Immunologic determinants of susceptibility to ... - Kidney International

Kidney International, Vol. 51 (1997), pp. 646—652

Immunologic determinants of susceptibility to experimental

glomerulonephritis: Role of cellular immunity SANDRA N. COELHO1, SOHAIL SALEEM, BOGUMILA T. KONIECZNY, KANNAN R. PAREKH,

Fw K. BADDOURA, and FASI G. Ljiuus Renal Division, Emoty University School of Medicine and Veterans Affairs Medical Center, Atlanta, Georgia, and Pathology and Laboratoty Medicine, Veterans Affairs Medical Center, Buffalo, New Yorlç USA

mechanisms contribute to glomerular injury. For example, induction of accelerated anti-glomerular basement membrane (GBM) mechanisms that influence susceptibility to GN, we compared the severity of accelerated anti-glomerular basement membrane (GBM) nephritis nephritis in the rat is dependent on the host's antibody response between Lewis (LEW) and Brown Norway (BN) rats and analyzed to heterologous immunoglobulin planted in the glomerulus [5, 6] differences in their immune responses to the nephritogenic immunoglob- and on infiltration of the kidney with mononuclear cells [7, 8]. ulin. Lewis (LEW) rats preimmunized with sheep IgG developed prolif- Animals depleted of either monocytes or T lymphocytes fail to erative GN with marked proteinuria [peak protein excretion (mean develop significant proteinuria, glomerular cell proliferation, or sEM) = 85.3 15.3 mg124 hr; normal = 6.4 0.8 mg124 hr] after receiving a subnephritogenic dose of sheep anti-rat GBM antiserum. Identically crescent formation [7, 8]. Susceptibility to GN varies widely among humans and could be treated Brown Norway (BN) rats, on the other hand, had minimal renal pathology and minimal proteinuria (peak protein excretion = 22.6 3.1 determined by the nature of immune responses to nephritogenic mg/24 hr; normal = 13.0 0.6 mg124 hr). Serum titers of rat anti-sheep antigens. This is suggested by studies demonstrating an associaIgG isotypes and intraglomerular binding of sheep IgG, rat IgG, and rat complement (C3) were comparable in both strains. In contrast, only LEW tion between certain forms of human GN and immune response rats developed a strong cellular immune response to sheep IgG repre- genes of the major histocompatibility (MHC) complex [9—li]. sented by intrarenal T lymphocyte (OX19) and monocyte (ED1) Similarly, susceptibility of an inbred rat strain to experimental GN accumulation [LEW vs. BN (mean sEat): OX19 = 0.60 0.10 vs. is influenced by its MHC background. For example, strains of the 0.14 0.01 cells/glomerulus, control = 0.02 0.01; ED1* = 4.0 0.4 vs. RT-11 haplotype such as Lewis (LEW) and Wistar Kyoto (WKY) 1.0 0.2 cells/glom., control = 0.8 0.3] and a significant cutaneous rats are highly susceptible to experimental autologous immune delayed-type hypersensitivity (DTH) reaction [LEW versus BN (mean ear thickness = 0.22 0.02 vs. 0.05 0.03 mm; control = 0.04 complex glomerulonephritis, while Brown Norway (BN) and SEM): 0.02 mm]. Upon rechallenge with sheep IgG in vitro, LEW splenocytes LEW.BN rats (RT-1 haplotype) are relatively resistant [12—14]. expressed a T helper 1 (Thi) cytokine pattern (IFNy and IL-2 mRNA, but The specific immunologic pathways that confer susceptibility to little IL-4 mRNA) which is associated with delayed-type hypersensitivity GN are not well defined. Recent data suggest that the nature of T reactions. BN splenocytes, on the other hand, expressed IL-4 in addition to IL-2 and IFNY mRNA that is consistent with an undifferentiated (ThO) helper (Th) lymphocyte responses to disease-inciting antigens cytokine profile. These studies suggest that humoral immunity to heteroldetermines the outcome of immunologically-mediated disorders ogous immunoglobulin planted in the kidney is not sufficient for full [15]. Thi-type responses characterized by IL-2 and IFNy producexpression of accelerated anti-GBM nephritis, and that additional cellular tion and delayed-type hypersensitivity (DTH) have been shown to immune mechanisms are required. We conclude that susceptibility to accelerated anti-GBM nephritis is strongly influenced by the host's mediate tissue injury, while Th2-type responses identified by propensity to mount a Thi-type response and DTH reaction to the production of anti-inflammatory cytokines such as IL-4, IL-b, Immunologic determinants of susceptibility to experimental glomeru-

lonephritis: Role of cellular immunity. To identify the immunologic

disease-inciting immunoglobulin.

