MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient...

11
Molecular Basis of Complement C3 Deficiency in Guinea Pigs Harvey S. Auerbach, Reinhard Burger,* Alister Dodds,* and Harvey R. Colten Edward Mallinckrodt Department ofPediatrics, and tThe Department of Genetics, Washington University School ofMedicine, St. Louis, Missouri 63110; *The Department ofImmunology, Robert Koch Institut, Berlin, Federal Republic of Germany Abstract In experiments to ascertain the biochemical basis of a geneti- cally determined deficiency of the third component of comple- ment (C3) in guinea pigs, we found that C3-deficient liver and peritoneal macrophages contain C3 messenger RNA of normal size (- 5 kb) and amounts, that this mRNA programs synthe- sis of pro-C3 in oocytes primed with liver RNA and in primary macrophage cultures. In each instance, heterodimeric native C3 protein was secreted with normal kinetics but the C3 pro- tein product of the deficient cells failed to undergo autolytic cleavage and was unusually susceptible to proteolysis. These data and a selective failure of C3 in plasma of deficient animals to incorporate 1"4Cjmethylamine suggested either a mutation in primary structure of the C3 protein or a selective defect in co- or postsynthetic processing affecting the thiolester bridge, a structure important for C3 function. A mutation in the primary structure of C3 was ruled out by comparison of direct sequence analysis of C3 cDNA generated from two C3 deficient and two C3 sufficient guinea pig liver libraries. Three base pair differ- ences, none resulting in derived amino acid sequence differ- ences were identified. Finally, restriction fragment length polymorphisms were identified in the C3 gene that are inde- pendent of the deficiency phenotype. This marker of the C3 gene permits testing of these hypotheses using molecular bio- logical and classical genetic methods. (J. Clin. Invest. 1990. 86:96-106.) Key words: immunodeficiency * complement com- ponent C3 Introduction The importance of the complement (C)' system in host de- fenses and immunopathology has been fully appreciated in the study of congenital and acquired complement deficiencies. The third component of complement (C3) in particular plays a pivotal role in many facets of the inflammatory response. Alper et al. (1, 2) first described homozygous deficiency of C3 in a child with a history of recurrent infections. The principal Address reprint requests to Dr. Auerbach, Department of Pediatrics, Washington University School of Medicine, 400 S. Kingshighway Blvd., St. Louis, MO 63110. Received for publication 16 June 1989 and in revised form 13 March 1990. 1. Abbreviations used in this paper: C, complement; C3, third compo- nent of complement; C3D, C3 deficient; DOC, sodium deoxycholate; PCR, polymerase chain reaction. pathogens affecting this child, as in most C3-deficient individ- uals (3-6), were encapsulated bacteria, thus helping to estab- lish in vivo the importance of C3 in opsonization, phagocyto- sis, and intracellular killing of these organisms. However, sub- sequent reports of C3-deficient individuals have presented conflicting data on defects in other C3-dependent functions such as immune adherence, development of leukocytosis in response to infection, vasopermeability, and antibody re- sponse (1-1 1). One possible explanation for these conflicting data is that, in C3 deficient individuals, C3 is synthesized and secreted, albeit at markedly reduced rates. Some C3 functions might therefore be retained, especially in extravascular sites. This suggestion is supported by observations that monocytes isolated from individuals with < 0.0 1% normal serum levels of C3, produced C3 at a rate - 25% of normal (12). Thus, al- though the hepatocyte is the primary source of plasma C3, studies of C3 biosynthesis in monocytes, fibroblasts and syno- vial tissue from patients with rheumatoid arthritis (13) sug- gested that regulation of C3 synthesis is independent in hepa- tocytes and other cells in extrahepatic sites. Ethical consider- ations and other technical constraints, however, did not permit further exploration of these questions in C3-deficient humans. An isolated genetically determined deficiency of C3 in the guinea pig has recently been described (14). These animals were derived from an inbred strain 2 colony. Sera from C3 deficient (C3D) guinea pigs contain 5% of normal C3 as assessed functionally and antigenically, i.e., the C3 appeared to have approximately normal specific hemolytic activity. Pre- liminary studies (14) suggested that peritoneal macrophages and hepatocytes from the C3D animals produced normal amounts of C3 and studies of the clearance of 1251I-radiolabeled normal guinea pig C3 in C3D guinea pigs demonstrated a normal catabolic rate. It is interesting that C3-deficient guinea pigs mounted only a minimal primary antibody response to X X 174, a T cell-dependent antigen, and failed to undergo iso- type switching from IgM to IgG with booster immunizations (15). This inadequate antibody response has also been found in deficiencies of complement components of the classical pathway (16, 17); all are believed to be due to deficiency of activated C3. The availability of a small animal model deficient in C3 prompted further studies of the molecular basis for C3 defi- ciency. In this report we show that cells from C3D animals express C3 mRNA similar in size and amount to that found in normal cells. The C3D mRNA programs synthesis of a C3 protein that fails to undergo autolytic cleavage after heating in SDS, a property associated with the presence of a thiolester bridge in the a chain. In addition, a restriction length poly- morphism of the C3 gene has been identified in C3D and some, but not all, C3-sufficient guinea pig strains, providing a marker for the C3 gene independent of the deficiency pheno- type. 96 Auerbach, Burger, Dodds, and Colten J. Clin. Invest. © The American Society for Clinical Investigation, Inc. 0021-9738/90/07/096/1 1 $2.00 Volume 86, July 1990, 96-106

Transcript of MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient...

Page 1: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

Molecular Basis of Complement C3 Deficiency in Guinea PigsHarvey S. Auerbach, Reinhard Burger,* Alister Dodds,* and Harvey R. ColtenEdward Mallinckrodt Department of Pediatrics, and tThe Department of Genetics, Washington University School of Medicine,St. Louis, Missouri 63110; *The Department of Immunology, Robert Koch Institut, Berlin, Federal Republic of Germany

Abstract

In experiments to ascertain the biochemical basis of a geneti-cally determined deficiency of the third component of comple-ment (C3) in guinea pigs, we found that C3-deficient liver andperitoneal macrophages contain C3 messenger RNAof normalsize (- 5 kb) and amounts, that this mRNAprograms synthe-sis of pro-C3 in oocytes primed with liver RNAand in primarymacrophage cultures. In each instance, heterodimeric nativeC3 protein was secreted with normal kinetics but the C3 pro-tein product of the deficient cells failed to undergo autolyticcleavage and was unusually susceptible to proteolysis. Thesedata and a selective failure of C3 in plasma of deficient animalsto incorporate 1"4Cjmethylamine suggested either a mutation inprimary structure of the C3 protein or a selective defect in co-or postsynthetic processing affecting the thiolester bridge, astructure important for C3 function. A mutation in the primarystructure of C3 was ruled out by comparison of direct sequenceanalysis of C3 cDNAgenerated from two C3 deficient and twoC3 sufficient guinea pig liver libraries. Three base pair differ-ences, none resulting in derived amino acid sequence differ-ences were identified. Finally, restriction fragment lengthpolymorphisms were identified in the C3 gene that are inde-pendent of the deficiency phenotype. This marker of the C3gene permits testing of these hypotheses using molecular bio-logical and classical genetic methods. (J. Clin. Invest. 1990.86:96-106.) Key words: immunodeficiency * complement com-ponent C3

Introduction

The importance of the complement (C)' system in host de-fenses and immunopathology has been fully appreciated in thestudy of congenital and acquired complement deficiencies.The third component of complement (C3) in particular plays apivotal role in many facets of the inflammatory response.Alper et al. (1, 2) first described homozygous deficiency of C3in a child with a history of recurrent infections. The principal

Address reprint requests to Dr. Auerbach, Department of Pediatrics,Washington University School of Medicine, 400 S. KingshighwayBlvd., St. Louis, MO63110.

Received for publication 16 June 1989 and in revised form 13March 1990.

1. Abbreviations used in this paper: C, complement; C3, third compo-nent of complement; C3D, C3 deficient; DOC, sodium deoxycholate;PCR, polymerase chain reaction.

pathogens affecting this child, as in most C3-deficient individ-uals (3-6), were encapsulated bacteria, thus helping to estab-lish in vivo the importance of C3 in opsonization, phagocyto-sis, and intracellular killing of these organisms. However, sub-sequent reports of C3-deficient individuals have presentedconflicting data on defects in other C3-dependent functionssuch as immune adherence, development of leukocytosis inresponse to infection, vasopermeability, and antibody re-sponse (1-1 1). One possible explanation for these conflictingdata is that, in C3 deficient individuals, C3 is synthesized andsecreted, albeit at markedly reduced rates. SomeC3 functionsmight therefore be retained, especially in extravascular sites.This suggestion is supported by observations that monocytesisolated from individuals with < 0.0 1%normal serum levels ofC3, produced C3 at a rate - 25% of normal (12). Thus, al-though the hepatocyte is the primary source of plasma C3,studies of C3 biosynthesis in monocytes, fibroblasts and syno-vial tissue from patients with rheumatoid arthritis (13) sug-gested that regulation of C3 synthesis is independent in hepa-tocytes and other cells in extrahepatic sites. Ethical consider-ations and other technical constraints, however, did notpermit further exploration of these questions in C3-deficienthumans.

