J Am Soc Nephrol 9: 638-164-4. 1998

All Four Putative Ligand-.Binding Domains in Megalin

Contain Pathogenic Epitopes Capable of Inducing Passive

Heymann Nephritis

HAJIME YAMAZAKI,* ROBERT ULLRICH,�� MARKUS EXNER,t AKIHIKO SAITO,*

ROBERT A. ORLANDO,* DONTSCHO KERJASCHKI,1 and

MARILYN GIST FARQUHAR’�*Ditisio,i of Cellular and Molecular Medicine and Department of Pathology, University of California San

Diego, La Jolla, (‘alifornia: (md tDivision of Ultrast,uctural Pathology and Cell Biology, institute of Clinical

Pathology, University of Vienna, Vienna, Austria.

Absti-act. Megalin (gp330) is the main target antigen involved

in the induction of Heymann nephritis (HN), a rat model of

human membranous nephropathy. Its large extracellular region

contains four putative ligand-binding domains separated by

spacer regions. Previously, it was reported that the second

ligand-binding domain (LBD II) of megalin is involved in the

pathogenesis of passive HN because it is capable of binding

antibodies in t’lt’() and initiating formation of immune deposits

(ID). This study explores the possibility that pathogenic

epitopes might also be present in the other putative ligand-

binding domains. Recombinant fragments of ligand-binding

domains (LBD) I through IV expressed in a baculovirus system

were used to generate polyclonal domain-specific antibodies.

Antibodies raised against each of the recombinant megalin

fragments reacted preferentially with its respective antigen and

with whole megalin by immunobbotting. Each ofthe antibodies

also gave a characteristic brush-border staining for megalin by

indirect immunofluorescence on rat kidney. When rats were

injected with the domain-specific antibodies to test their ability

to produce passive HN. gbomerular ID were present in kidneys

of all injected animals. The staining pattern in glomeruli of rats

injected with LBD I, III. or IV was similar to that obtained with

antibodies to LBD II. It is concluded that passive HN can be

induced with antibodies against LBD I. III, and IV, as well as

LBD II, and that each of the ligand-binding domains contains

a pathogenic epitope. These findings provide further evidence

for the multiple epitope model of HN.

Heymann nephritis (HN) is a well characterized experimental

rat model of human membranous nephropathy ( I ). The gb-

merular lesions that represent the hallmark of this disease

consist of subepithelial immune deposits (ID) formed mainly

by shedding of antigen-antibody complexes from podocytes

(2). The major pathogenic antigens of HN are megalin (gp330)

and the receptor-associated protein (RAP) (2,3). Megalin is a

multiligand-binding endocytic receptor expressed in clathrin-

coated pits at the surface of a number of epithelia. including

those of the kidney glomerulus (4), proximal tubule (5,6). yolk

sac. type II cells of the lung, etc. (7). Sequence analysis of the

eDNA for megalin (8) reveals that it is a large (>600 kD)

transmembrane protein containing an extracellular region

(4400 amino acids). a single transmembrane domain (22 amino

acids), and a C-terminal cytoplasmic tail (213 amino acids).

The extracellular domain of megalin contains structural motifs

characteristic of members of the LDL receptor family: namely.

Received E)ecember I I. 1997. Accepted April 9. 1998.

Correspondence to Dr. Marilyn Gist Farquhar. Division of Cellular and Mo-

lecular Medicine. ()6S I . University of California. San Diego. 9500 Gilman

Drive. La Jolla. CA 92093-0651.

1(146-6673/0909- 1638503.()0/()

Jotirn�tl of the American Society of Nephrology

Copyright f) l998 by the American Society of Nephrology

four clusters of cysteine-rich LDL receptor ligand-binding

repeats (LBD I through IV). growth factor repeats, an epider-

mal growth factor repeat, and YWTD spacer regions. Megalin

binds RAP (9), a specialized chaperone that assists in the

folding of megalin in the endoplasmic reticulum (10) and in its

delivery to the cell surface ( I 1 ). Together, megalin and RAP

constitute what we have called the Heymann nephritis anti-

genie complex (HNAC) (2,9). HN can be induced by immu-

nization with either megalin or RAP, and ID can be passively

induced by injection ofeither antimegalin (2.4,12) or anti-RAP

( I 2-16) antibodies. Thus, both niegalin and RAP contain

pathogenic epitopes capable of binding antibodies and induc-

ing HN.

