Local extrahepatic expression of complement genes C3, factor B, C2, and C4 is increased in murine lupus nephritis

Local extrahepatic expression of complement

genes C3, factor B, C2, and C4 is increased in

murine lupus nephritis.

J Passwell, … , H U Beuscher, H R Colten

J Clin Invest.

1988;

82(5)

:1676-1684.

https://doi.org/10.1172/JCI113780

.

Systemic lupus erythematosus (SLE) is associated with the presence of complement

proteins and immune complexes in affected organs. Since complement proteins are

synthesized in hepatic and extrahepatic sites, we studied a murine model of SLE to

ascertain the relative importance of local and humoral (liver) synthesis of complement. C3,

C4, and C2 mRNA increase in kidney coincident with the development of nephritis in the

MRL lpr/lpr mouse, a strain that spontaneously develops SLE. Two factor B messenger

RNA transcripts are expressed in kidney and intestine; SLE nephritis is associated with

decrease in the long factor B mRNA and increase in the short form. Increased local

synthesis of C3 and B protein and a concomitant glomerular and renal interstitial

macrophage infiltrate paralleled the increase in mRNA content in the (lpr/lpr) mice. In

addition to kidney, an increase in C3, C4, C2 and factor B mRNA was noted in the lung,

heart and intestine and to a lesser extent in liver of (lpr/lpr) in comparison to the MRL (+/+)

animals. These results suggest that in SLE local expression of complement genes plays a

role in the pathogenesis of chronic glomerulonephritis and in the autoimmune arteritis of

other organs.

Research Article

Find the latest version:

Local Extrahepatic Expression

of

Complement

Genes

C3,

Factor

B, C2,

and

C4

Is

Increased

in Murine

Lupus Nephritis

Justen Passwell, GeorgeF.Schreiner,* MasaruNonaka,Horst U.Beuscher,andHarveyR.Colten

EdwardMallinckrodtDepartment ofPediatrics, and *Departments ofInternalMedicine andPathology, Washington University School ofMedicine, St. Louis, Missouri 63110

Abstract

Systemic lupus erythematosus (SLE) is associated with the presence of complement proteins and immune complexes in

affectedorgans. Sincecomplementproteinsaresynthesizedin

hepatic andextrahepatic sites,we studiedamurine model of

SLEtoascertain therelativeimportanceof local and humoral

(liver)synthesis of complement. C3, C4, and C2 mRNA

in-crease inkidney coincidentwith thedevelopment ofnephritis

in the MRL lpr/lpr mouse, a strain that spontaneously de-velops SLE. Two factor B messenger RNA transcripts are

expressed inkidneyandintestine;SLEnephritisis associated with decrease in the long factor BmKNAandincrease in the shortform. Increased localsynthesisof C3 and I1

protein

anda

concomitant glomerular and renal interstitial macrophage in-filtrate paralleled the increasein mRNA contentinthe(lpr/

lpr) mice. Inadditiontokidney,anincrease inC3, C4,C2 and factor B mRNAwasnoted in thelung,heart andintestine and

toa lesserextentinliver of(Ipr/lpr)incomparisontothe MRL

(+/+)

animals.

These results suggest that in SLElocal expression of

com-plement genesplaysarole in thepathogenesis ofchronic

glo-merulonephritis and in the autoimmune arteritis ofother

organs.

Introduction

The MRL-LP lpr/lpr inbred mouse strain

consistently

de-velops a spontaneous murine

systemic

lupus

erythematous

(SLE)'

syndrome withsevere

glomerulonephritis

and evidence

of arteritisin other organs.

Lymph gland

hyperplasia

and

ar-thritis are also common clinical features.

Serologic

studies

showaprogressiveincreasewithage,of

immunoglobulin

con-centrations,

circulating

immune

complexes,

anti-DNA and other

autoantibodies,

anda decrease in totalserum

comple-menthemolyticactivity

(1,

2).

Immune cellinfiltrationintoaffected organs is particularly

prominent in this murine model of SLE. The expression of

Dr. Passwell was on sabbatical leave from the Sackter School of

Medi-cine,TelAviv,Israel.

Address reprintrequests to Dr. Colten, Department ofPediatrics, WashingtonUniversity School of Medicine, 400SouthKingshighway, St.Louis,MO 63110.

Received forpublication 8February 1988 and in revisedform 20 June1988.

1. Abbreviations used in thispaper:SLE,systemic lupus erythema-tosus.

autoimmunity is associated with theemergenceofalymphoid

cell population of unusual phenotype (Lyl+-, dThy-l+,

L3T4-) (3). Abnormal immunoregulation by T cells is clearly implicated in the pathogenesis ofthe diseasesince depletion of

T cells by monoclonalanti-T cellantibody (4), thymectomy (5),allogeneic bone marrowtransplantation (6),or treatment

with cyclosporin (7) can prevent or ameliorate the systemic

autoimmune disease.

Inthe MRL lpr/lplr mice,asin the human disease, deposits

ofimmunoglobulin andcomplement components, particu-larlyC3, are identified in the kidney with progression of

dis-ease. Activated complement proteins maycause tissueinjury directlyorrecruitinflammatory effectorcells(8).

There is increasing evidence that macrophages have an

important role in the pathogenesis ofglomerulonephritis. In

ratand rabbit experimental models of glomerulonephritisand

in the NZB model ofmurinelupus, mononuclear phagocytes

are recruited intotheglomerulus duringthe development of inflammation (9-1 1). The functional role of these cells may be regulationoflocal immune response (they express the Ia

anti-genand react with lymphocytes in a genetically restricted

manner

[12]),

clearance of immune complexes (13),

enhance-ment of tissueinjury byrelease ofproducts suchas prostaglan-dins and leukotrienes (14) and coagulant factors (15), all of

which have been identified in isolated glomerular cultures fromratswith glomerulonephritis.

