protein level of CD59a and GAAGAGACAA-TAMRA-3Ј, and CD59b, 5Ј-FAM-TAATTCCAAC Ј CD59b. TATATTATGAGCCGATTAGACGTGGCA-TAMRA-3 ). Primers and probe were designed to ␤-actin as an internal control for normalization of starting cDNA levels (ACGGCCAGGTCATCACTATTG, forward; Materials and Methods AGTTTCATGGATGCCACAGGAT, reverse; TaqMan probe, 5Ј-VIC Generation of novel mAbs against mouse CD59b TCCGATGCCCTGAGGCTCTTTTCC-TAMRA-3Ј). Quantitative PCR was performed using TaqMan Universal PCR Master Mix according to the CD59aϪ/Ϫ mice were immunized with a fusion protein, comprising murine manufacturer’s instructions (Applied Biosystems), with the exception that (m)CD59b attached to the Fc portion of human IgG4, following a standard 25-␮l reaction volumes were used, with 45 cycles of amplification. The procedure (23). CD59aϪ/Ϫ mice were chosen for immunization because of concentrations of each of the primer pairs and the TaqMan probes were the known high homology between CD59a and CD59b (63% identity at the optimized to ensure amplification of the specific product and the absence of level) that would in wt mice have restricted the immune re- primer/probe dimers. PCR was performed on the ABI PRISM 7000 (Ap- sponse to CD59b (20). The spleen was removed from an immunized plied Biosystems) using the following primer concentrations: 900(for- mouse, and cells were harvested and fused with SP2/0-Ag14 myeloma ward)/300(reverse) nM for the cd59a and cd59b primer pair; 300(forward)/ cells. The hybrid clones were cultured and the supernatants screened for 300(reverse) nM for the ␤-actin primers. Each of the TaqMan probes was anti-CD59b Abs in an ELISA as described previously (23). Briefly, alter- used at 200 nM. The real-time PCR results were analyzed using the se- nate rows of 96-well plates were coated with 100 ␮l of either CD59b-Fc, quence detection system software version 1.9 (Applied Biosystems). RNA or a human MCP-Fc as a control for the IgG4 Fc portion. After blocking, expression levels were calculated using the comparative Ct method (⌬⌬Ct) culture supernatant from the hybridoma well was added to one of each (30). ⌬⌬Ct validation experiments showed similar amplification efficiency coated well type and incubated. Plates were washed and incubated with for all templates used (difference between line slopes for all templates HRP-conjugated goat anti-mouse Ig (Bio-Rad, Hertfordshire, U.K.). Sub- Ͻ0.1). Expression levels of the mRNAs were normalized to those in testis. sequently, the plates were washed and developed with substrate (orthophe- At least two independent experiments were performed for each mRNA. 3686 CD59a IS THE PRIMARY REGULATOR OF MEMBRANE ATTACK IN MOUSE

Flow cytometric quantitation of CD59a and CD59b expression lized fusion proteins. Ab-sensitized erythrocytes were washed into CFD on mouse erythrocytes and incubated with C8-depleted human serum (1:20 final dilution) for 20 min at 37°C. The EAC5b-7 cells were washed into PBS/10 mM EDTA, mAb against mouse CD59a (mCD59.4) (23) and CD59b were labeled with and incubated with or without function-blocking mAb (10 ␮g/ml) against PE molecules at a molar ratio of 1:1 using the Phycolink PE conjugation kit CD59a and/or CD59b on ice for 10 min. To complete the lytic pathway, rat (Prozyme, Cambridge, U.K.) as instructed by the manufacturer. Erythro- or mouse serum, at an appropriate dilution in PBS/10 mM EDTA, was Ϫ Ϫ cytes from wt and CD59a / mice were stained for CD59a and CD59b added to the cells and incubated at 37°C for 30 min. Zero and 100% lysis with the PE-labeled Abs at a dilution determined by titration. QuantiBRITE controls were included in all assays and percentage lysis was calculated as Beads (BD Biosciences, Oxford, U.K.) were used to determine the number described earlier (15). of PE molecules bound per cell (31, 32). These beads, conjugated with four different levels of PE molecules (863, 8612, 31,779, or 66,408 PE mole- Results cules per bead; lot no. 68094), were used to calibrate the FL-2 axis in terms of number of PE molecules. The beads were acquired in the flow cytometer Generation and characterization of mAb against mouse CD59b (FACSCalibur; BD Biosciences) with setting adjusted for mouse erythro- To generate new mAb against mouse CD59b, CD59aϪ/Ϫ mice cytes under the quantitation acquisition document within the CellQuest folder. The CellQuest software (BD Biosciences) was used to perform the were immunized with CD59b-Fc and hybridomas screened first, regression analysis and to display the slope (m), intercept (c), and corre- for binding to CD59b, second, for lack of binding to CD59a, and 2 ϭ ϩ lation coefficient (r ) for the equation y mx c, where y equals log10 third, for function-blocking capacity for CD59b. A strong immune fluorescence and x is equal to PE molecules per bead. Optimally stained response against the CD59b component of the fusion protein was erythrocytes were acquired using the same machine settings. The number obtained in all immunized mice. A single mAb of the IgG1 iso- of molecules of CD59a and CD59b, expressed as Ab bound per cell, was determined by substituting the log FL2 geometric means in the above equa- type, designated mCD59b.2, was selected based upon these pa-

