Background

Renal parenchymal cells produce cytokines, colony-stimulating factor-1 (CSF-1), granulocyte-macrophage colony-stimulating factor (GM-CSF), and tumor necrosis factor- (TNF-), which recruit autoreactive T cells and, in turn, elicit renal injury in MRL-Fas^lpr mice.

Methods

To determine whether select T-cell populations regulate intrarenal nephritogenic cytokines (CSF-1, GM-CSF, and TNF-) and renal disease, we compared MRL-Fas^lpr mice that are genetically deficient in T-cell receptor (TCR) T cells, CD4 T cells, and major histocompatibility complex class I (MHC class I), lacking CD8 and double negative (DN) T cells, with wild-type mice. To identify the T cells instrumental in downstream (effector) events, we delivered CSF-1 or GM-CSF into the kidney via gene transfer in these select T-cell–deficient and wild-type strains.

Results

Intrarenal CSF-1, GM-CSF, and TNF- were absent or dramatically reduced in TCR , CD4, and class I-deficient MRL-Fas^lpr strains as compared with wild-type mice. In addition, the decrease in CSF-1, GM-CSF, and TNF- was associated with a reduced kidney leukocytic infiltrates and spontaneous autoimmune nephritis. Intrarenal ex vivo retroviral gene transfer of CSF-1 and GM-CSF failed to elicit nephritis in these T-cell–deficient MRL strains (TCR , CD4, CD8/DN) as compared with wild-type mice.

Conclusions

Multiple T-cell populations initiate renal disease by increasing intrarenal nephritogenic cytokines, CSF-1, GM-CSF, and TNF-. CSF-1 and GM-CSF recruit additional CD4 and CD8 and DN T cells, which augment downstream events, resulting in progressive autoimmune renal disease. We suggest that MRL-Fas^lpr kidney disease is driven by a T-cell amplification feedback loop dependent on multiple T-cell populations.

METHODS

Mice

Autoimmune MRL-Fas^lpr (H-2^k) and MRL-++ (H-2^k) were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). TCR -deficient MRL-Fas^lpr mice were derived through a series of six generations of backcrosses as previously reported^9,12. Similarly, CD4-deficient MRL-Fas^lpr mice were derived through a series of seven to eight generations of backcrosses¹¹. The ₂-microglobulin (₂m), MHC class I, deficient MRL-Fas^lpr mice at the 10th backcross generation were purchased from Jackson Laboratory¹⁰. Mice were maintained in our animal facility on standard laboratory chow.

Reagents

Tissue culture media and supplements were purchased from GIBCO/BRL Life Technologies (Grand Island, NY, USA), and chemicals were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Polyclonal rabbit anti-mouse L cell-derived CSF-1 and polyclonal rabbit anti-human CSF-1 were provided by Dr. R. Shadduck (Montefiore Hospital, Pittsburgh, PA, USA) and the Genetics Institute (Cambridge, MA, USA), respectively. Rabbit antimurine TNF- antibody was a gift from Dr. A. Cerami (Picower Institute for Medical Research, Manhasset, NY, USA)¹³. Monoclonal antibody for F4/80 antigen (American Tissue Culture Collection, Rockville, MD, USA) was purified by affinity chromatography using protein A-Sepharose CL-4B columns (Pharmacia, Piscataway, NJ, USA). Monoclonal antibodies to CD4, CD8, B220, biotinylated antimurine GM-CSF antibody, and fluorescence-conjugated murine IgG were purchased from PharMingen (San Diego, CA, USA).

Cytokine expression

Intrarenal transcripts

We evaluated intrarenal CSF-1, TNF-, and GM-CSF transcripts in the renal cortex by Northern blot analysis. Total RNA was extracted from the renal cortex using RNAzol B (Tel-Test, Friendswood, TX, USA), a modification of the guanidium thiocyanate-phenolchloroform method¹⁴. RNA was washed with 70% ethanol, resuspended in diethylpyrocarbonate (DEPC)-treated H₂O and stored at -80°C. Total RNA (20 g) was electrophoresed through a 1% agarose-formaldehyde gel, blotted to nylon membrane, and hybridized in 50% formamide with ³²P-labeled nick-translated probes at 42°C. Hybridized membranes were washed in 2 standard saline citrate (SSC), 0.1% sodium dodecyl sulfate (SDS) at room temperature and then washed in 0.2 SSC, 0.1% SDS at 60°C. The CSF-1, TNF-, and GM-CSF probes, kindly provided by Dr. R. Stanley (Albert Einstein College of Medicine, New York, NY, USA), Dr. K. Matsushima (University of Tokyo, Tokyo, Japan), and Dr. G. Dranoff (Dana-Faber Cancer Institute, Boston, MA, USA), respectively, consisted of 594, 500, and 400 bp fragments of the plasmid containing the cDNA. Blots were reprobed with -actin (Pst-1 fragment of pBA-1) as an internal control to assess the RNA quantity and integrity. The blots were analyzed by densitometry to quantitate mRNA.

Intrarenal proteins

We detected CSF-1, TNF-, and GM-CSF in kidney sections using polyclonal rabbit anti-human CSF-1 antibody (10 g/mL), rabbit anti-murine TNF- antibody (1:150 dilution), and biotinylated rat antimurine GM-CSF antibody (50 g/mL) using the immunoperoxidase technique^3,15. Tissue sections were incubated with CSF-1, TNF-, and GM-CSF antibodies in 1% bovine serum albumin (BSA) overnight at 4°C. Specificity controls included (1) replacing the primary antibody with either normal rat IgG or rabbit serum and (2) antibody neutralization by incubating antimurine CSF-1, TNF-, and GM-CSF antibodies with a 20-fold molar excess of recombinant murine CSF-1, TNF-, and GM-CSF, respectively (R&D Systems, Minneapolis, MN, USA). We evaluated> 50 glomeruli and> 20 interstitium/perivascular areas containing> 400 tubules per kidney in randomly selected microscopic fields (magnification 400). The numbers of cells expressing CSF-1, TNF-, and GM-CSF were graded on a scale from 0 to 3 (0, 0 cells per field; 1, mild = <10 cells per field; 2, moderate = 10 to 100 cells per field; 3, maximum => 100 cells per field) and determined the mean grade as shown in Figure 1. Scoring was performed on coded slides.

Figure 1.

