Kidney International, Vol. 58 (2000), pp. 17971804

TECHNICAL NOTE

Progressive adriamycin nephropathy in mice: Sequence ofhistologic and immunohistochemical events

YANG WANG, YI PING WANG, YUET-CHING TAY, and DAVID C.H. HARRIS

Department of Renal Medicine, University of Sydney at Westmead Hospital, Westmead, Australia

Progressive adriamycin nephropathy in mice: Sequence of his- the administration of aminonucleoside [1, 2], proteintologic and immunohistochemical events. overload [3, 4], partial nephrectomy [5, 6], and the ad-

Background. As an experimental analogue of human focal ministration of adriamycin (ADR) [79], have been de-glomerular sclerosis (FGS), adriamycin (ADR)-induced ne-scribed in rats. Among these, the ADR-induced ne-phropathy is well-characterized in rats. Previously, this modelphropathy model is considered to be an experimentalhas not been fully established in mice. The extension of this

model to the mouse would be useful in the application of analogue of human minimal lesion nephrotic syndromegenetic and monoclonal antibody technology to characterize and FGS [7, 9]. Furthermore, a long-term study of thismechanisms of progressive renal disease. Herein, we describe

model in rats demonstrated severe renal damage witha stable and reproducible murine model of chronic proteinuriafeatures characteristic of chronic progressive renal dis-induced by ADR.

Methods. Male BALB/c mice were intravenously injected ease in humans [9, 10]. Although ADR-induced chronicwith a single dose of ADR (10 to 11 mg/kg). Seven mice were nephropathy has been well characterized in rats for somesacrificed at weeks 1, 2, 4, and 6. Renal function, quantitative time, this model has not been fully established in themorphometric analysis, and electron microscopic studies were

mouse [11, 12]. A recent report from Chen et al describedperformed. Peripheral CD41 and CD81 T cells were evaluatedexperimental FGS in BALB/c mice induced by ADR,using flow cytometric analysis of splenocytes. The leukocytic

inflammatory pattern was analyzed by immunohistochemical with an observation period extended only to 18 daysexamination. [13].

Results. Overt proteinuria was observed from day 5 and The present study was designed to establish and char-remained significantly elevated throughout the study period.

acterize a stable and reproducible murine model ofA focal increase in reabsorption droplets in tubular cells ap-chronic progressive nephropathy with significant andpeared at weeks 1 and 2, and numerous intraluminal casts were

present after two weeks. Glomerular vacuolation and mild FGS persistent proteinuria using ADR. Such a murine modelappeared at week 4. At week 6, extensive focal and even global carries the special advantages of economy and the poten-glomerular sclerosis, associated with moderate interstitial tial for application of genetic and monoclonal antibodyexpansion and severe inflammation, were observed. A promi-

manipulation to study pathogenesis.nent macrophage infiltration appeared within both interstitiumand glomeruli at week 2, followed by accumulation of bothCD41 and CD81 T cells in interstitium but not glomeruli.

METHODSThere were almost no B lymphocytes seen at any time.Conclusion. This model should be useful in unraveling the Male BALB/c mice weighing 20 to 25 g and aged eight

pathogenesis of glomerular and interstitial inflammation and weeks were obtained from the Animal Care Facility,fibrosis in chronic proteinuric renal disease.Westmead Hospital (NSW, Australia). Dose-findingstudies defined an optimal dose of 10 to 11 mg/kg bodywt of ADR (Doxorubicin Hydrochloride; Pharmacia &Experimental focal glomerular sclerosis (FGS) can beUpjohn Pty Ltd., Perth, Australia). ADR was injectedinduced by both immunologic and nonimmunologicalonce via the tail vein of each nonanesthetized mouse.methods. Most nonimmunological procedures, such asSeven age-matched male BALB/c mice were injectedwith same volume of isotonic saline. All control andexperimental mice were housed individually, and onceKey words: glomerulosclerosis, proteinuria, tubulointerstitial fibrosis,

inflammation. a week, a 12-hour collection of spontaneously voidedurine was made from each animal. Body weight, urineReceived for publication November 22, 1999volume, and urinary protein were measured weekly.and in revised form March 7, 2000