and IL-13 were protective [151. In this study, we tested the possibility that Thi-mediated cellular immune mechanisms confer susceptibility to experimental GN by: (1) comparing the susceptibilities of LEW and BN rats to accelerated anti-GBM nephritis, and (2) analyzing differences in antibody production, delayed-type hypersensitivity, and Thi-type cytokine expression in response to

The immune response to a circulating antigen or an antigen planted in the kidney can lead to intraglomerular immune complex deposits and subsequently, glomerulonephritis (GN) [1]. Studies in patients with proliferative GN [2—4] and in experimen- the nephritogenic immunoglobulin (sheep anti-rat GBM) between the disease-susceptible and disease-resistant strain. tal animals [4—8] suggest that humoral and cellular immunologic Methods Current address: Department of Medicine, Renal Division, Center of Health Sciences, Federal University of Pernambuco at Recife, Brazil. Received for publication May 28, 1996 and in revised form October 3, 1996 Accepted for publication October 3, 1996

© 1997 by the International Society of Nephrology

Experimental protocol

Inbred LEW and BN male rats, weighing 150 g to 175 g, were obtained from the Jackson Labs (Bar Harbor, ME, USA). Accelerated anti-GBM nephritis was induced in 10 LEW and 10 BN rats as described [7]. Briefly, rats were immunized i.p. with 4 mg 646

Coelho et al: Immunologic determinants of susceptibility to GN

Immunohistochemistiy Deparaffinized sections mounted on silanized slides were incubated with 1:100 dilution of EDI monoclonal antibody (Harlan/ Serotec, Indianapolis, IN, USA), which recognizes a cell surface antigen specific for rat mononuclear phagocytes [16]. Using an

100

75

alkaline phosphatase anti-alkaline phosphatase (APAAP) antibody method (DAKO, Carpinteria, CA, USA), cells that bound ED1 were identified by bright red staining [16]. Positive cells within 100 randomly selected glomeruli per animal were enumerated by light microscopy in a blinded fashion. T lymphocyte infiltrate was quantitated in a similar fashion except that fresh frozen 4 xm renal sections were fixed in acetone and incubated with 1:100 dilution of OX-19 (monoclonal mouse anti-rat CD5) antibody (Harlan/Serotec, Indianapolis, IN, USA) [8]. Extraglomerular leukocyte accumulation was quantitated by counting extraglomerular ED1+ and OX-19+ cells in a cortical interstitial area containing 100 glomeruli.

50

I

647

25

0 0

1

2

3

4

Immunofluorescence Kidney tissue was mounted in OCT compound (Miles DiagnosFig. 1. Twenty-four-hour urinaty protein excretion by LEW (U) and BN (0)tics, Elkhart, IN, USA), frozen in liquid isopentane cooled on dry ice, and 4 jxm cryostat sections were prepared (Leica, Germany). rats following induction of accelerated anti-GBM nephritis. *signiflcantly greater than LEW baseline (protein excretion on day 0) and than Direct immunofluorescence studies were performed by fixing day-matched BN rats (N = 10/group, P < 0.05). frozen sections in acetone followed by blocking with 10% normal Day (post-NTS injection)

goat serum and addition of FITC-conjugated rabbit antibody directed against sheep IgG (Zymed, South San Francisco, CA, sheep IgG (Sigma, St. Louis, MO, USA) mixed 1:1 (vol:vol) with USA), rat IgG (Zymed), rat IgGi (Zymed), rat IgG2a (Zymed), complete Freund's adjuvant (CFA; Gibco-BRL, New Haven, CT, rat IgE (Zymed) or rat complement (C3) (Pharmingen, San USA). Seven days later, a subnephritogenic dose (50 u1) of Diego, CA, USA) at 1:100 dilution. After washing with PBS + 1% heat-inactivated sheep anti-rat GBM nephrotoxic serum (NTS; Tween 20, sections were examined under a fluorescence microprovided by Dr. Kamal F. Badr, Emory University, Atlanta, GA,

scope by a pathologist (FKB) in a blinded fashion.