Previously we mapped the pathogenic epitope in RAP to a

14-amino acid sequence near its N terminus (16). We also

mapped a major pathogenic epitope in megalin to 46 amino

acids (amino acids 1 160-1205) that correspond to the fifth

ligand-binding repeat of the second ligand-binding domain

(LBD II) (17). Similar findings were obtained independently

by Raychowdhury et al. ( I 8). who reported that amino acids

I I 14-1250 from the second LBD of megalin are involved in

induction of HN. It has not yet been determined whether other

LBD of megalin contain pathogenic epitopes. It is important to

map the pathogenic epitopes in megalin to elucidate the mo-

lecular mechanisms of ID formation in HN.

NIJIIIIIU.....I.”s.IIIIIIII

LBD I LBD II LBD III LBD IV

IIElI

EGF repeat

growth factor repeat

J Am Soc Nephrol 9: 1638-1644. 1998 Mapping Pathogenic Epitopes of Rat Megalin 1639

For this study, we generated recombinant fragments repre-

senting the four clusters of ligand-binding repeats (LBD I

through IV) in megalin, using a baculovirus expression system,

and raised domain-specific polycbonal antibodies against these

fragments. We determined that each of the antibodies is capa-

ble of causing ID formation (passive HN), indicating that

pathogenic epitopes are present in all four LBD of megalin.

Materials and MethodsBaculovirus Expression citid Purificatioii of Megalin

Fragments

Recombinant baculoviruses expressi ng eDNA fragments encoding

the individual clusters of ligand-binding repeats (LBD I through IV)

(Figure 1 ) were prepared and used to infect Sf9 cells. The proteins

expressed in Sf-9 cells were affinity-purified by Ni2� chromatography

as described ( 17). In brief, fragment I (nucleotides 106-1364, amino

acids -25-389), fragment II (nucleotides 2886-4916, amino acids

897-1573), fragment III (nucleotides 7834-10134, amino acids

2547-3312), and fragment IV (nucleotides 10315-12405. amino acids

3374-4069) were cloned into pSPH vector. Fragment I and fragment

II inserts with a hexa-histidine tag sequence were subcboned into

pAcSB2 and pAcGP67B (PharMingen. San Diego. CA), respectively.

Fragment III and fragment IV inserts with the hexa-histidine tag

sequence were subcloned into pAcGP67A (PharMingen). Sf9 cells

were transfected with the constructs using the BaculoGold transfec-

tion kit (PharMingen) to generate recombinant baculoviruses. The

viruses were amplified, and cells were infected and cultured for 72 h,

after which they were lysed in 6 M guanidine HC1. 20 mM sodium

phosphate, 500 mM NaCI, pH 7.8. for 30 mm at 23#{176}Cfollowed by

sonication (three times, for 5 s each). Lysates were cleared by cen-

trifugation (10,000 X g for 15 mm), and the supernatants were

incubated for 15 mm with nickel-Sepharose (Invitrogen. Carlsbad,

CA). The resin was washed stepwise using 8 M urea. 20 mM sodiumphosphate, 500 mM NaCI at decreasing pH values of 7.8, 6.0, 5.3, and

4.0. Bound proteins were eluted with a 10 to 320 mM step gradient of

imidazole in the urea-containing buffer, pH 4.0. Baculovirus-derived

megalin fragments were visualized by silver staining after 7.5%

sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Individ-

ual fragments were then pooled. dialyzed against 10 mM Tris. pH 7.7.

150 mM NaCI, 0.1% Triton X-lOO, and concentrated using Ultra-

free-15 (Millipore. Bedford, MA).

Antibodies

Antimegalin LBD domain-specific antisera were raised in rabbits

by immunization with 50 p.g of each purified recombinant niegalin

fragment (LBD I through IV) in complete Freund’s adjuvant. kllowed

by two boosts with 50 p�g of protein in incotiiplete Freund�s adjutant

at 3 and 6 wk. Polyclonal antisera against full-length megalin were

prepared by immunization with megalin purified from the rat renal

cortex by gel filtration and lentil lectin chromatography (4). All

procedures on animals were conducted in accord with National Insti-

tutes of Health Guide for the Care and Use of Laboratory Animals.

Iminunoblot Analysis

Rat kidney cortex proteins and the purified recombinant megalin

fragments were boiled for 5 mm in Laemmli sample buffer sup-

plemented with 5% 2-mercaptoethanol and separated by sodium

dodecyl sulfate-polyacrylamide gel electrophoresis. Proteins were

transferred to polyvinylidene difluoride membranes (Millipore).