T-cell receptor (TCR) , CD4, and class I-selected double negative (DN) and CD8 T cells are required for intrarenal colony stimulating factor-1 (CSF-1), tumor necrosis factor- (TNF-), and granulocyte-macrophage colony-stimulating factor (GM-CSF). CSF-1 and GM-CSF were not detected in MRL-Fas^lpr strains deficient in TCR or CD4. CSF-1 was reduced and GM-CSF was absent in ₂m-deficient MRL-Fas^lpr kidneys. In comparison, the amount of TNF- was reduced in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m. N = 2 to 4 per group at three and six months of age. Values are mean SEM. *MRL-Fas^lpr mice versus MRL-++ and T-cell–deficient strains, P < 0.05; +₂m-deficient MRL-Fas^lpr mice vs. MRL-++, TCR -deficient MRL-Fas^lpr, CD4-deficient MRL-Fas^lpr mice.

Full figure and legend (26K)

Renal disease

Renal pathology

We examined kidneys from wild-type MRL-Fas^lpr mice, T-cell–deficient MRL-Fas^lpr strains, and MRL-++ mice at three and six months of age. Kidney sections evaluated by histopathology were fixed with 10% buffered formalin and then paraffin embedded and stained with hematoxylin and eosin. Coded slides were analyzed by two individuals. The entire renal pathology (consisting of glomerulonephritis, interstitial/perivascular nephritis and tubular damage) was evaluated using a grading system ranging from 0 to 3: 0 = none, 1 = mild, 2 = moderate, and 3 = severe to generate data in Figure 4. The extent of renal pathology was assessed by determining (1) the percentage of crescents (defined as thickening of Bowman's capsule wall with 2 or more cell layers); (2) the percentage of segmental lesions (exhibiting at least one of the following: necrosis, proliferation, hyalinosis) in glomeruli; (3) the percentage of damaged tubuli (consisting of at least one of the following: dilatation, atrophy, necrosis) in randomly selected microscopic fields (magnification 400) to generate data in Table 1. We determined the cell number in the glomeruli, interstitial, and vascular areas to generate Figure 5. Specifically, we counted cells in the intraglomerular (glomerular resident cells, leukocytic infiltrates, and cells in crescents within Bowman's capsule) and periglomerular areas (outside Bowman's capsule) of 50 randomly selected glomeruli and evaluated the mean enumerated cells per glomerulus Figure 5a,b. Kidney leukocytic infiltrates within the cortical interstitium were counted in 20 randomly selected fields of cortical interstitium and mean cells per field calculated Figure 5c. The perivascular leukocytic cells were evaluated in 20 interlobular and intralobular arteries and graded by counting the number of cell layers surrounding each vessel: 0 = none; 1 = <5 layers surrounding <50% of the vessel; 2 = 5 to 10 layers surrounding> 50% of the vessel; 3 => 10 layers. Finally, the mean grade was determined Figure 5d.

Figure 4.

TCR , CD4, and class I-selected DN and CD8 T cells are required for spontaneous kidney disease in MRL-Fas^lpr mice. (1) Renal pathology was decreased in T-cell–deficient MRL-Fas^lpr mice, but remained elevated as compared with MRL-++ mice. (2) Glomerulonephritis in MRL-Fas^lpr mice at six months of age (A). In contrast, renal pathology was reduced in TCR -deficient MRL-Fas^lpr (B) in CD4-deficient MRL-Fas^lpr (C) and in ₂m-deficient MRL-Fas^lpr (D) mice (hematoxylin eosin, 500).

Full figure and legend (130K)

Figure 5.

TCR , CD4, and class I-selected CD8 and DN T cells contribute to spontaneous renal disease in MRL-Fas^lpr mice. T-cell–deficient MRL-Fas^lpr strains had fewer cells in the kidney within the intraglomerular (A), periglomerular (B), interstitial (C), and perivascular (D) areas than in the corresponding areas in MRL-Fas^lpr kidneys, but more than in the MRL-++ kidneys. Values are mean SEM; N = 3 to 5 per group at six months of age.

Full figure and legend (33K)

Table 1 - T cell-deficient MRL strains were protected from kidney injury (proteinuria, IgG deposition and renal pathology).

Full table

Proteinuria

Urine protein levels in T-cell–deficient MRL-Fas^lpr mice were assessed semiquantitatively using albumin reagent strips (0 = none, 1 = 30 to 100 mg/dL, 2 = 100 to 300 mg/dL, 3 = 300 to 1000 mg/dL, 4 => 1000 mg/dL; Albustix; Miles Scientific, Naperville, IL, USA), as shown in Table 1.

Deposition of IgG

Semiquantitative evaluation of the glomerular deposition of mouse IgG was performed by direct immunofluorescence staining using a semiquantitative grading system ranging from 0 to 3 (0 = none, 1 = mild, 2 = moderate, 3 = severe), as shown in Table 1.

Identifying kidney infiltrating leukocyte phenotypes

To evaluate the leukocytic phenotypes that infiltrate into the kidney, we identified M, CD4, CD8, and DN T cells. Cryostat-sectioned kidneys were stained for the presence of M and T cells (CD4, CD8, DN) with monoclonal antibodies using the avidin-biotin complex immunoperoxidase technique¹⁵. The intrarenal M and T cells (CD4, CD8, and DN) were enumerated both in glomeruli (within glomeruli and adjacent to glomeruli) and interstitium (adjacent to cortical tubules and perivascular areas) in 20 randomly selected microscopic fields (magnification 400). The intrarenal M and T cells were graded on a scale from 0 to 3 (0 = none; 1 = mild, <50 per field; 2 = moderate, 50 to 100 per field; 3 = severe,> 100 per field), and the mean grade per field was determined Figure 6. T-cell intrarenal phenotypes CD4, CD8, and DN T cells were graded on a scale from 0 to 3 (0 = none; 1 = mild, <10 per field; 2 = moderate, 10 to 50 per field; 3 = severe,> 50 per field), and the mean grade per field was determined Figure 7. Scoring was performed on coded slides by two blinded observers.

Figure 6.

MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m have a reduction in intrarenal T cells, but M remain. M (A) and T cells (CD4, CD8, and DN; B) were abundant in the kidney of MRL-Fas^lpr mice. The intrarenal M were minimally reduced in strains deficient in TCR , CD4, and ₂m (CD8 and/or DN T cells) as compared with the T-cell populations. Although T cells were reduced, they were elevated as compared to MRL-++ normal kidneys (B). Values = mean SEM. N = 3 to 5 per group at six months of age.