USA) was injected i.v. Control rats received 50 p1 of heatMeasurement of circulating rat anti-sheep IgG antibody titers

inactivated, non-immune sheep serum (NSS; Sigma). Twentyfour-hour urine was collected on four consecutive days from individual metabolic cages (Nalgene Inc., Rochester, NY, USA). Baseline 24-hour urine was collected three days prior to NTS

Microtiter wells (Nunc, Denmark) were coated with 1.0 rg/ml normal sheep IgG (Sigma) in carbonate/bicarbonate buffer pH 9.5

following disease induction, and serum, splenic, and renal tissue were saved for analysis. Urinary protein was quantitated by the Bradford method (BioRad Laboratories, Melville, NY, USA).

for one hour at room temperature [8]. Wells were washed twice in wash buffer prior to incubation with triplicate serial dilutions of rat serum for two hours. After three washes in wash buffer, wells

Histopathology

were incubated with horseradish peroxidase-conjugated rabbit anti-rat IgG (Sigma), IgGi, IgG2a, IgG2b, or IgE (Zymed) at the

Kidney sections were fixed in B5 solution (Great Lakes Diagnostics, Troy, MI, USA) followed by 10% neutral-buffered forma-

optimal dilution for two hours. Wells were finally washed six times in wash buffer and incubated with 0.1 M of the substrate 2,2'-

by incubation overnight at 4°C, washed twice in wash buffer injection. Rats were sacrificed at the end of the fourth day (PBS/0.1% Tween 20), and blocked with 2% BSA in wash buffer

azino-di-3-ethylbenzthiazoline sulphonate (ABTS; Boehringer paraffin wax. Hematoxylin and cosin (H&E), periodic acid Schiff Mannheim, Indianapolis, IN, USA) in 0.02% H202 for one hour

lin (Fisher Scientific, Pittsburgh, PA, USA) and embedded in

(PAS), Masson trichrome (MT), and silver staining were performed on 4 /Lm sections. Light microscopic examination was performed in a blinded fashion by a pathologist (FKB). Glomerular pathology was assessed by examining at least 100 glomeruli per animal in a blinded fashion. Extent of extracapillary hypercellularity in glomeruli was categorized into four grades [7]: 0, normal glomerulus; 1+, partial obliteration of capillary lumina

at room temperture. Reaction was stopped by adding an equal volume of 0.1 M citric acid/0.01% sodium azide and absorbance at

405 nm (O.D.4) was read immediately in a microplate reader (BioRad Laboratories). Control serum was obtained from nonimmunized LEW rats. Delayed-type hypersensitivity

due to extracapillary nuclear crowding/overlapping without oblitRats were immunized i.p. with 4 mg sheep lgG (Sigma) mixed eration of Bowman's space; 2+, subtotal obliteration of capillary 1:1 (vol:vol) with complete Freund's adjuvant (CFA) (Gibcolumina due to extracapillary nuclear crowding/overlapping with or BRL). Seven days later, 50 p.g sheep IgO in 50 p1 PBS was without obliteration of Bowman's space; 3+, total obliteration of injected intradermally in the left ear lobe. Control, nonimmunized capillary lumina and Bowman's space due to extracapillary nuclear LEW rats were also injected in the left ear with sheep lgG. crowding/overlapping; and 4+, presence of epithelial crescents. Twenty-four and 48 hours later, left and right ear lobe thickness

648

A

Coelho et a!: Immunologic determinants of susceptibility to GN

t) r'

B

_ft4 itL D D

C

Fig. 2. Photomicrographs of glomendi (PAS stain, X400 magnification) illustrating differences in pathology of accelerated anti-GBM nephritis between LEW and BN rats. Control rats received NSS while experimental rats received NTS seven days after immunization with sheep IgG. All rats were sacrificed four days after NTS or NSS injection. Control LEW kidney (A) with normal glomerular morphology and patent Bowman's space (°); experimental LEW kidney (B) with increased glomerular cellularity, collapsed Bowman's space, and PAS+ deposits (arrow); control (C) and experimental (D) BN kidneys without significant glomerular pathology.

Table 1. Leukocyte infiltration of LEW and BN glomeruli four days after induction of accelerated anti-GBM nephritis

Group LEW BN ControlC

N

Monocytes (EDI+)C

6 6 4

4.0 04b

0.60

0•10h

1.0 0.8

0.14 0.02

0.01 0.01

0.2 0.3

T lymphocytes (0X19+)"

+Cells/glomerular cross section (mean sEM) Significantly greater than BN and control groups (P < 0.05) LEW rats injected with normal sheep serum (NSS) seven days after immunization with sheep lgG

was measured in a blinded fashion using a dial gauge (Starrett; Small Parts Inc., Miami Lakes, FL, USA). DTH response was determined by subtracting right ear lobe thickness (baseline) from that of the left ear (experimental) [17]. Splenocyte isolation and stimulation