Membranes were blocked with phosphate-buffered saline, 0.2e/

Triton X-l00, 5% calf serum for 1 h at 23#{176}C,and incubated with

primary antibodies for 2 h at 23#{176}C.They were then washed three

times ( 10 mm), incubated with goat anti-rabbit horseradish perox-

idase conjugate (BioRad, Hercules, CA), and bound primary anti-

bodies were detected by SuperSignal chemiluminescence (Pierce.

Rockford, IL).

I,iduction of Passive HN and Iminunofluorescence

Analysis

Female Sprague Dawley rats (200 to 250 g) were injected intrave-

nously with 5 to 10 mg of IgG specific for megalin LBD I. II. III. or

IV, purified on protein A agarose (BioRad). Kidneys of normal and

3-d passive HN rats were perfused with paraformaldehyde-lysine-

periodate fixative. Samples of renal cortex were cryoprotected and

frozen in liquid nitrogen for cryosectioning as described ( I 2). Semi-

thin (0.5 j.tm) sections of normal rat kidneys were incubated with

antimegalin LBD I-lV antisera, followed by FITC-conjugated goat

anti-rabbit lgG (DAKO, Carpinteria, CA). Sections of kidneys from

rats with passive HN were incubated directly in FITC-conjugated goat

anti-rabbit lgG, and examined and photographed using a Zeiss Axio-

phot microscope equipped for epifluorescence.

u�i.I.....I.....I.....IIIIIII11I......III1II1IIIIu.....I:II��

ligand-binding repeat #{149}‘ s s e YWTD spacer region

I transmembrane domainc:� cytoplasmic tail

Figure 1. Structure of rat iiiegalin indicating the regions of niegalin against which antibodies were generated. Four cDNA tlagmciits

(ligand-binding domains I through 1V [LBD I-lVl). encoding the first. second. third. and fourth LBD. n�spectively. ttete e.�p�c�.cd in a

haculovirus system. purified to honiogeneity. and used to raise polyclonal. domain -specific aatibodies against L.BD 1 through IV i � uively

I 640 Journal of the American Society of Nephrology J Am Soc Nephrol 9: 1638-1644, 1998

kDa I II Ill IV I II III IV

230-

126-

80-

48-

I II III IV I II III IV I II III IV

anti-megalin anti-LBD I anti-LBD U anti-LBD III anti-LBD IV

Figure 2. Immunoblot analysis demonstrating the specificity of anti-LBD I, II. III, or IV antibodies. Each of the four clusters of ligand-bindingrepeats, LBD I through IV, was separated by 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to

polyvinylidene difluoride (PVDF), and immunoblotted with polyclonal antibodies raised against whole megalin (antimegalin) or recombinant

niegalin fragments, (anti-LBD I-IV). The antibody raised against whole megalin recognized all four ligand-binding domains (LBD I-IV)migrating with relative electrophoretic mobilities of 80, 126, 124, and 130 kD, respectively. The antisera raised against megalin fragments LBD

I through IV reacted preferentially with the recombinant fragment used as immunogen, indicating that each antibody is domain-specific.

Urinalysis

Rats were housed overnight in metabolic cages, and 16-h urinespecimens were collected and analyzed for protein content by the

biuret method.

Homology Search

Amino acid homology searches were performed using MacVector

(version 6.0, International Biotechnologies, Inc., New Haven, CT).

ResultsCharacterization of Polvclonal Antibodies Against

Recombinant Megalin Fragments

Rat megalin eDNA fragments I through IV encoding the first

through fourth clusters of ligand-binding repeats (LBD I-IV)

were used to generate recombinant bacuboviruses, and Sf9 cells

were infected with these recombinant bacuboviruses. The ex-

pressed proteins were affinity-purified by Ni2� chromatogra-

phy and used to raise four domain-specific polyclonal antibod-

ies against LBD I, II, III, and IV. The specificity of these

antibodies for each cluster of ligand-binding repeats was de-

termined by immunobbot analysis. Each of the domain-specific

antisera reacted preferentially with its respective recombinant

protein used as immunogen, whereas all four domains were

recognized by a polycbonal antibody raised against whole

megalin (Figure 2).