Thecomplement proteins inserumaresynthesized by

he-patocytes but extrahepatic expression ofthese proteins in

mononuclearphagocyteshasbeen demonstratedaswell(16).

Inseveralexperimental models, hepatic and extrahepatic regu-lation ofcomplement gene expression have been shown to be underindependent control (17, 18). To ascertain whether local expressionofcomplement proteins may also playarole in the

immunopathology oflupus nephritis, weexamined the

con-centration of mRNA forC3,factorB,C4 and C2inthe kidney

andother organsof the MRLlpr/lpr and the congenic MRL

+/+ mousestrains attimed intervals. Amarked increase in mRNA forthese complement proteins in the lpr/lpr kidney

and inother organs affected by the autoimmune disease was observed concomitant with thedevelopment ofSLE. The in-creased mRNA content isreflected in increased synthesisof complementinthenephritickidney. Finally,using the F4/80 monoclonal antibody that identifies a specific mouse macro-phage antigen (19), a glomerular and interstitial macrophage

infiltratecorrelated with progression of the kidney disease.

Methods

Mice. MRL(lpr/lpr)and MRL(+/+)male micewerepurchasedfrom the JacksonLaboratory (Bar Harbor, Maine). Theyweremaintained

on aregulardiet andgivenwaterad lib. Urine forprotein analysiswas

collected fromgroupsof mice in metaboliccages.At four timepoints, 1676 Passwelletal.

J.Clin. Invest.

©TheAmericanSocietyforClinical Investigation, Inc.

0021-9738/88/11/1676/09 $2.00

i.e.,6, 12, 18,and 22wk, micewerekilled by cervical dislocationand bledby cardiac puncture, and theliver,kidneys, spleen,lungs, intes-tine,and heartwererapidlyremoved.Smallslices ofliverandkidney were obtained and stored in 10% formalin for histologyand quick frozeninliquid nitrogen for immunohistology.Theremainder ofthe organswerefrozeninliquid nitrogen andstoredat-70'Cuntil pro-cessedfor isolation oftotal RNA. Bloodwascollected intotubeswith

EDTAandkeptonice,and the separatedplasmawasstoredat-70'C untilanalyzedforC3,anti-DNAantibody,andcreatinine concentra-tions. Urineprotein content wasdetermined byaquantitative

sulfo-salicylicmethod(20).C3 concentration intheplasmawasquantitated by immunodiffusion usinga rabbit anti-mouseC3 antibody kindly

providedby Dr.Natsuume-Sakai (CancerResearchInstitute,

Kana-zawaUniversity,Japan). Plasmafrom 8-wk-oldBalb/c mice (C3 50

mg/dl)was used asstandard. Plasma creatininewasmeasured bya routinemethodusinganautoanalyzer.Kidneyandliversectionswere

embedded inparaffin, stained withhematoxylinand eosin and

peri-odic acid-Schiffandreviewedina"blinded" fashion.

Topermitgreaterspecificityinscoring progressionofrenaldisease, histologywasevaluatedfortheseverity of changesin threecategories: (a)acuteglomerulonephritis (GN),(b) glomerulosclerosis (GS),and(c)

interstitialinflammation (IN)accordingtothefollowing scoringscale:

Score: + ++ +++ ++++

GN Focal segmen- Diffuse mes- Diffuse necrotiz- Diffuse GN tal mesan- angiopro- ingprolifera- with fibro-gioprolifer- liferative tiveGNwith cellular ative disease focalcrescents crescents in

changes >50%of

glomeruli

GS Focal segmen- Focal glomer- Diffuse,seg- Diffuse glo-talsclerosis ularhya- mentalglo- merular

linosis meruloscler- hyalinosis osis

IN Patchy non- Diffuse inter- Necrotizing in- Diffuse inter-necrotizing stitial infil- terstitialinfil- stitial fi-mononu- trate trateswithin- brosis and clear cell terstitial tubular

at-infiltrate edemaand rophy

focaltubular

atrophy

TheF4/80hybridoma (kindlyprovided byDr.HowardGray, Uni-versity ofColorado, Denver) was culturedinRPMI1640+10%

heat-inactivated fetal calfserum. The monoclonal antibody was

precipi-tatedfromthe culture supernatantby50%ammonium sulphateand

dialyzed for48 hagainstPBS.

Immunoperoxidase labeling ofrenaltissue. Frozensections (6

Ism)

ofkidneys harvested from control and experimental animals were fixedinchilled(40C)acetonefor 10 min andair-dried. Endogenous peroxidase activitywasinhibitedby immersion for 10 min ina

solu-tionconsistingof 40partsmethanol, 9 partsH20, and I partH202

(30%). Endogenousbiotinactivitywasinhibited bysequential30-min exposurestoavidinDandbiotinblocking solutions (Vector

Laborato-ries, Burlingame, CA).Thesectionswerethenplacedinrabbitserum blocking solution (Zymed Laboratories,SanFrancisco,CA),followed byF4/80 antibody (diluted1:10 in PBScontaining3.5%bovineserum albumin) for30 min. Afterwashing, biotinylated rabbit anti-ratIgG

(Vector Laboratories) affinity purified and adsorbed againstmouse IgG,wasaddedfor 15 min.Afterwashing,theslidesweresequentially exposedtostreptavidin-peroxidasefollowed byaminoethylcarbazole andhydrogenperoxide,assupplied by Zymed Laboratories.The sec-tionswerecounterstained withhematoxylinand mountedunder cov-erslips.