Accepted for publication April 25, 2000 Seven mice treated with ADR were sacrificed at weeks1, 2, 4, and 6. Blood samples for serum albumin and 2000 by the International Society of Nephrology

1797

Wang et al: ADR nephropathy in mice1798

Table 1. Monoclonal antibodies (mAbs) used in FACS and immunohistochemical staining

Clone Target molecule Specificity Reference

RM45 Murine L3T4 MHC class II-restricted T lymphocytes, including [15]most T helper cells

536.7 Murine Ly-2 MHC class I-restricted T lymphocytes, including most [16]T suppressor/cytotoxic cells

M3/84 Murine Mac-3 Tissue macrophages, some myeloid cell lines, but not [17]lymphocytes or monocytes

RA3-6B2 Murine CD45R/B220 Pan-B lymphocytes marker, at all stages from pro-B [18]through mature and activated B cells

All antibodies were purchased from Pharmingen Company (Becton Dickinson and Company, New Jersey, USA). FACS is fluorescence-activated cell sorter.

creatinine were obtained by cardiac puncture under an- staining procedures used were modified slightly fromthose previously reported [19]. Briefly, spleen cell sus-esthetic (ketamine/xylazine, 100/8 mg/kg body weight),pensions were made in RPMI 1640 medium (GIBCOand then animals were completely exsanguinated via theBRL Life Technologies Inc., Grand Island, NY, USA)abdominal aorta to minimize the number of circulatingwith 10% (vol/vol) fetal calf serum (FCS; Trace Biosci-inflammatory cells remaining in harvested kidney andences Pty Ltd., Castle Hill, NSW, Australia) and thenspleen. Kidney weight and body weight were measuredseparated by centrifugation in 48C. Red blood cells wereimmediately. All kidney and spleen specimens were pro-lyzed, and the number of viable cells was determined.cessed without delay.Three 3 105 cells in 100 mL of medium were incubatedPlasma and urinary creatinine were analyzed using awith 10 mL of the relevant antibodies at room tempera-BM/Hitachi 747 analyzer (Tokyo, Japan). Total urinaryture for 10 minutes in the dark. Washing twice usingprotein was quantitatively determined using the Bio-phosphate-buffered saline (PBS) with 5% FCS, cellsRad protein assay (Bio-Rad Laboratories, Hercules, CA,were fixed with 2% (wt/vol) paraformaldehyde andUSA) according to the manufacturers instructions in astored in the dark at 48C until the cells were analyzedDU-68 Spectrophotometer (Beckman Instrument, Inc.,using a FACScan (Becton Dickinson). The immunohis-Irvine, CA, USA), for which purified bovine serum albu-tochemical (IHC) staining was performed by the follow-

min (BSA; Sigma Chemical Company, St. Louis, MO, ing protocol: 5 m cryostat sections of both kidney andUSA) at concentration ranges 0.2 to 1.5 mg/mL was spleen were placed on slides previously coated withused as a reference standard. Urine samples were also 0.01% (wt/vol) of poly-l-lysine (Sigma). Sections wereexamined for hematuria and leukocyturia using the Re- fixed in Zambonis fixative [20] for 30 minutes at 48Cagent Strips for Urinalysis (Bayer Australia Ltd., Pymble and then passed through methanol for five minutes andNSW, Australia). acetone for one minute at 2208C. An avidin-biotin com-

A piece of renal tissue was placed in 10% neutral- plex (ABC) technique was used. The primary mAbs arebuffered formalin fixative at room temperature for 12 also listed in Table 1. The secondary antibody was ahours, embedded in paraffin, evaluated using 4 m sec- biotinylated rabbit anti-rat immunoglobulin (Dako Cor-tions, and stained with periodic acid-Schiff (PAS), hema- poration, Carpinteria, CA, USA). For negative controls,toxylin and eosin (HE), and Massons trichrome. Quanti- primary antibodies were replaced with equivalent con-tative evaluation of FGS, tubular and interstitial lesions centrations of normal rat immunoglobulin. All speci-

mens were stained in duplicate. To avoid false negativewas performed using an image analysis system (Optimasstaining, a spleen section was placed and stained on every5.23; Optimas Corporation, Seattle, WA, USA) as de-slide as a positive control. Brown staining of cells wasscribed previously [14]. Mean values were calculatedregarded as positive immunoreactivity. The number offrom each of 20 glomeruli or tubules or 10 random corti-positive cells was evaluated from the mean of five corticalcal regions per section.and medullary fields (magnification 3400) in each sec-For electron microscopy, kidney specimens were fixedtion.in modified Karnovskys fixative buffer for two hours,