To confirm the specificity of the anti-LBD l-lV antisera for

megalin, we performed inimunobbotting on rat renal microvil-

lar extract and immunolocalization on sections of rat kidney.

Each of the domain-specific antisera raised against recombi-

nant megalin fragments reacted with native megalin in the

microvillar extract (Figure 3). By immunofluorescence, mega-

lin was detected at the base of the proximal tubule microvilli

with all four antibodies (Figure 4, A through D). Thus, each of

the domain-specific antibodies recognizes intact megalin in

situ as well as by immunoblotting.

Polyclonal Antibodies Specific for Megalin LBD I

Through IV Induce Passive Heymann Nephritis

To determine whether the domain-specific antibodies are

capable of inducing passive HN, Sprague Dawley rats were

injected intravenously with antimegalin LBD l-IV IgG, and

3 d later the kidneys were examined by direct immunofluores-

cence. Granular subepithelial ID were observed in gbomeruli of

rats injected with all four domain-specific IgG (Figure 5). The

ID induced by anti-LBD I, II, and III IgG were similar in size,

but those induced by anti-LBD IV IgG were smaller. This may

reflect the somewhat lower titer of this anti-LBD IV antibody.

An additional finding was that staining of proximal tubule

brush border was seen in the specimen with anti-LBD I IgG,

but not the others. This suggests that anti-LBD I IgG may pass

through the gbomerulus in greater amounts. Rats injected with

all four of the domain-specific IgG had urinary protein excre-

tion within the normal range (data not shown). We conclude

Ml 234

kDa �,�1p*megaIin

230-

126-

0

Figure 3. Immunoblot demonstrating that domain-specific antibodies

against LBD I through IV recognize megalin in a renal microvillar

extract. Rat kidney microvillar proteins were separated by 5% SDS-

PAGE, transferred to PVDF membranes, and immunoblotted using

anti-whole megalin polyclonal antibody (lane M), anti-LBD I (lane I),

anti-LBD II (lane 2), anti-LBD III (lane 3), and anti-LBD IV (lane 4).

All antibodies react strongly with megalin (arrow).

J Am Soc Nephrol 9: 1638-1644. l998 Mapping Pathogenic Epitopes of Rat Megalin 1641

Figure 4. Indirect immunofluorescence demonstrating that anti-LBD I, II, III, and IV antibodies recognize megalin in rat kidney. Semithin

cryosections were prepared from a normal rat kidney and incubated with domain-specific antimegalin antibodies. A. anti-LBD I: B, anti-LBD

II: C, anti-LBD III: and D. anti-LBD IV. Megalin is detected in coated pits located at the base of the proximal tubule brush border with all

of the antibodies. Magnification. X400

that antibodies against all four LBD are capable of producing

passive HN.

Sequence Homology Between the Fijiii Repeat of LBD

II and Repeats in LBD I, II!, and IV

Previously, we defined the fifth ligand-binding repeat in

LBD II as containing a major pathogenic epitope of megalin

( I 7). To investigate whether there are structurally similar

regions in other LBD, we compared the amino acid se-

quence of LBD I, III, and IV with that of the fifth ligand-

binding repeat of LBD II. Amino acid identity percentages

are 30 to 47% among the ligand-binding repeats in LBD I,

III, and IV. As shown in Figure 6, the third ligand-binding

repeat of LBD I (amino acids 80-126), the third ligand-

binding repeat of LBD III (amino acids 2753-2802), and the

tenth ligand-binding repeat of LBD IV (amino acids 3857-

3906) showed the highest homology, with 45 to 47% iden-

tity and 13 to 15% similarity.

DiscussionThe extracellular domain of megalin contains four clusters

of cysteine-rich regions, LBD I through IV, composed of 7, 8,

10, and I I ligand-binding repeats, respectively. with adjacent

growth factor repeats. To date only the second cluster, LBD II,

of ligand-binding repeats has been shown to contain a patho-

genie epitope involved in HN (17,18). In this study, we gen-

erated domain-specific polycbonal antibodies against recombi-

nant fragments of all four LBD of megalin to determine

whether additional pathogenic epitopes are present in megalin.

We found that passive HN can be induced by injecting rats

with each of the domain-specific antibodies, indicating that all

four putative LBD contain pathogenic epitopes.