Cellswerescoredpositiveforexpressionof theF4/80 antigen if

theydemonstratedarim_patternof membranelabelingwith the chro-magen.Interstitialcellsareexpressedaspositivecells per 100 tubule

cells,which wereuniformly negative,in ordertominimizesampling

error. 15high-powerfieldswereread per section. Allglomerulion a

section(10-20 glomerulipersection)werescored forpositivelabeled

cellsand theresults calculatedon aperglomerularbasis.

mRNAisolation.Atearlytimepoints (6and 12_wk)twoanimals and at 18 and 22 wk four animals from each strainwere used for mRNAstudies.Organswerepulverizedin_{liquid nitrogen}witha

mor-tarandpestleandlysedusing4 M_{guanidium thiocyanate}asdescribed

previously (21).RNAwasseparatedfromthelysates by

ultracentrifu-gation (35,000 rpmat 20'Cforatleast 12 husingaSW50.1 rotor, BeckmanInstruments,Inc.,Fullerton,CA) througha5.7 M cesium

chloridegradient. The RNA pelletwaswashed in 70% ethanol, dis-solved in water, reprecipitated in sodium acetate andethanol, and

redissolvedinwater.RNAcontent wasdeterminedbymeasurementof

absorptionat260nm.

Northern blotanalysis.RNAwasseparated by 1%agarose

formal-dehyde gelelectrophoresisrun forat least 12 h,then transferred to

nylonmembranes(Hybond-N, AmershamCorp.,ArlingtonHeights,

IL)byblottingandfixedbyexposureto anultravioletlightsourcefor5 min. cDNA probeswerelabeledwith [32PjdCTP bynicktranslation

and Northern blot hybridization was performed as described

pre-viously (22). Prehybridization and hybridizationwith specific RNA

probes (seebelow)weredoneat55-60'C in 50%formamide,50 mM

sodiumphosphate pH 6.5,0.8%NaCl,1 mMEDTA,0.5%Denhardt's

solution, 500 ug/ml denatured salmon spermDNA,and 500/Ag/ml

yeast RNA(SigmaChemical_Co.,St.Louis, MO).Filterswerewashed

at650Cin50mMNaCl,20 mMsodiumphosphate pH 6.5, 1 mM

EDTA, and0.1% SDS. Semiquantitative analysisofthe amountof increasein mRNAwasquantitated bylaser_densitometryof the

auto-radiographs.

Probes. ForC2 mRNAthe 800bp PstI _fragment ofthepC2M

mousecDNAclonewasused(23).The C4probe,a 1.4-kbfragmentof

the clonepFC4/5.4 (24), and achickena-actin cDNA (25), which

hybridizesto,B and yactin,wereused.Synthetic oligomersofmouse

C4 (CCTGCAGCATGTTGTGGTTG) and Slp

(AATTCAC-CATGTCCCGGACC) werekinaselabeled with [32P]ATP. A 1.3-kb

Pst I fragmentfrom mouse factorB cDNA clonepBmB2 (26)was

subclonedintoa_PGEM4standardcloningvectorandtransformed into

Escherichiacoli. Transcriptionofthe RNA probe wasachievedby

linearizingthevectorwith SmaI andlabelingwith[32P]UTPfollowing

the protocolofthesupplier (Promega, Madison, WI). A mouseC3 cDNAclone(kindly provided byDr. M.Takahashi,Kanazawa,Japan)

containinga2.7-kbinsertwas_digestedwithPst I.Fivedistinctbands

wereidentified.An800-bpfragmentwassubclonedintoPGEM4and

usedfortranscriptionof labeledRNAasdescribedabove.

Biosynthesis,immunoprecipitation,and SDS PAGE.Kidneysfrom

6- and 18-wk-old MRL(+/+)and MRL(lpr/lpr)micewereharvested

intocoldPBS with 1.5 mMKClandheparin (10 U/ml)andkepton ice.Coronalkidney slicesapproximately0.2 X 1 cm were cutwitha

razorblade and washed twiceandincubated for20 minat37°Cin

methionine-free DME (Gibco, Grand Island, NY) tissue culture

media.Thetissue slicesweretransferredtothe wells in separate 16-mm

tissuecultureplates, medium wasremoved, and 0.5 ml of medium

containing [35S]methionine (0.5 mCi/ml, sp act - 1,000 Ci/mmol;

AmershamCorp.) wasadded. After 1 h thekidneysliceswere har-vested inlysisbuffer (10 mMEDTA, 1% TritonX100,0.5%

deoxycho-late inPBS, pH 7.4,in the presence of 2mM

phenylmethylsulfonyl-fluoride and 250 gg/ml soybean trypsin inhibitor[Sigma Chemical

Co.])and frozenat -70°C.Intracellular lysatesofkidneywere ob-tainedby repeatedfreezethawingandvigorousvortexingof the organ

slices.TotalproteinsynthesiswasdeterminedbyTCA_{precipitation}of

aliquotsof the intracellular lysatesandextracellular media.Samples

werepreclearedwithformalin-fixedStaphylococcusaureusand

succes-siveimmunoprecipitationswith anti-mouseC3(Cappel Laboratories,

Table I. Urineand Plasma FindingsintheMRLStrainswithIncreasing Age

MRL (+/+) MRL (Ipr/lpr)

Age (wk) 6 12 18 22 6 12 18 22

Urineprotein(mg/ml) 0.85 1.73 3.15 1.77 4.87 6.39 9.67 8.17 Plasmacreatinine* (mg/ml) 0.2 0.3 0.2 0.2 0.2 0.2 0.2 0.4 PlasmaC38(mg/dl) 70±5 54±18 80±7 106±15 61±6 48±6 77±3 76±4