Two-way analysis of variance was used in all statisticalwashed two times in 0.1 mol/L MOPS buffer, and thentests. Bonferronis correction was used for multiple com-postfixed in 2% buffered osmium tetroxide. After dehy-parison. Results are represented as mean 6 SD. A valuedration in a graded ethanol series, specimens were em-of P less than 0.05 was considered to indicate statisticalbedded in epoxy resin. Ultrathin sections were stainedsignificance.with uranyl acetate and lead citrate and examined in a

Philips CM201s electron microscope.RESULTSThe sources and specificities of monoclonal antibodies

(mAbs) used for fluorescence-activated cell sorter In a series of preliminary experiments, we attemptedto induce nephropathy in two different strains of mice:(FACS) analysis are listed in Table 1 [1518]. The FACS

Wang et al: ADR nephropathy in mice 1799

Fig. 1. Mean body weight (A), total urinary protein (B), ratio of kidney weight/body weight (C ), serum albumin (D), creatinine clearance (E),relative interstitial volume (F ), mean percentage of glomerulosclerosis (G), and mean number of nuclei per glomerulus (H ) for groups with saline-treated (r) and adriamycin (ADR)-treated (j) mice plotted against time. Values are expressed as means 6 SD. *P , 0.05; **P , 0.01.

BALB/c and C57BL/6J. C57BL/6J mice either failed to with wide individual variations. A high dosage of ADR(more than 12 mg/kg body weight) by single injectiondevelop any significant proteinuria or histologic change,

even at eight weeks after ADR injection of 15 mg/kg caused dehydration, gastrointestinal bleeding, or severehematuria and, in about 25% of experimental mice, se-body weight or died within two weeks of ADR injection

of 20 mg/kg body weight or more. In contrast, in BALB/c vere cachexia, causing death before week 3. A similarphenomenon has also been described by Chen et al [13].mice, a single injection of ADR less than 10 mg/kg body

weight or a double injection with a total dosage of up However, in the present study, using an average dosageof 10.5 mg/kg body weight in a 2 mg/mL solution (rangeto 14 mg/kg body weight produced an unstable model

Fig. 5. Typical immunostaining of kidney cor-tex. The frozen sections from saline-treated(a) and ADR-treated (b) kidneys at six weeks.Stained with MAb 53-6.7 (discussed in the text).

Wang et al: ADR nephropathy in mice 1801

b

Fig. 2. (a) Saline-treated control mice showed morphologically normal glomeruli and tubules. (b) ADR-treated mice at week 2 showed glomerularhypertrophy and hyaline deposits (arrow). Resorption droplets (arrowhead) in tubule cells, as well as intratubular casts, were also found. Sometubules were normal. (c) At week 4, mesangial expansion (arrow) in glomeruli, tubular vacuolization (arrowhead), and mild interstitial proliferationwere found. Well-developed exudative (fibrin-cap) lesions were also easily seen in glomeruli. (d) Global glomerular sclerosis (arrow), tubularcollapse (arrowhead), and severe interstitial expansion were seen at week 6. Micrographs ad were stained by PAS. ( f ) The number of nucleiwas reduced in glomeruli and accompanied by a significant interstitial infiltration at week 6, as compared with control (e). Both were stained byHE (3400).