We have previously proposed a multiple pathogenic epitope

model of HN in which both megalin and RAP contain at least

one pathogenic epitope ( 12, 19), and antibody binding to either

epitope can trigger deposition of immune complexes. We sub-

sequently mapped these two pathogenic epitopes to a 14-amino

acid sequence in RAP ( 1 4, 1 6) and 46 amino acids (amino acids

1 160-1205) constituting the fifth ligand-binding repeat of

LBD II in megalin ( 17). The present study indicates that all

four putative LBD contain pathogenic epitopes involved in

passive HN, thus providing further support for the multiple

epitope model of HN. Raychowdhury ci al. ( 1 8) reported that

a 137-amino acid sequence (amino acids 1 1 14-1250) within

LBD II produces active HN but noted that the ID were smaller

than those found in gbomeruli of rats immunized with whole

megalin. It seems likely that antibody binding to multiple

pathogenic epitopes present in megalin or RAP are required for

enlargement of ID.

The fact that the third ligand-binding repeat of LBD I, the

third ligand-binding repeat of LBD III, and the tenth ligand-

binding repeat of LBD IV showed the highest homology (45 to

47% identity and 13 to 15% similarity) to the fifth ligand-

binding repeat of LBD II suggests that these ligand-binding

1642 Journal of the American Society of Nephrology J Am Soc Nephrol 9: l638-1644. 1998

Figure 5. Injection of rats with each of the four antimegalin domain-specific lgG resulted in formation of glomerular immune deposits (ID).Rats were injected with antimegalin LBD 1, 11, III, or IV IgG, and 3 d later the kidneys were removed and processed for direct

immunofluorescence. Subepithelial ID, the hallmark of passive Heymann nephritis (HN), are seen in rats injected with each of the antibodies.

Staining patterns on semithin cryosections are similar among the four specimens, except that the ID induced by anti-LBD IV are somewhat

smaller than the others. Also. specimens from rats injected with LBD I, but not anti-LBD II through IV, bound to the proximal tubule brush

border. Magnification: x400 and X 1000.

5thofLBDII V LNCT SA�1FK .AD�JS S .iNSR�Y:RCDG�V�Y CR�NSDEAGC - - PTR P PGM

3rdofLBDI G I TCS A Q�jM I S N�JQ - � I P S: E�YR.CD H�JS C P�G SD#{149}ER N C H Y PT C D Q L I

5thofLBDII V LN,�TSI- A Q!IK . A D�S S � I N S R�fl R D V Y D�C RDNl- . SDEA GCPl. - TR !�G

3rdofLBDIII F Rfr�CN�I I E�jT � S N�JR . I P L s�jv #{149}� N � I N N�.H DN]D I S D E K N C P1� H T C p:P D

5thofLBDII V L N�I S� A�- FK C A G S�S I N S R Y R�D V Y ;C� R�D N SD.E . . A G C - - �IP.T R P 1� G

lOthofLBDIV N I P�JE S� P�JR FR C - N SJR :-v y G H Q L�JN V D C G�G S:DE K E E H R KIPj�H K PC

Figure 6. Comparative homology between ligand-binding repeats in LBD I, III. and IV and the fifth ligand-binding repeat of LBD II. The third

ligand-binding repeat of LBD I (amino acids 80-126), the third ligand-binding repeat of LBD III (amino acids 2753-2802), and the tenth

ligand-binding repeat of LBD IV (amino acids 3857-3906) showed the highest homology. i.e., 45 to 47� identity and 13 to l59� similarity.

Gaps have been introduced to optimize the alignment. Identical residues are shaded. and conservative changes are boxed.

repeats could contribute to the immunogenicity and pathoge- epitopes. Additional studies are needed to determine whether

nicity in passive HN. However, we cannot rule out that other these homologous regions are involved in the pathogenesis of

ligand-binding repeats of lower sequence homology might HN.

have conformational homology and present pathogenic LBD II stands out as an important region in megalin from