Anti-DNAantibody8(1/titer) neg 40 40 40 20 320 10,240 5,120

*_{Averageoftwo to threeindividual miceateachtime point. $ Averageofthreetosevenindividual miceateach timepoint. § Samples from}

onemouseateachtimepoint.

mouse antigen (Atlantic Antibodies, Scarsborough, ME)were done exactly as described previously (17). The immune complexes were precipitated with formalin-fixed Staphylococcus aureus (IgSorb En-zyme Center, Malden, MA), washed extensively, and subjected to SDS-PAGE underreducing conditions (5%f3-mercaptoethanol). The

gelsweretreated withEn3hance, dried undervacuumafter which auto-radiographywasperformed.

Results

Clinicalfindings.

By 18 wk of age the mice of the lpr/lprstrain

manifested ulcerated skin lesions, lymphadenopathy, and

lacked their normal coat lustre. There was a progressive

in-crease in proteinuria (Table I). Plasma C3 concentrations,

which were higher than that of the BALB/c mouse strain, did notdecreasewith the development ofrenaldisease, but were less than the C3 levels in the+/+ strainat22 wkof age. There was a markedprogressiveelevationofanti-DNA antibodies in plasmaofthelpr/lpr strainwithonly a modest increase in the

+/+ strain.Thecreatinineconcentrationwasslightlyelevated

onlyatthelasttimepoint (22 wk)for thelpr/lprmice.

Renal histology.At6 and 12wk,control MRL

(+/+)

kid-neys showed no evidence ofpathology. At 18 and 22 wk, a mild mesangioproliferative glomerulonephritis without evi-dence ofcrescents, necrosis, sclerosis, orinterstitial nephritis wasobserved.Incontrast, the MRL lpr/lpr mice developed a mild increase in mesangial hypercellularity as early as 6 wk, which evolved into a diffuse, moderate mesangioproliferative

glomerulonephritis by age 12 wk. Enlarged, reactive parietal

epithelial cells were observed but there were no crescents.

There was noevidence of interstitial nephritis. By age 18 wk,

thelpr/lprmicedevelopedadiffuse, proliferative, necrotizing

glomerulonephritis with crescents involving at least 50% of

glomeruliand a moderate interstitialnephritis.By age 22wk,

virtually allglomerulidisplayed crescentformationwith

peri-glomerularfibrosisandworsening interstitial nephritis (Table

II). Scattered sclerosed glomeruli were observed.

Immunohistology with F4/80 antibody.Control

(+/+)

and

experimental(lpr/lpr)kidneyswerelabeledfor the presence of

macrophages at 6 and 18 wk ofage.Control kidneysat6 wk

hadvirtuallynoglomerularmacrophagesandasmall number

ofresident, interstitial macrophages (Table II). Aged control

miceshowed amildglomerularinfiltration withmacrophages

and an insignificant increase in interstitial macrophages. In

contrast, lpr/lpr mice displayed increased glomerular

macro-phages at 6wkandamarkedincrease in both glomerular and interstitial macrophages by 18 wk (Fig. 1 and Table II). At this

timepoint, theinterstitial infiltrate wasdiffusely peritubular,

confined to the cortex, and not associated with tubular

ne-crosis.

Expression

of

mRNA in

kidneys

and other organs

by

Northern blot analysis. As early as 12 wk there was a slight,

then

progressive,

increase in

expression

ofboth C3 and C4

mRNA in kidneys ofthelpr/lpr strain; in the control MRL

+/+ strain,therewas asmall,butdefinite increase in C3 and

C4 mRNA with

advancing

age

(Fig.

2).Both C2 and factor B

mRNA are constitutively expressed in thekidneys in young

animals ofboth MRLstrains. A clearprogressive increasein

C2 mRNAduringthedevelopment of glomerulonephritiswas seenin thelpr/lpr

strain,

withonlyaminimal changenoted in

the +/+ animal. FactorB mRNAis expressedin themurine

kidneyas adoublet(reference17andFig. 2),thesmallerofthe

Table II. RenalHistologyintheMRLStrains

MRL(+/+) MRL(lpr/lpr)

Age(wk) 6 12 18 22 6 12 18 22

Glomerulonephritis* - - + + + ++ ++++ ++++

Glomerular sclerosis* - - - + - - ++ ++

Interstitial inflammation* - - - - + ++ +++ ++++

F4/80Stain

Glomerular

macrophagesi

(mean±SE) 0.1±0.1 ND ND 0 2.3±0.7 ND 12.0±1.5 ND

InterstitialmacrophagesO(mean±SE) 0.3±0.1 ND ND 0.9±0.4 0.8±0.2 ND 29.0±2.5 ND

ND,notdone. *Description of scoringsystemis given inMethods. *Macrophages (F4/80positive)/glomerulus(allglomeruliin asection

[10-20]

werecounted). § Macrophages (F4/80 positive)/100tubule cells.

a0

_t

f

.

;~~~ ~ ~

-~~~~~~~~

4

.7

4~~4

a

_

. 7

4',

*: Jts

'-*'s r

,# ., _., r ,t , b. Zs_ ': s w tt'a!;-t 5

*4 s i .: ^

5*Ati*r ';. >4 ,!t ,.4 g*ft;t ; C.S' * a,:

*.'t eriitE...,+-*/.StRtff ;,,Zc'- T;}

,.,vX,,4 ,tr<^,Xi jtC, ,..,, .^,..v....

