Enhanced susceptibility to end-organ disease in the lupus-facilitating NZW mouse strain.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 48, No. 4, April 2003, pp 1080–1092 DOI 10.1002/art.10887 © 2003, American College of Rheumatology Enhanced Susceptibility to End-Organ Disease in the Lupus-Facilitating NZW Mouse Strain Chun Xie,1 Xin J. Zhou,2 Xuebin Liu,1 and Chandra Mohan1 kines (monocyte chemoattractant protein 1, RANTES, KC), and significantly accelerated mortality. Importantly, these changes occurred within a few days after NTS administration. Finally, (B6 ⴛ NZW)F1 mice were as susceptible as the NZW parents, which indicates dominant NZW contributions. Conclusion. Collectively, these findings support the notion that a lupus-facilitating genome may contribute to disease susceptibility by modulating the degree of immune-mediated end-organ damage. The availability of B6-based congenic strains bearing individual NZWderived lupus susceptibility loci will permit future genetic dissection of end-organ susceptibility in murine lupus. Objective. Although the NZW mouse strain is phenotypically normal, fulminant lupus glomerulonephritis (GN) develops when NZW mice are bred to several other strains, such as NZB, BXSB, B6.Sle1, and B6.Yaa. Based on the observation that aging NZW mice exhibit histologic evidence of GN, we sought to test our hypothesis that NZW mice may be more susceptible to immune-mediated renal damage. Methods. NZW mice, as well as C57BL/6 (B6) and BALB/c control mice, were challenged with rabbit anti– glomerular basement membrane nephrotoxic sera (NTS), to induce renal disease. The different mouse strains were monitored for the degree of clinical disease, renal pathology, chemokine profiles, and cellular infiltrates. Results. Although the NZW and control strains showed similar glomerular deposits of rabbit Ig and exhibited similar levels of anti-rabbit xenogeneic immune response, the NZW mice had significantly worse pathologic changes and disease. Compared with the control strains, the NTS-injected NZW mice demonstrated significantly increased proteinuria, elevated blood urea nitrogen levels, more severe histologic GN and tubulointerstitial nephritis, increased glomerular crescent formation with macrophage and neutrophil infiltrates, elevated expression of CC and CXC chemo- Over the last 4 decades, studies with the New Zealand strains of mice have contributed significantly to our understanding of the pathogenesis of lupus. Of particular interest are the specific genetic and immunologic contributions of the NZW and NZB strains to disease. The NZW strain is fairly healthy, but contributes to lupus nephritis, in epistasis with several other strains. When NZW mice are bred to the NZB strain, either as an F1 hybrid or as a recombinant inbred strain, this leads to highly penetrant lupus nephritis (1–3). Likewise, when NZW mice are bred to the BXSB strain, the F1 offspring exhibit lupus nephritis and coronary vascular disease (4). Moreover, when NZW mice are bred to B6 mice, either in the context of the lupus susceptibility locus Sle1 (5) or the Yaa locus (6), the F1 offspring also exhibit severe proliferative glomerulonephritis (GN) and accelerated mortality. Thus, although the NZW genome by itself does not lead to the development of clinical nephritis, it clearly has the capacity to interact with several other genomes to precipitate endorgan disease. Although NZW mice are clinically healthy, a few reports have documented subclinical histologic changes in aging NZW kidneys (7–9). Importantly, these histo- Supported in part by grants from the NIH (P01-AI-39824), The Lupus Research Institute, and the National Arthritis Foundation. Dr. Mohan is the recipient of the Robert Wood Johnson Jr. Arthritis Investigator Award. 1 Chun Xie, MD, Xuebin Liu, MD, PhD, Chandra Mohan, MD, PhD: Simmons Arthritis Research Center, University of Texas Southwestern Medical Center at Dallas; 2Xin J. Zhou, MD: University of Texas Southwestern Medical Center at Dallas. Address correspondence and reprint requests to Chandra Mohan, MD, PhD, Simmons Arthritis Research Center, Department of Internal Medicine/Rheumatology, University of Texas Southwestern Medical Center, Mail Code 8884, Y8.204, 5323 Harry Hines Boulevard, Dallas, TX 75390-8884. E-mail: Chandra.mohan@ utsouthwestern.edu. Submitted for publication July 8, 2002; accepted in revised form January 3, 2003. 1080 INTRINSIC SUSCEPTIBILITY TO RENAL DISEASE IN NZW MICE logic changes occur in the absence of any prominent serologic phenotypes or clinical signs of disease. These observations raise the possibility that one critical avenue by which the NZW genome may be contributing to disease in the (NZB ⫻ NZW)F1, NZM2410, and related lupus models, is by facilitating immune-mediated endorgan damage. The NZW genome could be effecting this either by affecting the systemic immune system (e.g., by promoting the formation of pathogenic antinuclear antibodies) or by facilitating any of the several local renal processes (e.g., local inflammatory cascades) that lead to end-organ damage. Certainly, these 2 mechanisms are not mutually exclusive. One approach to determining whether the NZW genome does indeed confer increased susceptibility to end-organ damage is to challenge these mice with preformed nephrophilic antibodies that are potentially pathogenic. We aimed to achieve this in the present study with the use of the nephrotoxic serum (NTS) nephritis model. In the nephrotoxic serum nephritis model, also known as Masugi nephritis (10), preformed rabbit antiglomerular sera are injected into recipient mice. Such an abrupt exposure to nephrophilic (or nephrotoxic) antibodies is known to compromise renal function in a brisk and reproducible manner. Thus, in this model, one essentially “short-circuits” the chain of pathogenic events that is otherwise required for the formation of pathogenic autoantibodies, in order to directly assess how kidneys of different genotypes might handle the same immunologic insult. Indeed, this model has been very useful in demonstrating the essential roles played by several different molecules in mediating GN. For instance, studies using this approach have elegantly dissected out the respective contributions of complement, Fc receptors, cytokines, chemokines, adhesion molecules, and several other key players in renal disease (11–19). In the present study, we compared the NZW strain with 2 normal strains of mice, C57BL/6 (B6) and BALB/c, with respect to how they handle the same immunologic insult, in this case, rabbit antiglomerular nephrotoxic sera. MATERIALS AND METHODS Nephrotoxic rabbit sera. NTS was purchased from Lampire Laboratories (Pipersville, PA). Briefly, renal cortices of B6 mouse kidneys were minced and then pressed through a series of sieves of decreasing pore size (250-m, 150-m, and 75-m mesh). The glomeruli were collected on the finest sieve, washed with cold phosphate buffered saline, and