I Am Soc Nephrol 9: 1638-1644, 1998 Mapping Pathogenic Epitopes of Rat Megalin 1643

both the structural and the functional standpoint. This region

not only contains a pathogenic epitope for passive HN, but it

also contains a binding site for several ligands, including

apolipoprotein E-f3 very low density lipoprotein (ApoE-

/3VLDL), lipoprotein lipase, aprotinin, lactoferrmn, and RAP

(20). The fact that this region binds circulating antibodies

indicates that this site is exposed in vito on the basal surface of

the glomerular epithelium where it is competent to bind mul-

tiple ligands (2 1 ), as well as pathogenic antibodies. This study

demonstrates that not only LBD II, but also LBD I, III, and IV

are exposed in viva and can bind circulating antibodies. The

finding that LBD I, III, and IV are exposed in vivo raises the

possibility that LBD other than LBD II may contribute to

ligand-binding by megalin. LBD II is also an important domain

in the LDL receptor-related protein/a2-macroglobulin receptor

(LRP). which is the closest relative to megalin and has four

structurally similar LBD (8,22,23). LRP is also capable of

binding multiple ligands (lipoprotein lipase, urokinase-type

plasminogen activator: plasminogen activator inhibitor type 1

complex, pro-urokinase-type plasminogen activator, a2-mac-

roglobulin and RAP), and all of these ligands have been shown

to bind to LBD II (24-26), but not to LBD I, III, or IV. Only

RAP has been shown to bind to other LBD (III and IV) (25,27).

Thus, the functional contributions of LBD I, III, and IV in

megalin and LRP are not yet clear.

In summary, in this study we show that all four clusters of

ligand-binding repeats in megalin contain pathogenic epitopes

capable of inducing passive HN. Additional studies are needed

to determine the nature of the epitopes in LBD I, III, and IV.

It is hoped that this information will provide insight into the

mechanism of passive HN and human membranous nephro-

pathy that can eventually lead to rational therapeutic interven-

tions.

AcknowledgmentsThis work was supported by National Institutes of Health Grant

DK17724 (to Dr. Farquhar) and the SFB 05, Project 07 from the

Fonds zur Forderung der Wissenschaftlichen Forschung (to Dr. Ker-jaschki). Dr. Yamazaki was supported by a fellowship from the

Uehara Memorial Foundation (Toshimaku, Tokyo, Japan). We thank

Nicki Watson for technical assistance with immunofluorescence.

ReferencesI . Heymann W, Hackel DB, Harwood 5, Wilson SOF, Hunter JLP:

Production of the nephrotic syndrome in rats by Freund’s adju-

vants and rat kidney suspensions. Proc Soc E.vp Biol Med 100:

660-664, 1959

2. Farquhar MG, Saito A, Kerjaschki D, Orlando RA: The Hey-mann nephritis antigenic complex: Megalin (gp330) and RAP.J Am Soc Nephrol 6: 35-47, 1995

3. Farquhar MG, Kerjaschki D. Lundstrom M, Orlando RA: gp330

and RAP: The Heymann nephritis antigenic complex. Ann NY

Acad Sci 737: 96-1 13, 1994

4. Kerjaschki D, Farquhar MG: Immunocytochemical localization

of the Heymann nephritis antigen (gp330) in glomerular epithe-

hal cells of normal Lewis rats. J Exp Med 157: 667-686. 1983

5. Kerjaschki D, Farquhar MG: The pathogenic antigen of Hey-

mann nephritis is a membrane glycoprotein of the renal proximal

tubule brush border. Proc Natl Ataci Sci USA 79: 5557-5561,

I 982

6. Kerjaschki D, Noronha-Blob L, Sactor B, Farquhar MG: Mi-

crodomains of distinctive glycoprotein composition in the kidney

proximal tubule brush border. J Cell Biol 98: 1505-1 5 13, 1984

7. Zheng 0, Bachinsky DR. Stamenkovic I: Organ distribution in

rats of two members of the low-density lipoprotein receptor gene

family, gp330 and LRP/a2MR, and the receptor-associated pro-

tein, (RAP). J Histochen, Cytochem 42: 53 1-542, 1994

8. Saito A, Pietromonaco S. Loo A, Farquhar MG: Complete don-

ing and sequencing of rat gp330/”megalin,” a distinctive member

of the low density lipoprotein receptor gene family. Proc Nail

AcadSci USA 91: 9725-9729, 1994

9. Orlando RA, Kerjaschki D, Kurihara H, Biemesderfer D, Far-quhar MG: Op330 associates with a 44 kDa protein in the rat

kidney to form the Heymann nephritis antigenic complex. Proc

NatlAcadSci USA 89: 6698-6702, 1992

10. Biemesderfer D, Dekan 0, Aronson P. Farquhar MG: Biosyn-thesis of the gp330/44-kDa Heymann nephritis antigenic corn-

plex: Assembly takes place in the ER. Am J Phvsiol 264: FlOl 1-

Fl020, 1993

1 1 . Orlando RA, Farquhar MG: Cellular trafficking of megalin(gp330) and the receptor associated protein (RAP). Mol Biol Cell

5[Suppl]: l87a, 1994

12. Orlando RA, Kerjaschki D, Farquhar MG: Megalin (gp330)

possesses antigenic epitopes capable of inducing passive Hey-

mann nephritis independent of the nephritogenic epitope in RAP.