tion and solving for x. rameters for further analysis. Flow cytometric analysis on EL4 Downloaded from cells transfected with CD59a or CD59b or the empty vector Analysis of expression of CD59a and CD59b on mouse showed the mAb to be specific for CD59b (Fig. 1A). In Western erythrocytes and testis lysates by Western blotting and blot analysis, the mAb specifically detected a broad band of mo- immunoprecipitation lecular mass 17–19 kDa in lysates of cells transfected with CD59b Lysates of mouse erythrocyte ghosts and EL4 cells, either untransfected or alone but was negative in lysates of vector control or CD59a- transfected with CD59a or CD59b were prepared as described earlier (23).

transfected EL4 (Fig. 1B). Preincubation of Ab-sensitized CD59b- http://www.jimmunol.org/ Lysates were resolved on 12.5% SDS-PAGE under nonreducing condi- tions, and then transferred onto nitrocellulose. The membrane was blocked transfected EL4 cells with the mAb increased lysis by rat and with 5% skimmed milk powder in PBS, probed with either anti-CD59a mouse C by 25–50%, demonstrating that the mAb had function- (mCD59.4) (23), or anti-CD59b mAb at an appropriate dilution, washed, blocking activity (data not shown). then incubated with HRP-conjugated rabbit anti-mouse IgG (Bio-Rad) at an appropriate dilution. The membrane was washed and immunoreactive Tissue expression of CD59b protein is restricted to mouse testis proteins were visualized by ECL (Pierce). For analysis by immunoprecipi- tation, mouse erythrocytes ghosts were generated from 2 ml of packed Acetone-fixed frozen sections of the principle organs were stained erythrocytes, washed by centrifugation, and then solubilized in lysis buffer as described in Materials and Methods. Directly FITC-labeled (50 mM Tris-HCl, pH8. 0, 150 mM NaCl, 25 mM NaF, 1 mM Na3VO4, mAb mCD59b.2 was used to detect CD59b. The mAb (mCD59.1) ␮

0.5% Nonidet P-40) containing protease inhibitor mixture (10 l/ml) (Sig- by guest on September 27, 2021 ma-Aldrich). The lysate was precleared by incubation with 30 ␮l of protein A-Sepharose beads (Amersham Pharmacia, Bucks, U.K.) for1hat4°C with mixing. Immunoprecipitation was performed by mixing lysate with 10 ␮g of mouse mAbs against CD59b and 50 ␮l of protein A-Sepharose beads for2hat4°C. The immunoprecipitates were washed twice in lysis buffer, then boiled for 5 min in 50 ␮l of nonreducing SDS-PAGE sample buffer. Beads were removed by centrifugation and the supernatants resolved on SDS-PAGE and blotted as described above. Lysates obtained from whole mouse testis and CD59b-transfected EL4 cells were also immunoprecipi- tated and analyzed as positive controls. To examine the extent of N-glycosylation of CD59a and CD59b, ali- quots of testis lysate (10 ␮l) were incubated with or without N-glycosidase (0.5 U; Roche, Basel, Switzerland) at 37°C overnight. Samples were then separated on SDS-PAGE, Western blotted, and probed with mAb mCD59b.2 or mCD59.4. Functional assay for CD59a and CD59b on mouse erythrocytes wt and CD59aϪ/Ϫ mice were exsanguinated by cardiac puncture and blood collected into EDTA (10 mM). Erythrocytes were separated by centrifu- gation (1300 ϫ g) for 5 min at room temperature and washed twice with PBS. A 1% suspension of erythrocytes was made from packed, washed cells in PBS, and 100-␮l aliquots were placed in the wells of a 96-well plate. Cells were then incubated with or without function-blocking mAb (10 ␮g/ml) against CD59a (mCD59. 3) (23) and/or CD59b on ice for 10 min. Cells were washed into C fixation diluent (CFD; Oxoid, Basingstoke, U.K.) and incubated with rat or mouse serum at an appropriate dilution (0.1 ml in CFD) and cobra venom factor (CVF; 1.5 ␮g/ml final) for 30 min at 37°C. Zero and 100% lysis controls were included in all assays. Plates were centrifuged, and the absorbance of hemoglobin at 412 nm of the superna- tant was measured as an index of lysis. Percentage of lysis was calculated FIGURE 1. Characterization of anti-CD59b mAb. A, Flow cytometric as described earlier (15). analysis of EL4 cells transfected with CD59a (gray line), CD59b (black To examine specifically the effects of the mAb on MAC activity, mouse erythrocytes bearing C5b-7 sites were first generated. Erythrocytes were line), or with empty vector (dashed line), and stained with mAb Ab sensitized by incubating a 1% suspension in PBS with a specific rabbit mCD59b.2. B, Western blots of EL4 cells transfected with CD59a (lane i), anti-mouse erythrocyte antiserum generated in-house, and depleted of anti- CD59b (lane ii), or empty vector (lane iii). Cell lysates were separated by CD59a and anti-CD59b reactivity by adsorption with the relevant fusion SDS-PAGE, blotted to nitrocellulose, and probed with the mAb proteins. Efficiency of adsorption was confirmed by ELISA on immobi- mCD59b.2. The Journal of Immunology 3687