An isolated genetically determined deficiency of C3 in theguinea pig has recently been described (14). These animalswere derived from an inbred strain 2 colony. Sera from C3deficient (C3D) guinea pigs contain 5% of normal C3 asassessed functionally and antigenically, i.e., the C3 appeared tohave approximately normal specific hemolytic activity. Pre-liminary studies (14) suggested that peritoneal macrophagesand hepatocytes from the C3D animals produced normalamounts of C3 and studies of the clearance of 1251I-radiolabelednormal guinea pig C3 in C3D guinea pigs demonstrated anormal catabolic rate. It is interesting that C3-deficient guineapigs mounted only a minimal primary antibody response to XX 174, a T cell-dependent antigen, and failed to undergo iso-type switching from IgM to IgG with booster immunizations(15). This inadequate antibody response has also been foundin deficiencies of complement components of the classicalpathway (16, 17); all are believed to be due to deficiency ofactivated C3.

The availability of a small animal model deficient in C3prompted further studies of the molecular basis for C3 defi-ciency. In this report we show that cells from C3D animalsexpress C3 mRNAsimilar in size and amount to that found innormal cells. The C3D mRNAprograms synthesis of a C3protein that fails to undergo autolytic cleavage after heating inSDS, a property associated with the presence of a thiolesterbridge in the a chain. In addition, a restriction length poly-morphism of the C3 gene has been identified in C3D andsome, but not all, C3-sufficient guinea pig strains, providing amarker for the C3 gene independent of the deficiency pheno-type.

96 Auerbach, Burger, Dodds, and Colten

J. Clin. Invest.© The American Society for Clinical Investigation, Inc.0021-9738/90/07/096/1 1 $2.00Volume 86, July 1990, 96-106

Page 2: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

200-

92-

68- _. b

46-

30-NN C3DC4D1 N 03G C4D

IC XC

Figure 1. Biosynthesisof C3 by adherent peri-toneal macrophagesfrom normal (N), C3deficient (C3D), and C4deficient (C4D) guineapigs. Autoradiograph ofreduced SDS-PAGEanalysis of intracellular(IC) pro-C3 and extra-cellular (XC) native C3precipitated from cul-ture media.

68- - -

46-

30-

kD cIiv-,6,-',200 - 11-a

I>aS

", ,"

IC XC ICXCI

Figure 3. Biosynthesisof C4 in normal andC3Dguinea pig perito-neal macrophages.Analysis of intracellular(IC) pro-C4 and extra-cellular (XC) native C4on SDS-PAGE.

Methods

[IC X CXC

N C3D

Figure 2. Biosynthesisof C3 in Xenopus oo-

cytes primed with livermRNAfrom normal(N) or C3Dguinea pigs.Pro-C3 (intracellular)and C3 (extracellular)are analyzed by SDS-PAGEunder reducingconditions.

Cell culture. Peritoneal cells were harvested in HBSSwithout calciumand magnesium from normal Hartley, C3D, and C4D guinea pigs(C4D guinea pig plasma contains C3 at normal concentrations). Thecells were isolated by centrifugation and resuspended in medium 199(Gibco Laboratories, Grand Island, NY) with 50 Upenicillin and 50 Agstreptomycin (Gibco Laboratories) per ml supplemented with 10%heat inactivated (56°C for 2 h) fetal calf serum (Irvine Scientific Co.,Santa Ana, CA) to 4 X 106/ml. 0.5 ml of the cell suspension was placedin 16-mm flat-bottom tissue culture wells (Coming Glass Works,Coming, NY) and the cells were allowed to adhere for 2 h. The mono-layers were extensively washed with HBSS(viability was assessed bytrypan blue exclusion, differential counts by Wright's stain and stainedfor esterase) and incubated for specified time periods at 37°C in Dul-becco's modified essential medium (DMEM; Gibco) lacking methio-nine and containing [35S]methionine (- 1,000 Ci/mmol; ICN Bio-medicals, Irvine, CA). At timed intervals media were collected, clearedof cell debris by centrifugation at 600 g for 15 min, then stored at-70°C. Macrophage cultures were then washed extensively with HBSSand lysed by freeze-thawing in a solution of 50 mMTris-HCI (pH = 7)containing 100 mMKC1, 0.5% Triton X- 100 (Sigma Chemical Co., St.Louis, MO) 0.5% sodium deoxycholate (DOC; Sigma), 2 mMPMSF(Sigma), and 10 mMEDTA.

Immunoprecipitation and analytical gel electrophoresis. Radiola-beled proteins were assayed as previously described (18). Briefly, ali-quots of cell lysate or medium were incubated overnight at 4°C in 1%Triton X-100/1% SDS/0.5% DOCwith excess antibody. A suspensionof formalin fixed Staphylococcus aureus bearing protein A (IgSorb;The Enzyme Center, Boston, MA) was then added to each sample andincubated for 1 h at 4°C. After extensive washing, the antigen-antibodycomplex was released by boiling in sample buffer (0.05 Tris, pH = 6.8,1% SDS, 10% glycerol, 5% 2-mercaptoethanol, and 0.01% bromo-phenol blue), and applied to 9% SDS-PAGEunder reducing condi-tions. After electrophoresis, gels were stained in Coomassie brilliant

Genetic Deficiency of Complement C3 Protein in Guinea Pigs 97

.. . .^.

92-.. .:

.... ..

.......

.. ..

.... .......

..... ......... .i .Xs

.;

*:^:.,ys'(e Q_ | Wl:-3SQg .XX X S:':''';X' l=iJE |

EX V |.$ Gv5rOii !_ S! i:.. S s u N | L| A:N 'e Xg

:X W=a:: s¢L at ac: SX.:

3 k @ei i ?aa 3 jrtg;i >XAsW S EAn X | l S.,#<,

S D-.Si5 >. | .l, W

1. :..,.i

Page 3: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

Normal

kD

200- "92-. /J

92- t

68-

46-

30-

j 0 30 60 120 240_1 10 30(min)

ICC3D

kD200 -

60 120 240i(min)xc

OM92-

Freund's adjuvant. Blood was drawn 5 d after immunization, and IgGantibody was purified by ammonium sulfate precipitation followed byion exchange column chromatography (DE52; Pharmacia Fine Chem-icals, Piscataway, NJ). Antibody specificity was determined by inhibi-tion of complement lysis and ouchterlony analysis.

C3 proteolysis. Media were harvested from normal Hartley or C3deficient guinea pig macrophage cultures that had been incubated with[35S]methionine for 4 h. Aliquots were incubated with appropriateconcentrations of TPCK-treated trypsin (Sigma) for 4-5 min at roomtemperature. The reaction was stopped by addition of DFP(final con-centration 1 M) at 370C for 20 min. Radiolabeled C3 protein was thenisolated and applied to SDS-PAGEas described above. For autolyticcleavage studies, 50-Ml aliquots of media from normal Hartley or C3Dmacrophages previously incubated with [35S]methionine were added to750 ul of 0.1 MTris acetate, pH = 8.5, and 0.1% SDS. After 20 minincubation at 750C, 200 ,d of a stock solution to yield final concentra-tions of 1%Triton X-100, 1.0% SDS, 0.5% DOC, was added. Radiola-beled C3 was then analyzed by immunoprecipitation, SDS-PAGE, andfluorography, as described above. 10 ug of RNAwere subjected toNorthern blot analysis as previously described and probed with 32P-la-beled mouse C3 cDNAgenerously donated by M. Takahashi and M.Nonaka (Kanazawa University, Kanazawa, Japan) or with a 32P-la-beled full length guinea pig C3 cDNAprobe.

Xenopus oocytes were injected with 50 ng of mRNAper oocyte.The oocytes were incubated in microtiter wells (Corning Medical In-struments) containing 40 Ml of Barth's medium/4 oocytes per well at25°C for 6 h to allow puncture healing and recruitment of mRNA;theoocytes were then refed with DMEMcontaining 1 mCi/ml of [35S]-methionine. After 50 h of incubation at 25°C, the media were re-moved. The oocytes were rinsed with PBS and homogenized with aloose fitting Dounce homogenizer (Fisher Scientific, Pittsburgh, PA) in0.5 ml of 15 mMTris, pH = 7.5, 50 mMNaCl, and 2 mMPMSF. Thehomogenates were centrifuged at 100,000 g for 5 min and aliquots thenimmunoprecipitated for C3, as described above.