LEW and BN rats were immunized with 4 mg sheep IgG in CFA (1:1, vol:vol) i.p. on day 0, received 4 mg sheep IgG (without CFA) i.p. on day 7, and sacrificed four days later. Splenocytes

were isolated from freshly harvested spleen tissue by first lysing RBC's in 0.15 M NH4CI, 1 mrvi KHCO3, 0.1 mivi Na2EDTA, pH 7.4

followed by a one-step Ficoll gradient purification (Lympholyte-

Rat, Cedar Line-Hornby, Ontario, Canada) [18]. After three washes in Hanks balanced salt solution (HBSS), splenocytes were

resuspended at 5 X 106/ml in complete medium (RPMI 1640 medium supplemented with 5% fresh syngeneic rat serum, 1% L-glutamine, 0.5% 2-ME, 1% penicillin/streptomycin mix, 1% HEPES buffer, 1% nonessential amino acids, and 1% pyruvate) [18], stimulated for 24 hours at 37°C, 5% CO2 with either sheep IgG (200 .tg/ml), BSA (200 j.rg/ml), or no additive [6]. RNA extraction was performed immediately as described below. RNA isolation and RT-PCR

Total cellular RNA was isolated according to the Chomczynski and Sacehi method [19] using RNAzo1 reagent (Biotecx Laboratories, Houston, TX, USA). RNA samples with 0D2601250 < 1.6 were rejected. 2.5 p.g of total RNA was reverse transcribed using oligo-dT primers and Superscript® reverse transcriptase according to the manufacturer's instructions (Gibco-BRL). Ten percent

649

Coelho et a!: Immunologic determinants of susceptibility to GN

LEW

CTRL

BN

A

B

C

Fig. 3. Direct immunofluorescence studies demonstrating comparable sheep IgG (A) rat IgG (B), and rat C3 (C) linear deposits in LEW and BN glomeruli four days after induction of accelerated anti-GBM nephritis. Abbreviation CTRL is the representative control LEW rat injected with NSS seven days after immunization with sheep IgG.

A

0.2

0.2

0.2

E

E 0.15

E

C:

0.15

0.15

a)

a)

a)

0C:

C.)

C

0

C

B

0.25

0.25

Ca

0.1

.0

0

Cl,

C)

C 0.1

0(I,

C,)

.0

.0

-D

0.05

0.05

0

0 10000 1000

100

Serum dilution factor

10

0.1

Ca

-D

0.05

0

10000 1000

100


10

10000 1000

100

10


Fig. 4. Circulating rat IgG anti-sheep IgG (autologous antibody) measurements. ELISA was performed on serial serum dilutions from LEW (, N = 7) and BN (0, N 7) rats four days after induction of accelerated anti-GBM ncphritis. Titration of normal LEW serum is also shown (V, N 2).

of eDNA was then subjected to 40 cycles of PCR amplification (total volume of reaction = 100 jil) in a Perkin Elmer Thermo-

cycler 480 (Perkin Elmer, Foster City, CA, USA) using rat cytokine-specific primers [16]. For semiquantitative PCR analysis of cytokine mRNA levels, eDNA was diluted 1:10- and 1:20-fold

prior to amplification [20, 211. PCR conditions and primer

sequences for IFNy, IL-2, IL-4, and HPRT have been previously published [16, 21]. RT-PCR controls included "no RNA" (blank) and "no reverse transcriptase" reactions. Fifteen microliters of each RT-PCR reaction were electrophoresed on 2% SeaKem LE agarose (American Bioanalytical, Natick, MA, USA) gels and stained with ethidium bromide.

650


Statistical analyses

Table 2. Delayed-type hypersensitivity (DTH) response to sheep IgG

Analysis of variance was used to compare multiple groups. Confidence interval was set at 95%.

Results Difference in susceptibility to accelerated anti-GBM nephritis between LEW and BN rats

There were no deaths following injection of NTS. At the end of the experiment, body wts (mean SD) were similar: 214 5 g and 221 6 g for LEW and BN rats, respectively. LEW rats developed significantly more severe form of accelerated anti-GBM nephritis than BN rats as measured by the following parameters: (1.) Proteinuria. Twenty-four-hour urinary protein excretion in

the LEW group (N 10) on days one through four after NTS injection was significantly higher than baseline (day 0) and than day-matched BN rats (N = 10; P < 0.05; Fig. 1). Peak proteinuria (mean SEM: 85.3 15.6 mg/day) in the LEW strain occurred on day 2 while peak proteinuria in BN rats (22.6 3.1 mg/day) was not significantly higher than baseline. Urinary protein excretion did not increase in control LEW animals injected with NSS.