(23) against CD59a was biotinylated to permit double staining for showed a remarkable concentration of staining in a granular mem- the two proteins where appropriate. Staining of testis sections re- branous pattern in the head region (Fig. 2, G and H). vealed strong staining for both CD59a and CD59b in wt mice, although the distribution patterns of the two proteins were mark- Quantitative analysis confirms restricted expression of mRNA edly different (Fig. 2, A and E). In CD59aϪ/Ϫ mice, testis staining for cd59b for CD59b was identical with that in wt mice, whereas CD59a Quantitative PCR was performed to determine precisely the levels staining was absent (Fig. 2, A, B, and F). All the other tissues of mRNA for cd59a and cd59b in the various tissues. The results examined were strongly positive for CD59a in wt mice and neg- are summarized in Table I. Consistent with immunofluorescence ative in CD59aϪ/Ϫ mice, as previously reported (23), but were analyses, cd59a was expressed in all the tissues studied. Expres- negative for CD59b in both wt and CD59aϪ/Ϫ mice (Fig. 2, C and sion, normalized to that in testis, was highest in liver (8-fold that D; data shown for kidney, other negative tissues not shown). In the in testis) and lowest in brain (17% of that in testis). Other tissues testis, CD59a showed a strong staining for Leydig cells between expressed intermediate amounts of mRNA. In contrast, cd59b the seminiferous tubules and a weak diffuse staining at the center mRNA was detected only in testis and bone marrow, with similar of the tubules (Fig. 2E). In contrast, CD59b strongly stained sper- expression levels in wt and CD59aϪ/Ϫ mice. Testis showed the matids, the late precursors of germ cells, and maturing spermato- highest level of expression for cd59b, while expression in bone zoa in the center of seminiferous tubules (Fig. 2, A and B). Staining marrow was low (9% compared with testis). Because identical of isolated spermatozoa showed that CD59a was expressed pre- primers to fully conserved regions in cd59a and cd59b were used dominantly on neck and tail, with essentially no staining on head, in the PCR, it was possible to directly compare expression levels whereas CD59b was diffusely expressed in the neck and tail, and of the two mRNA species. In testis, expression of cd59b was 23% Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 2. Immunofluorescence analysis of CD59a and CD59b expression in wt and CD59aϪ/Ϫ mice. Tis- sue sections of testis (A and B) and kidney (C and D) from wt mice (A and C) and CD59aϪ/Ϫ mice (B and D) stained for CD59b. In the testis, CD59b was expressed close to the lumen of the seminiferous tubules in both wt and CD59aϪ/Ϫ mice. In the kidney, shown as an exam- ple of peripheral organs, listed in text, CD59b was not expressed in either wt or CD59aϪ/Ϫ mice. E, In testis from wt mouse, CD59a staining was more diffuse throughout the seminiferous tubule and strong on sup- port cells between the seminiferous tubules. F, No stain- ing for CD59a was observed in CD59aϪ/Ϫ testis. Smears of acetone-fixed spermatozoa from wt mouse were stained for CD59b (G) and CD59a (H). CD59a and CD59b were diffusely expressed on all regions of the spermatozoa, but strong granular staining for CD59b was restricted to the head region. Magnification, A–F; ϫ400, G and H; ϫ1000. 3688 CD59a IS THE PRIMARY REGULATOR OF MEMBRANE ATTACK IN MOUSE that of cd59a, and in bone marrow, cd59b expression was just 1.7% that of cd59a.

Quantitative analysis of expression of CD59a and CD59b on mouse erythrocytes The above data implied that CD59b might be expressed on bone marrow-derived cells. Preliminary flow cytometric analyses using conventionally labeled mAb detected no staining on leukocytes or platelets. However, a small but consistent shift was observed on erythrocytes (data not shown). This prompted us to examine quan- titatively the expression of CD59a and CD59b on erythrocytes using mAb labeled with PE at a defined molar ratio and Quan- tiBRITE PE beads. Representative histograms from wt and CD59aϪ/Ϫ erythrocytes are shown in Fig. 3. Table II lists the cal- culated numbers of molecules of CD59a and CD59b on erythro- cytes from wt and CD59aϪ/Ϫ mice. Male and female wt mice expressed similar numbers of CD59a molecules per erythrocyte (2286 Ϯ 127 and 2467 Ϯ 135, respectively), while CD59aϪ/Ϫ erythrocytes were completely negative. wt mice expressed low Downloaded from numbers of CD59b molecules on erythrocytes (177 Ϯ 10 male, and 184 Ϯ 11 female), and similar numbers were expressed on CD59aϪ/Ϫ erythrocytes (171 Ϯ 8 male, and 180 Ϯ 7 female).