10 to 11 mg/kg body weight), all BALB/c male mice of tubular epithelial cells, a loss of brush border, andvacuolization. The interstitial volume increased mildlytreated with ADR developed nephrotic syndrome and

remained alive throughout the study period of up to and focally, but there was a moderate interstitial mono-cyte infiltration. Glomeruli were reduced in size witheight weeks.several vacuoles, collapse of tuft, as well as expansion

General characteristics of the mesangium caused by an increase in PAS-positivematerial. At week 6, tubular atrophy and intratubularAll experimental animals developed nephropathy

characterized by proteinuria, hypoalbuminemia, hyper- cast formation with tubulointerstitial expansion werewidely observed in the cortex. Extensive FGS and severecreatininemia, and progressive renal injury. As shown

in Figure 1A, the mean body weight of ADR groups fell interstitial fibrosis and inflammation were present. Inaddition, global sclerosis was observed in many glo-quickly after treatment and reached its nadir at week 2,

after which body weight increased gradually and more meruli.slowly than control. The surface of all kidneys of ADR-

Electron microscopic studytreated animals was granular and pale in color in compar-ison to control. The kidneys were hypertrophied during Ultrastructural examination showed fusion of foot

processes of the glomerular epithelial cells in ADR-early nephrosis and atrophied later (Fig. 1C). Overt pro-teinuria appeared at day 5 and was maximal at day 7. treated mice. The fusion was segmental at week 1 (Fig.

3) and widespread at week 6 (data not shown). NormalAt all time points after one week, high levels of protein-uria were observed in all experimental animals (Fig. 1B). control mice failed to show any epithelial cell abnor-

mality.As determined by the Reagent Strips for Urinalysis, 10of 28 mice (35.7%) had hematuria (graded from 2 to

FACS and IHC analysis31, equivalent to 25 to 80 red blood cells/mL), whereas15 mice (53.6%) had leukocyturia (graded from 2 to 31, Peripheral T lymphocytes in spleen were analyzed us-

ing a FACScan. The mean percentage of CD41 andequivalent to 70 to 125 leukocytes/mL). No control micehad hematuria or leukocyturia at any time point during CD81 cells among total leukocytes in spleen was 37 and

21%, respectively (Fig. 4a), in agreement with the resultsthe study period. ADR treatment produced a markedfall of serum albumin by a median of 39% after one of other researchers [21, 22]. There were no significant

changes in either CD41 or CD81 cell numbers at anyweek, after which levels rose gradually but remainedpersistently lower than control at all times. Creatinine time point in experimental animals (Fig. 4 b, c). IHC

examination of the renal tissue was performed at weeksclearance continued to decline with time and was signifi-cantly different in comparison with control after week 2, 4, and 6 after ADR injection. Lymphocytes were rarely

seen within interstitium and were never identified in4 (Fig. 1E).glomeruli of control animals, whereas a small population

Morphometric and histopathologic studies of macrophages could be found. CD41, CD81 T cells,and macrophages was significantly increased in the kid-The mean number of cells (Fig. 1H) and relative capil-

lary area in each glomerulus fell throughout the experi- neys of ADR mice. In the earlier stages, macrophageaccumulation in the interstitium was prominent, andmental period, coincident with a significant increase in

mesangial matrix area. The degree of interstitial volume infiltration in the glomeruli could be readily found. Inter-stitial macrophages appeared to originate in a sheathmarkedly increased with time (Fig. 1F), indicating the

severity of tubular interstitial expansion. Analysis by surrounding many glomeruli. However, macrophage in-filtration dramatically decreased in later stages, accom-light microscopy demonstrated the serial changes in ex-

perimental kidneys (Fig. 2). At weeks 1 and 2, there was panied by a gradual increase in the number of T lympho-cytes and development of renal injury, including tubulara focal increase in resorption droplets in tubular cells

and intraluminal casts. Glomeruli were increased in size, atrophy and interstitial expansion. The number of infil-trating T lymphocytes within the interstitium increasedwith marked hyaline deposits. At week 4, tubules dis-

played severe changes, including a decrease in height throughout the evolution of the experiment (Table 2).

Wang et al: ADR nephropathy in mice1802

characterize the mechanisms of progressive renal dis-ease, and the real advantage of cost savings in terms ofanimal purchase and maintenance, and use of expensiveinterventions such as mAbs.

In preliminary experiments, we showed that the mu-rine ADR model could be established in BALB/c butnot C57BL/6 mice. The reason for this difference is un-known. Using eight inbred murine strains, Kimura et alreported that BALB/c mice were highly susceptible tonephrosis induced by daunomycin, which has a quitesimilar structure to ADR, while C57BL/6J was com-pletely resistant to it [23]. They reported that the strainspecificity in susceptibility to nephrosis is genetically con-trolled, involving at least three genes (or clusters ofgenes). Our experiment showed that BALB/c is a suit-able but strictly dose-dependent strain for establishinga stable and reproducible ADR nephropathy model.