F S>;,usltziS ..r

,,, ,E,,4,wej8j ts, tws :^8zi ,f

...

,.,,*

;, ,.

_Lt--:

_*w,

&*t FX

_..-.

v} e

' iA * " .' sb. +

, u .,

;J) Ss b 4,t

- 'a'= s:

t iX

:'_.,,\:

't4^! _,.

t, .:, .J

4._."

.1

,

..1

j4

a -~

Z

i

o

(C~~~~~~~~~l

~~~~t~~r

e

cis

.Is

+

~v;~- -.00

J

-

i

CIS

A-

<t g * , I r

eQ Z,

+00

4^D

3

I-cd

r)c

-XQ

*~ *

8l a

4 0 c-Ywj

o a

o * CuQ

c

~-I-.CA

soc 0CuL

o 3

XL £

to,

4

.11_*.A"I"

.41 _....A :, `

.-t.

t

*jt ..:.i

k:''

t.

,4~ .,W,3L

.:.Zl.,C, W.""'

.% _1. 4

I

ss, ,o,. ..

.Ap-Ai&,

"* .- 'PWWI 4

C3

Sto

s

0

Factor B

6 12 18 22

6 12 18 22

MRL

MRL

(+/+)

(Ipr/I

pr)

Figure2.Northernblot analysis of C4, C3, C2, and factor B mRNA inkidney of MRL+/+and MRL lpr/lpratages6, 12, 18, and 22

wks. 10,gof total RNA loaded in each lane. Representative

experi-mentfromtwoanimals (6, 12 wk) and four animals (18, 22 wk) two mRNA species being - 0.3 kb shorter. Both factor B mRNAbandsarepresent inapproximately equal

concentra-tionsin theyoung animals. Inthe MRL +/+ strain asmall

6 12 18 22

MRL

(+/+)

6 12

18 22 weeks

MRL

(

Ipr/lpr)

from each strain blotted and probedonreplicate (threeto fourtimes) Northernanalyses. Both by ethidium stain and actinmRNA con-tent, there were nosignificantdifferences in amount of RNA loaded in each lane.

increase in eachband is notedasthe miceage.However,inthe

MRL lpr/lpr strain there is a significant and consistent

in-crease in concentration of the lower band, while the upper

10

---.--- _---o

0.9.

0.5

32

10

_-

I--II

5

~~~~~~~~~~~~~~12

1.0

0.Y

0.8

0.7

0.4

,.

---2I1 '7

Figure 3. Tracing of laser densitometry of factor B mRNA in thekidney of the MRL+/+and MRLlpr/lpr strainatages6, 12, 18,and 22 wk

(lefttoright). ,ugof RNAwasloaded in each lane.Upper tracingsfrom(+/+)andlowertracingsfrom(lpr/lpr).

1680 Passwelletal

C4

C48s

28s-C3

18s-Afib.

_

C2

28s-

18s-Factor B

a P@ _S

1.1

1.S

0.9

S.,

0.7

0.6

0.5

0.4

0.1

0.7

0.0

0.5

I

band decreases. It was not possible to accurately separate the bands for scintillation spectrometry, but laser densitometry clearly showed the presence of the double band, a twofold increase of the lower band and a > 80% fold decrease of the upperband withadvancingage inthelpr/lprstrain (Figs. 2 and

3). Ofall theother organsexamined,only theintestine displays

the double-bandedfactorB mRNA.

Synthetic oligomers homologous to C4 and Slp sequence were used todetermine whether the Slp sequence of the C4 genewasexpressed in thekidney of this mousestrain. Only the sequencehomologous to C4 hybridized to kidney mRNA at 18 and22 wkin the lpr/lpr strain. No hybridization to kidney mRNA wasobservedwith the Slp oligomer, but a strong signal wasobtained when,for a control, a plasmid containing the Slp cDNA wasprobed with the Slpspecificoligonucleotide (results notshown).

Theexpressionof complement genes in other organs of the

+/+ and lpr/lpr strain was estimatedat 18 wk by Northern

blotanalysisasshown inFig.4. Forliver,the sourceofplasma

complement, the mRNA content of C2, C3, C4, and factor B in lpr/lpr was approximately twofold that of each in the

+/+

strain(Fig.4). There were no markeddifferences between the

twostrainsinexpressionof these mRNA species in the spleen.

However,acomparisonof lung, heart and intestine in the two

strainsrevealedsubstantially more mRNA in the lpr/lpr than

in the+/+ strainat 18 wk.

Loading ofthegels for Northernblotting was initially

esti-mated by comparing ethidium bromide staining of the 28S and 18S RNAbands. Thechicken a-actincDNAprobe which

cross-hybridizes with both

_#

and

T-actin

was used to more

C4

04,.

28s-

18s-L

V

K

S

Vv

Lu

V

I

\

H

precisely estimate comparability of the data. Actin mRNA concentrations varied among the different organs, however, similar amounts were present for each organ in comparing the two mousestrains(Fig. 5).

Proteinsynthesisinkidney. Miceages6 and 18 wkofthe congenic strains were used to study the biosynthesis of factor B and C3. Tissues were culturedfor 1 h afteraninitial 20-min labeling period. Only the kidney specimens derived from the 18-wk oldlpr/lpr strain synthesizedsignificantamountsof C3 and factor B protein (Fig. 6). The majority of labeled C3 pro-teinwasexpressedaspro-C3 withtrace amountsof native C3 detectable,asexpected, in the intracellular compartment.