Enrichment by immunoprecipitation detects CD59b on mouse

erythrocytes http://www.jimmunol.org/ To confirm the low level expression of CD59b detected on mouse erythrocytes by quantitative flow cytometry, we undertook West- ern blot analyses. Under standard conditions, CD59a and CD59b were strongly stained in lysates of EL4 cells transfected with the corresponding vector and CD59a was readily detected in mouse erythrocyte ghosts (Fig. 4A). However, CD59b was not detected in erythrocyte ghosts under standard conditions (Fig. 4B). Therefore, we enriched for CD59b by immunoprecipitation from a large FIGURE 3. Representative histograms for the quantitation of CD59a by guest on September 27, 2021 quantity of erythrocyte ghosts (from 2 ml of packed erythrocytes) and CD59b. Erythrocytes from wt (A) and CD59aϪ/Ϫ mice (B) were ana- and analyzed the lysate in Western blots. Using this enrichment lyzed as described in Materials and Methods. Mouse erythrocyte staining strategy, we identified a band of ϳ16 kDa in mouse erythrocyte with directly PE-conjugated mAb to CD59a (solid black line), CD59b ghost. Testis lysate was run as a positive control and gave a strong (dashed line), and isotype-matched control Ab (gray line). QuantiBRITE band of molecular mass, ϳ18 kDa (Fig. 4C). N-deglycosylation of beads with four different levels of PE molecules, detailed in text, are shown in filled profiles. testis lysate was performed to examine molecular mass changes in CD59a and CD59b (Fig. 4D). N-deglycosylation reduced the ap- parent molecular mass of CD59a by 4–5 kDa and that of CD59b by 2 kDa, suggesting that the latter molecule contained substan- blockade of CD59a and/or CD59b using specific mAb. The first tially less N-linked carbohydrate. assay used CVF to trigger fluid-phase C activation causing by- stander lysis of surrounding erythrocytes (CVF reactive lysis). In CD59b is irrelevant to complement resistance in mouse this assay, both mouse serum (Fig. 5A) and rat serum (Fig. 5B) erythrocytes caused lysis of mouse erythrocytes that was significantly greater To examine the relative roles of CD59a and CD59b in protecting for CD59aϪ/Ϫ erythrocytes compared with wt. Blocking CD59a mouse erythrocytes from complement-mediated lysis, two differ- on wt erythrocytes significantly increased lysis to levels approach- ent hemolytic assays were used in association with functional ing those on CD59aϪ/Ϫ cells without mAb. Anti-CD59a had no

Table I. mRNA levels of cd59a and cd59b by quantitative RT-PCR

wt Mouse CD59aϪ/Ϫ Mouse cd59b Compared with cd59b in CD59aϪ/Ϫ Mouse Compared with Tissues cd59aa cd59ba cd59ba cd59a in wt Mouse cd59b in wt Mouse

Testis 100 Ϯ 3.5 100 Ϯ 3.4 100 Ϯ 3.1 23 Ϯ 2.7 102 Ϯ 3.3 Bone marrow 121 Ϯ 4.8 9 Ϯ 1.2 9 Ϯ 1.4 1.7 Ϯ 0.4 97 Ϯ 4.3 Brain 17 Ϯ 3.2 Lungs 87 Ϯ 4.9 Heart 110 Ϯ 3.7 Liver 800 Ϯ 4.6 Spleen 264 Ϯ 4.1 Kidney 400 Ϯ 3.8

a Data are presented as percent compared to testis. The Journal of Immunology 3689

Table II. Number of molecules of CD59a and CD59b on mouse erythrocytes

No. of CD59a No. of CD59b Molecules on Molecules on Erythrocytes Erythrocytes Mouse n (mean ϩ SD) (mean ϩ SD)

wt male 5 2286 Ϯ 127 177 Ϯ 10a wt female 5 2467 Ϯ 135 184 Ϯ 11a CD59aϪ/Ϫ male 5 171 Ϯ 8 CD59aϪ/Ϫ female 5 180 Ϯ 7

a Statistically significant ( p Ͻ 0.0001) as compared to CD59a.

effect on the lytic susceptibility of CD59aϪ/Ϫ erythrocytes. Block- ing CD59b on wt erythrocytes did not significantly increase lysis, but when CD59b was blocked on CD59aϪ/Ϫ erythrocytes there was a small increase in lysis that was significant. When both