sonicated for 7 minutes. The glomerular sonicates were then used to immunize rabbits (2 mg per rabbit; 3 injections administered 21 days 1081 apart). Sera obtained from these rabbits 50 days following the primary immunization stained glomeruli strongly by immunofluorescence, and these were the NTS used in the present study. In addition, preimmune rabbit sera were used as a negative control (placebo). Mice and nephrotoxic serum nephritis. B6, BALB/c, and NZW mice were purchased from The Jackson Laboratory (Bar Harbor, ME). (B6 ⫻ NZW)F1 mice were bred at our animal facility. All mice were maintained in a specific pathogen–free colony. Female mice ages 2–3 months were used for all studies. Nephrotoxic serum nephritis was induced as described elsewhere (10), with some modifications. Briefly, mice were first sensitized on day 0 with an intraperitoneal injection of rabbit IgG (250 g per mouse) in adjuvant. On days 4, 5, and 6, the mice received either NTS or placebo. In both cases, 25 g of total Ig in a 300-l volume was administered intravenously per mouse. The same dosage of NTS was used for all 3 mouse strains since these strains do not differ significantly in weight. This dosing regimen was selected because it was sufficient to induce proteinuria but not mortality in the control strains. On days 0, 4, 7, 11, 14, 18, and 21, we collected 24-hour urine samples from all mice with the use of metabolic cages; mice had free access to drinking water. Urinary protein concentrations were determined using the Coomassie Plus protein assay kit (Pierce, Rockford, IL). Sera were collected on days 0, 4, 7, 14, and 21. Blood urea nitrogen (BUN) levels were determined using a urea nitrogen kit (Sigma, St. Louis, MO) according to the manufacturer’s instructions. All surviving animals were killed on day 21, and the kidneys were harvested and processed for light microscopy and immunofluorescence microscopy, as described below. For the abbreviated 11-day protocol (adopted for the experiments shown in Figure 9), a modification of the above approach was used. Briefly, mice were sensitized and challenged with NTS or placebo as described above, but instead of a 21-day followup, they were killed on day 11 (5 days after the last injection of NTS). Kidneys were harvested and processed for histopathologic examination as described below. Histopathologic examination. Three-micrometer sections of formalin-fixed, paraffin-embedded kidney tissues were cut and stained with hematoxylin and eosin and with periodic acid–Schiff. These sections were examined by 2 investigators (CX and XJZ), in a blinded manner, for any evidence of pathologic changes in the glomeruli, tubules, or interstitial areas. The glomeruli were screened for evidence of hypertrophy, proliferative changes, crescent formation, hyaline deposits, fibrosis/sclerosis, and basement membrane thickening. The severity of GN was graded on a 0–4 scale, where 0 ⫽ normal, 1 ⫽ mild increase in mesangial cellularity and matrix, 2 ⫽ moderate increase in mesangial cellularity and matrix, with thickening of the glomerular basement membrane (GBM), 3 ⫽ focal endocapillary hypercellularity with obliteration of capillary lumina and a substantial increase in the thickness and irregularity of the GBM, and 4 ⫽ diffuse endocapillary hypercellularity, segmental necrosis, crescents, and hyalinized endstage glomeruli. Similarly, the severity of tubulointerstitial nephritis was graded on a 0–4 scale, based on the extent of tubular atrophy, inflammatory infiltrates, and interstitial fibrosis, as detailed previously (20). The numbers of infiltrating polymorphonuclear leukocytes were directly enumerated 1082 XIE ET AL Table 1. Primer amplification* Mouse gene GAPDH MCP-1 RANTES IP-10 KC sequences for polymerase chain reaction Primer sequences 5⬘: 3⬘: 5⬘: 3⬘: 5⬘: 3⬘: 5⬘: 3⬘: 5⬘: 3⬘: ACC-ACA-GTC-CAT-GCC-ATC-AC TCC-ACC-ACC-CTG-TTG-CTG-TA ACT-GAA-GCC-AGC-TCT-CTC-TTC-CTC TTC-CTT-CTT-GGG-GTC-AGC-ACA-GAC CCC-TGC-TGC-TTT-GCC-TAC-CTC-TCC TGG-GTT-GGC-ACA-CAC-TTG-GCG GGA-TGG-CTG-TCC-TAG-CTC-TG ATA-ACC-AAT-TGG-GAA-GAT-GG GGA-TTC-ACC-TCA-AGA-ACA-TCC-AGA-G CAC-CCT-TCT-ACT-AGC-ACA-GTG-GTT-G * MCP-1 ⫽ monocyte chemoattractant protein 1; IP-10 ⫽ interferon␥–inducible 10-kd protein. based on their typical polymorphonuclear morphology, by examining 50 glomeruli per kidney section per mouse. Immunohistochemistry and immunofluorescence studies. For the immunofluorescence studies, 3-m–thick cryostat sections of the kidneys obtained from all mice were fixed for 10 minutes in acetone and air-dried. To detect any deposited rabbit antibodies (from the NTS or preimmune rabbit sera), the kidney slides were incubated with pretitrated fluorescein isothiocyanate–coupled mouse anti-rabbit Ig and then quantitated by fluorescence microscopy, as described elsewhere (21). Briefly, the fluorescence intensity in 10 full-sized glomeruli from each section was measured using the autoexposure feature of the fluorescence microscope imaging system (Axioskop System; Carl Zeiss Instruments, Dallas, TX). The duration of exposure, as indicated by the autoexposure system, was considered to reciprocally reflect staining intensity. Autoexposure units (AU) were calculated by dividing 1,000 by the autoexposure time (in seconds). On this scale, background staining uniformly yielded 12–14 AU, whereas maximal fluorescence typically yielded 25–30 AU. Macrophages were identified and enumerated by serially treating deparaffinized kidney sections with anti–Mac-2 antibody (Cedarlane, Westbury, NY) and rabbit anti-rat Ig coupled to biotin (Vector, Burlingame, CA), followed by amplification using a Vectastain ABC kit (Vector), and finally, with diaminobenzidine (Sigma) substrate. For the negative control, a monoclonal mouse IgG1 antibody (Bethyl Laboratories, Montgomery, TX) was used at equivalent concentrations. Analysis of chemokine expression. Total RNA was isolated from renal cortices using TRIzol reagent (Life Technologies, Grand Island, NY) and quantitated spectrophotometrically. Reverse transcription–polymerase chain reaction (RT-PCR) was performed using the enhanced avian myeloblastosis virus RT-PCR kit (Sigma), according to the manufacturer’s protocol. The primer sequences used for the RT-PCR are listed in Table 1. For each pair of primers, PCR was performed over a range of cycles to define the conditions under which a linear relationship held between the quantity of RNA substrate and the final PCR product obtained. The PCR products were electrophoresed using a 2% agarose gel and then stained with ethidium bromide. The density of the bands was ascertained with an AlphaImager 2000 Documentation and Analysis System (Alpha Innotech, San Leandro, CA), and analyzed using ImageJ software (National Institutes of Health/National Center for Biotechnology Information, Bethesda, MD; online at: http://rsb.info.nih.gov/ ij/). Results are expressed as relative units (RU), which represent the ratio of the chemokine band intensity