68-

46-

30-

10 30 60 1202401L0 30 60 120 240j(min) (min)IC xc

Figure 4. Kinetics of C3 secretion from normal and C3Dmacro-phages. Cells were incubated with [35S]methionine-containing mediafor 30 min, then washed and refed with fresh unlabeled media. C3was immunoprecipitated from intracellular and extracellular samplesat indicated time points.

kD

92-__ _N

68- to

46 -

blue, destained, impregnated with 2.5-diphenyloxazole (EN3HANCE;New England Nuclear, Boston, MA), and dried for fluorography onXARfilm (Eastman Kodak Co., Rochester, NY).

Antibodies. Antiserum to guinea pig C4 was raised in C4-deficientguinea pigs (19) by previously described methods (18). Antiserum toguinea pig C3 was raised in albino rabbits by subcutaneous injection of1.0 ml of a mixture containing equal volumes of purified guinea pigC3, 1.0 mg/ml (supplied by M. Thomas, Washington UniversitySchool of Medicine, St. Louis, MO) and complete Freund's adjuvant,followed by two injections at 2-wk intervals of 0.5 mgC3 in incomplete

30-

,N C3D N C3D,Before After

Figure 5. Autolytic cleavage ofC3 synthesized in normal andC3Dmacrophage cultures.SDS-PAGEof C3 immunopre-cipitated from extracellularmedia before and after incuba-tion in 0.1% SDS, 0.1 Mtrisacetate, pH = 8.5, at 75°C for20 min.

98 Auerbach, Burger, Dodds, and Colten

4

Page 4: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

Figure 6. Incorporation of ["4C]-methylamine into guinea pig a2macroglobulin, and the a chain ofC3 and C4. Normal and C3Dguinea pig sera were fractionated ona QSepharose column. Fractionswere incubated with [14C]-methylamine and analyzed by SDS-PAGEand autoradiography. (A)Normal, (B) C3 deficient guinea pigserum. No ['4C]methylamine wasincorporated into the C3 in theC3Dserum fractions (B).

DNA analysis. 1 g liver fragments from C3D, C4D, C2D, andnormal Hartley guinea pigs were digested in 0.5% SDS, 0.5 MEDTA,pH = 8, and 100 Ag/ml proteinase K for at least 6 h at 50'C. DNAwas

extracted from digested solutions by phenol/chloroform, dialyzedagainst 1,000 times the volume of 10 mMTris, pH 8, 1 mMEDTA,and stored at 4VC. Restriction endonuclease digestion of I0-ug aliquotswas performed for more than 18 h, and the samples were subjected toagarose gel electrophoresis, transferred to nitrocellulose paper, andhybridized with a 32P-labeled full length guinea pig C3 cDNAprobe.

Methylamine incorporation in serum C3. A 4 cm X 1 cm QSepha-rose Fast Flow (Pharmacia Fine Chemicals) column was equilibratedwith 10 mMTris, 5 mMEDTA, 50 mMe-aminocaproic acid, 0.2% Naazide, 0.2 mMPMSF, 100 mMNaCl, pH 7.4. 2 ml of normal or C3Dguinea pig serum was applied and eluted with a 20-ml gradient to 500mMNaCl in the same buffer. -ml fractions were collected. For me-

thylamine incorporation, 22.5-,ul aliquots of each fraction were incu-bated with 2.5 of 2 M Tris, pH 8.8, containing 50 mM[14C]-methylamine, 48 mCi/mmol (New England Nuclear) for 30 min at37°C, then were reduced and subjected to gel electrophoresis on a

modified Laemmli gel (20). The gel was stained with Coomassie blue,dried, and exposed to x-ray film at -70°C.

a-

a'- _p

go.ta

.

RNA. Total liver RNAwas isolated from three C3-sufficient (Hart-ley, one animal, C4-deficient subline of NIH multipurpose guinea pigs,two animals) and two C3-deficient guinea pig livers according to pre-

viously described techniques (21). The polyadenylated RNAfractionwas isolated by affinity chromatography on oligo-dT cellulose (22).

Construction of cDNA libraries. The cDNA synthesis system was

purchased from Bethesda Research Laboratories (Bethesda, MD); re-

striction enzymes, T4 DNApolymerase, and T4 DNAligase were pur-

chased from New England Biolabs (Boston, MA); dNTPs were pur-

chased from Pharmacia; S-adenosyl methionine and Eco RI methylasewere purchased from Promega Co. (Madison, WI). Radionuclides[32P]a-dCTP (800 Ci/mmol) and [35S]a-dATP were purchased fromNew England Nuclear; oligo-dT cellulose type 3 was purchased fromCollaborative Research (Bedford, MA); nitrocellulose HAWfilterswere purchased from Millipore Corp. (Bedford, MA); phosphorylatedEco RI linkers, lambda ZAP, XL- I Blue, and Gigapak were purchasedfrom Stratagene (La Jolla, CA); sequenase DNAsequencing kit was

purchased from U. S. Biochemicals (Cleveland, OH); SP6/T7 Tran-scription kit was purchased from Boehringer-Mannheim (Mannheim,West Germany).

The C4- and C3-deficient poly A' was used to direct cDNAsynthe-sis, which was then cloned into lambda ZAP by a modification of themethods described by Okayama and Berg (23, 24). Oligo-dT (12-18)or specific oligonucleotides designed from the guinea pig C3 sequence

were utilized as primers for first strand synthesis in generating theprimer extension libraries from one additional C3-sufficient (C4 defi-cient) and C3-deficient animal.

Isolation of C3 cDNA clones. cDNA libraries were plated at a den-sity of 8,000 plaque-forming units per 100-mm dish. Developed

plaques were transferred to nitrocellulose filters. Filters were dena-tured, neutralized, baked in vacuo at 80°C for 2 h, prehybridized for 2h at 42°C in 6x SSC, 1X Denhardt's solution, 0.5% SDS, 1% dextran

A.

18S-

Figure 7. Trypsin cleavage ofC3 from normal (N) and C3Dguinea pig peritoneal macro-

phages. Native C3 immuno-precipitated before and afterincubation with trypsin. Tryp-sin cleavage generates an a'

N C3D N C3D fragment

no immunoprecipitable C3Before Af ter protein in the C3Dsample.

28s-

B.

Figure 8. Northern blotanalysis of equivalentamounts of RNAiso-lated from normal andC3D liver visualizedunder ethidium stain(A) and autoradiogra-phy (B). The blot was

probed with a fullN C3D length C3 cDNA.

Genetic Deficiency of Complement C3 Protein in Guinea Pigs 99

a2M -- -

__.w C304

NORMAL C3 D

US&46'..

#M.. qmbo,...-

Page 5: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

A. PST 1

23 -9.4 -'6.5-4.3 .-

4.3 -2.0-

0.5-

A B C D E FB. ECO RI

23-94:

6.5- _4.

2.3 -2.0-

Construction ofpolymerase chain reaction (PCR) fragments. Oligo-nucleotides (29") containing guinea pig C3 sequences with 5' restric-tion endonuclease recognition sites were used to prime first strandcDNAsynthesis from C3-deficient and C3-sufficient (Hartley) guineapig liver mRNAfrom additional animals of each type. The RNA-DNAhybrid was amplified in a buffer containing 50 mMKCI, 35 mMTris(pH = 8.3), 2.5 mMMgCl2, bovine serum albumin (100 ,ug/ml), 1.25mNdNTPs, sense and antisense primers (500 ng each) and 5 U Taqpolymerase in a total volume of 100 til. The reactions were carried outfor 40 cycles of denaturation (2 min at 950C), primer annealing (3 minat 61 0C) and primer extension (7 min at 720C), in an automatedthermal cycler (Perkin Elmer-Cetus, Norwalk, CT). The product wassubjected to electrophoresis on a 1% agarose gel (Seaplaque agarose;FMCBioproducts, Rockland, ME), and the appropriate sized frag-ment isolated, phenol extracted and ethanol precipitated. The DNAwas digested with the appropriate restriction enzyme, ligated intobluescript phagemid, and sequenced as above.