DTH responsc (mm)a 48 hrs

Group

N

24 hrs

LEW BN Control

6 6 3

0.22 002b 0.05

0.04

0.03 0.02

0.16 0.04 0.03

O.02' 0.03 0.01

Rats were immunized i.p. with 4 mg sheep IgG in CFA, except for controls, and rechallenged in the left ear lobe with 50 jsg sheep lgG seven days later. Left and right ear lobe thickness was measured 24 and 48 hours later. DTH response (mean 5EM) (mm) was determined by subtracting right ear lobe thickness (baseline) from that of the left (experimental). b

Significantly greater than BN and control groups (P < 0.05) Nonimmunized (naive) LEW rats

Systemic humoral immune response to sheep IgG

To quantitate the secondary antibody response to sheep IgG,

we measured rat IgG anti-sheep IgG serum titers four days following NTS injection (N = 7/group). Titration curves of LEW and BN sera were identical (Fig. 4A), and there were no differences in IgG isotype production (rat IgGi and rat IgG2a; Fig. 4 B,

(2.) Histopathological changes. Four days following NTS injec- C). Rat IgG2b and IgE directed against sheep IgG were not tion, the majority of LEW rats studied (7 out of 9) developed detected in either strain (data not shown). Control sera from diffuse renal changes characterized by severe extracapillary gb- non-immunized LEW rats (N = 2) had minimal antibody to sheep

merular hypercellularity (3+), increased glomerular size, narrow-

IgG (Fig. 4A). ing of Bowman's space, and partial capillary collapse (Fig. 2). Delayed-type hypersensitivity (DTH) response to sheep IgG Moderate glomerular hypercellularity (2+) was observed in the remaining LEW kidneys. All BN rats (N = 10) had normal renal To investigate whether susceptibility to accelerated anti-GBM histology or minimal glomerular hypercellularity (0 to 1+) (Fig. nephritis is related to cell-mediated mechanisms, we measured 2). Intraglomerular PAS+ deposits were present in two out of cutaneous (ear lobe) DTH responses to sheep IgG in preimmunine LEW rats but in none of the BN rats (Fig. 2B). Intracapillary thrombi or epithelial crescent formation were not observed.

nized LEW, preimmunized BN, and nonimmunized, control LEW rats. Preimmunized LEW rats mounted a significant DTH reac-

(3.) Mononuclear cell infiltration of renal tissue. Four days tion, measured at either 24 or 48 hours after rechallenge with following NTS injection, LEW glomeruli exhibited significantly sheep IgG, while BN rats and control LEW rats did not (Table 2). higher monocyte (ED1+) and T lymphocyte (OX-19+) accumulation than BN or control rats that received NSS (Table 1; P < Sheep IgG-induced cytokine mRNA expression in LEW and BN 0.05). Monocyte and T lymphocyte counts in BN glomeruli splenocytes following induction of nephritis were not significantly increased over

the control group (Table 1). Significantly greater extraglomerular accumulation of mononuclear cells was observed in LEW kidney sections compared to BN and control renal tissue (data not shown). Antibody binding to renal tissue

Because generation of a DTH response is largely determined by the nature of cytokines secreted after antigenic challenge [22], we studied mRNA expression of IL-2, IL-4, and IFNy by LEW and BN splenocytes following restimulation with sheep IgG in vitro. Semiquantitative RT-PCR analysis (Fig. 5A) revealed approxi-

mately 10-fold higher IL-4 mRNA levels in BN splenocytes

Because glomerular pathology in accelerated anti-GBM nephri- compared to LEW splenocytes. No apparent differences in IFNy tis is initiated by binding of autobogous rat antibodies to hcterol- and IL-2 mRNA expression were detected (Fig. 5A). An irreleogous sheep IgG planted along the GBM [6, 7], we investigated vant antigen, BSA, induced minimal expression of IL-4 mRNA by

the presence of sheep and rat TgG in LEW and BN gbomeruli

following disease induction. As shown in Figure 3 A and B, linear sheep and rat IgG deposits were detected along the GBM of LEW and BN gbomeruli. Despite significant differences in proteinuria, glomerular pathology and mononuclear cell infiltration between the two strains, no apparent differences in the amounts of bound lgG could be detected by immunofluorescence (total of 4 kidneys per strain were analyzed). Similarly, comparable binding of rat C3

BN splenocytes (Fig. SB). Equal loading of eDNA in the PCR reactions was confirmed by amplification of HPRT mRNA (Fig. 5A). DNA contamination was excluded by running "blank" and "no RT" PCR control samples. Discussion We have observed a significant difference in the susceptibilities of disparate rat strains to accelerated anti-GBM nephritis. While

to LEW and BN glomeruli was observed (Fig. 3C). Equal LEW rats developed proliferative GN, BN rats had minimal intraglomerular binding of rat IgG isotypes (IgOl and IgG2a) was

proteinuria and minimal glomerular pathology. Binding of heter-

detected in both rat strains while rat IgE was absent (data not ologous (sheep anti-rat GBM) and autologous (rat anti-sheep shown). Pre-immunized LEW rats injected with NSS did not IgG) antibodies to glomerular tissue, and high circulating titers of develop sheep IgG, rat IgG, or rat C3 deposits in their glomeruli autologous antibody were detected in both strains. Only LEW rats, however, mounted a strong cellular immune response to (Fig. 3).