to that of GAPDH assayed in parallel. Urine levels of monocyte chemoattractant protein 1 (MCP-1) and KC were measured using commercially available enzyme-linked immunosorbent assay (ELISA) kits (Becton Dickinson, San Jose, CA, and R&D Systems, Minneapolis, MN, respectively). ELISA. Mouse antibodies to rabbit Ig were assayed by ELISA. Briefly, purified rabbit Ig (Sigma) was coated onto Immulon I plates (Dynatech, Chantilly, CA) and then blocked. Serially diluted mouse sera were added to these plates, and any bound anti-rabbit Ig was detected using alkaline phosphatase– conjugated goat anti-mouse IgG (that was not cross-reactive with rabbit Ig; Roche, Indianapolis, IN), and paranitrophenyl phosphate substrate (Sigma). For the measurement of isotypespecific antibodies, enzyme-conjugated anti-mouse IgG1 (PharMingen, San Diego, CA), IgG2a (Zymed, Burlingame, CA), or IgG2b (PharMingen) was used instead. All reagents were tested using serum from uninjected (B6 ⫻ NZW)F1 mice to ensure that the second antibodies detected both the B6 and NZW Ig allotypes equally well. To facilitate interstrain comparisons, sera from the different mice were all assayed within the same ELISA plate. Mouse IgG or isotype standards were also included on each plate for standardization. Results are expressed as micrograms per milliliter, based on the standard curves generated. Statistical analysis. Intergroup comparisons were performed using Student’s t-test, unless indicated otherwise. Results are expressed as the mean ⫾ SEM. All statistical analyses were performed using SigmaStat (Jandel Scientific, San Rafael, CA). RESULTS Since the goal was to ascertain if NZW mice exhibited any increased sensitivity to potentially pathogenic (i.e., nephritogenic) autoantibodies, batches of NZW mice as well as B6 and BALB/c controls were challenged with rabbit NTS or preimmune sera as detailed in Materials and Methods. In order to first establish that kidneys from both strains were exposed to similar amounts of rabbit NTS, they were examined for the extent of rabbit Ig deposition in their kidneys. As indicated in Figures 1A and C, the NZW kidneys and the kidneys of the control mice exhibited equivalent amounts of rabbit Ig deposits following exposure to NTS, as quantitated by fluorescence autoexposure measurements. Intense linear staining of IgG along the glomerular capillary walls (but not the tubules) was noted in both the NZW and control kidneys. The controls that received placebo (i.e., preimmune rabbit sera) did not exhibit rabbit Ig deposits in their kidneys INTRINSIC SUSCEPTIBILITY TO RENAL DISEASE IN NZW MICE Figure 1. Rabbit Ig deposits in NZW and control mouse kidneys. The presence and specific location of rabbit Ig deposits in A, a nephrotoxic rabbit sera (NTS)–injected NZW mouse, B, a preimmune rabbit sera (placebo)–injected NZW mouse, C, an NTS-injected C57BL/6 (B6) control mouse, and D, a placebo-injected B6 control mouse were determined by immunofluorescence analysis using fluorescein isothiocyanate–conjugated anti-rabbit Ig. Images are representative of kidney sections obtained from 4–6 mice from each of the 4 experimental groups. The immunofluorescence autoexposure times for the kidneys from the NTS-injected NZW mice (mean ⫾ SEM 47.4 ⫾ 1.6 seconds) did not differ significantly from the corresponding values for the BALB/c (48.9 ⫾ 2.0 seconds) or the C57BL/6 (46.3 ⫾ 1.7 seconds) mice. (Figures 1B and D). In addition, the 3 strains did not differ in the extent of complement deposition in their kidneys (data not shown). Despite receiving the identical immunologic insult and despite equivalent amounts of rabbit Ig deposited in their glomeruli, the NZW strain differed significantly from the normal control strains in responding to this challenge. First and foremost, the NZW strain exhibited accelerated mortality, as illustrated in Figure 2. Although all the B6 and BALB/c mice injected with NTS and all the placebo-injected mice lived for at least 21 days postinjection, NZW mice injected with NTS had a cumulative mortality rate of 40% by day 21, which was significantly higher than that in the other NTS-injected groups (P ⬍ 0.035, by Fisher’s exact test). 1083 To ascertain the likely cause of the increased mortality, we next examined these mice for evidence of compromised renal function. In our colony, the control strains rarely exhibit 24-hour urinary protein excretion levels in excess of 1 mg (22). With the dose of NTS used, BALB/c mice did not develop any significant proteinuria until day 21 (Figure 3A). B6 mice injected with NTS began excreting significantly more protein from day 18 onward (mean ⫾ SEM 2.38 ⫾ 0.54 mg/24 hours on day 18) compared with the placebo-injected B6 mice (0.77 ⫾ 0.26 mg/24 hours on day 18) (P ⬍ 0.027). The findings were very different in the NZW strain (Figures 3A and B). NTS-injected NZW mice exhibited a dramatic (⬃10-fold) increase in urinary protein excretion between day 7 and day 11. Since NTS was injected on days 4, 5, and 6, this rapid deterioration in renal function in the NZW mice was clearly occurring precipitously, within a couple of days following NTS administration. At the conclusion of the experiment, among the mice surviving on day 21, the NTS-injected NZW mice still excreted significantly more urinary protein (mean ⫾ SEM 14.93 ⫾ 2.59 mg/24 hours) than did the placebo-injected NZW mice (1.02 ⫾ 0.13 mg/24 hours) (P ⬍ 0.002), the NTS-injected B6 mice (2.17 ⫾ Figure 2. Accelerated mortality in nephrotoxic rabbit sera (NTS)– injected NZW mice. NTS-injected NZW mice (n ⫽ 10) and control mice (n ⫽ 5 preimmune rabbit sera [placebo]–injected NZW mice, 6 NTS-injected C57BL/6 [B6] mice, 8 placebo-injected B6 mice, 5 NTS-injected BALB/c mice, and 5 placebo-injected BALB/c mice) were monitored for survival (y-axis) over a period of 21 days after initiation of experiments (x-axis), as detailed in Materials and Methods. The 5 groups of mice that exhibited no mortality are represented as a single group. The NTS-injected NZW mice had significantly higher mortality rates compared with the NTS-injected B6 and BALB/c controls (P ⬍ 0.035, by Fisher’s exact test). 