Full figure and legend (19K)

Figure 7.

A deficiency in several select T-cell populations in MRL-Fas^lpr mice leads to a reduction in the entire intrarenal T-cell populations. TCR - and CD4-deficient MRL-Fas^lpr mice were spared from an intrarenal influx of CD4, CD8, and most DN T cells. ₂m-deficient MRL-Fas^lpr mice were spared from an intrarenal influx of DN and CD8 T cells and most CD4 T cells. Values are mean SEM; N = 3 to 5 per group at six months of age.

Full figure and legend (21K)

Ex vivo gene transfer system delivering M growth factor "carrier cells" into the kidneys

We delivered CSF-1 and GM-CSF into the kidney using an ex vivo gene transfer system. TEC from MRL-Fas^lpr mice at one to two months of age was isolated and these TEC infected with recombinant retroviruses encoding CSF-1 and GM-CSF as previously described^{4,16,17,18,19}. To ensure that genetically modified TEC, termed CSF-1 or GM-CSF "carrier cells," produced these M growth factors, the release of CSF-1 or GM-CSF was measured in these culture supernatants, respectively²⁰. The viability of TEC immediately before implantation was> 90% assessed by trypan blue staining. We implanted CSF-1 or GM-CSF "carrier cells" or uninfected TEC (1 10⁶ TEC in 50 L of HBSS) under the left renal capsule of recipient mice at two to three months of age⁴. At 14 or 28 days postimplant, the kidney implanted with CSF-1 or GM-CSF "carrier cells" was excised, and the accumulation of cells in the subcapsular implant site and adjacent renal cortex was assessed by enumerating the maximum number of cell layers within these areas as shown in Figures 8 and 9-1. The percentage of M, CD4, CD8, and DN T cells were expressed as an index [maximal subcapsular cell layers leukocytic phenotypes (%)] as in Figure 9-2. To determine that CSF-1 and GM-CSF "carrier cells" produce CSF-1 or GM-CSF, kidneys were stained for the presence of either cytokine using the immunoperoxidase technique¹⁵.

Figure 8.

M growth factor "carrier cells" do not incite an intrarenal leukocytic infiltrates in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m. M growth factor "carrier cells" did not incite an intrarenal accumulation of leukocytic infiltrates in the MRL-Fas^lpr strains deficient in TCR , CD4, and ₂m as compared with the wild-type strain. Values are mean SEM; N = 3 to 5 per group. Recipients were two to three months of age.

Full figure and legend (12K)

Figure 9.

GM-CSF "carrier cells" elicit an enhanced M accumulation in the renal capsule in T cell deficient MRL-Fas^lpr strains. (1) GM-CSF "carrier cells" caused less renal pathology (50%) in the TCR , CD4, and ₂m-deficient MRL-Fas^lpr strains as compared with the wild-type strain. (2) The majority of kidney leukocytic infiltrates elicited by GM-CSF "carrier cells" in wild-type mice were CD4 and DN T cells (A–D). In contrast, the majority of kidney leukocytic infiltrates in T-cell–deficient MRL-Fas^lpr strains were M (A). There were more M in the T-cell–deficient MRL-Fas^lpr strains than in the wild-type strain (A). Values are mean SEM. Recipients were two to three months of age (N = 3 to 5 per group). Index = maximum cell layers cell phenotypes (%).

Full figure and legend (30K)

Circulating proteins

To evaluate the amount of circulating cytokines, recipient mice were bled 7 days prior to implanting CSF-1 or GM-CSF "carrier cells" and at 14 and 28 days follow this procedure. Biologically active CSF-1 and GM-CSF in TEC supernatants and serum samples were measured using the colony-stimulating assay (CSA)²⁰, as shown in Table 2.

Table 2 - Retrovirally infected GM-CSF "carrier cells" implanted under the renal capsule increase circulating GM-CSF in MRL mice.

Full table

Statistical analysis

The data represent the means SEM. Statistical significance was determined by analysis of variance (ANOVA).

RESULTS

Multiple T-cell populations are required for CSF-1, TNF-, and GM-CSF in MRL-Fas^lpr mice

Intrarenal transcripts and proteins

We analyzed CSF-1, TNF-, and GM-CSF expression in the kidney of MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m using immunostaining Figures 1 and 3 and Northern blot analysis Figure 2. CSF-1, TNF-, and GM-CSF were up-regulated between three and six month of age in wild-type MRL-Fas^lpr kidney as compared with normal kidneys from age matched MRL-++ mice Figure 1. We localized CSF-1 Figures 1a and 3a, TNF- Figures 1b and 3e, and GM-CSF (Figures 1c and 3i,j) within glomeruli, the interstitium, TEC, and perivascular areas and within kidney infiltrating leukocytes. In contrast, we did not detect CSF-1 in TCR -deficient MRL-Fas^lpr or CD4-deficient MRL-Fas^lpr mouse kidneys at three and six months of age using Northern blot analysis Figure 2a and immunoperoxidase techniques (Figures 1a and 3b, c). Although CSF-1 was detected in glomeruli, TEC, and kidney-infiltrating leukocytes in ₂m-deficient MRL-Fas^lpr mice, the level was far less than in the wild-type strain Figures 1a and 3d. The CSF-1 transcript/-actin densitometry ratios were 1.1 in the wild-type strain, 0.4 in TCR -deficient and CD4-deficient MRL-Fas^lpr mice and 0.7 in ₂m-deficient MRL-Fas^lpr mice Figure 2. The TNF-/-actin densitometry ratios were 1.0 in the wild-type strain, 0.8 in the TCR -deficient and CD4-deficient MRL-Fas^lpr mice, and 0.9 in the ₂m-deficient MRL-Fas^lpr mice. The GM-CSF/-actin densitometry ratios were 0.4 in the wild-type strain and 0.2 in the TCR -deficient, CD4-deficient, and ₂m-deficient MRL-Fas^lpr mice.

Figure 3.

Multiple T-cell populations are required for promoting the intrarenal production of CSF-1, TNF-, and GM-CSF of MRL-Fas^lpr mice. CSF-1 was detected in glomeruli (arrowheads), interstitium, and infiltrating cells in MRL-Fas^lpr mice at six months of age (A). CSF-1 was not detected in MRL-Fas^lpr strains deficient in TCR , or CD4 (B and C). CSF-1 was detected in the kidney of ₂m-deficient MRL-Fas^lpr mice, but was less intense (D; arrowheads in a glomerulus). TNF- was strongly detected in TEC (arrows) and glomeruli (arrowheads) in MRL-Fas^lpr mice at six months of age (E). TNF- was reduced in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m (F–H; arrows in TEC). GM-CSF was detected in glomeruli (I; arrowheads) and interstitial infiltrating leukocytes (J; arrows) in MRL-Fas^lpr mice at six months of age, but was absent in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m. G, glomeruli (immunoperoxidase staining, A–D 500; E–J 1000).