Results

C3 biosynthesis. Net biosynthesis of C3 was assessed in normalHartley, C3D, and C4D guinea pig peritoneal cell culturespulsed with [35S]methionine for 4 h. The amount of radiola-beled pro-C3 (intracellular) and native C3 protein (extracellu-lar) produced was quantitatively similar, and the apparent mo-lecular mass of a chain (1 15 kD) and ,B chain (65 kD) wereindistinguishable in cells from the normal and C3D guineapigs (Fig. 1). In this experiment, pro-C3 produced by the C4Dcells was apparently present in reduced amounts but theamount of extracellular native C3 was similar to that producedby the normal and C3D cultures. This decreased amount ofpro-C3 in the C4Dcells was not a reproducible finding since infour other experiments, the amount of pro-C3 synthesized inC4D cells was similar to that produced in normal and C3Dcells. In each culture (normal, C4D, and C3D), the net incor-

0.5 -

A B C D E FFigure 9. Southern blot analysis of Pst I (A) and Eco RI (B) digests ofDNAfrom (A) normal outbred Hartley, (B) strain 2, (C) strain 13,(D) C2-deficient, (E) C3-deficient, and (F) C4-deficient guinea pigs.Additional 2.3 kb (Pst 1) and 1.6 kb (Eco RI) bands are visualized inthe C3Dand C4Dlanes.

sulfate, 100 Ag/ml, and hybridized in the same solution at 42°C using aradiolabeled 32P-labeled mouse C3 cDNAprobe kindly supplied by M.Nonaka and M. Takahashi. Filters were washed at 55°C in 1X SSC,0.1% SDSfor 1 h X 2, and subjected to autoradiography using KodakXAR-5 film. Positive clones were purified by two subsequent rounds ofplating and screening. After purification, candidate C3 clones weredigested with Eco RI and subjected to 1% agarose gel electrophoresis.

cDNA sequencing. 10 gg of cDNA was self-ligated, sheared ran-domly by sonication (six bursts of 40 s), subcloned into Ml 3mp18according to methods described by Messing (25), and subjected todideoxy-sequencing. In other experiments, double-stranded bluescriptphagemids were denatured in 0.2 MNaOH, 1 mMEDTAfor 5 min,ethanol precipitated, and then sequenced as above.

A

B

5' 3'0 1000 2000 3000 4000 5000

base pairs

C3D..._+ _ +5 C4D(NL)

5.1E83T

1630 2789G1 C3D

i.] 3

NL(C4D)

665 1165 1760 2440 3440 4240 5010C .

C60 1235 1690 2500 3373 4273

Figure 10. (A) Sequence strategy for C3D- and C3-sufficient (C4-de-ficient, C4D) C3 cDNAclones. Solid arrows represent individual gelreadings from randomly sheared (shotgun) fragments. Dashed arrowsrepresent additional sequence obtained from oligonucleotide-directedsequencing. (B) Solid bar represents sequence of C3D (above) and C3sufficient (C4D, below) C3 cDNA, with base pair differences identi-fied. Arrow indicates origin of primer extension libraries. (C) PCR-generated fragments derived from C3Dand normal guinea pig livermRNAfor further sequence analysis.

100 Auerbach, Burger, Dodds, and Colten

-rC

Page 6: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

0 0 00 0 0 0 4j00000000 00 00 00 00 00 ON 0~~~~~~~~~~~~~~~~~~ 00 ~~00c 00 V

ON0~~~~NO 00 00 ~~~~~N o~~~ No 000 ONON olv Now 00

o 1- 1-~~~~ d 0 0 0ec ol 4c0 0 0 0 * c0 l0 04 <V 0- C'

0.0 00I I.I >1 0u- -1- 0<0.0) 0< 0< >1- <0 <0W0 co CI - 0 1-0 CY < 0 0 0 <

00U0A0(0 00.0>0 0 0 00C 0>C 0 r.

U U-0< 1 < 0 0 1 0 <0 1- 00

0 0 1 0 0 0 0 .0

0 0A 0 0 (AO0 0 0L0Z 0 Z- <a

*<0 0 < 1- 0 0 0 0 0 0 < 0 I-~~~~ <

LI)- 0 0 0 0 0 0 0I).LI)1- < 0 0~~~~L <L 0L < 0U. ( I 0L0 ) 0>

0 0 0<LI 0 < ( 0 0

I- 0 - 0 a1 0I 0I 0 0I < <

)0 0 > 0 - 0 00 OOL I- I 0 8zO

< 0I 0 0 0 0 ) <O 0 0I 0 < ~0000 0 < 0 0 0 0 ~~~~~~~~~~~~~~~~~~<0

0 - < 0I 0 00> x< <90oCI 0<

0 0LI00 ) 0 1 0 < 0C0D0<0.0< 1-0 < >1 Z ~0 00<0 0< .iI-< W< 0<

0 > < < 0 <> Z< 0 C 0 CL0 0

1 0C0D 0 0 0 0 0 (a 0 1-

0 1- 0~~LI 1- 0 < CD0 0 0 0I I00 0 0 0 ~~~~~~~00 0F 0< CL 1-0 0..db 0 0

P

00 0 0 1-~~~~~~~0 0 0 < < <

0 0 < 0 1- 0LI) 0 0 00D< 0 00 0 1-~~~0 0 1-- 0 0c 00 0< 0 0 0 <

0 0 0~~~~< 0 0 0 0 0 C< 0O0I- 00 0 0 0 0 0 0 0 00~~~~~~~~~~~~LI 0 1 a

00I) 0 0 0Z 0 0 < 0 0 >- 0 0 cc- 0

00 < < 0 1-~~~~~~< 0 < < 0 1- << 0~~~~0 0 0D

001- IL1 1- 0 1 <0 00I) 1-< >..I- 0 -1 0<'< C - 0 (I0 < < <0I)JC.0 U 0 0 < U00LI00 ) 0O

<0U >1 W< IL< 0< 0 . --1-.00 00 >1- -0cc O <.J1- ~00 0o <0.)< 0 0 0 < 0I 0

0D I- 0 0 0 0LI) 1- 0..J- 0.0- ~< >w<--~1 <0 <.1 1-0 0.1-i Z A1- -< C)

0 0 <O < <) 0 0qC 0(1O < < C)>0 0 0 0 ~~~~~~~00o 0.0- < 0L0 1-i.-0< 0. 0 a

-1-00 1-0 0.0 0.0 Z< W< 0 >1Z< < LI-)< 0< v

> 0a<<U.0 0 < c0 1- C0 (A 1- < 1-t 0<

~~.J1-0< <0 0< -1-00~~c >1 -1 < <1 .0Z -1 )- 0< o

0. 0 0 0 0 < 1-~~~~0 C 0 CD 0 a < 1- 00 0 0. 0 0 0o 0 0 0 0- 0 00-0 <

<.1 000--0 <W >1- <.11 1-0 Z<) I)- <0 000 0 0 < 0 < 0 00c< <<L00D 0Z'LI

0 < 0 0 0e 00< c 0o 0 < 0I 0 01<~0j1- 0< -I-~~C <--0 < o.1 1 -1- WLI) 00 ><>10<o

0 0 <~~~1- 0 0 0D 0 < 0 < 1- 1 0 r

0 00X 0 0 0O 0 < 0 0 0 000<<Q*0 0 ~~~~<0I 0 1- 0 0 0C-D < < 0 < >-0 co 0L < 0 < 0 0o 00> 0 0 0 0<

0 W< <I -< 10L- < 1- - 000 )< < O00< 0 <<<o 1- > < < 0- <L 1- 1- 0 0

0' 0 <0 < 0 < 1- 1 1

CL 0 0 0 1- 01--

- 1 -- 0 0 0 0 0o U

0 0 < < 0 1- 1 < < 0 -< - 0o 0D 0 0 0 0 0 0

<0 W< 1-0 0<<10--0 0< >- 0.0 <0 1-0 -1 0. 000 0 ~~~~<0 < 0 0 0 CD< 00 0 0~~~( 0 < 0 0 0 1- U

0> 0 0- C 0 < D 0 I> <L 0Z1 <<

< 0 1- 0- 0q 0 0 1- 0

< 1- 1 0 1- 0C'00s o

(a 0 1 0 0 0 0 0 L0 CO0 0 < (C00 0 CI00 >1<<>---0 . 00<<0 .a1< UW >1- Cd

CL0 0 < 0 0 0 CL0 0 0 0 1< < 0~~~~0 - < < <

0 0 0 0 0 0 c 0 0A 0 < 0CrC r 0 U)0 0 '

< C0 <D 0< < < < O

I0 0 0 000 W< <0 >1 00Z < 0 < >1 1- 0 1- 210 C-1 OWLI

1- 0 0~~ ~0 < 0I 0 O < 0 LI0)n

0 < < < 0~~~~~~~0 < < < 0 < 0 0 <

1-0 < 0 0~~~~~~I 1- 0 <LI) < < < 1-

0L 00> 00co0 0 0i 0L 0. 0 < 0 0-00 .l1 .1- .1- l1-W< < 0 W<u-I 0 -1- I) 0. 0

0 0 01 0 0-O'.L) I) > < 0> 1-00 >1 0< 0. In>1 <0 <J LJ1 0<>

0I 0 0 000 0 00D 0 E0 0 0~~ ~CD 0 8I LI) 0CD00 C'