In the present studies, overt proteinuria appearedshortly after an intravenous injection of ADR and re-mained significantly elevated throughout the experimen-tal period. Furthermore, renal pathological, ultrastruc-tural, and functional changes progressed during a longobservation period. The renal injury varied with time,and the histologic hallmarks resembled those seen inchronic renal diseases in humans. Interstitial inflamma-tory cell infiltration is an early event in the developmentof this murine ADR nephropathy, and the infiltrationconsists of macrophages and T lymphocytes. The demon-stration of initial interstitial infiltration by macrophagesat week 2, followed by accumulation of both CD41 andFig. 3. Ultrastructure of glomerular epithelial cells showed segmental

fusion of foot processes (arrow) in ADR-treated mice at week 1 CD81 T cells and the disappearance of macrophage in-(317,800). filtration by week 6 as well as a virtual absence of B

lymphocytes, is consistent with previous observationsin other FGS models [4, 2429]. The ultimate roles ofmacrophages and T cells in glomerular and tubulointer-Both CD41 and CD81 cells were greatly increased atstitial injury are still unclear. Macrophages have beenweek 6 compared with at week 4 (P , 0.01), both inshown to mediate acute glomerular injury and to becortex and medulla. In contrast, T cells were never identi-involved in the stimulation of mesangial cell proliferationfied within the glomerulus at any stage of the experiment.and induction of FGS in many experimental modelsFigure 5 shows the IHC staining of CD81 cells in kidney[2528]. Macrophage-derived products, for example in-cortex. CD81 cells appeared to migrate from blood ves-terleukin-1, interleukin-6, tumor necrosis factor-a, plate-sels and infiltrated the tubulointerstitium (Fig. 4b). Nolet-derived growth factor, and transforming growthCD81 cells were seen in glomeruli even though glomerulifactor-b, have been associated with glomerular injurywere closely surrounded by CD81 cells. CD41 cells fol-and stimulation of mesangial cell proliferation and ma-lowed a similar distribution (data not shown). In contrasttrix formation [29]. The immune-activated macrophagesto T cells and macrophages, few if any B cells were foundmay also play an important role in the immunopathogen-at any time.esis of interstitial fibrosis in glomerulonephritis via adelayed type hypersensitivity mechanism [29]. T lympho-

DISCUSSION cytes also appear to be important in the progression ofDespite an abundance of literature describing chronic renal injury. Accumulation and activation of macro-

nephropathy in rats, no model of chronic glomerular phages can be directed by T cells. Tipping, Neale, andand tubulointerstitial changes induced by ADR has been Holdsworth proposed that glomerular macrophage infil-convincingly established in mice. Extension of this model tration is dependent on T cells [24]. In preliminary exper-to the mouse has the potential benefits of application iments, we found that a few CD41 T cells appeared in

juxtamedullary glomeruli in first 24 hours and disap-of gene knockout and transgenic mouse technology to

Wang et al: ADR nephropathy in mice 1803

Fig. 4. (A) Flow cytometry analysis of peripheral CD41 and CD81 cells (in spleen) with FITC-conjugated monoclonal antibodies (mAbs). TheX axis shows fluorescence intensity and the Y axis shows the relative cell number. Age-matched saline controls are shown in the upper panels,and ADR-treated animals at week 6 are shown in the lower panels. Column 1: Background staining with rat IgG 2a, k isotype standard; column 2:mAb RM4-5 (anti-L3T4); column 3: MAb 53-6.7 (anti-Ly-2). Stained cells were analyzed using the FACS-II with a logarithmic amplifier. (B andC) Percentage of peripheral CD41 and CD81 cells measured by FACS at weeks 0, 2, 4, and 6 after ADR treatment. Symbols are: (j) ADR-treated and (h) saline-treated age-matched groups.