Full figure and legend (239K)

Figure 2.

CSF-1, TNF-, and GM-CSF transcripts are reduced in T-cell–depleted MRL-Fas^lpr mice. Total kidney cortex RNA from MRL-Fas^lpr mice (lane 1), TCR -deficient MRL-Fas^lpr (lane 2), CD4-deficient MRL-Fas^lpr (lane 3), and ₂m-deficient MRL-Fas^lpr mice (lane 4) at six months of age, CSF-1 (A), TNF- (B), GM-CSF (C), and -actin (D), respectively. N = 3 to 6 per lane. Data are representative of three experiments.

Full figure and legend (124K)

Intrarenal TNF- in MRL-Fas^lpr mice is primarily expressed by TEC. The level of TNF- in TCR /, CD4, and ₂m-deficient MRL-Fas^lpr mice was substantially reduced as compared with wild-type mice Figures 1, 2, 3. We localized the TNF- reduction in comparison to age- and sex-matched wild-type mice mostly within TEC and glomeruli in T-cell–depleted strains (Figure 1b and 3f,g,h).

In contrast to CSF-1 and TNF-, which are evident prior to renal disease (1 to 2 months of age), no GM-CSF in MRL-Fas^lpr kidneys was detected at one and two months of age (data not shown). However, intrarenal GM-CSF expression did increase during renal disease in MRL-Fas^lpr mice (Figures 1c, 2c, 3i, J). Thus, an up-regulation of intrarenal GM-CSF is downstream from CSF-1 and TNF-. By comparison, TCR , CD4, and ₂m-deficient MRL-Fas^lpr strains at three and six months of age did not express GM-CSF Figures 1c and 2c.

Circulating proteins

The pattern of cytokine production in the kidneys closely paralleled the cytokine production profile in the circulation. Specifically, CSF-1 was absent from the sera of MRL-Fas^lpr strains deficient in TCR and CD4 at six months of age [4 5 and 2 1 colony-forming unit (CFU)], respectively, compared with the wild-type strain (49 5 CFU, P < 0.01, N = 3 to 4 per group). Similarly, CSF-1 was diminished in ₂m-deficient MRL-Fas^lpr mice (10 1 CFU) compared with the wild-type strain (49 5 CFU, P < 0.01, N = 3 per group).

Reduced renal diseases in T-cell–deficient MRL-Fas^lpr strains

T cells play an important role in long-term spontaneous renal injury in MRL-Fas^lpr mice

To determine the impact of select T-cell populations on spontaneous autoimmune renal injury in MRL-Fas^lpr mice, we evaluated the extent of renal pathology in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m in comparison to wild-type MRL-Fas^lpr mice and MRL-++ mice at six months of age (Figures 4 and 5, and Table 1). We selected this age point since MRL-Fas^lpr mice have advanced renal pathology and MRL-++ kidneys remain normal. Renal pathology in the wild-type MRL-Fas^lpr strain consisted of severe proliferative glomerulonephritis, interstitial/perivascular nephritis, and tubular damage. Renal pathology was decreased in T-cell–deficient MRL-Fas^lpr mice in comparison to the wild-type counterparts, but remained higher than MRL-++ mice (Figures 4-1, 4-2, and 5). We noted reductions in glomerular crescents and segmental lesions, tubular damage Table 1, and the numbers of glomerular, interstitial, and perivascular cells Figure 5. The numbers of cells in the kidney were counted in four separate areas: intraglomerular, periglomerular, interstitial, and perivascular. There were fewer kidney cells in the glomerular, interstitial, and perivascular areas in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m compared with the wild-type strain Figure 5. However, the number of cells in the kidney was not reduced to normal values; there were more cells in the kidney of T-cell–deficient strains than in MRL-++ mice Figure 5. For example, interstitial cells decreased substantially (59.3, 62.5, and 53.1%) in the TCR -deficient strain, CD4-deficient strain, and the ₂m-deficient strain, respectively, as compared with wild-type mice Figure 5. However, the decrease in the kidney cells in these T-cell–deficient strains was not reduced to the baseline levels of the MRL-++ normal kidney Figure 4 and 5. Similarly, the T-cell–deficient MRL-Fas^lpr strains had less proteinuria and glomerular IgG deposits as compared with wild-type mice Table 1. Again, the extent of proteinuria and glomerular IgG deposits remained higher than in the MRL-++ strains. The decrease in autoimmune kidney disease was greater in the TCR and the CD4 MRL-Fas^lpr strains as compared with the ₂m-deficient MRL-Fas^lpr mice Table 1.

M alone is not sufficient to induce autoimmune renal injury in MRL-Fas^lpr mice

Long-term spontaneous autoimmune renal injury in MRL-Fas^lpr mice

The number of kidney leukocytic infiltrates in the kidneys of wild-type MRL-Fas^lpr mice, MRL-++ mice, and MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m was evaluated at six months of age. The ₂m-deficient (class I) mice cannot select CD8 or DN T cells Figures 6 and 7^9,21,22. It is notable that the number of M was barely reduced in T-cell–deficient MRL-Fas^lpr strains as compared with the wild-type Figure 6a. In fact, the numbers of M were far more abundant in the T-cell–deficient MRL-Fas^lpr strains as compared with the normal MRL-++ kidneys Figure 6a. CD4 T cells were, as anticipated, absent from the kidneys of MRL-Fas^lpr mice that were deficient in CD4 and TCR Figure 7a. However, CD4 T cells were greatly reduced (>50%) in ₂m-deficient MRL-Fas^lpr kidney Figure 7a. Since class I selects for CD8 and DN T cells while class II selects for CD4 T cells, we anticipated a dramatic decrease in CD8/DN T cells in the class I-deficient MRL-Fas^lpr strain²². Similarly, CD8 T cells were, as expected, absent in MRL-Fas^lpr deficient in TCR or ₂m and were substantially reduced (50%) in the CD4-deficient MRL-Fas^lpr kidneys Figure 7b. DN T cells were absent from the kidney in the ₂m-deficient MRL-Fas^lpr strain and were greatly diminished in the TCR -deficient MRL-Fas^lpr strain Figure 7c. This suggests that the majority of intrarenal DN T cells bear TCR /. Interestingly, DN T cells were also diminished (>50%) in CD4-deficient MRL-Fas^lpr mice compared with the wild-type strain Figure 7c. Taken together, this indicates an interdependency of CD4 and CD8/DN T-cell populations in MRL-Fas^lpr kidney disease. In addition, our data indicate that even in the presence of an abundance of M, T-cell populations play an important role in promoting kidney pathology.