CD59a and CD59b were blocked on wt erythrocytes, lysis was Downloaded from significantly increased compared with blockade of CD59a alone. The second assay used preformed C5b-7 sites to interrogate spe- cifically the terminal pathway, using EDTA mouse serum (Fig. 5C) or EDTA rat serum (Fig. 5D) to complete MAC formation. The results from this assay were supportive of those obtained in FIGURE 4. Detection of erythrocyte and testis CD59b by immunopre- cipitation and Western blotting. A, Lysates from cell-expressing CD59a the CVF-triggered assay described above. Blocking CD59a on wt http://www.jimmunol.org/ erythrocytes increased lysis significantly, while blocking CD59b (lane i) and mouse erythrocyte ghosts (lane ii) were separated by SDS- did not. Blocking CD59b on CD59aϪ/Ϫ erythrocytes increased PAGE, blotted to nitrocellulose, and probed with the mAb mCD59.4. B, Lysates from mouse erythrocyte ghost (lane i) and cell-expressing CD59b lysis only by rat C8/C9 but mAb against CD59a were without (lane ii) were separated by SDS-PAGE, blotted to nitrocellulose, and effect. Blocking both CD59a and CD59b caused the greatest in- probed with the mAb mCD59b.2. C, Lysates from cells expressing CD59a crease in lysis of wt erythrocytes. (lane i), CD59b (lane ii), mouse erythrocyte ghost (lane iii), and mouse testis (lane iv) were immunoprecipitated with anti-CD59b Ab. Immuno- Discussion blots were then probed with anti-CD59b Ab, specifically detecting a band C regulators play important roles in tissue homeostasis in health within the range of 16–20 kDa. D, Lysates from mouse testis were incu- and disease (1–12). The recent availability of mice in which indi- bated with (lanes ii and iv) or without N- glycosidase overnight at 37°C, by guest on September 27, 2021 vidual C regulators have been deleted by homologous recombina- separated on SDS-PAGE, blotted and probed with mAb mCD59b.2 (lanes tion has provided a powerful tool for examining the roles of the C i and ii) or mCD59.4 (lanes iii and iv). Molecular mass markers are indi- cated to the right of each frame. regulators in various organs and contexts, but the analysis has been complicated by differences in C regulation between mouse and man. Mice possess additional regulators absent in man, modified regulators with distinct functions and duplicated regulators with immune responses when immunized with CD59b, from which spe- differential tissue expression (18, 21, 22). CD59 falls into this last cific mAb can be derived. category with gene duplication giving rise to two CD59 proteins, In this study, we have used quantitative analytical methods to both of which have the capacity to regulate C when expressed comprehensively characterize the expression patterns and func- either on cells or as soluble molecules (20, 23). Our original data tional roles of CD59a and CD59b in wt and CD59aϪ/Ϫ mice. We describing the phenotype associated with deletion of the first de- first generated new highly specific mAb against CD59b, selected to scribed form of mouse CD59 (CD59aϪ/Ϫ) strongly implicate exclude any cross-reactivity with CD59a, for function-blocking CD59a as a key homeostatic molecule in the mouse (15). This activity and for IgG1 isotype. These reagents were then used to implication is supported by more recent studies of C-mediated dis- confirm our previous observation, from analyses using a single ease models in the CD59Ϫ/Ϫ mouse, which suggest that CD59a IgM mAb and polyclonal anti-peptide Ab that, of all the solid regulates MAC injury in many tissues (26, 27, 28). Early analyses tissues examined, CD59b protein was expressed at detectable level of the second form of mouse CD59, CD59b, reported expression only in testis (23). In testis, expression was restricted to mature only in testis, ruling out a more global homeostatic role for this spermatozoa and their late precursors, spermatids. Qin, et al. (24), molecule (20, 23). However, Qin, et al. (24, 25) have recently using an anti-peptide Ab targeting a different region in CD59b reported that CD59b is broadly expressed, contradicting their first reported a broad expression closely mirroring that of CD59a in the reports, and have gone on to generate a CD59b gene-deleted tissues tested. We have not attempted to duplicate these studies mouse. The remarkable phenotype in this mouse, with severe with the precise reagent used by Qin, et al. (24, 25), and have no spontaneous hemolysis and platelet activation in vivo and progres- explanation for their findings with this polyclonal antiserum. The sive deterioration of male fertility, implies a much wider role and patterns of expression of CD59a and CD59b on spermatozoa are of indeed has led these authors to conclude that CD59b is the major particular interest. Both CD59a and CD59b were broadly and dif- regulator of the MAC in vivo. This conclusion is impossible to fusely expressed on all regions of the sperm cell, but CD59a was reconcile with our analyses in wt and CD59aϪ/Ϫ mice and those of essentially absent from the head, whereas CD59b was densely ex- others in murine models of PNH. It is also difficult to square with pressed on this region in a markedly granular pattern that appeared our demonstration, in this study and previously, that CD59aϪ/Ϫ to be membrane associated. Spermatozoa were permeabilized by mice (and, in our unpublished observations, wt mice) mount strong fixation in acetone before staining with mAb; it is thus possible 3690 CD59a IS THE PRIMARY REGULATOR OF MEMBRANE ATTACK IN MOUSE Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 5. Roles of CD59a and CD59b in protecting mouse erythrocytes against C lysis. Mouse erythrocytes from wt (Ⅺ) and CD59aϪ/Ϫ (f) were either blocked with Abs against CD59a or CD59b, or left untreated. The cells were then subjected to alternative pathway-mediated lysis with CVF in either mouse serum (A) or rat serum (B). Mouse erythrocytes from wt (Ⅺ) and CD59aϪ/Ϫ (f) were Ab sensitized and coated with c5b-8 sites as detailed in Materials and Methods. Cells were then either blocked with Abs against CD59a or CD59b, or left untreated. The cells were subjected to terminal p Ͻ ,ءء ;p Ͻ 0.001 vs unblocked ,ء .pathway-mediated lysis in EDTA-containing either mouse serum (C) or rat serum (D). Data shown are mean Ϯ SD .p Ͻ 0.03 vs CD59a blocked ,ءءء ;vs CD59a blocked 0.002