Discussion

We have demonstrated a progressive increase in mRNA cor-responding to the complement genes C3, C4, C2, and factor B in thekidney ofthe MRL lpr/lpr mousestrain coincident with the development of glomerulonephritis and interstitial nephritis. This mRNA is translated, since kidney organ cul-turesderived from micewith glomerulonephritis manifested

increasedbiosynthesis of C3 and factor B protein.Thisis the

first demonstration of local regulation of complement genes and in situ synthesis of complement protein in an organ

af-fected by a naturally evolving autoimmune disease. Similar

increases in concentration of mRNA of these complement

components in other organs thatmanifest autoimmune

arter-itiswereobserved, but there waslittle change in C3, C4, C2,

and factorB mRNAin liver, although liver is the source of

plasma complement proteins. These marked differences in

ex-C3

I

L

K

S

Lu

I

A AA A AA

.AA

AA A

*

0C3

H

A

C4,.

28s-C2Dw

18s-.^~~~~~~~~~~~~~~.

..

s :.

C2

Figure4. Northern blotanalysisofC4,C3,C2,andfactorB mRNA inorgansofMRL+/+and MRLlpr/lprat18wkofage. 10,g of

total RNAwasloaded in each lane exceptfor liverwhere 5gg/lane

ofRNAwasloadedineach.L,liver; K,kidney; S, spleen; Lu, lung;

-Factor B

Factor

B

I,intestine;H, heart. First sample of each pair from (+/+) and sec-ondfrom (Ipr/lpr)organs.Thisexperimentisoneof fourseparate

Northernblotsperformedoneach of threesetsof animals.

0

28s-

18s-L W JL JLJ11 A L I L L W I

L

K

S

Lu

I

H

pression ofcomplement mRNA at localsites of inflammation

suggestthatlocalcomplement productionmay play a rolein

thepathogenesis of theautoimmunepathology andthat local

regulatoryfactors modulate this effect.

Theidentificationof two forms offactorB mRNAin

kid-ney and intestine, but not in other organs, is an instance of

tissue specificityinexpressionof

this

complement gene. Only

C3

Factor B

Mr(x

10-3)

- -

200

I

-92.5

-

69

... ..

..

..

.4

...;

46

- i-

a

6

18

weeks

....

.e- W.

*...-.A..

...A...

. ....

...

_*....

.. ..

.. ... tbstz

....Of.-.

is

-L

JL

--a

6

18

Figure6.SDS-PAGE of immunoprecipitated[35S]methionine 1-h

pulse-labeledC3 andfactorBfromkidneyorgan culturelysatesof 6-and 1 8-wk old MRL+/+(left lane in each pair) andMkL lpr/lpr

mice(right lane in each pair). Molecular mass markers (D) are

indi-cated.

Actin

_.

ma

Figure5. Northern blotanalysis of actin

mRNA(probedwith chicken a-actin cDNA) in organs of MRL +/+ and MRL lpr/lpr mice strainsat18 wkofage.All lanes loaded with 10jigperlaneexcept

for liver with 5 jig per lane. L, liver; K, kidney;S,spleen; Lu, lung; I, intestine; H,heart.Leftlaneof eachpair (+/+)

andright(lpr/lpr).

the smaller ofthetwofactorB mRNAspecieswasupregulated, coincidental with advancing kidney disease, while the larger form decreased by >80%. A similar increase ofthe smaller

mRNA speciesin mouse kidney has been induced by intra-peritoneal injection of endotoxin (unpublished results).Other

genesexpressing different forms of mRNA inatissue specific mannerhave been described. One example of this is the ex-pression of a-amylase inmouseliverandsalivary gland. In this instance differences inexpression ofmRNAhave been attrib-uted tovariationin5'untranslated regions whichwerethought

to arisebyeither differentialtranscription and/orprocessingof

the mRNA (27).

Although increases in complement-specific mRNA and

protein synthesis duringtheevolution ofnephritiscorrelated

with increased numbers of macrophages, attribution of the observedchangesto aspecific cell type must remain specula-tive. That is, it is possible that the different factorB mRNA

speciesarederived from different cells within thesameorgan.

Indeed, such a mechanism has recently been described for

expression

of the humana-1

proteinase

inhibitorgene.

Macro-phagesand hepatocytes express ac-l proteinaseinhibitor, but

two mRNA species are found in macrophages (28), both of

whicharelarger than thehepatocellulara- 1 PI mRNA. In each

celldifferent promoter sequences have been identified.An al-ternativesplicing mechanism apparently accountsforthe two forms ofmRNA found in the macrophage (29). In immune

complex glomerulonephritis, hypercellularity can be

attrib-utedtomacrophageinfiltrationand toproliferationof

intrin-sicglomerularcells,particularly mesangial cells.Macrophages

(30)andfibroblasts(Y. Katz,F.S.Cole,R.C.Strunk,

unpub-lished results) synthesize factorB intissue culture, however, both express only the smaller of thetwofactorBmRNA

spe-cies. Further studies using in situ hybridization or isolated

kidney cell preparations will be needed to determine whether thetwo factorB mRNA speciesarederived from asingleor

multiple cell types. Finally,the two factorB mRNA species

expressedinintestinearealso modulatedduringprogression of SLEwithapatternqualitativelydistinct from that observedin

kidney. This observationsuggestseither differentcells express

1682 Passwelletal.

.... *: ::.

factor B in kidney and intestineor a more complex mecha-nism ofregulationof the two forms of B mRNA must exist.