1084 XIE ET AL 0.79 mg/24 hours) (P ⬍ 0.002), and the NTS-injected BALB/c mice (1.3 ⫾ 0.8 mg/24 hours) (P ⬍ 0.006). Since the excreted protein in NZW mice was predominantly of high molecular weight (as determined by sodium dode- Figure 3. Increased proteinuria in nephrotoxic rabbit sera (NTS)– injected NZW mice. A, NTS-injected NZW (n ⫽ 10), preimmune rabbit sera (placebo)–injected NZW (n ⫽ 5), NTS-injected C57BL/6 (B6) (n ⫽ 6), placebo-injected B6 (n ⫽ 8), NTS-injected BALB/c (n ⫽ 5), and placebo-injected BALB/c (n ⫽ 5) mice were monitored for 24-hour urinary protein excretion (y-axis) over a period of 21 days after initiation of experiments (x-axis), as detailed in Materials and Methods. The NTS-injected NZW mice exhibited significantly elevated proteinuria compared with that in all other groups of mice. P values are for the differences between the NTS-injected NZW mice and the NTS-injected control mice (B6 and BALB/c). The numbers of dead NTS-injected NZW mice shown in parentheses are not represented at the respective time points. Values are the mean ⫾ SEM. B, Levels of 24-hour proteinuria in the 10 NZW mice injected with NTS, as measured at 3–4-day intervals. Broken line indicates the cutoff for determining positivity, which was set at 2 SD above the mean levels observed in placebo-injected NZW mice. Each solid line represents a single NZW mouse; ⴛ indicates death. Statistical comparison of these data with those in the controls is summarized in A. Figure 4. Elevated blood urea nitrogen (BUN) levels in nephrotoxic rabbit sera (NTS)–injected NZW mice. A, NTS-injected NZW (n ⫽ 10), preimmune rabbit sera (placebo)–injected NZW (n ⫽ 5), NTSinjected C57BL/6 (B6) (n ⫽ 6), placebo-injected B6 (n ⫽ 8), NTSinjected BALB/c (n ⫽ 5), and placebo-injected BALB/c (n ⫽ 5) mice were monitored for BUN levels (y-axis) over a period of 21 days after initiation of experiments (x-axis), as detailed in Materials and Methods. The NTS-injected NZW mice exhibited significantly elevated BUN levels compared with those in all other groups of mice. P values are for the differences between the NTS-injected NZW mice and the NTS-injected control mice (B6 and BALB/c). The numbers of dead NTS-injected NZW mice shown in parentheses are not represented at the respective time points. Values are the mean ⫾ SEM. B, BUN levels in the 10 NZW mice injected with NTS, as measured at regular intervals. Broken line indicates the cutoff for determining positivity, which was set at 2 SD above the mean levels observed in placeboinjected NZW mice. Each solid line represents a single NZW mouse; ⴛ indicates death. Statistical comparison of these data with those in the controls is summarized in A. cyl sulfate–polyacrylamide gel electrophoresis; data not shown), this most likely reflects glomerular proteinuria. Whereas increased urinary protein is a relatively early indication of impaired renal function, elevated INTRINSIC SUSCEPTIBILITY TO RENAL DISEASE IN NZW MICE 1085 Figure 5. Severe glomerulonephritis (GN) in nephrotoxic rabbit sera (NTS)–injected NZW mice, as determined histologically. Shown are periodic acid–Schiff–stained, formalin-fixed, paraffin-embedded renal sections from a, an NTS-injected NZW mouse, b, a preimmune rabbit sera (placebo)–injected NZW mouse, c, an NTS-injected control C57BL/6 (B6) mouse, and d, a placebo-injected B6 mouse. Shown in a are some of the typical light microscopic features seen in NTS-injected NZW mice, including intracapillary hypercellularity, with obliteration of capillary lumina, and tubular dilatation with casts. Shown in b is mild mesangial proliferative GN typical of the NTS-injected control mice. Both the placebo-injected NZW and control mice showed essentially normal renal histologic findings, as indicated in b and d, respectively. Images are representative of sections from at least 4 mice in each study group. The results are quantitatively summarized in Table 2. (Original magnification ⫻ 400.) BUN levels appear later, as a consequence of impaired glomerular clearance, often preceding death. As illustrated in Figure 4, elevated BUN levels were also prominent in the NTS-injected NZW mice. Thus, on day 14, the NTS-injected NZW mice exhibited significantly increased BUN levels (mean ⫾ SEM 83.56 ⫾ 23.60 mg/dl), in stark contrast to the placebo-injected NZW (19.24 ⫾ 2.71 mg/dl) (P ⬍ 0.0003) and the NTS-injected B6 and BALB/c mice (P ⬍ 0.036, by Mann-Whitney rank sum test). Although the BUN levels dipped beyond this time point, the differences remained until the conclusion of the experiment on day 21 (Figures 4A and B), when the NTS-injected NZW mice had a mean BUN level of 44.37 ⫾ 3.0 mg/dl, compared with 19.24 ⫾ 2.71 mg/dl in the placebo-injected NZW mice. Thus, the 3 strains clearly differed in their clinical response to the administered NTS. Whereas the BALB/c strain experienced hardly any proteinuria or mortality, the B6 strain developed mild proteinuria toward the conclusion of the experiment. In contrast, the NZW strain clearly exhibited significantly worse clinical nephritis and mortality rates in response to the same immunologic insult. To gain a better understanding of the underlying renal disease, we next examined the kidneys of these mice. The kidneys of the NTS-injected NZW mice exhibited more severe histologic changes compared with the kidneys of the NTS-injected controls (Figure 5). Following administration of NTS, the kidneys from the NZW mice exhibited significantly more severe GN (characterized by focal to diffuse intracapillary hypercellularity with obliterated capillary lumina and thickened capillary walls), tubulointerstitial nephritis, and increased glomerular crescent formation, with more 1086 XIE ET AL Table 2. Features of nephritis in 3 mouse strains on day 21 after injection of nephrotoxic rabbit sera (NTS) or preimmune rabbit sera (placebo)* C57BL/6 mice Glomerulonephritis grade Tubulointerstitial nephritis grade % of 100 glomeruli with crescents No. of macrophage infiltrates No. of polymorphonuclear leukocytes BALB/c mice NZW mice Placebo (n ⫽ 8) NTS (n ⫽ 6) Placebo (n ⫽ 4) NTS (n ⫽ 4) Placebo (n ⫽ 5) NTS (n ⫽ 6) 0⫾0 0⫾0 1.6 ⫾ 0.2† 0.3 ⫾ 0.2† 0.1 ⫾ 0.1 0⫾0 0.6 ⫾ 0.1‡ 0.3 ⫾ 0.1† 0.4 ⫾ 0.2 0.2 ⫾ 0.2 2.3 ⫾ 0.4 1.6 ⫾ 0.4 0⫾0 1.5 ⫾ 0.5§ 0⫾0 0 ⫾ 0§ 0⫾0 9.2 ⫾ 3.6 20.5 ⫾ 4.7 31.2 ⫾ 4.0§ 5.5 ⫾ 1.8 15.7 ⫾ 4.2‡ 9.8 ⫾ 0.8 64.4 ⫾ 13.5 11.5 ⫾ 1.0 14.3 ⫾ 1.3‡ 13.7 ⫾ 1.2 13.5 ⫾ 1.0‡ 16.0 ⫾ 1.4 31.3 ⫾ 2.3 * Glomerulonephritis and tubulointerstitial nephritis were graded on a 0–4 scale (see Materials and Methods). Macrophages were enumerated per 100 glomeruli, after immunohistologic staining. Polymorphonuclear leukocytes were enumerated per 50 glomeruli. Values are the mean ⫾ SEM. † P ⬍ 0.01, versus NZW kidneys, by Student’s t-test. ‡ P ⬍ 0.001, versus NZW kidneys, by Student’s t-test. § P ⬍ 0.05, versus NZW kidneys, by Student’s t-test. prominent polymorphonuclear leukocyte and macrophage infiltrates compared with the control kidneys from the B6 and BALB/c mice (Table 2). Given the prominence of macrophage and neutrophil infiltrates in the NTS-challenged NZW kidneys, it then became important to determine whether these differences were driven by coordinate differences in chemokine profiles. Among the extensive array of chemokines expressed in kidneys, MCP-1, RANTES, interferon-␥–inducible 10-kd protein (IP-10), and KC represent key mediators implicated by other investigators as playing a role in end-organ pathology. Therefore, we next ascertained the expression of these selected chemokines in NZW kidneys, using B6 kidneys as controls. Interestingly, the placebo-injected NZW kidneys (without nephrotoxic immunologic insult) exhibited increased levels of MCP-1 (mean ⫾ SEM 0.40 ⫾ 0.04 RU versus 0.22 ⫾ 0.05 RU; P ⬍ 0.015), IP-10 (0.90 ⫾ 0.04 RU versus 0.67 ⫾ 0.11 RU; P ⬍ 