that the granular staining observed represents intercellular rather levels of cd59a and cd59b in mouse tissues. Expression of cd59a than membranous elements. Further analyses using unfixed and was abundant in all tissues tested, the highest expression being acrosome-reacted spermatozoa are needed to further characterize found in liver and lowest in brain. In contrast, cd59b transcript was this intriguing distribution pattern. detected only in testis and, at very low level, bone marrow. The Despite numerous attempts using multiple primer pairs we have expression levels in wt and CD59aϪ/Ϫ mice were identical. Be- consistently failed to PCR amplify mRNA encoding cd59b from cause the primers used in the PCR were identical, it was possible any mouse tissue with the sole exception of testis, positive in all directly to compare expression levels of cd59a and cd59b in tis- attempts (our unpublished data). Qin, et al. (24, 25) originally de- sues. The ratio of transcript expression in testis was 4:1 and in scribed similar results but later reported the presence of cd59b bone marrow 60:1 (cd59a:cd59b). mRNA in a wide range of tissues using PCR and Northern blot The mCD59b.2 mAb consistently showed a small shift of flu- analysis. However, our analysis of primers for cd59b these authors orescence on mouse erythrocytes, not seen with the previously have applied in their assay showed 75% and 100% homology with described IgM mAb mCD59b.1 (23). Comparison of staining of cd59a for the forward primer in the two studies. In this study, we CD59b-expressing EL4 demonstrated that mCD59b.2 gave fluo- designed a primer pair complementary to sequences conserved in rescence intensities almost 10-fold greater than mCD59.1, reflect- the two mRNA species that, with specific labeled TaqMan probes, ing differences in isotype (IgM vs IgG1) and affinity. We further could be used in real-time PCR to measure the relative expression explored this weak but consistent signal using quantitative flow The Journal of Immunology 3691 cytometry and were able to show that mouse erythrocytes did in- in C regulation on erythrocytes, it is unlikely that even a complete deed express CD59b, albeit at very low copy number; fewer than absence of CD59a would fully explain the observed phenotype in 200 molecules per cell. In contrast, CD59a was present at around the CD59bϪ/Ϫ mice. Nevertheless, resolution of the current con- 2500 copies per cell. Erythrocyte expression of CD59b was the fusion should enable the available mice and reagents to be used same in CD59aϪ/Ϫ and wt mice and did not differ between males logically to address the roles of CD59a and CD59b in tissue ho- and females. Expression was confirmed by Western blotting, but to meostasis, pathology, and reproduction. obtain a positive result for erythrocyte CD59b, enrichment by im- munoprecipitation was essential. CD59b ran on SDS-PAGE with Acknowledgments an apparent molecular mass of 16 kDa, some 4 kDa smaller than We thank Dr. Tim R. Hughes and Ruth Davies for the help provided in this the apparent molecular mass of CD59a. Although CD59b is larger study, and Prof. Marino Botto (Royal Postgraduate Medical School, Lon- by 11 aa than CD59a, one of the two potential N-glycosylation don, U.K.) for collaboration to generate the CD59a knockout mice. sites in the CD59a sequence (at N71) is lost in CD59b, perhaps explaining the lower apparent molecular mass. Enzymatic N-deg- References lycosylation of CD59a and CD59b in testis extracts supported this 1. Meri, S., B. P. Morgan, A. Davies, R. H. Daniels, M. G. Olavesen, H. Waldmann, interpretation. Testis CD59a had a molecular mass of 19 kDa, and P. J. Lachmann. 1990. Human protectin (CD59), an 18,000–20,000 MW complement lysis restricting factor, inhibits C5b-8 catalysed of C9 into reduced to 15 kDa by N-deglycosylation, whereas CD59b had a lipid bilayers. Immunology 71:1. molecular mass of 18 kDa, reduced to 16 kDa by N-deglycosyla- 2. Meri, S., H. Waldmann, and P. J. Lachmann. 1991. Distribution of protectin tion. Finally, we analyzed the functional relevance of the low level (CD59), a complement membrane attack inhibitor, in normal human tissues. Lab. Invest. 65:5. of CD59b expressed on erythrocytes. Qin, et al. (24) reported that 3. Davies, A., and P. J. Lachmann. 1993. Membrane defence against complement Downloaded from CD59b was ϳ6-fold more active than CD59a in protecting against lysis: the structure and biological properties of CD59. Immunol. Res. 12:258. lysis by human C but did not test activity against rodent C. Our 4. Morgan, B. P., and C. L. Harris. 1999. Complement Regulatory Proteins. Aca- demic Press, London. own published analyses indicate that CD59a and CD59b are ap- 5. Rosse, W. F. 1992. Paroxysmal nocturnal hemoglobinuria. Curr. Top. Microbiol. proximately equipotent as inhibitors, regardless of the source of C Immunol. 178:163. (23). In the current study, two different hemolytic assays and two 6. Telen, M. J., and A. M. Green. 1989. The Inab phenotype: characterization of the membrane protein and complement regulatory defect. Blood 74:437. sources of C were used, and in each case, while mAb blockade of 7. Merry, A. H., V. I. Rawlinson, M. Uchikawa, M. R. Daha, and R. B. Sim. 1989. http://www.jimmunol.org/ CD59a markedly enhanced lytic susceptibility in wt erythrocytes, Studies on the sensitivity to complement-mediated lysis of erythrocytes (Inab phenotype) with a deficiency of DAF (decay accelerating factor). blockade of CD59b had no effect. However, when CD59b was Br. J. Haematol. 