Table 2. Frequency of interstitial CD41 and CD81 cells mation and fibrosis in chronic proteinuric renal disease.(cell number/4003 field) by immunostaining

At present, a project using mAbs for in vivo manipula-CD41 CD81 tion of T cell subsets is underway to investigate the link-

Week(s) Cortex Medulla Cortex Medulla age between T-cell infiltration and progressive interstitialfibrosis in murine ADR nephropathy.0 1.861.5 1.661.1 0.260.5 0.460.6

2 12.464.3a 12.862.6a 9.961.7a 12.363.3a

4 25.867.7a 35.064.9a 20.464.1a 37.666.0a ACKNOWLEDGMENTS6 148.1619.6a,b 121.0611.7a,b 63.369.3a,b 101.3612.0a,b

This study was supported by a grant from the National Health andValues are means 6 SD.Medical Research Council of Australia (No. 970721). The authorsa P , 0.01 compared with age-matched normal controlthank Dr. Ross Boadle, Department of Electron Microscopy at West-b P , 0.01 compared with week 4mead Hospital, for his invaluable assistance.

Reprint requests to Dr. Yang Wang, Department of Renal Medicine,Westmead Hospital, Westmead, NSW 2145, Australia.peared within two days, followed later by macrophageE-mail: yangw@westgate.wh.usyd.edu.au

infiltration (unpublished data). This is also consistentwith the finding that glomerular T-cell infiltration is only REFERENCESa transient feature of the first three days following the

1. Grond J, Weening JJ, Elema JD: Glomerular sclerosis in ne-induction of experimental crescentic glomerulonephritisphrotic rats: Comparison of the long-term effects of adriamycin

in the rat [25, 30]. Furthermore, immune-activated T and aminonucleoside. Lab Invest 51:277285, 19842. Diamond JR, Anderson S: Irreversible tubulointerstitial damagecells have been shown to play an important role in the

associated with chronic aminonucleoside nephrosis. Am J Patholimmunopathogenesis of interstitial fibrosis in glomerulo-137:13231332, 1990

nephritis [24, 31]. 3. Lalich JJ, Burkholder PM, Paik WC: Protein overload nephropa-thy in rats with unilateral nephrectomy: A correlative light immu-In conclusion, a stable and reproducible murine ADRnofluorescence and electron microscopical analysis. Arch Patholmodel has been established and characterized in present99:7279, 1975

study. This model is likely to prove useful in unraveling 4. Eddy AA: Interstitial nephritis induced by protein-overload pro-teinuria. Am J Pathol 135:719733, 1989the pathogenesis of interstitial and glomerular inflam-

Wang et al: ADR nephropathy in mice1804

5. Grond J, Schilthuis MS, Koudstaal J, Elema JD: Mesangial 18. Ballas ZK, Rasmussen W: Lymphokine-activated killer cells. VII.IL-4 induces an NK1.11CD8, a1b- TCR-ab B2201 lymphokine-function and glomerular sclerosis in rats after unilateral nephrec-activated killer subset. J Immunol 150:1730, 1993tomy. Kidney Int 22:338343, 1982

19. Goronzy J, Weyand CM, Fathman CG: Long-term humoral unre-6. Olson JL, Hostetter TH, Rennke HG, Brenner BM, Venkata-sponsiveness in vivo, induced by treatment with monoclonal anti-chalam MA: Altered glomerular permselectivity and progressivebody against L3T4. J Exp Med 164:911925, 1986sclerosis following extreme ablation of renal mass. Kidney Int

20. Stefanini M, De Martino C, Zamboni L: Fixation of ejaculated22:112126, 1982spermatozoa for electron microscopy. Nature 216:173174, 19677. Bertani T, Poggi A, Pozzoni R, Delaini F, Sacchi G, Thoua Y,

21. Ghobrial RR, Boublik M, Winn HJ, Auchincloss HJ: In vivoMecca G, Remuzzi G, Donati MB: Adriamycin-induced nephroticuse of monoclonal antibodies against murine T cell antigens. Clinsyndrome in rats: Sequence of pathologic events. Lab Invest 46:16Immunol Immunopathol 52:486506, 198923, 1982