00 .J1-~~( 1-0iI. 1-0 0< -v1- .J1 <1~ <1 Z< 0< Z< >1< 0Q1- < < 0 < 0 < < 0 <*0 0 0 0 0 0~~~~~~~~LI 0 0 C

<0 -1<00 00 Z< 1 00 00 0< 00<0 u-I- >1- >

0 < < < 0 0 co 0 1- 0 I- 0 1-

LI)0 0 0 0 0 0D LI 0 0 0 00E0.0 Z< w< 00>I (v I)Z w 0<u->0.0 ~-< 00 0 Z<C I1- 0< C

O0 0 O~)1 0 0 < 0 0

0 0 W0 v- - -- I

Genetic Deficiency of Complement C3 Protein in Guinea Pigs 101

0 -.14 0 co w p- %W r-

Page 7: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

00

co0 0 0 400V 0 0 0 00 00c 00

4

co0~ 0C 00 0CtdW c 0~ 0Y 00o 0 0

V0 C4 V ~0)

o~~ *c~ ?%C~0 I%. N 0co 0 01 00co 0 04 Cb 014 V'00) - 0)V (I)C

04 0 0 0 0 0 1- 0 0 0 0:coOU a- -1 -at- 0 1- 00 < < O< 0 1 0 0 <0C0 C.)1

0.) 1- 00C 0 0s 0 0>O~~ea<O >1- W< -1- ~~~~~~0 1- >1 <0

CD

O- L1- ><

0 0

>o. 0 0 100 <00 z1Z< 1-0 1-> -1 00< <0- z0<><0< < < 0< < < <-

0O 0 00 0 0 0<0 O 0 0

a < 6 M CI- Z w< w<0> > C0-0 <0 0< - Z< C<0<.< < < < 0 < 0 0 0<0 < 0oc0 0 0 0 0 0 0 0 00< 0 0

0< 0.6 0. 0< W<CL0 U - <1 0. >1 1- - 00 00

O~C0 0 0 0 00 0 0< 0 0 0

< 0 < 0 < 0<0C< 1< < (C0 < O

<0 <

0<< 0 o00x

0 0

<

1- 0

<

0

O< 0<CD0 ~t0C) -a

.0

< 0i c0 0r 0

n< oC > *Z < IC<0

U- 0-0D 0a 0 < 0 < 0 < < < 10 1- 0 0 0D 0 0 0 0 0 CID0

>1 0D0 21E- Z< --<0 0 < )< >1- 1-0 21~- <0uW<I0 < 0 0a01 0 < < 00

0D < 0 < 0 a 0 < 0D 0. 0. 0

>1 UJ< 00 0< Wi< > I-. <0( C0.0 cc CD W< 00O C- 000 0 0I 0 0 0 0O 0 < 0I 0 0 1

0 ~~~~00 0 0 0D 0 1-0 D0< 00 < 0 < W< 00W< >1 21 00-- 0< 0<

0D 0 0 0 0 1- 00 < 1- < 0. 00o 0 0 < 0 0 0 1 0 < 0

~0 0 < w< Cc0 CC- 0 <0 00 C0< 2 ~- 0 <-11- 0 0 ~~~~~0< < 0 0 < 0 <0 0 0 0 0 00~~C 0 < 0a < 0

--<0 00 -iJ- 100)< >1 Wi< 10 Z < 10 00O< 0D 1-a < 0 -00 < < < 0

00 0 0 < 0 -0 0 1

< C) 0 0 < x < Z < (0 0.0L 0a< 0 < ><-1< < 1 0 1- < < 1- .0 0 0 1- <

CD 0 0 0 0 0 <. 0 0 0

C4O 0 < Cc0 <0C.)-< )< co00 >1L00W< 0 0 -

< 0 < 0C 0 < 11-0000 0 00 0 10 O. < 0 1

0 < -I 0 21~- ~0 0< .1 < >1l- >I... 0 0 cc0 Z <0 0 0 ~~~~~~<F- 0 0 < 0 CD 0 < <1- 0 0 0~~~~~t 0 < 0 00 <1 0 0 0

1-000 .J1- >~~~<C(00 (00 >1 0.0o < 0 <.200 CDZ< < 0< 0 C0. 1 < 1- 0 < C.) 1- 0

00 0 0 ~~~~~~~00 0 < 0 01- 0 m...i1-<0W< 1-0 W< ~~~~~~J1- <0 00~(C J1- < 00 -1 <0C.

C0D < 0 - 0 C0 0I < 0

< 0 0 0 < 0 0 0a 0 0 0~< NC< ~< 00 0.C0 co) ( CC0 -- 10 (00 Ca O J-i J-i

< < a- a < < < 1-0 0

0. 0 0 0 1- 0 0 0D < 0C0)Z < CO0 (00CC 00C U. I.. >1- -1 - 0 x < ..1 00 Z < (00

0 < < t- 1- < < < 1- 0 < 1< 0 0 0. 00 0 0 00 0C 0

~< (0 < 0< 0 < <0W.U< >1 < 0< 10 J-i u1-L< 1- < 0I 0 0 < 0 < 0 10 00< 0 0 0 0 -0 < 0 0>< < 1-0< 00< 0D 0D < 0 0D0 00 0 < 1- D 0D 0C 1- 0 0D 0

Z < U. 1- NC< UL- cc -1- 0< 0 < 00C <0 -1- W< x<< 1- < 1- < < 0 0 0 0 < 0 <

0 0~~~D 0 Ca0 0i0<0 < 0L 0

0

0 1- 0 0~~~~C 1- 0 <

0 0 0 ~~~~~<< 1- < 0< < 0 0 0

1- 0 < 0 0

.

1-c

0c 0D-t <~0tx 0 0

1- 0 ~~~<0 < < 0 0 < 0 < < 10 0 ~~~00 1-< 0D 0 0 0D 0

> <0 0< 00< 1- >1- x< LL1 W< -1- W< <0O00 0 1- ~~ ~~~~~0< 0 < 1- D 0. 0 0D1-0 0 0 0~~~~~~C 0 0 0 0 I0 0D 0 0

007 0a< 21 < 0 W< -1- >1- x < >-< cc0 cc .J- -0 0D < 0D 0 < 0 < 1-< 0 0 <0 0 0C0 < 0 0 0 0 0 0 C0)

~< 10 0 W< WU< (00 (00C <0C)Z< 0 < W< 00..< < CD0 0 < 1- < 0 C0 0I 0.

a 0 a < 0 0 0 0 1- 0 0D0< u1 A.0 00 0< -< 1- >- < >1 <0 U.1

0I 1- 0

<0 1-C1-0 0) - 1

0 0~~ 0 0 < < 0000 00 1>1- >1-~~~W< -1 1-L 00 <0 > 00 0< NC< 1-

0D 0. 00D < < 0 0 0C0D < <

0. 02C0)0E1- 0 < 0D 0C0D 0D 0< 0 00 M- c 0 -1 I... -11F- -11~- <0 0 >1 Wi< (00

0I o-2 <0< < 0 0 0 0<LI Oi0 0 0 0 0I 0 1- < 1

>I < i .j-> -. a0< U.I. c (D (00<0a 0 < <0 0 (0

0 0. 05C 0 0 - 0< 0 < 0 0 1- -

0 0 0

-< 0 0 0NC<ODa< 00 0 0

0 1-~~~~~0 1- 0~~~~ < 0 - < 0C 0 0 0I0 0 0 0 1- 0 ~~~~~~~~00 0 0D 0 C00 0' 0 ~~~~~00D < CD- 0c0 o001 0 0 ~~~~~~~00L00 ) 0L00)

>i )< yO 02-2--1- <0aW< 0 < <0 C10Z <<0< < < 0D 0 < <CD0 0 0 0 0 < 0C0D 0D

CD0 w00 0a< 1-0 Il 00 >1 W< 21 0< <

O < 1-0D < 0 <0 0CD- < 0 0

< 0 ( O r.0 w to co)0 x < 1.1 Z < Wu< 0. 0 0 < 0< U.-0 < 0 --...< < 1- < 0 0 0 00 0 00 C)

0 a-0000 0 1

000 0 (0 1-0 0< < 0 < Z < < 0 .-<0000 0 0 < 00~~~~~~~~~C 0 < 0 0 0000 000 0 0~~~~~~~~~~C 0 0 C00D

0<0.0~~~~~0 000<0 00~~~~ 0< -1-.J1 1- >1 -< <0t~~~I.- o~~c C) 0 0 0f1 %

102 Auerbach, Burger, Dodds, and Colten

u %J %ql

Page 8: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

00000 00 00 00 00 00 00 00 00 00 00 0*00toleCM c O cmQ *0 0co 0 W*CMN q* 0100 0 00 Vc04 V'- q oio oir4 1. cm o 0)a0)oV 4 lVoci w W) ) 1) 0 00co 0 0