73:248. Ϫ/Ϫ blocked in CD59a erythrocytes, lysis was significantly en- 8. Reid, M. E., G. Mallinson, R. B. Sim, J. Poole, V. Pausch, A. H. Merry, hanced. These data, supported by results of double blockade with Y. W. Liew, and M. J. Tanner. 1991. Biochemical studies on red blood cells from a patient with the Inab phenotype (decay-accelerating factor deficiency). Blood mAb in wt cells, demonstrate clearly that the small numbers of 78:3291. CD59b molecules expressed on erythrocytes play no significant 9. Lublin, D. M., G. Mallinson, J. Poole, M. E. Reid, E. S. Thompson, role in the protection of these cells from C lysis in the presence of B. R. Ferdman, M. J. Telen, D. J. Anstee, and M. J. Tanner. 1994. Molecular basis of reduced or absent expression of decay-accelerating factor in Cromer blood CD59a. Only when the numerically dominant MAC regulator is group . Blood 84:1276. eliminated by gene deletion or by mAb neutralization is a protec- 10. Okada, N., R. Harada, R. Taguchi, and H. Okada. 1989. Complete deficiency of by guest on September 27, 2021 tive role for CD59b revealed. 20 KDa homologous restriction factor (HRF20) and restoration with purified HRF20. Biochem. Biophys. Res. Commun. 164:468. The overall conclusion from this work and our previous studies 11. Yamashina, M., E. Ueda, T. Kinoshita, T. Takami, A. Ojima, H. Ono, H. Tanaka, is that CD59a is the physiologically relevant regulator of the MAC N. Kondo, T. Orii, N. Okada, et al. 1990. Inherited complete deficiency of 20- kilodalton homologous restriction factor (CD59) as a cause of paroxysmal noc- in the majority of tissues, while expression of CD59b is restricted turnal hemoglobinuria. N. Engl. J. Med. 323:1184. and likely only of relevance in the male reproductive system. This 12. Acosta, J., J. Hettinga, R. Fluckiger, N. Krumrei, A. Goldfine, L. Angarita, and conclusion fits with our own observations of spontaneous pheno- J. Halperin. 2000. Molecular basis for a link between complement and the vas- Ϫ/Ϫ cular complications of diabetes. Proc. Natl. Acad. Sci. USA 97:5450. type and disease enhancement in the CD59a mouse, and with 13. Powell, M. B., K. J. Marchbank, N. K. Rushmere, C. W. van den Berg, and the findings of others in murine models of PNH (15–17, 26–28). B. P. Morgan. 1997. Molecular cloning, chromosomal localization, expression, However, they are at odds with the published description of the and functional characterization of the mouse analogue of human CD59. J. Im- Ϫ/Ϫ munol. 159:1692. phenotype in the CD59b mouse (25). One possible explanation 14. Rushmere, N. K., C. W. Van Den Berg, and B. P. Morgan. 2000. Production and is that expression of CD59a is also compromised in these mice; the functional characterization of a soluble recombinant form of mouse CD59. Im- combined deficiency of CD59a and CD59b would be predicted munology 99:326. 15. Holt, D. S., M. Botto, A. E. Bygrave, S. M. Hanna, M. J. Walport, and from our data to cause a hemolytic phenotype more severe than B. P. Morgan. 2001. Targeted deletion of the CD59 gene causes spontaneous that seen in the CD59aϪ/Ϫ mice. The two genes are closely linked intravascular hemolysis and hemoglobinuria. Blood 98:442. 16. Tremml, G., C. Dominguez, V. Rosti, Z. Zhang, P. P. Pandolfi, P. Keller, and on 2, region E3, separated by 11.6 kb of genomic M. Bessler. 1999. Increased sensitivity to complement and a decreased red blood DNA. There are numerous examples in the literature where tar- cell life span in mice mosaic for a nonfunctional Piga gene. Blood 94:2945. geting by homologous recombination of a specific gene has led to 17. Murakami, Y., T. Kinoshita, Y. Maeda, T. Nakano, H. Kosaka, and J. Takeda. 1999. Different roles of glycosylphosphatidylinositol in various hematopoietic deletion or down-regulation of a closely linked gene. In the C field, cells as revealed by a mouse model of paroxysmal nocturnal hemoglobinuria. targeted disruption of the factor B gene caused marked down- Blood 94:2963. regulation of expression of the two flanking genes in the murine 18. 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 MHC encoding C2 and D7H6S45 (33). Down-regulation of C2 specific mechanisms of both human decay-accelerating factor and membrane was essentially complete and the resultant mice were effectively cofactor protein. J. Exp. Med. 181:151. double knockouts, lacking activity in classical (C2 deficiency) and 19. Cochrane, C. G., and D. Koffler. 1973. Immune complex disease in experimental animals and man. Adv. Immunol. 16:185. alternative (factor B deficiency) C pathways. Qin, et al. (25) com- 20. Qian, Y. M., X. Qin, T. Miwa, X. Sun, J. A. Halperin, and W. C. Song. 2000. pared CD59a expression in wt and CD59bϪ/Ϫ mice by flow cy- Identification and functional characterization of a new gene encoding the mouse terminal complement inhibitor CD59. J. Immunol. 165:2528. tometry using an undefined polyclonal antiserum and found no 21. Song, W. C., C. Deng, K. Raszmann, R. Moore, R. Newbold, J. A. McLachlan, difference. It is now essential that expression be further tested us- and M. Negishi. 1996. Mouse decay-accelerating factor: selective and tissue- ing the available well-characterized mAb against CD59a to elim- specific induction by estrogen of the gene encoding the glycosylphosphatidyli- nositol-anchored form. J. Immunol. 157:4166. inate the possibility that CD59a has been substantially down-reg- 22. Garnier, G., A. Circolo, X. U. Yuanyuan, and J. E. Volanakis. 2003. Complement ulated. In of our demonstration that CD59b plays a minor role C1r and C1s genes are duplicated in the mouse: differential expression generates 3692 CD59a IS THE PRIMARY REGULATOR OF MEMBRANE ATTACK IN MOUSE