Multiple T-cell populations are essential in M growth factor "carrier cells" elicited autoimmune renal injury in MRL-Fas^lpr mice

To evaluate the impact of selective T-cell deletion on the downstream events (effector phase), we implanted CSF-1 and GM-CSF "carrier cells" into the kidneys using an ex vivo gene transfer approach. As previously established, CSF-1 or GM-CSF "carrier cells," infused under the renal capsule, incited a substantial accumulation of leukocytes in the implant site in MRL-Fas^lpr recipients as compared with control "carrier cells" (Figures 8, 9-1, and 10A)⁴. The MRL-++ strain was resistant to GM-CSF elicited renal injury (Figures 8 and 9-1). CSF-1 "carrier cells" did not induce intrarenal injury in CD4-deficient MRL-Fas^lpr mice (10 3 cell layers per intrarenal and 22 3 per subcapsule) as compared with the wild-type strain (48 8 per intrarenal and 40 10 per subcapsule, day 28 postimplant, P < 0.05). Similarly, GM-CSF "carrier cells" did not incite intrarenal injury in TCR and CD4 T-cell–deficient MRL-Fas^lpr mice as compared with the wild-type strain (Figures 8, 9-1, and 10 b, c). GM-CSF "carrier cells" elicited a modest intrarenal injury in ₂m-deficient MRL-Fas^lpr mice, which was dramatically less than in wild-type strain (Figures 8, 9-1, and 10d). Thus, renal pathology was not readily elicited by CSF-1 or GM-CSF in TCR , CD4, and ₂m-deficient MRL-Fas^lpr strains.

Figure 10.

GM-CSF "carrier cells" did not incite intrarenal injury in MRL-Fas^lpr strains deficient in TCR , CD4, or ₂m. In the TCR -deficient MRL-Fas^lpr (B), CD4-deficient MRL-Fas^lpr (C) and ₂m-deficient MRL-Fas^lpr strains (D), GM-CSF "carrier cells" elicited fewer kidney infiltrating cells than in the wild-type strain. These are evaluated at 28-days postimplant (N = 3 to 5 per group). Data are representative of six separate experiments. sc, subcapsular space (hematoxylin eosin, 200).

Full figure and legend (350K)

To document that genetically modified TEC implanted under the renal capsule produced M growth factors locally and systematically, we evaluated the circulating cytokines using a bioassay (CSA) and probed for cytokines in the implant site by immunostaining. Serum CSF-1 was elevated in the CD4-deficient (11 1 CFU) and wild-type (12 1 CFU) at 14 days postimplant. Similarly, we detected serum GM-CSF after implanting GM-CSF "carrier cells" Table 2. In addition, CSF-1 and GM-CSF were detected in CSF-1 and GM-CSF "carrier cells," respectively, by immunostaining and remained in the subcapsule throughout the experimental period (day 28 postimplant, data not shown). In contrast, CSF-1 and GM-CSF were not detected in kidneys implanted with uninfected TEC (data not shown). Thus, using a gene transfer approach, M growth factors were delivered into the kidney and circulation throughout experimental period.

We analyzed the phenotype of the leukocytic infiltrates under the renal capsule in the GM-CSF–implanted kidneys in T-cell–deficient MRL-Fas^lpr mice. We previously noted that GM-CSF "carrier cells" elicited an initial accumulation of primarily M (3 to 7 days), followed by an influx of T cells (14 to 28 days postimplant) in the MRL-Fas^lpr kidney⁴. The majority (80%) of the kidney leukocytic infiltrates were CD4 or DN, and only a few M (10%) were detected (28 days postimplant; Figure 9-2). In contrast, the kidney leukocytic infiltrates in MRL-Fas^lpr mice that lacked select T-cell populations were primarily M (AFigure 9-2a); T cells were notably absent (Figure 9-2b, c, d). In fact, there was an increase in the absolute number of M in the T-cell–deficient strains compared with the wild-type MRL-Fas^lpr mice, suggesting that T cells release molecules that prevent M accumulation.

Age did not affect M growth factor elicited, T-cell–dependent nephritis

To evaluate the influence of age on the intrarenal leukocytic accumulation, GM-CSF "carrier cells" were implanted into CD4-deficient and intact MRL-Fas^lpr recipients at different ages. We selected an age range during mild, moderate, and severe renal pathologies (3, 5, and 7 months of age, respectively) in MRL-Fas^lpr mice. GM-CSF elicited a similar amount of kidney infiltrating cells in MRL-Fas^lpr mice at three, five, and seven months of age (20 8, 20 1, 21 7 cell layers, respectively, N = 2 to 4 per point). In contrast, GM-CSF "carrier cells" did not elicit an intrarenal accumulation of kidney infiltrating cells in age-matched CD4-deficient MRL-Fas^lpr mice (3 2, 4 1, and 2 1 cell layers, respectively, N = 2 to 4 per point). Thus, M growth factor elicited nephritis is dependent on multiple T-cell populations in MRL-Fas^lpr mice within a broad age range during progressive renal disease.

DISCUSSION

In this report, we have tested the role of select T-cell populations in intrarenal cytokine expression and autoimmune renal injury in MRL-Fas^lpr mice. We now report that TCR , CD4, and CD8 and/or DN T cells (1) contribute to the intrarenal induction of nephritogenic cytokines (CSF-1, GM-CSF, and TNF-) and autoimmune renal disease in MRL-Fas^lpr mice, and (2) are required for the downstream events following cytokine (CSF-1, GM-CSF)-elicited renal pathology. We also note that these select T-cell populations reduce the intrarenal M accumulation in response to M growth factors. However, despite an abundance of M, these select T-cell populations play an important role in autoimmune nephritis. We conclude that multiple T-cell populations are required to promote intrarenal cytokines, which in turn recruit additional T cells that foster a feedback amplification loop culminating in lethal autoimmune destruction in MRL-Fas^lpr kidneys.