0.07), and KC (0.92 ⫾ 0.11 RU versus 0.41 ⫾ 0.13 RU; P ⬍ 0.01) compared with the kidneys of the placebo-injected B6 mice (Figure 6). However, the level of RANTES expression was similar in both of these groups of placebo-injected mice. NTS administration boosted the levels of most chemokines in the kidneys of both strains. Notably, the kidneys from the NTS-injected NZW mice exhibited significantly higher levels of message for RANTES (0.69 ⫾ 0.12 RU versus 0.37 ⫾ 0.005 RU; P ⬍ 0.028) and KC (1.68 ⫾ 0.10 RU versus 1.04 ⫾ 0.10 RU; P ⬍ 0.001) compared with the kidneys from NTS-injected B6 mice (Figure 6). Although an elevation was also noted in the message levels of MCP-1, no increases were seen in the levels of Figure 6. Elevated chemokine expression in nephrotoxic rabbit sera (NTS)–injected NZW mice. Levels of expression of monocyte chemoattractant protein 1 (MCP-1), RANTES, KC, and interferon-␥– inducible 10-kd protein (IP-10) in the renal cortices of NTS-injected or preimmune rabbit sera (placebo)–injected NZW and C57BL/6 (B6) control mice were semiquantitated by reverse transcription– polymerase chain reaction, as detailed in Materials and Methods. All kidneys were obtained on day 21, at the conclusion of the experiment, with 1 exception (the NTS-injected NZW group includes 6 kidneys harvested on day 21 and 1 kidney rescued from the mouse that died on day 18). Bars show the mean for each group. Chemokine expression levels are expressed as relative units, representing the ratio of chemokine band intensity to that of a housekeeping gene, GAPDH. P values shown at the bottom right of each plot are for the differences between the NTS-injected NZW mice and the placebo-injected NZW mice (top value) as well as the NTS-injected B6 mice (bottom value). INTRINSIC SUSCEPTIBILITY TO RENAL DISEASE IN NZW MICE Figure 7. Increased urinary excretion of monocyte chemoattractant protein 1 (MCP-1) and KC over 24 hours in nephrotoxic rabbit sera (NTS)–injected NZW mice, as measured by enzyme-linked immunosorbent assay on day 21 after NTS injection. Bars show the mean for each group. P values are for the differences between the NTS-injected NZW mice and the NTS-injected controls, by Student’s t-test. B6 ⫽ C57BL/6. Figure 8. Anti-rabbit Ig response in the nephrotoxic rabbit sera (NTS)–injected NZW and C57BL/6 (B6) control mice. Serum levels of A, IgG1, B, IgG2a, and C, IgG2b mouse anti-rabbit Ig antibodies (Abs) were quantitated in the NTS-injected NZW mice (n ⫽ 6) and the NTS-injected control mice on day 21, at the conclusion of the experiment. D, Serum levels of (total) IgG mouse anti-rabbit Ig and 24-hour urinary protein excretion in the NTS-injected NZW mice on day 21. The correlation coefficient for these 2 parameters is indicated by the r value. E, Levels of mouse IgG deposited in the glomeruli of NTS-challenged mice, expressed in autofluorescence units (AU). For each kidney section, autofluorescence of 10 glomeruli was quantitated and averaged. No significant differences were observed between the groups. Broken line indicates the mean autofluorescence observed in preimmune rabbit sera (placebo)–injected control mice. 1087 1088 IP-10 (Figure 6) in these 2 groups. These message differences were paralleled by increased urinary excretion of these chemokines, as depicted for MCP-1 and KC in Figure 7. Collectively, the above data indicated that the NZW mice had worse renal disease than the B6 and BALB/c controls despite having received the identical immunologic insult. To exclude the possibility that the exaggerated disease response seen in the NZW strain might be the consequence of a stronger systemic immune response to the injected rabbit Ig, we compared the serum levels of mouse anti-rabbit antibodies in the NZW mice with those in the B6 controls. As shown in Figures 8A–C, the NTS-injected NZW mice did not exhibit a higher anti-rabbit xenogeneic response than that in the controls. Importantly, the isotype distribution of mouse antibodies to rabbit Ig did not differ significantly among the 3 groups. Moreover, when the degree of anti-rabbit immune response in the NTS-injected mice and the extent of proteinuria at the same time points were examined in parallel, the correlation between these 2 parameters was poor (Pearson’s product-moment coefficient r ⫽ –0.51) (Figure 8D). In addition, the NTS-injected NZW and control mice did not differ in the extent of mouse IgG deposits in their kidneys, as quantitated by autofluorescence (Figure 8E). These observations indicate that the accentuated renal disease seen in NZW mice is not simply the result of an exaggerated or altered anti-rabbit (xenogeneic) immune response against the administered rabbit NTS. Most of the above data were derived from the sera, urine, and tissues procured on day 21 of experimentation. Since several of the phenotypes appeared to dip toward the conclusion of the experiment (Figures 3 and 4), we were concerned that we may not be studying the pathologic changes in these mice at the peak of their disease. Hence, we repeated the above studies using an abbreviated 11-day protocol, where NZW mice and B6 controls were challenged with NTS as before, but were then killed on day 11 for examination. As illustrated in Figure 9, and as predicted, the clinical parameters (as measured by proteinuria and BUN) and the histologic features (as marked by the GN scores) were even more pronounced in NZW mice at this earlier time point. Ongoing studies are focused on evaluating groups of mice at daily intervals following NTS administration to elucidate the detailed kinetics of GN onset in these strains. Thus, both the conventional and abbreviated experimental GN induction protocols yielded similar XIE ET AL Figure 9. Features of anti–glomerular basement membrane disease at the peak of pathologic changes. To confirm our findings in the other studies and to ascertain the degree of pathologic changes at the peak of the disease, C57BL/6 (B6) and NZW mice were challenged with nephrotoxic rabbit sera (NTS) or preimmune rabbit sera (placebo) (n ⫽ 4–7 per group) and killed on day 11, as detailed in Materials and Methods. Shown are the levels of A, proteinuria, B, blood urea nitrogen (BUN), and C, glomerulonephritis (GN) scores on day 11. Although no premature deaths were observed in the NTS-injected B6 mice or the placebo-injected controls, 2 of the 7 NZW NTS-injected mice died before day 11. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 and ⴱⴱ ⫽ P ⬍ 0.01 versus B6 mice, by Student’s t-test. INTRINSIC SUSCEPTIBILITY TO RENAL DISEASE IN NZW MICE Figure 10. Nephrotoxic rabbit sera (NTS)–induced anti–glomerular basement membrane disease in (B6 ⫻ NZW)F1 mice. To determine if the enhanced susceptibility to renal disease in the NZW strain was a dominant or recessive phenotype, (B6 ⫻ NZW)F1 hybrids were bred and subjected to NTS challenge, using the 21-day protocol. Solid lines show the urinary protein levels observed in the “negative control” (C57BL/6 [B6]) and “positive control” (NZW) strains (see Figure 3), as indicated. Each data point represents an individual F1 mouse. On days 11, 14, 18, and 21, the mean levels of 24-hour proteinuria in the F1 mice were significantly higher than those in the B6 controls (P ⬍ 0.01). Placebo-injected F1 mice did not develop significant proteinuria (data not shown). conclusions: NZW mice were significantly more prone to renal disease following an immunologic insult than were B6 and BALB/c controls. As a first step in exploring the genetic basis of this difference, we next bred (B6 ⫻ NZW)F1 hybrids, and challenged them with NTS and placebo. As shown in Figure 10, (B6 ⫻ NZW)F1 mice were clearly as susceptible to renal disease as the NZW parents were, indicating the contribution of one or more dominant NZW disease genes toward this phenotype. It would clearly be important to elucidate the identity of these culprit genes in future studies. DISCUSSION The phenotypically normal NZW strain was first described by W. H. Hall at the Otago Medical School, Dunedin, New Zealand (1). Soon after, it was quickly noted that when this strain was bred to the NZB strain, the F1 offspring succumbed to lupus nephritis, with features that were strikingly similar to those seen in humans with systemic lupus erythematosus (2,3,23). Indeed, over the last 3 decades, studies with this hybrid strain and its derivatives have broadened our under- 1089 standing of the pathogenic mechanisms leading to lupus (3,24,25). Subsequent studies revealed that although the NZW strain was phenotypically normal, with aging the strain exhibited histologic changes suggestive of GN (7–9). However, the relative contributions of the systemic immune components versus the intrinsic renal factors to these observed pathologic changes have remained unexplored. In this context, the nephrotoxic serum nephritis (or, Masugi nephritis) model has been very useful in discriminating “systemic” factors from intrinsic renal factors in facilitating renal disease. Indeed, this very approach has been instrumental in demonstrating the respective roles of several different molecules in mediating end-organ pathology (11–19). If the NZW mice did not really differ from the B6 or BALB/c controls with respect to how they handled and responded to immunologic insults, then one would have expected the NZW and the control strains to have responded similarly when challenged with NTS (which clearly target the glomeruli of these strains in a similar manner, as demonstrated in Figure 1). This was not the case, however, since the NTS-injected NZW mice experienced greater renal pathology and higher mortality compared with the NTSinjected control strains. Indeed, the deterioration in renal function was noted to occur within a couple of days following NTS administration. As detailed in Materials and Methods, NTS administration was completed on day 6. On day 7, the mice in the different groups did not differ significantly from each other in terms of urinary protein excretion or BUN level. However, at the next time point of study 4 days later (day 11), the NTS-injected NZW mice differed sharply from the NTS-injected controls, both in the 21-day and the 11-day experiments: they exhibited significantly worse pathologic changes (Figure 9C), elevated BUN levels (Figure 9B), and higher proteinuria (Figures 3 and 9A). This was accompanied by death in several of these mice (Figure 2 and legend to Figure 9). Thus, the immunopathologic pathways triggered by the administration of NTS appear to act very rapidly, within a couple of days, in a manner sufficient to induce death. We believe that the dominant cause of death in these mice is compromised renal function, for several reasons. First, most mice that died had sharply diminished urine output but high urine protein concentration, indicative of acute renal shutdown (Figure 3B). Second, several histologic features of disease were observed in the NZW mice that may potentially account for the 1090 worse clinical outcome. This includes the more proliferative nature of the GN (GN scores of 3 or 4) accompanied by the significantly increased numbers of glomerular crescents, neutrophils, and macrophages (Table 2 and Figures 5 and 9). It was also interesting to note that these mice exhibited tubulointerstitial disease despite having no Ig deposits on their tubular basement membranes. It remains to be established whether the observed tubulointerstitial disease in the NTS-injected NZW mice was a secondary consequence of the massive proteinuria seen in these mice. Finally, no evidence of disease elsewhere was apparent by morphologic observation or histologic examination of the lungs from these mice. Thus far, lung involvement in anti-GBM disease has not been a prominent feature in animal models since additional pulmonary insults (such as cigarette smoking) appear to be required (to damage the alveolar endothelial lining) before this phenotype can ensue, as is the case even in patients with Goodpasture’s disease (26). Both CC chemokines that chemoattract monocytes and CXC chemokines that chemoattract neutrophils have been implicated by other investigators to play major roles in spontaneously arising, as well as experimentally induced, models of immune-mediated nephritis (14,15,27–38). In this study, we therefore ascertained the expression of 2 CC chemokines (MCP-1 and RANTES) and 2 CXC chemokines (KC and IP-10). The importance of MCP-1 and RANTES in immune-mediated nephritis has been demonstrated by several end-organ– expression studies, as well as by antibody-blocking and knockout studies (14,33,34,36). KC serves as an important chemoattractant for neutrophils and mesangial cells, and it also functions to amplify the production of other chemokines in the end organs (29). IP-10 production is also elevated in experimentally induced nephritis, and this CXC chemokine plays an important role in recruiting neutrophils (38). It is interesting to note that the placebo-injected NZW mice in the present study exhibited elevated basal levels of MCP-1, IP-10, and KC compared with the levels in the B6 controls. This might suggest that the NZW kidneys exist in a somewhat “triggered” state, even before they are challenged with nephrotoxic antibodies. The genetic basis for this increased basal state in the NZW kidneys remains unexplored. More interestingly, the NZW kidneys exhibited a further increases in RANTES and KC levels, compared with those in B6 kidneys, upon NTS administration (Figure 6); this indicates the potential relevance of these chemokines in inducing the more severe disease seen in these mice. In contrast, the difference in IP-10 expression between the XIE ET AL B6 and the NZW strains did not appear to correlate with the respective disease outcomes. It is likely, however, that the chemokine differences we found may represent only a minor fraction of the molecular differences between the end organs of these different