LEW

BN

BN

B C BSA

A Neat 1:10 1:20 Neat 1:10 1:20

IL-4

I

651

— 378 bp —378bp

FN'y

—405 bp —405bp

lL-2

— 409 bp —409bp

H PRT

— —200 bp —-200bp

sheep IgG represented by mononuclear cell accumulation in the kidneys and a significant cutaneous DTH reaction. These studies demonstrate that a humoral response to heterologous immunoglobulin planted in the kidney is not sufficient for initiating glomerular inflammation, and that additional cellular

immune mechanisms such as DTH are required. In fact, T lymphocyte and monocyte accumulation was not observed in BN

kidneys despite the presence of autologous antibody (isotype identical to that in LEW rats) and complement fixation in their glomeruli. This suggests that mononuclear cell infiltration in accelerated anti-GBM nephritis is not a passive consequence of antibody binding to the glomerulus but represents an independent DTH-like response to the heterologous immunoglobulin. Huang,

— 378 bp

Fig. 5. Semiquantitative RT-PCR analysis of cytokine mRNA expression by rat splenocytes.

Cells were isolated and stimulated as described in the Methods section. Undiluted (Neat), 1:10 dilution, or 1:20 dilution of reverse transcribed RNA was subjected to 40 cycles of PCR amplification. 2% agarose gel electrophoresis of PCR products is shown. DNA ladder 4X174 RF DNAJHaeIII fragments. A. Sheep IgGstimulated LEW and BN splenocytes. B. Unstimulated (C) or bovine serum albuminstimulated (BSA) BN splenocytes.

produce IFNy, IL-2, and TNFp, but little or no IL-4. These lymphocytes also promote inflammation by stimulating macro-

phages, cytotoxic T cells, natural killer cells, and antibodydependent cell cytotoxicity [22, 25]. Differentiation of Thi cells from precursor lymphocytes (ThO) is determined by the relative amounts of cytokines present [15, 22]. Predominance of IFNy promotes maturation of the Thi phenotype while IL-4 is inhibitory. In this study, sensitized LEW splenocytes rechallenged with sheep IgG expressed IL-2 and IFNy mRNA but much less IL-4, suggesting the presence of Thi-like lymphocytes. In contrast, significant IL-4 mRNA expression in addition to IL-2 and IFNY among sheep IgG-stimulated BN splenocytes suggests the presence of undifferentiated ThO cells. These findings support our in vivo data demonstrating a DTH response to sheep IgG among LEW but not BN rats. Although several studies have suggested that a Th2 immune response (identified by IL-4 production in the absence of IL-2 or IFNy) protects against organ-specific autoimmunity [15], we did

Holdsworth and Tipping have pointed out the histopathologic similarities between proliferative GN and cutaneous [)TH reactions and demonstrated that in vivo T cell depletion, which suppressed 1)TH but spared antibody production, prevented the induction of accelerated anti-GBM nephritis in WKY rats [8]. Similarly, Bhan et a! provided evidence for a cellular immune not observe such a response in the BN strain. In fact, sheep

reaction against glomerular-hound immune complexes by demon TgG-induced cytokine mRNA expression by BN splenocytes was strating that transfer of sensitized T lymphocytes to animals with consistent with a ThO rather than Th2 phenotype. In addition, IgE immune complexes planted in their kidneys resulted in accumu- production, which characterizes Th2-mediated immune responses lation of proliferating cells in the glomerulus [23, 24]. Our results [221, was not detected in sera or renal tissue of BN rats which extended these studies [8, 23, 24] by confirming the role of DTH received NTS (data not shown). The results of our studies in accelerated anti-GBM nephritis in proliferative GN without resorting to experimental manipulation of the immune system. are corroborated by other experimental models. For example, Generation of a DTH response is largely determined by the LEW rats have been shown to be susceptible to Thi lymphocytenature of cytokines secreted after antigenic challenge [15, 22]. mediated inflammatory disorders such as experimental allergic Evidence from mice [25], rats [26], and humans [27, 28], suggests encephalitis (EAE) and experimental autoimmune uveoretinitis that DTH is mediated by T helper 1 (ThI) lymphocytes which (EAU) while BN rats were resistant [29, 30]. In contrast, mercuric