We report a novel finding that TCR , CD4, and CD8 and/or DN T cells in MRL-Fas^lpr mice each contribute to the pathologic expression of intrarenal cytokines responsible for autoimmune renal disease. This finding implies that TCR , CD4, and CD8 and/or DN T cells either secrete or induce the production of these cytokines by renal parenchymal cells. Although CD4 T cells are capable of secreting CSF-1²³, the majority of CSF-1 in the kidney is produced by mesangial cells²⁴. Thus, we suggest that T cells release a cytokine that induces CSF-1 production by mesangial cells. Based on our recent study, IFN- receptor-deficient MRL-Fas^lpr mice fail to express intrarenal CSF-1 and TNF- and are spared from autoimmune kidney destruction²⁵. More recently, we determined that IL-12 elicited autoimmune injury by fostering the accumulation of IFN-–secreting T cells and nephritis in MRL-Fas^lpr mice²⁶. Therefore, we propose that multiple intrarenal T-cell populations dependent on IL-12 releasing IFN- are responsible for the inducing CSF-1 and TNF-. Of note, CD8 and DN T cells in MRL-Fas^lpr mice secrete more IFN- than CD4 T cells²⁷. Therefore, we might predict that the CD8 and/or DN T cells in the CD4-deficient MRL-Fas^lpr strain would cause renal damage. However, few CD8 and/or DN T cells infiltrate the kidney in the CD4-deficient MRL-Fas^lpr strain; consequently, there may be insufficient IFN- released locally to induce CSF-1 and TNF-. We suggest that IFN- delivered by kidney infiltrating TCR , CD4, and CD8 and/or DN T cells may be the stimulus that triggers the production of CSF-1 and a cascade of events resulting in fatal autoimmune kidney disease.

Understanding the mechanism responsible for the expression of TNF- in the kidney is complex. We previously identified two distinct mechanisms of TNF- regulation in the kidneys of MRL-Fas^lpr mice. One mechanism involves neonatal up-regulation of TNF- that is related to the Fas^lpr mutation; the other mechanism involves an increase in TNF- in mature mice that is proportional to the severity of lupus nephritis³. This up-regulation of TNF- prior to and during nephritis in MRL-Fas^lpr mice is largely generated by TEC. This is consistent with the present study; intrarenal TNF- was diminished, but not totally absent in TCR , CD4, or CD8 and/or DN T-cell–deficient MRL-Fas^lpr strains protected from progressive renal disease. Thus, multiple T-cell populations release cytokines that induce intrarenal TNF-. It is tempting to speculate that the neonatal expression of TNF- is a very proximal step in the development of nephritis; multiple, interdependent T-cell populations responding to TNF- may initiate kidney disease and subsequently stimulate the kidney to generate more TNF- with advancing renal disease.

We now report that intrarenal GM-CSF expression is downstream of CSF-1 and TNF- expression, which appear in advance of renal injury. We did not detect GM-CSF in the T-cell–deficient MRL-Fas^lpr strains. GM-CSF is up-regulated by cytokines, including TNF- and IL-1^28,29,30,31, and is secreted by CD4 T cells^28,29 and cytokine-stimulated glomerular epithelial cells³¹. Since IL-1 and TNF- are increased in proportion to the severity of kidney disease in MRL-Fas^lpr kidneys^3,32, it is possible that intrarenal GM-CSF is triggered by IL-1 and TNF-^28,29,30,31. We suggest that GM-CSF recruits M and T cells and assists in amplifying established renal disease⁴.

We have uncovered an important interplay between T-cell populations responsible for initiating and promoting autoimmune kidney disease. We now report in CD4-deficient MRL-Fas^lpr mice that far fewer CD8 and DN T cells are detected in the kidney, even though they are abundant in the lymph nodes or circulation¹¹. Similarly, ₂m-deficient MRL-Fas^lpr mice lacking CD8 and DN T cells have a few intrarenal CD4 cells, despite the availability of CD4 T cells in other tissues¹⁰. This interdependency of T-cell populations is even more striking in the downstream M growth factor-initiated studies. There are several possibilities. CD4 may induce chemokines that are required to attract CD8 and DN T cells, and vise versa, CD8 and DN T cells may induce chemokines that recruit CD4 cells into the kidney. In this regard, we determined that monocyte chemoattractant protein-1 (MCP-1) and regulated upon activation, normal T cell expressed and secreted (RANTES) are detected prior to and during renal injury and are required for autoimmune kidney diseases in MRL-Fas^lpr mice^33,34,35. In addition, newly identified CC chemokines may be involved in the trafficking and homing of select lymphocyte subsets into specific organs^36,37. It is alternatively possible that the cytokines and other molecules that are responsible for intrarenal T-cell adherence and proliferation fail to be released in the absence of these select T-cell populations. We are currently exploring the molecular interdependency of CD4, CD8, and DN T cells in the kidney.

Our studies support the concept that an influx of M alone into the kidney without T cells fails to promote autoimmune kidney disease. A dynamic interaction of M and T cells is required for nephritogenic cytokines and autoimmune renal injury in MRL-Fas^lpr mice. This is based on the finding that T-cell–deficient strains were spared from spontaneous lethal renal injury despite an appreciable intrarenal reduction in M. This is consistent with our previous finding that CSF-1 and GM-CSF "carrier cells" implanted under the renal capsule recruit M into normal (C3H-Hej) kidneys, but in the absence of autoreactive T cells, CSF-1 and GM-CSF do not elicit kidney injury⁴. In addition, M growth factors elicited more M in the kidneys of T-cell–deficient MRL-Fas^lpr mice compared with wild-type MRL-Fas^lpr mice. This finding suggests that kidney infiltrating T cells reduce the accumulation of M in the kidney. This is keeping with our findings that IFN- produced by kidney infiltrating T cells is responsible for halting M proliferation and enhancing apoptosis of M³⁸. Thus, it is plausible that IFN- released by T cells provides a negative regulatory pathway restricting M proliferation.

We now report that ₂m-deficient MRL-Fas^lpr mice deficient in CD8 and DN T cells are necessary in the initiation and effector phase of autoimmune kidney disease. Our conclusion differs with a prior study claiming that the requirement for class 1 proteins is restricted temporally to later stages of the disease¹⁰. This concept is based on similar increasing levels of total immunoglobulins in ₂m-deficient MRL-Fas^lpr and wild-type mice in the early phase (<75 days) and heightened increases in the wild-type as compared with the class I-deficient mice during the later phase. Since CSF-1, GM-CSF, and TNF- remained at baseline or diminished levels at three months of age and extended through six months of age, we suggest that class I is required throughout disease in MRL-Fas^lpr mice.