strains. The impending challenge is to identify any additional molecular differences that may account for the increased end-organ disease susceptibility exhibited by the NZW strain. One possible explanation for the more severe disease in the NZW strain may be an exaggerated systemic immune response to the administered rabbit Ig as compared with the response of the control strains. There are several reasons why this is unlikely. First, previous studies with the nephrotoxic serum–induced nephritis model have indicated that the host’s humoral immune response does not contribute significantly to the observed pathologic features, since host mice devoid of B cells developed a similar disease (39). Second, the NZW strain is not known to have heightened immune responsiveness (40). Third, the NZW mice receiving preimmune rabbit sera did not develop any pathologic changes or disease, suggesting that the observed endorgan pathologic changes were absolutely dependent upon the presence of a nephrotoxic specificity (i.e., NTS is absolutely required). Fourth, elevated proteinuria was noted in the NZW mice within a couple of days following the administration of NTS. This makes it unlikely for an immune response to the rabbit NTS to have been the cause of the differential pathologic changes observed. Finally, even when the mouse anti-rabbit Ig (xenogeneic) response following the NTS administration was measured, the NZW mice and the controls both exhibited a xenogeneic immune response that was quantitatively and qualitatively similar (Figure 8). It also appeared unlikely that the observed differences were due to Th1/Th2 differences between the strains. Both the B6 strain (which is Th1-skewed) and the BALB/c strain (which is Th2-skewed) developed minimal disease compared with the NZW strain. Furthermore, we could not discern any differences in the serum levels of IgG1 and IgG2a mouse anti-rabbit antibody in these different strains (Figures 8A–C). Given the above observations, we believe that the increased disease seen in the NZW strain is most likely a direct consequence of local pathogenic cascades triggered by the administered NTS in the kidneys. However, it should be stressed that the present findings do not conclusively establish whether the NZW genome encodes resident glomerular cells that are intrinsically abnormal or whether the observed phenotypes reflect INTRINSIC SUSCEPTIBILITY TO RENAL DISEASE IN NZW MICE intrinsic abnormalities in the infiltrating leukocytes. Thus, with respect to one of the phenotypes observed in NZW mice, increased chemokine expression, for example, 2 non–mutually exclusive scenarios are possible. First, it is possible that the NZW genome may be affecting chemokine elaboration by intrinsic glomerular cells. Alternatively, the NZW genome may be dictating the responsiveness of infiltrating leukocytes to these mediators. Renal transplant–based approaches as well as studies with NZW-derived glomerular cell lines are in progress to distinguish between these possibilities. The results of these studies raise the possibility that a lupus-facilitating genome may contribute to disease susceptibility by encoding end organs that are potentially more “sensitive” to immune-mediated damage. Indeed, it is becoming increasingly apparent that enhanced intrinsic susceptibility to other varieties of end-organ disease may also be genetically determined (41,42). Given the observation that the NZW genome confers increased susceptibility to immune-mediated GN in a dominant manner, it next becomes important to ascertain which loci/genes within the NZW genome might actually be responsible for the observed differences. Indeed, the NZW genome is notoriously riddled with several lupus susceptibility loci, including Sle1, (proximal part of) Sle2, Sle3, Sle5, and Sle6, many of which are linked to GN susceptibility and function in a dose-dependent manner (5,43). Studies of B6 mice rendered congenic for these different NZW-derived intervals have recently revealed how several of these loci have the potential to affect systemic immunity, either by themselves or in epistasis with other loci (44–51). It remains to be determined whether any of these loci might serve to directly modulate susceptibility to endorgan disease. Dissecting out the genetic origins and molecular mediators of acute experimental nephritis, as well as spontaneous lupus nephritis, using these novel NZW-derived genetic models is the challenge that lies ahead. ACKNOWLEDGMENTS The authors would like to acknowledge Drs. Christopher Lu and Michael P. Madaio for critical reading of the manuscript. REFERENCES 1. Bielschowsky M, Goodall CM. Origin of inbred NZ mouse strains. Cancer Res 1970;30:834–6. 2. Helyer BJ, Howie JB. Renal disease associated with positive lupus erythematosus in cross-bred strains of mice. Nature 1963;197:197. 1091 3. Theofilopoulos AN, Dixon FJ. Murine models of systemic lupus erythematosus. Adv Immunol 1985;37:269–390. 4. Hang LM, Izui S, Dixon FJ. (NZW ⫻ BXSB)F1 hybrid. A model of acute lupus and coronary vascular disease with myocardial infarction. J Exp Med 1981;154:216–21. 5. Morel L, Tian XH, Croker BP, Wakeland EK. Epistatic modifiers of autoimmunity in a murine model of lupus nephritis. Immunity 1999;11:131–9. 6. Izui S, Higaki M, Morrow D, Merino R. The Y chromosome from autoimmune BXSB/MpJ mice induces a lupus-like syndrome in (NZW ⫻ C57BL/6)F1 male mice, but not in C57BL/6 male mice. Eur J Immunol 1988;18:911–5. 7. Braverman IM. Study of autoimmune disease in New Zealand mice. I. Genetic features and natural history of NZB, NZY and NZW strains and NZB-NZW hybrids. J Invest Dermatol 1968;50: 483–99. 8. Hahn BH, Shulman LE. Autoantibodies and nephritis in the white strain (NZW) of New Zealand mice. Arthritis Rheum 1969;12: 355–64. 9. Kelley VE, Winkelstein A. Age- and sex-related glomerulonephritis in New Zealand white mice. Clin Immunol Immunopathol 1980;16:142–50. 10. Masugi M. Glomerulonephritis durch das spezifische antinierenserum: ein Beitrag zur Pathogenese der diffusen Glomerulonephritis. Beitr Pathol Anat Allgemeine Pathol 1934;92:429–429. 11. Sheerin NS, Springall T, Carroll MC, Hartley B, Sacks SH. Protection against anti-glomerular basement membrane (GBM)mediated nephritis in C3- and C4-deficient mice. Clin Exp Immunol 1997;110:403–9. 12. Park SY, Ueda S, Ohno H, Hamano Y, Tanaka M, Shiratori T, et al. Resistance of Fc receptor-deficient mice to fatal glomerulonephritis. J Clin Invest 1998;102:1229–38. 13. Suzuki Y, Shirato I, Okumura K, Ravetch JV, Takai T, Tomino Y, et al. Distinct contribution of Fc receptors and angiotensin IIdependent pathways in anti-GBM glomerulonephritis. Kidney Int 1998;54:1166–74. 14. Tesch GH, Schwarting A, Kinoshita K, Lan HY, Rollins BJ, Kelley VR. Monocyte chemoattractant protein-1 promotes macrophagemediated tubular injury, but not glomerular injury, in nephrotoxic serum nephritis. J Clin Invest 1999;103:73–80. 