J A,iz Soc Nephrol 6: 61-67, 1995

1 3. Pietromonaco SD, Kerjaschki S. Binder R, Ullrich R. Farquhar

MG: Molecular cloning of a cDNA encoding a major pathogenicdomain of the Heymann nephritis antigen gp330. Proc Nail Acad

Sci USA 87: 1811-1815, 1990

14. Kerjaschki D, Ullrich R, Diem K, Pietrornonaco 5, Orlando RA,

Farquhar MG: Identification of a pathogenic epitope involved in

the initiation of Heymann nephritis. Proc Nail Acad Sci USA 89:

11179-11183, 1992

15. Huang I, Makker SP: Role of receptor-associated 39/45 kD

protein in active Heymann nephritis. Kidney Jut 47: 432-441,

I 995

16. Kerjaschki D, Ullrich R, Exner M, Orlando RA, Farquhar MG:

Induction of passive Heymann nephritis with antibodies specific

for a synthetic peptide derived from the receptor-associated pro-

tein. J Exp Med 183: 2007-2015, 199617. Saito A, Yamazaki H, Rader K: Mapping rat megalin: The

second cluster of ligand binding repeats contains a 46 amino acid

pathogenic epitope involved in the formation of immune deposits

in Heymann nephritis. Proc NatI Acad Sci USA 93: 860 1-8605,

1996

18. Raychowdhury R, Zheng 0, Brown D, McCluskey RT: Induction

of Heymann nephritis with a gp330/megalin fusion protein. Am JPathol 148: 1613-1623, 1996

19. Farquhar MG: Molecular analysis of the pathological autoim-

mune antigens of Heymann nephritis. Am J Pathol 148: 133 1-

1337, 199620. Orlando RA, Exner M, Czekay R-P: Identification of the second

cluster of ligand-binding repeats in megalin as a site for receptor

ligand interactions. Proc Nail Acad Sci USA 94: 2368-2373,

1997

21. Kerjaschki D, Exner M, Ullrich R, Susani M, Curtiss LK, Wit-

ztum JL, Farquhar MG, Orlando RA: Pathogenic antibodies

1644 Journal of the American Society of Nephrology

inhibit the binding of apolipoproteins to megalinlgp330 in pas-

sive Heymann nephritis. J Cliii Itivest 100: 2303-2309, 1997

22. Kounnas MZ, Steingrimur 5, Loukinova E, Argraves KM.

Strickland DK, Argraves WS: An overview of the structure and

function of glycoprotein 330, a receptor related to the a2-mac-

roglobulin receptor. Aizii NY Acad Sci 737: 1 14-1 23, 1994

23. Krieger M. Herz I: Structures and functions of multiligand Ii-poprotein receptors: Macrophage scavenger receptors and LDL

receptor-related protein (LRP). Anna Rev Biocliem 63: 601-637,

1994

24. Moestrup 5K, Holtet TL, Etzerodt M: Alpha 2-macrogbobulin-proteinase complexes, plasminogen activator inhibitor type- I -

plasminogen activator complexes, and receptor-associated pro-

tein bind to a region of the alpha2-macroglobulin receptor

J Am Soc Nephrol 9: 1638-1644, 1998

containing a cluster of eight complement-type repeats. J Biol

Chein 268: 13691-13696, 1993

25. Willnow TM, Orth K, Herz I: Molecular dissection of ligand

binding sites on the low density lipoprotein receptor-related

protein. J Biol Chein 269: 15827-15832, 1994

26. Horn IR, Moestrup 5K, van den Berg BM, Pannekoek H, Nielsen

MS. van Zonneveld Al: Analysis of the binding of pro-urokinaseand urokinase-plasminogen activator inhibitor- I complex to the

low density lipoprotein receptor-related protein using a Fab

fragment selected from a phage-displayed Fab library. J BiolCheat 270: 11770-1 1775, 1995

27. Bu 0, Rennke 5: Receptor-associated protein is a folding chap-erone for low density lipoprotein receptor-related protein. J Biol

Cheat 271: 22218-22224, 1996