652

Coelho et at: Immunologic determinants of susceptibility to GN

chloride (HgCI2) induced IL-4-mediated polyclonal B cell activation and autoantibody production in BN rats but failed to do so in the LEW strain [31—33]. In a series of in vitro experiments Prigent et al demonstrated that HgCI2 triggered IL-4 production by BN but not LEW T cells while IFNy expression was stimulated in both

12. STENGLEIN B, THOENES GH, GUNTHER E: Genetically controlled

autologous immune complex glomerulonephritis in rats. J Immunol 115:895—897, 1975

13. ZANETrI M, BELLON B, VERROUST P, DRUET P: A search for circulating immune complex-like material during the course of autoimmune complex glomerulonephritis in Lewis and Brown Norway

rats. Clin Exp Immunol 42:86—94, 1980 [I• These studies underscore the differences between LEW and 14. B0LT0N WK, MAY Wi, STURGILL BC: Proliferative autoimmune BN immune responses to disease-inciting agents, mainly that glomerulonephritis in rats: A model for autoimmune glomeruloneLEW rats are prone to Thi-type responses while BN rats are phritis in humans. Kidney mt 44:294—306, 1993 biased towards IL-4 production. Our results suggest that these 15. MOSMAN TR, SAD 5: The expanding universe of T-cell subsets: ThI, Th2 and more. Immunol Today 17:138—146, 1996 immunologic differences could account for disparities in susceptibility to accelerated anti-GBM nephritis between the LEW and 16. LAKKIS FG, BADDOURA FK, CRUET EN, PAREKH KR, FUKUNAGA M, MUNGER KA: Anti-inflammatory lymphokine mRNA expression in BN strains. antibody-induced glomerulonephritis. Kidney mt 49:117—126, 1996 We conclude that susceptibility to accelerated anti-GBM ne- 17. HANCOCK WW, KHOURY Si, CARPENTER CB, SAYEGH MH: Differenphritis is strongly influenced by the host's propensity to mount a tial effects of oral versus intrathymic administration of polymorphic major histocompatibility complex class IT peptides on mononuclear ThI-mediated DTH response to the nephritogenic immunogloband endothelial cell activation and cytokine expression during a ulin. This could be pertinent to immunologically-mediated GN in delayed-type hypersensitivity response. Am J Pathol 144:1149—1158, humans, among whom significant differences in the immune 1994 18. H0I:r PG, LEt VERS 5, WARNER LA: Optimal culture conditions for in response to foreign antigens are observed [28].

vitro antigen-induced proliferation of rat lymph node cells. J Immunol

Acknowledgments This work was supported by a VA Merit Review grant (FGL) and by the Conselbo Nacional de Desenvolvimento Cientifico e Tecnologico, Brazil

(SNC). The authors thank Drs. William E. Mitch and Kamal F. Badr, Emory University School of Medicine, Atlanta, Georgia, USA, for their excellent advice.

Meth 44:205—209, 1981 19. CHOMCZYNSKI P, SACCHI N: Single-step method of RNA isolation by

acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156—159, 1987 20. STUMBLES P, MASON D: Activation of CD4+ T cells in the presence of

a nondepleting monoclonal antibody to CD4 induces a Th2-type response in vitro. J Exp Med 182:5—13, 1995 21. MURPHY F, HIENY 5, SHER A, O'GARRA A: Detection of in vivo

expression of interleukin-lO using a semiquantitative polymerase

Reprint requests to Fadi G. Lakkis, M.D., J/AMC - Atlanta, Research 151, chain reaction method in Schictoma mansoni infected mice.Jlmmunol Meth 162:211—223, 1993 1670 Clairmont Road, Atlanta, Georgia 30033, USA. 22. PAUL WE, SEDER RA: Lymphocyte responses and cytokines. Cell E-mail: [email protected]

References 1. WILSON CB: Renal response to immunologic glomerular injury, in The Kidney (5th ed, vol. 2), edited by BRENNER BM, Philadelphia, Saunders, 1996, p 1253

2. Hoo DH, GEE DC, ATKINS RC: Leukocyte analysis using monoclonal antibodies in human glomerulonephritis. Kidney mt 3 1:961—972, 1987 3. NOLASCO FEB, CAMERON iS, HARTLEY B, COELHO A, HILDRETH G,

REUBEN R: Intraglomerular T cells and m000cytes in nephritis: Study with monoclonal antibodies. Kidney mt 31:1160—1166, 1987 4. TAM FWK, PUSEY CD: The role of T lymphocytes in extracapillary glomerulonephritis. J Nephrol 8:305—316, 1995 5. UNANUE ER, DIXON FJ: Experimental glomerulonephritis. IV. The autologous phase of nephrotoxic serum nephritis. J Exp Med 121:715— 725, 1965 6. LOWRY RP, CLARKE FORBES RD, BI,ACKBURN JH: Immune reactivity and immunosuppressive intervention (TLI) in experimental nephritis.