0 0 0 CD U-- 0 0 0 C) <UL.) 0.0 x < < IL.0 < 0 J-i 0< L)- < < ?- >)- 0)- 0 < I- 0 0 CD ) < < 0 I< 0 0 CD 0 0 0 0 0<0 0 0< 00 0 >) -- >- IL) 00 )0 2< ~< 0 - 0 < 0 - < 0 )0 CD0-C C) 0 C ) C-0 0 cc c CD'.(A)0 >1 ~I- 00C 0 < -1- < 00O -< C) <.< 0 < < < CCD 0 0 0 0 0 <0 0 <

0>

0d co1 <

. < 0 < <- <0000 C) 0 0 0 0 0 0 C)~~~ ~~~~~~~~~~~~~~~0 <CD0 0 < W< I- < 00 -)00 < 0 -I- < >1 U.) ~ <I- 0 - < I< 0 < < 0 - < 1< 0D 0 0 0 0o 0 < 0 ) < 01>1 < W< >< 0.0U 1-C) < ..J- x< 1) < ccco C 001 <0 0~~~~1- 0 < < 0 < C) - < 0 <)- 0 0~~~< 09 0 0 C0D 0 0 0 010 < >< 3.0 cr C I-i 00 < 0 -iI U.) 21 < cc -I- CL0.0 I I- < 0 0 0 1 < 0 01- < 0 0 0 0 00C0D 0

<c0 cc0 cc0 0 < 0 < 0 < W< x< C)0000c > P- w< --0

< <( 00 0 0o 0 <

k I-j 0 0D0Dx<0 < 0 < 0 0 C 0 0 0 0< C 0 0DO oQ45L -

W< 00 2< ~~<< <1-0 0< CD) ..- 0 00 O <0 - < 0 0~~~~00 < 0D 0 <0 0- 0 >0t0 0 <1- 0 1- 0 0 < 0 l-c- a 1>0< W << (<D~ 0.00< 0 <<0.0toji0 ~< < 0 < 0 0< - 0 C 0 C.) CD< 0 0 0 0 0< 1- < 0 0 0< z < 1>

<0 -- w< -cc >1 00 >< 0in< W< CL w< <o>0~ I >1a0 < 0 00 0 <D00 <D0 <200<00 < 0< < 0 0 000 0>- CID 10< 0 > W< < 0 I-C) (A W <00 u0< cc >-< 1- < 1- 0 <~~~< < 0 I D <0 1- 1- ~~~~~00 co000 000< <0 1~0 )-i <0 > 0< W . <0 <9. 0. 2<0 0 O 09 0D << < <- 0 0 < 0

0 i C0 00 < < <L0 LL c0 1- 0 <~ 0a < <9.-1-00 0 0 1- 1 0 0 ~~ ~~~ ~~~~~~~~~~~0I- 000 <0 2<I >1- -3-< U.J 0< C-3 >< > < c 0< W< w<I0 0 <- 0 0 0C0D I 0 0t0 -

1- < 0D 0 co005<000 <<) 0. 0< < ( 0.CL0 IL) 0 IL) .<) c 00 0 00 < 0 0 <5 0 <- 1- 0 I

I- 0 <- I- 0 0 0 0 00--) 2 <I) 0. 0 0< IL<)- <0 00 a><- 0< a <0 00< << ) 0< <) 0 0 0 CD0 0 < 09L0 0 0 0 CO 0 )- -

29---0 ) < 0 . -0< )- W <0 .a- 1<< 00 0 0 << < 0 I0 < 0- 00. 00< I-< )- 0- 0 04

8D 0 0)--0< 0< < C CS~~~~< 0 0 00 0 0K 0 < > a < 0 < 1- 0

0 < D < 1 0 <0 0 0 02< W<)0I) 1- 0 - 0 ' W< 1- - 00 00 0 < I< 0L 0 x-I < 0 CD 0D< 0< 1 < 0<- 0-)I .) 0 < 0 < >) ) )0<U-- 0< ..) 0<<L 0 0 0 0 < 0t Z 0 < 0 0 0<' 00 0 < n i 0 0

< 0 0 < 0 0 )0. Z< <2<0

0q 0<0 << 0< c0< 0w 0 0 0 > < 0 < 10 0 000<0D0 0 1-0 00 000< >) WJ) .) < W<C0Dz< -)~- )-0 ( < 0.0 )- - 0< t00 0 0 1- 0~~~ ~~ ~0 < < < - 0 < < 000 0 0 0~~~~~~~c < 0 0 0 0 0 0 00 < cc0< -0 -1)C-a0<-I- < x<

0. j 0< 0 IL <00 ><I-1) W)< x)< 0<q u0D CW> <0 W . < CO) 2<> 2<. a <000 CD a >< 2a< <

0 0 ~~~~~~ ~~~~0I0 < <- 0 OI0< 0 0- 0<- 0 < 0 00DO < 0 0 >< << > 1- < 1-W0 0 0 00<0< < ~~~00<L- 0 < 1- <) <0 <) 00 a<0 IL)- >)- IL)-W<~0 0a < 0<) 0D 0<0 0 I>)-a -)- <0 co1W-0< 00 >1- 0< 3I0< D 00< 0a< 00 0< <0 ~~ ~~ ~~~~~~<0 00 < 0 0 0< 1 0 0 01<1- 0 0 ~~ ~~~~~0)- ) < 0 0 0 0 < 0

00< < .) <0 IL) 00 >) 0< < >)- 2< IL)- w < 0<- 1-0 0<- 0 1 - 0 00 < 0 0< 0 0<a 0 :0- 0 < 00(D~~~~~~~~~~~~~~~

0< 0 <- < 0 < 0cO < - I.( 0 0 u0 - 000<a- < 0 0 1- 0 10.0 <0 >> 00 2)- cW0 00 2)0W4< 0< -1- w )-

0 -i CaIU.I- <0 0<< 0 0 30 0~~<~~~ 00~~~ 0 II < .J. < 00 <) 0 2< W

~~ <~ <~ ~0 0 0 0 0 00 0 0 0~~~ ~ ~Ca 0 < <- I 0<-000 < 0 0 0 0 0~~~~~~~~~~~~~1~ 0 00 8<

A< .3 )- W <0 1) < 0 co -)- xz WA <0 W< 0< co<00 < < 0 0 0 0 0

Genetic Deficiency of Complement C3 Protein in Guinea Pigs 103

Page 9: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

poration of [35S]methionine in C3 of a chain was substantiallygreater than for ft chain. Two additional polypeptides (- 70kD and 48 kD) were detected in the normal and C4Dbut notthe C3Dextracellular culture media (see below).

Isolated liver poly A' selected mRNAfrom C3Dand nor-mal Hartley guinea pigs generated comparable amounts ofpro-C3 and secreted native C3 when injected into Xenopusoocytes pulse labeled for 72 h with [35S]methionine (Fig. 2).The size and subunit composition of the C3D and normalguinea pig liver-derived native C3 were similar, but again the70- and 48-kD products were only visualized when the oocyteswere injected with normal mRNA(even overexposed gelsfailed to demonstrate 70- and 48-kD products in C3D-injectedoocytes [data not stown]). Immunoprecipitation of C4, acomplement protein homologous with C3, from the normaland C3D cultures (Fig. 3) revealed comparable amounts ofpro-C4 intracellularly and native C4 in the extracellular me-dium.

Secretion of C3. To ascertain the rates of C3 secretion innormal and C3-deficient cells, peritoneal cell cultures werepulse labeled for 30 min with [35S]methionine (500 MCi/ml),washed, and refed with medium containing unlabeled methio-nine. Cell lysates and media were then assayed for radiolabeledC3 protein (Fig. 4). These data showed that the kinetics of C3secretion and net recovery of C3 protein were similar in theC3Dand normal peritoneal cells.

Autolytic cleavage and thiolester bridge. Since the 70- and48-kD polypeptides precipitable with antisera to C3 observedin Fig. 1 are similar in size to the previously reported C3autolytic cleavage products, we attempted to generate thesefragments from 35S-labeled native C3 protein produced inC3D and normal peritoneal cell cultures by incubating themedia in dilute alkaline SDS at 75°C for 20 min (21, 26).Again (Fig. 5), the 70- and 48-kD bands were visualized only inthe normal but not in the C3D-derived culture media.