The F4/80 membrane antigen is a specific marker of mousemacrophages (19). We have used monoclonal antibody to this antigen to demonstrate that in this mouse model of immune complex glomerulonephritis, macrophages infiltrate theglomerulus and the interstitium, where theyare the pre-dominant inflammatory cell. Macrophages produce comple-mentlocally and monokines mayact on adjacent cells. The regulation of complementbiosynthesisby the monokines IL- 1 and TNF in mouse and human hepatocytes has been well

documented. The effect of these monokines is exerted ata

pretranslational level,selectively upregulatingfactorB,C3 and C4 but not C2 (31-33). However, in vivo administration of rIL-1 increases renal concentrations of C2, C3 and factor B

mRNA (17). The distinct pattern in this SLE model of in-creased mRNAofall thecomplement componentswestudied suggests a common mediator of this upregulation. In vitro studies with recombinantIFN-yhave shown thatitenhances

themRNAandsynthesisof C2,B,C3and C4 in mononuclear

phagocytes (33). In this regard, it is of interest that in vivo

administration ofIFN-y aggravates, while ablation of T

lym-phocyte function ameliorates, progression ofmurine SLE (34, 35).

Inacute immunecomplex kidney disease, circulating im-munecomplexes aredeposited withinthesubendothelialand

mesangialregionsoftheglomerulus. Serum complement

con-centration decreases and local activation of the complement

cascade occurs (36). Activation of complement leads to

re-cruitment ofother inflammatory cells and enhanced

opsoni-zationofimmunecomplexes. Inaddition,components of the

classical complementpathwayinhibit immunecomplex

depo-sition and enhance solubilization ofimmune

complexes (37,

38). Hence, localproductionofcomplementmaysignificantly

modifytheevolution ofimmunenephritis. Our studies

dem-onstrate that increased local production ofcomplementalso

occursin thekidney coincident withthedevelopmentof

glo-merulonephritis and interstitial nephritis. Thus, insitu renal

synthesis of complement

proteins

may exacerbate orprotect

againstimmunecomplexmediated tissueinjury. Acknowledgments

We thank Ms. Barbara Pellerito for assistance in preparation of the

manuscript.

This work was supported in part by grants A124836, A124739,

DK36277,and HL37591 from the U. S. Public Health Service and a grantfromthe Samuel Jared Kushnick Foundation.

References

1.Andrews, B.S.,R. A.Eisenberg,A. N.Theofilopoulos, S. Izui, C. B.Wilson,P. J.McConahey, E. D. Murphy, J. B. Roths, and F. J.

Dixon. 1978.Spontaneous murinelupus-likesyndromes: clinical and

immunopathological manifestations in several strains. J. Exp. Med.

148:1198-1215.

2.Theofilopoulos,A.N., and F. J. Dixon. 1985. Murine model of

systemiclupuserythematosus.Adv.Immunol.37:269-390.

3. Weston, K. M., E. T. H. Yeh, and M.-S. Sy. 1987. Autoreactivity accelerates the developmentof autoimmunity and

lymphoprolifera-tioninMRL/Mp-lpr/lpr mice. J. Immunol. 139:734-742.

4.Seaman,W.E., D.Wofsy,J.S.Greenspan,and J. A.Ledbetter. 1983. Treatmentof autoimmune MRL/lpr mice with monoclonal

an-tibody to Thy-1.2: a single injection has sustained effects on lympho-proliferation and renaldisease. J. Immunol. 130:1713-1718.

5.Steinberg, A. D., J. B. Roths, E. D. Murphy, R. T. Steinberg, and E. S. Raveche. 1980. Effects of thymectomy or androgen administra-tion upon theauto-immune disease of MRL/MP-lpr/lpr mice. J. Im-munol. 125:871-873.

6.Theofilopoulos, A. N. 1982. Role of the thymus inmurine lupus and cellulartransfer of the disease. Arthritis Rheum. 25:726-733.

7. Mountz, J. D., H. R. Smith, R. L. Wilder, J. Patton Reeves, and A. D.Steinberg. 1987. CS-A therapy in MRL-lpr/lpr mice: ameliora-tion of immunopathology despite auto-antibody producameliora-tion. J. Im-munol. 138:157-163.

8. Couser, W. G. 1985. Mechanisms of glomerular injury in im-mune complex disease. KidneyInt. 28:569-583.

9. Schreiner, G. F., R. S. Cotran, U. Pardo, and E. R. Unanue. 1978. Amononuclear cell component in experimental immunological glomerulonephritis. J. Exp. Med. 147:369-384.

10.Holdsworth, S. R., T. J. Neale, and C. B. Wilson. 1981.

Abroga-tion ofmacrophage-dependent injury in experimental glomerulone-phritis in therabbit. J.Clin. Invest. 68:686-698.

11. Kimura, M., M. Nagase, A. Hishida, and N. Honda. 1987. Intramesangial passage of mononuclear phagocytes in murine lupus glomerulonephritis. Am. J. Pathol. 127:149-156.

12.Schreiner, G. F., J. M. Kiely, R. S. Cotran, and E. R. Unanue. 1981.Characterization of resident glomerular cells in the rat express-ingIadeterminants and manifesting genetically restricted interactions withlymphocytes.J. Clin.Invest. 68:920-931.

13. Boyce, N. W., and S. R. Holdsworth. 1986. Quantitation of the intrarenal uptake of immunoglobulin aggregates by macrophages in

diffuseproliferative glomerulonephritis. Clin. Exp. Immunol. 64:638-645.

14. Stork, J. E., and M. J. Dunn. 1985. Hemodynamic roles of thromboxane A2 and prostaglandin E2 in glomerulonephritis. J.

Pharmacol. Exp. Ther.233:672-678.

15.Wiggins, R. C., A. Glatfelter, and J. Brukman. 1985. Procoagu-lant activity in glomeruli and urine of rabbits with nephrotoxic nephritis.Lab.Invest. 53:156-165.