alternative isomorphs in the liver and in the male reproductive system. Biochem. J. 28. Yamada, K., T. Miwa, J. Liu, M. Nangaku, and W. C. Song. Critical protection 371:631. from renal ischemia reperfusion injury by CD55 and CD59. J. Immunol. 172: 23. Harris, C. L., S. M. Hanna, M. Mizuno, D. S. Holt, K. J. Marchbank, and 3869. B. P. Morgan. 2003. Characterization of the mouse analogues of CD59 using 29. Williams, A. S., M. Mizuno, P. J. Richards, D. S. Holt, and B. P. Morgan. novel monoclonal antibodies: tissue distribution and functional comparison. Im- Deletion of the gene encoding CD59a in mice increases disease severity in a munology 109:11. murine model of rheumatoid arthritis. Arthritis Rheum. In press. 24. Qin, X., T. Miwa, H. Aktas, M. Gao, C. Lee, Y. M. Qian, C. C. Morton, A. Shahsafaei, 30. 1997. Relative quantification of . In ABI Prism 7700 Sequence W. C. Song, and J. A. Halperin. 2001. Genomic structure, functional comparison, and Detection System Livak Applied Biosystems User Bulletin #2 (part number tissue distribution of mouse Cd59a and Cd59b. Mamm. Genome 12:582. 4303859). 25. Qin, X., N. Krumrei, L. Grubissich, M. Dobarro, H. Aktas, G. Perez, and J. A. Halperin. 2003. Deficiency of the mouse complement regulatory protein 31. Iyer, S. B., J. E. Bishop, B. Abrams, V. C. Maino, A. J. Ward, T. P. Christian, and mCd59b results in spontaneous hemolytic anemia with platelet activation and K. A. Davis. 1997. QuantiBRITE: A New Standard for Fluorescence Quantita- progressive male infertility. Immunity 18:217. tion. Becton Dickinson Immunocytometry Systems, San Jose, CA. 26. Turnberg, D., M. Botto, J. Warren, B. P. Morgan, M. J. Walport, and H. T. Cook. 32. Davies, K. A., and J. E. Bishop. Determination of the number of fluorescent 2003. CD59a deficiency exacerbates accelerated nephrotoxic nephritis in mice. molecules on calibration beads for flow cytometry. U.S. Patent 5,620,842. April J. Am. Soc. Nephrol. 14:2271. 15, 1997. 27. Mead, R. J., J. W. Neal, M. R. Griffiths, C. Linington, M. Botto, H. Lassmann, 33. Taylor, P. R., J. T. Nash, E. Theodoridis, A. E. Bygrave, M. J. Walport, M. Botto. and B. P. Morgan. 2004. Deficiency of the complement regulator CD59a en- 1998. A targeted disruption of the murine gene resulting in hances disease severity, demyelination and axonal injury in murine acute exper- loss of expression of three genes in close proximity, factor B, C2, and D17H6S45. imental allergic encephalomyelitis. Lab. Invest. 84:21. J. Biol. Chem. 273:1699. Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021