Methylamine incorporation into plasma C3, C4, and a2macroglobulin. Previous reports have shown that C3Dguineapig plasma contains C3 both functionally and antigenically ataround 5% of the level found in normals. To examine thepossibility that residual C3 in plasma from C3D guinea pigsdiffered from that produced in tissue culture the followingexperiment was performed. Plasma from C3D and normalguinea pigs were fractionated on an ion exchange column andaliquots of each fraction examined for the ability to incorpo-rate '4C-labeled methylamine, a function of an intact thiolesterbridge. Fig. 6 demonstrates that methylamine was incorpo-rated into a2-macroglobulin and C4, two proteins that containthiolester bonds, in both normal and C3D plasma; methyl-amine incorporation into C3 was detected only in normal butnot C3D plasma even when the C3D plasma fractions wereconcentrated - 20-fold (data not shown) to account for thedifference in content of C3 in'the deficient plasma.

Proteolysis. Trypsin cleavage of native C3 protein pro-duced by normal guinea pig peritoneal cells resulted in genera-tion of C3b, as expected (Fig. 7). However, incubation of theC3D-derived native C3 protein with between 0.01 and 0.1 gof trypsin (10-fold less than the minimal concentration re-quired for C3b generation from normal C3) resulted in com-plete loss of immunoprecipitable C3 protein.

C3 transcription. As shown in Fig. 8, Northern blot analy-sis of total RNAisolated from C3D and C3 sufficient guinea

pigs revealed mature C3 mRNAtranscripts (- 5 kb) compara-ble in size and amount. That is, C3 mRNAfrom normal andC3Dguinea pig liver appeared qualitatively and quantitativelyindistinguishable. The ethidium-stained gel indicated thatcomparable amounts of RNAwere analyzed.

C3 gene structure. DNAwas isolated from Hartley, C2-de-ficient (strain 13), C4-deficient (strain 13), and C3-deficient(strain 2) animals, digested with several restriction enzymes(Pst I, Eco RI, BamHI, Hind III, Bgl I, and Sma I) and sub-jected to Southern blot analysis (27) using a full length normalguinea pig C3 cDNA probe. Fig. 9 shows that digestion witheither Pst I or Eco RI resulted in restriction fragment lengthpolymorphisms (additional 2.3 kb [Pst I] and 1.6 kb fEco RI]bands) in the C3Dand C4Dsamples. Since C3 protein synthe-sis and C3 serum concentrations are normal in C4-deficientguinea pigs, this polymorphism is not a marker for the C3deficiency phenotype.

Comparison of C3 cDNAfrom C3-sufficient and deficientliver libraries. 17 C3 clones from the C3-deficient library and25 clones from the C4-deficient library were detected; the larg-est insert, 3.8 kb in the C3-deficient and 4.2 kb in the C4-defi-cient libraries. Inserts from the three largest clones of eachlibrary were isolated and sequenced. As shown in Fig. 10,

70% of the cDNAwas analyzed using this sequencing strat-egy, with most regions sequenced several times on bothstrands. To complete the sequence, oligonucleotides wereconstructed and used as primers for dideoxy sequencing di-rectly from phagemids. Using this initial strategy, the C3cDNA sequence extended from the poly(A) tail through the750 and 1150 most 5' nucleotides of the ,B chain for C3-defi-cient and C3-sufficient clones, respectively.

To complete the C3 cDNA sequence, primer extensioncDNA libraries were constructed using a C3 specific oligonu-cleotide (20 mer) corresponding to positions 2469-2488 ofcomplete guinea pig C3 cDNAsequence. Clones isolated fromeach library contained sequences that extended 29 nucleotides5' of the C3 signal peptide and overlapped by 1312 and 1712nucleotides, the largest clones from the C3-deficient and C3-sufficient libraries. Confirmation of the normal and C3 defi-cient guinea pig C3 sequence was accomplished by direct se-quence analysis of PCR fragments generated from C3 suffi-cient (Hartley) and C3 deficient mRNA(Fig. 10).

The complete C3-sufficient (C4 deficient) guinea pig C3cDNAsequence and derived amino acid sequence are shownin Fig. 1 1. The sequence spans 5081 nucleotides, from 29 basepairs upstream of the signal peptide to the beginning of thepoly(A) tail. A putative polyadenylation signal, AATAAA, islocated 16 nucleotides upstream from the polyadenylationsite, and is identical to the polyadenylation signal described formurine C3 and Slp. The deduced C3 protein sequence consistsof 1666 amino acids in a (3-a chain orientation with four argi-nines interposed between the # and a chains. The coding re-gion is, therefore, equal in size to murine C3 and three aminoacid residues longer than human C3. Overall, 77-80% nucleo-tide and amino acid identity was observed for human, mouse,and guinea pig C3. Complete nucleotide identity betweenhuman and guinea pig C3 sequences was found at the thioles-ter 1 5-base pair segment within the flanking 45 base pairs ineach direction. Guinea pig C3 shares other structural andfunctional characteristics with human and murine C3, includ-ing two asparagine carbohydrate attachment sites at positions

104 Auerbach, Burger, Dodds, and Colten

Page 10: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

944 and 1620, a trypsin cleavage site between lysine at position1006 and histidine at 1007, and two Factor I cleavage sites atpositions 1310 and 1327.

C3-sufficient (C4 deficient) and C3-deficient guinea pig C3cDNA sequences were compared. Only three nucleotide dif-ferences, at positions 83, 1630, and 2789 were found (Fig. 10).These nucleotide differences are "silent" in that they do notresult in a difference in the derived amino acid sequences. Thesequence of PCRfragments from an additional C3-deficientliver mRNAwas identical to the sequence generated from theC3-deficient cDNA and primary extension libraries. TheHartley C3 PCR fragments showed identity with the otherC3-sufficient (C4 deficient) C3 cDNAsequence except at po-sitions 83 (T to C); i.e., a substitution identical at this positionwith the C3-deficient sequence.

Discussion

Several possible explanations were considered to account fordiscrepancies between serum concentration of C3 and rate ofC3 biosynthesis in previous studies of humans with geneticallydetermined C3 deficiency (1-11). However, those experimentsnecessarily monitored C3 biosynthesis only at an extrahepaticsite (blood monocytes) since ethical considerations precludeda direct test of the hypothesis that C3 synthesis rate in liver wasmore closely related to the deficiency than C3 synthesis inother tissues. One possibility that was not considered at thattime was that a defective C3 protein susceptible to acceleratedcatabolism was produced by the deficient cells even though thenewly synthesized C3 protein appeared grossly normal by esti-mates of size and subunit composition. Moreover, a near nor-mal rate of catabolism of normal purified radiolabeled C3protein was observed in a C3-deficient patient (3). Similarfindings were obtained in the initial studies of C3 catabolism inC3-deficient guinea pigs (14). Preliminary data from thosestudies also suggested that C3 synthesis is normal in C3-defi-cient guinea pig hepatocytes.

In the present report we established that liver and perito-neal cells from C3-deficient, normal, and C4-deficient guineapigs (a) contain C3 specific mRNAindistinguishable in sizeand quantity by Northern blot analysis, and (b) synthesizepro-C3 and secrete native C3 protein at similar rates. Althoughthe C3 protein produced by the C3D peritoneal cells or se-creted by oocytes primed with C3D liver mRNAappears nor-mal in size and subunit composition it differs from normal C3in at least two respects. That is, the C3D-derived C3 proteinfails to undergo autolytic cleavage and is unusually susceptibleto proteolytic digestion.

C3, C4, and a2 macroglobulin are proteins that share anunusual structural feature, a thiolester bridge, important intheir respective functions. The phenomenon of autolytic cleav-age in dilute alkaline SDSis associated with the presence of anintact thiolester bridge. The C3 protein produced in C3-defi-cient guinea pigs, by this criterion, appears to display a defectin the thiolester bridge or its accessibility. Support for thisconcept and for the selectivity of the defect is provided bydemonstrating incorporation of methylamine into C4 and a2macroglobulin, but not into C3 of C3 deficient plasma evenwhen the samples are concentrated - 20-fold to increase theC3 content in the C3D sample to near normal levels. An ab-normality in primary or higher order structure of the C3-defi-

cient C3 protein is also suggested by the unusual susceptibilityof this protein to proteolysis. The rate of in vivo catabolism ofthis abnormal C3 protein is unknown because previous esti-mates of C3 metabolism in the deficient animals as in studiesof C3Dhumans were obtained using normal purified radiola-beled C3 protein (14).

These data are consistent with (a) a mutation in the codingregion of the C3 gene leading to a defect in primary or higherorder structure affecting the thiolester bridge or its accessibil-ity, (b) a C3 specific co- or post-translational processing defectthat affects the thiolester bridge; i.e., a genetic defect unlinkedto the C3 gene. The finding of a restriction fragment lengthpolymorphism (RFLP) that does not correlate with the C3deficiency phenotype lends support to the latter hypothesis.That is, the RFLPprovides a marker for the C3 structural genedistinct from the deficiency phenotype. Sequence analysis ofC3 cDNAfrom C3 deficient guinea pig liver libraries ruled outa mutation in primary structure of the C3 gene as the basis forthis genetic defect.