16. Alpert, S. E., H. S. Auerbach, F. S. Cole, and H. R. Colten. 1983. Macrophage maturation differences in complement secretion by marrow, monocyte, and tissue macrophages detected with an im-provedhemolyticplaque assay.J. Immunol. 130:102-107.

17. Falus, A., H. U. Beuscher, H. S. Auerbach, and H. R. Colten. 1987. Constitutiveand IL-1 regulated murine complement gene ex-pression is strain and tissue specific. J.Immunol. 138:856-860.

18. Ramadori, G., J. D. Sipe, and H. R. Colten. 1985. Expression andregulation ofthemurine serum amyloid A(SAA)gene in

extrahe-patic sites. J.Immunol. 135:3645-3647.

19. Lee, S. H., P. M. Starkey, and S. Gordon. 1985. Quantitative

analysis oftotal macrophage content in adult mouse tissues:

Immuno-chemicalstudies with monoclonal antibody F4/80. J. Exp. Med.

161:475-489.

20.Schreiner,G.,Cotran,R., Pardo, V., and Unanue, E. R. 1978. A mononuclearcell component in experimental immunological

glo-merulonephritis.J. Exp.Med. 147:369-384.

21. Chirgwin, T. M., A. E. Przybyla, R. J. MacDonald, and W. J. Rutter. 1979. Isolation of biologically active ribonucleic acid from

sourcesenriched in ribonuclease. Biochemistry. 18:5294-5299.

22. Thomas, P. 1980. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA.77:5201-5205.

23. Falus, A., E. K. Wakeland, T. J. McConnell, J. Gitlin,A. S. Whitehead, and H. R. Colten. 1987.DNApolymorphism ofMHCIII

genes in inbred andwildmouse strains.Immunogenetics. 25:290-298. 24. Nonaka, M., K. Nakayama, Y. D. Yuel, and M.Takahashi.

1985.Complete nucleotide and derivedamino-acidsequencesofthe

fourth component ofmouse complement (C4). J. Biol. Chem.

260:10936-10943.

Dugaic-zyk. 1980. Regulation of muscledifferentiation: cloning of sequences from alpha-actin messenger ribonucleic acid. Biochemistry.

19:5883-5890.

26. Sackstein, R., H. R. Colten,and D. W. Woods. 1983.

Phylo-genetic conservation ofa class IIImajor histocompatibility complex

antigen, factorB:isolationandnucleotide sequencing of mouse factor B cDNA clones. J.Biol. Chem. 258:14693-14697.

27. Young, R. A., D. Hagenbuchle, and U.Schibler.1981.A single mouse a-amylase genespecifies twodifferent tissue-specific mRNAs.

Cell. 23:451-458.

28.Perlmutter, D. H., R. M. Kay, F. S. Cole, T. H.Rossing,D. Van

Thiel,and H. R.Colten.1985.Thecellular defect in alpha 1 proteinase

inhibitor(alphaPI)deficiency is expressed in xenopusoocytesinjected with human liver RNA. Proc.Nati.Acad.Sc. USA. 82:6918-6921.

29. Perlino, E., R. Cortese, andG. Ciliberto. 1987. The human

a1-antitrypsin gene istranscribed from two differentpromoters in macrophages and hepatocytes. EMBO (Eur. Mol. Bio. Organ.) J.

6:2767-2771.

30.Perlmutter, D. H., and H. R.Colten. 1986. Molecular

immu-nobiology ofcomplementbiosynthesis:a modelof single-cell control

ofeffector-inhibitorbalance. Annu. Rev.Immunol.4:231-251.

31.Perlmutter, D. H., G.Goldberger, C.A.Dinarello,S. B. Mizel,

andH. R.Colten. 1986. Regulation of classIIImajor histocompatibil-ity complex gene products by interleukin 1. Science(Wash. DC).

232:850-852.

,32. Perlmutter, D. H., C. A. Dinarello, P. I. Punsal, and H. R. Colten. 1986. Cachectin/tumor necrosis factor regulates hepatic

acute-phasegeneexpression.J.Clin.Invest.78:1349-1354.

33. Strunk,R.C.,F.S.Cole,D. H.Perlmutter, and H. R. Colten.

1985.'y-Interferonincreasesexpressionof class IIIcomplementgenes

C2andfactorBin humanmonocytesand in murine fibroblasts trans-fected withhumanC2 and factorB genes.J.Biol. Chem.

260:15280-15285.

34. Adam, C., Y. Thoua, P. Ronco, P.Verroust,M.Tovey, andL. Morel-Maroger.1980.Theeffect ofexogenousinterferon: acceleration of autoimmuneand renaldiseases in (NZB/W) F1 mice. Clin. Exp. Immunol. 40:373-382.

35. Heremans, H., A. Billiau, A. Colombatti,J. Hilgers, and P.

DeSomer. 1978. InterferontreatmentofNZBmice: accelerated

pro-gressionof autoimmune disease.Infect.Immun.21:925-930. 36. Wener,M.H.,and M.Mannik. 1986.Mechanisms of immune deposit formation in renalglomeruli. Springer Semin. Immunopathol. 9:219-235.

37.Miller, G.W., and V.Nussenzweig. 1975. Anewcomplement function: solubilization ofantigen-antibody aggregates. Proc.Natl. Acad. Sci. USA. 72:418-422.

38.Schifferli,J. A., Y.C. Ng, andD.-K. Peters. 1986. Therole of complementanditsreceptorintheeliminationof immunecomplexes.

N.Engl.J.Med.315:488-495.