22. Pedrazzini T, Hug K, Louis JA: Importance of L3T41 and Lyt-8. ODonnell MP, Michels L, Kasiske B, Raij L, Keane WF: Adria-21 cells in the immunologic control of infection with Mycobacte-mycin-induced chronic proteinuria: A structural and functionalrium bovis strain bacillus Calmette-Guerin in mice: Assessmentstudy. J Lab Clin Med 106:6267, 1985by elimination of T cell subsets in vivo. J Immunol 139:20322037,9. Okuda S, Oh Y, Tsuruda H, Onoyama K, Fujimi S, Fujishima1987M: Adriamycin-induced nephropathy as a model of chronic pro-

23. Kimura M, Takahasi H, Ohtake T, Sato T, Hishida A, Nishimuragressive glomerular disease. Kidney Int 29:502510, 1986M, Honda N: Interstrain differences in murine daunomycin-in-10. Cameron J: The Natural History of Glomerulonephritis: Renal Dis-duced nephrosis. Nephron 63:193198, 1993ease. Oxford, Blackwell Scientific Publications, 1979, p 329

24. Tipping PG, Neale TJ, Holdsworth SR: T lymphocyte participa-11. Weiher H, Noda T, Gray DA, Sharpe AH, Jaenisch R: Trans-tion in antibody-induced experimental glomerulonephritis. Kidneygenic mouse model of kidney disease: Insertional inactivation ofInt 27:530537, 1985ubiquitously expressed gene leads to nephrotic syndrome. Cell 25. Floege J, Alpers CE, Burns MW, Pritzl P, Gordon K, Couser62:425434, 1990 WG, Johnson RJ: Glomerular cells, extracellular matrix accumula-12. Watanabe Y, Itoh Y, Yoshida F, Koh N, Tamai H, Fukatsu A, tion, and the development of glomerulosclerosis in the remnant

Matsuo S, Hotta N, Sakamoto N: Unique glomerular lesion with kidney model. Lab Invest 66:485497, 1992spontaneous lipid deposition in glomerular capillary lumina in the 26. Diamond JR, Ding G, Frye J, Diamond IP: Glomerular macro-NON strain of mice. Nephron 58:210218, 1991 phages and the mesangial proliferative response in the experimen-

13. Chen A, Sheu LF, Ho YS, Lin YF, Chou WY, Chou TC, Lee tal nephrotic syndrome. Am J Pathol 141:887894, 1992WH: Experimental focal segmental glomerulosclerosis in mice. 27. Pesek-Diamond I, Ding G, Frye J, Diamond JR: MacrophagesNephron 78:440452, 1998 mediate adverse effects of cholesterol feeding in experimental ne-

14. Rangan GK, Wang Y, Tay YC, Harris DC: Inhibition of nuclear phrosis. Am J Physiol 263:F776F783, 1992factor-kappaB activation reduces cortical tubulointerstitial injury 28. van Goor H, van der Horst ML, Fidler V, Grond J: Glomerularin proteinuric rats. Kidney Int 56:118134, 1999 macrophage modulation affects mesangial expansion in the rat

15. Wu L, Scollay R, Egerton M, Pearse M, Spangrude GJ, after renal ablation. Lab Invest 66:564571, 1992Shortman K: CD4 expressed on earliest T-lineage precursor cells 29. Nikolic-Paterson DJ, Lan HY, Hill PA, Atkins RC: Macro-in the adult murine thymus. Nature 349:7174, 1991 phages in renal injury. Kidney Int 45(Suppl 45):S79S82, 1994

16. Nakayama K, Nakayama K, Negishi I, Kuida K, Louie MC, Kana- 30. Lan HY, Paterson DJ, Atkins RC: Initiation and evolution ofgawa O, Nakauchi H, Loh DY: Requirement for CD8, b chain interstitial leukocytic infiltration in experimental glomerulonephri-in positive selection of CD8-lineage T cells. Science 263:11311133, tis. Kidney Int 40:425433, 19911994 31. Baldridge JR, Barry RA, Hinrichs DJ: Expression of systemic

17. Flotte TJ, Springer TA, Thorbecke GJ: Dendritic cell and macro- protection and delayed-type hypersensitivity to Listeria monocyto-phage staining by monoclonal antibodies in tissue sections and genes is mediated by different T-cell subsets. Infect Immun 58:654

658, 1990epidermal sheets. Am J Pathol 111:112124, 1983