I. Immunopathologic correlates of accelerated autologous form of nephrotoxic serum nephritis. J Immunol 132:1001—1006, 1984

7. SCHREINER GF, COTRAN RS, PARDO V, UNANUE ER: A mononuclear

cell component in experimental immunological glomeruloncphritis. J Exp Med 47:369—384, 1978

8. HUANO XR, HOLDsw0RrH SR, TIPI'ING PG: Evidence for delayedtype hypersensitivity niechan isms in glomerular crescent formation. Kidney mt 46:69—78, 1994

9. LI PK, PooN AS, LAI KN: Molecular genetics of MFTC class II alleles in Chinese patients with TgA nephropathy. Kidney mt 46: 185—1 90, 1994

76:241—251, 1994 23. BHAN AK, SCHNEEBERGER EE, COLLINS AB, MCCLUSKEY RT: Evi-

dence for a pathogenic role of a cell-mediated immune mechanism in experimental glomerulonephritis. J Exp Med 148:246—260, 1978 24. BHAN AK, COLLINS AB, SCI-INEEBERGER EE, MCCLUSKEY RT: A

cell-mediated reaction against glomerular-bound immune complexes. JExp Med 150:1410—1420, 1979 25. MOSSMAN TR, COFFMAN RL: THT and TH2 cells: Different patterns of lymphokine secretion lead to different functional properties. Annu Rev Immunol 7:145—173, 1989 26. SAYEGH MH, AKALIN E, HANCOCK WW, RUSSEL ME, CARPENTER CB, LINSLEY PS, TURKA LA: CD28—B7 blockade after alloantigenic

challenge in vivo inhibits Thi cytokines hut spares Th2. J Exp Med 181:1869—1874, 1995 27. SALGAME P, ABRAMS JS, CLAYBERGUR C, GOLDs1-EIN H, CoNvn-r i,

MODLIN RL, BI00M BR: Differing lymphokine profiles and functional

subsets of human CD4+ and CD8+ T cell clones. Science 254:279— 281, 1991 28. ROMAGNANI 5: Lymphokine production by human T cells in disease states. Annu Rev Immunol 12:227—257, 1994 29. CHUNG TY, NORRIS iG, BENVENISTE EN: Differential tumor necrosis factor expression by astrocytes from experimental allergic encephalomyelitis-susceptible and -resistant rat strains. J Exp Mcd 173:801—811, 1991 30. SAOUDI A, KUHN J, HUYGEN K, DL KOZAK Y, VEIl: T, GOLDMAN M,

DRUET F, BF,I.LON B: TTI2 activated cells prevent experimental autoimmune uveorelinitis, a TH1-dependent autoimmune disease. EurJ Immunol 23:3096—3103, 1993 31. DRUET E, SAPIN C, GUNIHER E, FI:INGOLD N, DRIJET P: Mercuric

chloride induced anti-glomerular basement membrane antibodies in the rat. Genetic control. Eon Immunol 7:348—351, 1977 32. OCHEL M, VOI-IR HW, PFEIFFER C, GIEICI-IMANN E: IL-4 is required

10. VAUGIIAN RW, TIGHE MR, B0KI K, ALEXOUPOI.OS 5, PAPADAKIS i, LANCHBURY iS, WELSH KI, WILLIAMS DG: An analysis of HLA class

for the IgE and TgG1 increase and IgGI autoantibody formation in

IT gene polymorphism in British and Greek idiopathic membranous nephropathy. EurJ Immunogen 22:179—186, 1995

33. PRIGENT P, SAOUDI A, PANNETIER C, GRABER F, BONNEFOY i-Y,

II. BURNS AP, FISHER M, LI P, PUSEY CD, REES AJ: Molecular analysis

helper cell (Th)2-mediated autoimmunity in Brown Norway rats,

of HLA class TI genes in Goodpasture's disease. Quart J Med 88:93—IOU, 1995

mice treated with mercuric chloride. J Immunol 146:3006—3011, 1991 DRUET P, HIRSCII F: Mercuric chloride, a chemical responsible for T directly triggers T cells to produce interieukin-4. J Clin Invest 96:1484— 1489, 1995