Sequencing of C3 cDNAclones from two separate C3 defi-cient liver libraries and from PCRfragments generated from athird C3-deficient liver mRNApreparation established thatthe C3 coding regions in C3-sufficient and deficient guineapigs is identical except for positions 83, 1630, and 2789. Noneof these substitutions result in an amino acid difference. Atposition 83 the Hartley (C3 sufficient) sequence is identical tothe C3-deficient sequence. These data thus support the hy-pothesis that C3 deficiency in guinea pig is due to a co- orpostsynthetic processing modification that renders the C3 pro-tein highly susceptible to proteolysis.

The biochemical mechanism by which the thiolester bridgeis generated in C3 is uncertain. Kahn and Erickson (28) sug-gested that isomerization of a lactam ring to generate the thio-lactone proceeds spontaneously under physiological condi-tions, though they could not rule out an enzymatically facili-tated process. lijima et al. (29) on the other hand providedtentative evidence for participation of a cytoplasmic factor(possibly an enzyme) in the formation of the C3 thiolester.Heretofore, investigators have assumed but not tested the as-sumption that the mechanisms for generation of the C3, C4and a2 macroglobulin thiolesters are identical. Since the mu-tation accounting for C3 deficiency is not in the coding regionof the C3 gene, this defect provides a probe to examine co- andpostsynthetic processing of this important group of proteins.Molecular biological and classical genetic studies are under-way to test this hypothesis.

Acknowledgments

Wethank Barbara Pellerito for excellent secretarial assistance.This work was supported by National Institutes of Health grants

AI-24836 and HL-3759 1.

References

1. Alper, C. E., H. R. Colten, F. S. Rosen, A. R. Rabson, G. M.McNab, and J. S. S. Gear. 1972. Homozygous deficiency of C3 in apatient with repeated infections. Lancet. ii: 1 179-1181.

2. Alper, C. E., H. R. Colten, J. S. S. Gear, A. R. Rabson, and F. S.Rosen. 1976. Homozygous human C3 deficiency. J. Clin. Invest.57:222-229.

3. Alper, C. E., R. P. Propp, M. R. Klemperer, and F. S. Rosen.

Genetic Deficiency of Complement C3 Protein in Guinea Pigs 105

Page 11: MolecularBasis Complement C3 Deficiency Guinea Pigs€¦ · deficient (C3D),andC4 deficient (C4D)guinea pigs. Autoradiograph of reducedSDS-PAGE analysis ofintracellular (IC)pro-C3andextra-cellular(XC)native

1969. Inherited deficiency of the third component of complement. J.Clin. Invest. 48:553-557.

4. Ballow, M., J. E. Shira, L. Harden, S. Y. Yang, and N. K. Day.1975. Complete absence of the third component. J. Clin. Invest.56:703-710.

5. Borzy, M. S., A. Gewurz, L. Wolff, D. Houghten, and E.Lovrein. 1988. Inherited C3 deficiency with recurrent infections andglomerulonephritis. Am. J. Dis. Child. 142:79-83.

6. Hsieh, K. H., C. Y. Lin, and T. C. Lee. 1981. Complete absenceof the third component of complement in a patient with repeatedinfections. Clin. Immunol. Immunopathol. 20:305-312.

7. Davis, A. E., J. S. Davis, A. R. Rabson, S. G. Osofsky, H. R.Colten, F. S. Rosen, and C. A. Alper. 1977. Homozygous C3 defi-ciency: Detection of C3 by radioimmunoassay. Clin. Immunol. Im-munopathol. 8:543-550.

8. Cochrane, C. G. 1984. The role of complement in experimentaldisease models. Springer Semin. Immunopathol. 7:263-270.

9. Osofsky, S. G., B. H. Thompson, T. F. Lint, and H. Gewurz.1977. Hereditary deficiency of the third component of complement ina child with fever, skin rash, and arthralgias: response to transfusion ofwhole blood. J. Pediatr. 90:180-185.

10. Sano, Y., H. Nishinukai, H. Kitamura, K. Nagaki, S. Inai, Y.Hamasaki, I. Maruyama, and A. Igata. 1981. Hereditary deficiency ofthe third component of complement in two sisters with systemic lupuserythematosis-like symptoms. Arthritis Rheum. 24:1255-1267.

11. Burger, M., J. E. Balow, C. B. Wilson, and M. M. Frank. 1983.Circulating immune complexes and glomerulonephritis in a patientwith congenital absence of the third component of complement. N.Engl. J. Med. 308:1009-1012.

12. Einstein, L. P., P. J. Hansen, M. Ballow, A. E. Davis, J. S.David, C. E. Alper, F. S. Rosen, and H. R. Colten. 1977. Biosynthesisof the third component of complement (C3) in vitro by monocytesfrom both normal and homozygous C3-deficient humans. J. Clin.Invest. 60:963-969.

13. Ruddy, S., and H. R. Colten. 1974. Rheumatoid arthritis bio-synthesis of complement proteins by synovial tissues. N. Engl. J. Med.290:1284-1288.

14. Burger, R., J. Gordon, G. Stevenson, G. Ramadori, B. Zanker,U. Hadding, and D. Bitter-Suermann. 1986. An inherited deficiency ofthe third component of complement, C3, in guinea pigs. Eur. J Im-munol. 16:7- 1.

15. Bottger, E. C., S. Metzger, D. Bitter-Suermann, and R. Burger.1986. Impaired humoral immune response in complement C3 defi-cient guinea pigs: absence of secondary antibody response. Eur. J.Immunol. 16:1231-1235.

16. Ochs, H. D., R. J. Wedgewood, M. M. Frank, S. R. Heller, andS. W. Hosea. 1983. The role of complement in induction of antibodyresponses. Clin. Exp. Immunol. 53:208-216.

17. Bottger, E. C., T. Hoffmann, U. Hadding, and D. Bitter-Suer-mann. 1985. Influence of genetically inherited complement deficien-cies on humoral immune response in guinea pigs. J. Immunol.135:4100-4107.

18. Auerbach, H. S., R. D. Baker, W. J. Matthews, Jr., and H. R.Colten. 1984. Molecular mechanism for feedback regulation of C4biosynthesis in guinea pig peritoneal macrophage. J. Exp. Med.159:1750-1761.

19. Frank, M. M., J. May, T. Gaither, and L. Ellman. 1971. In vitrostudies of complement function in sera of C4-deficient guinea pigs. J.Exp. Med. 134:176-187.

20. Roos, M. H., E. Mollenhauer, P. Demant, and C. Rittner. 1982.A molecular basis for the two locus model of human complementcomponent C4. Nature (Lond.). 298:854-856.

21. Chirgwin, J. M., A. E. Przybyla, R. J. McDonald, and W. J.Rutter. 1979. Isolation of biologically active ribonucleic acid fromsources enriched in ribonuclease. Biochemistry. 18:5294-5299.

22. Aviv, H., and P. Leder. 1972. Purification of biologically activeglobin messenger RNAby chromatography on oligothymidylic acidcellulose. Proc. Natl. Acad. Sci. USA. 69:1408-1412.

23. Okayama, H., and P. Berg. 1982. High efficiency cloning of fulllength cDNA. Mol. Cell. Biol. 2:161-170.

24. Gubler, U., and B. J. Hoffman. 1983. A simple and very effi-cient method for generating cDNA libraries. Gene. 25:263-269.

25. Messing, J., R. Crea, and P. H. Seeburg. 1981. A system forshotgun DNAsequencing. Nucleic Acids Res. 9:309-321.

26. Janatova, J., J. W. Prahl, and B. F. Tack. 1979. On the presenceand the availability of a single sulfhydryl group in both the third (C3)and fourth (C4) components of human complement. In Proceedings ofthe XIth International Congress of Biochemistry. Abstracts: 524 (No.08- I -R8 1). Toronto, Canada.

27. Southern, E. M. 1975. Detection of specific sequences amongDNA fragments separated by gel electrophoresis. J. Mol. Biol.98:503-5 17.

28. Khan, S. A., and B. W. Erickson. 1982. An equilibrium modelof the metastable binding sites of a2 macroglobulin and complementproteins C3 and C4. J. Biol. Chem. 257:11864-11867.

29. Iijima, M., T. Tobe, T. Sakamoto, and M. Tomita. 1984. Bio-synthesis of the internal thiolester bond of the third component ofcomplement. J. Biochem. 96:1539-1546.

106 Auerbach, Burger, Dodds, and Colten