These studies further support the concept that CD8-derived DN T cells are instrumental in autoimmune nephritis in MRL-Fas^lpr mice. It is difficult to determine the impact of CD8 T cells apart from DN T cells in the Fas^lpr strains. The unique DN T cells in the Fas^lpr strains are derived from the CD8 lineage; thus, the CD8 "knockout" MRL-Fas^lpr lack CD8 and DN T cells³⁹. To determine the impact of the DN T cells on autoimmune kidney disease, we propagated DN T cells from MRL-Fas^lpr nephritic kidneys and inserted them under the renal capsule⁴⁰. DN T cells induced T-cell activation molecules on the adjacent TEC. Thus, intrarenal DN T cells may provide signals required to stimulate TEC to participate in autoimmune kidney destruction. In addition, our studies are consistent with a prior report in which ₂m-deficient MRL-Fas^lpr mice were spared from glomerulonephritis¹⁰. On the other hand, survival was not prolonged in CD8-deficient T-cell MRL-Fas^lpr mice, lacking CD8 and the majority of DN T cells as compared with the wild-type strain³⁹. Of note, renal disease and the cause of death were not evaluated in these CD8-deficient MRL-Fas^lpr mice. Thus, it is possible that other class I-dependent events, in addition to the elimination of CD8 and DN T cells, may alter autoimmune kidney disease. Therefore, further studies are necessary to clarify the precise role of CD8 and DN T cells in the pathogenesis of kidney disease in MRL-Fas^lpr mice.

In conclusion, we have determined that autoimmune nephritis in MRL-Fas^lpr mice is dependent on multiple T-cell populations during the initiation and effector phases. We have established that M in the absence of select T-cell populations are incapable of promoting kidney disease. TCR , CD4, and class I-selected CD8/DN T cells up-regulate intrarenal CSF-1, TNF-, and GM-CSF production in the MRL-Fas^lpr mice. In turn, these cytokines recruit TCR , CD4, and CD8/DN T cells into the kidney, thereby perpetuating progressive autoimmune renal injury. Taken together, multiple T-cell populations are appealing therapeutic targets for combating autoimmune kidney disease.