15. Topham PS, Csizmadia V, Soler D, Hines D, Gerard CJ, Salant DJ, et al. Lack of chemokine receptor CCR1 enhances Th1 responses and glomerular injury during nephrotoxic nephritis. J Clin Invest 1999;104:1549–57. 16. Janssen U, Ostendorf T, Gaertner S, Eitner F, Hedrich HJ, Assmann KJ, et al. Improved survival and amelioration of nephrotoxic nephritis in intercellular adhesion molecule-1 knockout mice. J Am Soc Nephrol 1998;9:1805–14. 17. Tang T, Rosenkranz A, Assmann KM, Goodman MJ, GutierrezRamos JC, Carroll MC, et al. A role for Mac-1 (CD11b/CD18) in immune complex-stimulated neutrophil function in vivo: Mac-1 deficiency abrogates sustained Fc␥ receptor-dependent neutrophil adhesion and complement-dependent proteinuria in acute glomerulonephritis. J Exp Med 1997;186:1853–63. 18. Mayadas TN, Mendrick DL, Brady HR, Tang T, Papayianni A, Assmann KJ, et al. Acute passive anti-glomerular basement membrane nephritis in P-selectin-deficient mice. Kidney Int 1996;49: 1342–9. 19. Mayadas TN, Rosenkranz A, Cotran RS. Glomerular inflammation: use of genetically deficient mice to elucidate the roles of leukocyte adhesion molecules and Fc-␥ receptors in vivo. Curr Opin Nephrol Hypertens 1999;8:293–8. 20. Zhou XJ, Laszik Z, Wang XQ, Silva FG, Vaziri ND. Association of renal injury with increased oxygen free radical activity and altered nitric oxide metabolism in chronic experimental hemosiderosis. Lab Invest 2000;80:1905–14. 21. Kawachi H, Matsui K, Orikasa M, Morioka T, Oite T, Shimizu F. 1092 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. Quantitative studies of monoclonal antibody 5-1-6 induced proteinuric state in rats. Clin Exp Immunol 1992;87:215–9. Shi X, Xie C, Kreska D, Richardson J, Mohan C. Genetic dissection of SLE: Sle1 and Fas impact distinct pathways leading to lymphoproliferative autoimmunity. J Exp Med 2002;196:281–92. Burnet FM, Holmes MC. The natural history of the NZB/NZW F1 hybrid mouse: a laboratory model of systemic lupus erythematosus. Aust Ann Med 1965;14:185–91. Mohan C, Datta SK. Lupus: key pathogenic mechanisms and contributing factors. Clin Immunol Immunopathol 1995;77: 209–20. Kotzin BL. Systemic lupus erythematosus. Cell 1996;85:303–6. Donaghy M, Rees AJ. Cigarette smoking and lung hemorrhage in glomerulonephritis caused by autoantibodies to glomerular basement membrane. Lancet 1983;2:1390–3. DeLema GP, Maier H, Nieto E, Vielhauer V, Luckow B, Mampaso F, et al. Chemokine expression precedes inflammatory cell infiltration and chemokine receptor and cytokine expression during the initiation of murine lupus nephritis. J Am Soc Nephrol 2001;12:1369–82. Schadde E, Kretzler M, Banas B, Luckow B, Assmann K, Schlondorff D. Expression of chemokines and their receptors in nephrotoxic serum nephritis. Nephrol Dial Transplant 2000;15:1046–53. Luo Y, Lloyd C, Gutierrez-Ramos JC, Dorf ME. Chemokine amplification in mesangial cells. J Immunol 1999;163:3985–92. Kelley VR, Wuthrich RP. Cytokines in the pathogenesis of systemic lupus erythematosus. Semin Nephrol 1999;19:57–66. Tang WW, Yin S, Wittwer AJ, Qi M. Chemokine gene expression in anti-glomerular basement membrane antibody glomerulonephritis. Am J Physiol 1995;269:F323–30. Tang WW, Qi M, Warren JS. Monocyte chemoattractant protein 1 mediates glomerular macrophage infiltration in anti-GBM antibody GN. Kidney Int 1996;50:665–71. Lloyd CM, Minto AW, Dorf ME, Proudfoot A, Wells TN, Salant DJ, et al. RANTES and monocyte chemoattractant protein-1 (MCP-1) play an important role in the inflammatory phase of crescentic nephritis, but only MCP-1 is involved in crescent formation and interstitial fibrosis. J Exp Med 1997;185:1371–80. Fujinaka H, Yamamoto T, Takeya M, Feng L, Kawasaki K, Yaoita E, et al. Suppression of anti-glomerular basement membrane nephritis by administration of anti-monocyte chemoattractant protein-1 antibody in WKY rats. J Am Soc Nephrol 1997;8: 1174–8. Segerer S, Cui Y, Hudkins KL, Goodpaster T, Eitner F, Mack M, et al. Expression of the chemokine monocyte chemoattractant protein-1 and its receptor chemokine receptor 2 in human crescentic glomerulonephritis. J Am Soc Nephrol 2000;11:2231–42. Panzer U, Thaiss F, Zahner G, Barth P, Reszka M, Reinking RR, et al. Monocyte chemoattractant protein-1 and osteopontin differentially regulate monocyte recruitment in experimental glomerulonephritis. Kidney Int 2001;59:1762–9. XIE ET AL 37. Wada T, Furuichi K, Segawa-Takaeda C, Shimizu M, Sakai N, Takeda SI, et al. MIP-1␣ and MCP-1 contribute to crescents and interstitial lesions in human crescentic glomerulonephritis. Kidney Int 1999;56:995–1003. 38. Romagnani P, Beltrame C, Annunziato F, Lasagni L, Luconi M, Galli G, et al. Role for interactions between IP-10/Mig and CXCR3 in proliferative glomerulonephritis. J Am Soc Nephrol 1999;10:2518–26. 39. Li S, Holdsworth SR, Tipping PG. Antibody independent crescentic glomerulonephritis in mu chain deficient mice. Kidney Int 1997;51:672–8. 40. Milich D, Gershwin EM. The pathogenesis of autoimmunity in New Zealand mice. Semin Arthritis Rheum 1980;10:111–46. 41. McBrearty BA, Clark LD, Zhang XM, Blankenhorn EP, HeberKatz E. Genetic analysis of a mammalian wound-healing trait. Proc Natl Acad Sci U S A 1998;95:11792–7. 42. Kuan AP, Chamberlain W, Malkiel S, Lieu HD, Factor SM, Diamond B, et al. Genetic control of autoimmune myocarditis mediated by myosin-specific antibodies. Immunogenetics 1999;49: 79–85. 43. Morel L, Rudofsky UH, Longmate JA, Schiffenbauer J, Wakeland EK. Polygenic control of susceptibility to murine systemic lupus erythematosus. Immunity 1994;1:219–29. 44. Morel L, Mohan C, Yu Y, Croker B, Tian X-H, Deng A, et al. Functional dissection of SLE pathogenesis using congenic mouse strains. J Immunol 1997;158:6019–28. 45. Mohan C, Morel L, Croker B, Yang P, Wakeland EK. Genetic dissection of SLE pathogenesis: Sle2 on murine chromosome 4 leads to B-cell hyperactivity. J Immunol 1997;159:454–65. 46. Mohan C, Morel L, Yang P, Wakeland EK. Accumulation of splenic B1a cells with potent antigen-presenting capability in NZM2410 lupus-prone mice. Arthritis Rheum 1998;41:1652–62. 47. Mohan C, Alas E, Morel L, Yang P, Wakeland EK. Genetic dissection of SLE pathogenesis: Sle1 on murine chromosome 1 leads to a selective loss of tolerance to H2A/H2B/DNA subnucleosomes. J Clin Invest 1998;101:1362–72. 48. Mohan C, Yu Y, Morel L, Yang P, Wakeland EK. Genetic dissection of SLE pathogenesis: Sle3 on murine chromosome 7 impacts T cell activation, differentiation and cell death. J Immunol 1999;162:6492–502. 49. Mohan C, Morel L, Yang P, Watanabe H, Croker B, Gilkeson G, et al. Genetic dissection of lupus pathogenesis: a recipe for nephrophilic autoantibodies. J Clin Invest 1999;103:1685–95. 50. Morel L, Croker BP, Blenman KR, Mohan C, Huang G, Gilkeson G, et al. Genetic reconstitution of systemic lupus erythematosus immunopathology with polycongenic murine strains. Proc Natl Acad Sci U S A 2000;97:6670–5. 51. Mohan C. Murine lupus genetics: lessons learned. Curr Opin Rheumatol 2001;13:352–60.