References

1. Theophilopoulos, AN & Dixon, FJ: Murine models of systemic lupus erythematosus. Adv Immunol 1985 37: 269–390,  | PubMed | ISI | ChemPort |
2. Yui, MA, Brissette, WH & Brennan, DC, et al: Increased macrophage colony-stimulating factor in neonatal and adult autoimmune MRL-lpr mice. Am J Pathol 1991 139: 255–261,  | PubMed | ISI | ChemPort |
3. Yokoyama, H, Kreft, B & Kelley, VR: Biphasic increase in circulating and renal TNF- in MRL-lpr mice with differing regulatory mechanisms. Kidney Int 1995 47: 122–130,  | PubMed | ISI | ChemPort |
4. Naito, T, Yokoyama, H & Moore, KJ, et al: Macrophage growth factors introduced into the kidney initiate renal injury. Mol Med 1996 2: 297–312,  | PubMed | ISI | ChemPort |
5. Moore, KJ, Yeh, K & Naito, T, et al: TNF- enhances colony-stimulating factor-1-induced macrophage accumulation in autoimmune renal disease. J Immunol 1996 157: 427–432,  | PubMed | ISI | ChemPort |
6. Naito, T, Griffiths, RC & Coffman, TM, et al: Transplant approach establishes that kidneys are responsible for serum CSF-1 but require a stimulus in MRL-lpr mice. Kidney Int 1996 49: 67–74,  | PubMed | ISI | ChemPort |
7. Wada, T, Naito, T & Griffiths, RC, et al: Systemic autoimmune nephritogenic components induce CSF-1 and TNF- in MRL mice. Kidney Int 1997 52: 934–941,  | PubMed | ISI | ChemPort |
8. Jevnikar, AM, Grusby, MJ & Glimcher, LH: Prevention of nephritis in major histocompatibility complex class II-deficient MRL-lpr mice. J Exp Med 1994 179: 1137–1143,  | Article | PubMed | ISI | ChemPort |
9. Peng, SL, Madaio, MP & Highes, DP, et al: Murine lupus in the absence of T cells. J Immunol 1996 156: 4041–4049,  | PubMed | ISI | ChemPort |
10. Christianson, GJ, Blankenburg, RL & Duffy, TM, et al: 2-microglobulin dependence of the lupus-like autoimmune syndrome of MRL-lpr mice. J Immunol 1996 156: 4932–4939,  | PubMed | ISI | ChemPort |
11. Chesnutt, MS, Fink, BK & Killeen, N, et al: Enhanced lymphoproliferation and diminished autoimmunity in CD4 deficient MRL-lpr mice. Clin Immunol Immunopathol 1998 87: 23–32, 10.1006/clin.1997.4492 | Article | PubMed | ISI | ChemPort |
12. Philpot, KL, Viney, JL & Kay, G, et al: Lymphoid development in mice congenitally lacking T cell receptor -expressing cells. Science 1992 256: 1448–1452,  | PubMed |
13. Beutler, B, Milsark, IW & Cerami, AC: Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science 1985 229: 869–871,  | PubMed | ISI | ChemPort |
14. Chomczynski, P & Sacchi, N: Single-step method of RNA isolation by acid guanidium thiocyanate-phenol chloroform extraction. Anal Biochem 1987 162: 156–160,  | Article | PubMed | ISI | ChemPort |
15. Brennan, DC, Jevnikar, AM & Bloom, RD, et al: Cultured mesangial cells from autoimmune MRL-lpr mice have decreased secreted and surface M-CSF. Kidney Int 1992 42: 279–284,  | PubMed | ISI | ChemPort |
16. Wuthrich, RP, Glimcher, LH & Kelley, VE: MHC class II, antigen presentation and tumor necrosis factor in renal tubular epithelial cells. Kidney Int 1992 37: 783–792,
17. Danos, O & Mulligan, RC: Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proc Natl Acad Sci USA 1988 85: 6460–6464.,  | PubMed | ChemPort |
18. Wilson, JM, Jefferson, DM & Mulligan, RC: Retrovirus-mediated transduction of adult hepatocytes. Proc Natl Acad Sci USA 1988 85: 3014–3018,  | PubMed | ChemPort |
19. Dranoff, G, Jaffee, E & Mulligan, RC: Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc Natl Acad Sci USA 1993 90: 3539–3543,  | PubMed | ChemPort |
20. Bradley, TR & Metcalf, D: The growth of mouse bone marrow cells in vitro. Aust J Exp Biol Med 1966 44: 287–299,  | ISI | ChemPort |
21. Maldonado, MA, Eisenberg, RA & Roper, E, et al: Greatly reduced lymphoproliferation in lpr mice lacking major histocompatibility complex class I. J Exp Med 1995 181: 641–648,  | Article | PubMed | ISI | ChemPort |
22. Mixter, PF, Russell, JQ & Durie, FH, et al: Decreased CD4–CD8- TCR+ cells in lpr/lpr mice lacking 2-microglobulin. J Immunol 1995 154: 4986–4990,  | PubMed |
23. Satriano, JA, Hora, K & Shan, Z, et al: Regulation of monocyte chemoattractant protein-1 and macrophage colony-stimulating factor-1 by IFN-, tumor necrosis factor-, IgG aggregates, and cAMP in mouse mesangial cells. J Immunol 1993 50: 1971–1978,
24. Bloom, RD, Florquine, S & Singer, GG, et al: Pivotal role for colony stimulating factor-1 (CSF-1) in the induction of lupus nephritis. Kidney Int 1993 43: 1000–1009,  | PubMed | ISI | ChemPort |
25. Schwarting, A, Wada, T & Kinoshita, K, et al: IFN-gamma R signaling is essential for the initiation, acceleration and destruction of autoimmune kidney disease in MRL-Fas^lpr mice. J Immunol 1998 161: 494–503,  | PubMed | ISI | ChemPort |
26. Schwarting, A, Tesch, G & Kinoshita, K, et al: IL-12 drives IFN-gamma-dependent autoimmune kidney disease in MRL-Fas^lpr mice. J Immunol 1999 163: 6884–6891,  | PubMed | ISI | ChemPort |
27. Adamthwaite, D & Cooley, MA: CD8+ T cell subsets defined by expression of CD45 isoforms differ in their capacity to produce IL-2, IFN-gamma and TNF-beta. Immunology 1994 81: 253–260,  | PubMed | ISI | ChemPort |
28. al-Janadi, N & al-Dalaan al-Balla, S, et al: CD4+ T cell inducible immunoregulatory cytokine response in rheumatoid arthritis. J Rheumatol 1996 23: 809–814,  | PubMed | ChemPort |
29. Lai, CK, Ho, SS & Chan, CH, et al: Gene expression of interleukin-3 and granulocyte macrophage colony-stimulating factor in circulating CD4+ T cells in acute severe asthma. Clin Exp Allergy 1996 26: 138–146,  | PubMed | ISI | ChemPort |
30. Sterzel, RB, Schulze-Lohoff, E & Marx, M: Cytokines and mesangial cells. Kidney Int 1993 43(Suppl 39):S26–S31,  | ISI | ChemPort |
31. Budde, K, Neumayer, HH & Salant, DJ, et al: Glomerular epithelial cell products stimulate mesangial proliferation in culture. Kidney Int 1997 52: 733–741,  | PubMed | ISI | ChemPort |
32. Boswell, JM, Yui, MA & Endres, S, et al: Novel and enhanced IL-1 gene expression in autoimmune mice with nephritis. J Immunol 1988 141: 118–124,  | PubMed | ISI | ChemPort |
33. Moore, KJ, Wada, T & Barbee, SD, et al: Gene transfer of RANTES elicits autoimmune renal injury in MRL-Fas^lpr mice. Kidney Int 1998 53: 1631–1641,  | Article | PubMed | ISI | ChemPort |
34. Tesch, G, Schwarting, A & Kinoshita, K, et al: Monocyte chemoattractant protein-1 promotes macrophage-mediated tubular injury, but not glomerular injury, in nephrotoxic serum nephritis. J Clin Invest 1999 103: 73–80,  | PubMed | ISI | ChemPort |
35. Tesch, G, Maifert, S & Schwarting, A, et al: Monocyte chemoattractant protein 1-dependent leukocytic infiltrates are responsible for autoimmune disease in MRL-Fas^lpr mice. J Exp Med 1999 190: 1813–1824,  | Article | PubMed | ISI | ChemPort |
36. Yoshie, O, Imai, T & Nomiyama, H: Novel lymphocyte-specific CC chemokines and their receptors. J Leukoc Biol 1997 62: 634–644,  | PubMed | ISI | ChemPort |
37. Nagira, M, Imai, T & Yoshida, R, et al: A lymphocyte-specific CC chemokine, secondary lymphoid tissue chemokine (SLC), is a highly efficient chemoattractant for B cells and activated T cells. Eur J Immunol 1998 28: 1516–1523, 10.1002/(sici)1521-4141(199805)28:05&#60;1516::aid-immu1516>3.3.co;2-a | Article | PubMed | ISI | ChemPort |
38. Schwarting, A, Moore, K & Wada, T, et al: IFN-gamma limits macrophage expansion in MRL-Fas^lpr autoimmune interstitial nephritis: A negative regulatory pathway. J Immunol 1998 160: 4074–4081,  | PubMed | ISI | ChemPort |
39. Koh, DW, Ho, A & Rahemtulla, A, et al: Murine lupus in MRL/lpr mice lacking CD4 or CD8 T cells. Eur J Immunol 1995 25: 2558–2562,  | PubMed | ISI | ChemPort |
40. Kelley, VR, Diaz-Gallo, C & Jevnikar, AM, et al: Renal tubular epithelial and T cell interactions in autoimmune renal disease. Kidney Int 1993 43(Suppl 39):S108–S115,  | ISI |

Acknowledgments

This work was supported by National Institute of Health DK 36149 and DK 52369 (V.R.K.) and the Baxter Extramural Grant Program (V.R.K.), the Department of Veterans Affairs (D.W.), and the Rosalind Russell Center for Arthritis Research (D.W.). T.W. was a recipient of a grant from the Japan Society for the Promotion of Science. A.S. was supported by the German Ernst Jung-Foundation for Research and Science. We thank Dr. Joseph E. Craft for a kind gift of TCR -deficient MRL-Fas^lpr mice.