Tubulointerstitial inflammation is a consistent feature of chronic renal disease, regardless of the initiating insult. Parenchymal injury as a result of this inflammation is believed to contribute to the functional and pathologic changes observed. Proinflammatory effects of CD40 and CD40 ligand (CD40L) interaction are one of many factors involved in the initiation and maintenance of this inflammatory response and as such form a potential target for therapy.
Understanding of the biological role of CD40-CD40L interactions (reviewed in1) is in a state of evolution. CD40 was initially described as a transmembrane protein on B lymphocytes2 with a role in the activation of humoral immunity3. Subsequent studies have demonstrated constitutive and inducible CD40 expression on a large number of other cells, including mononuclear, dendritic, endothelial, and epithelial cells1. Not surprisingly, therefore, CD40-dependent effects on the immune system extend beyond B-lymphocyte activation. Ligation of CD40 upon other cell types induces a broad proinflammatory response, including the production of a number of proinflammatory cytokines and adhesion molecules. Of direct relevance to renal diseases, ligation of tubular cell CD40 in vitro resulted in the production of various inflammatory cytokines, including interleukin (IL)-1, IL-6, IL-8, IL-15, IL-17, monocyte-chemoattractant protein-1 (MCP-1) and regulated on activation normal T cell expressed and secreted (RANTES)4,5,6. In a similar manner, although binding of CD40 to membrane-associated CD40L on T lymphocytes is a crucial "co-stimulatory" signal for antigen recognition, it is now recognized that the soluble cytokine-like form of CD40L [which belongs to the tumor-necrosis factor (TNF) superfamily] is capable of inducing CD40-dependent responses remote from the site of its production7.
CD40L blockade has previously been shown to be protective in animal models of glomerular disease that are dependent upon the production of autoantibodies, namely murine lupus8 and membranous nephritis9. In light of the effects of CD40-CD40L interaction that extend beyond B-cell–dependent responses, we hypothesized that CD40-CD40L interactions may also have a pathogenic role in adriamycin nephrosis. Renal injury in adriamycin nephrosis is not autoantibody dependent and indeed it appears that B lymphocytes play a minimal role10.
MATERIALS
Animals
BALB/c mice, 8 weeks old and weighing around 24 g, were supplied by the Animal Care Facility, Westmead Hospital, Sydney, Australia. All animals were housed in standard conditions and allowed free access to standard food and water. Each animal subgroup consisted of six to seven mice.
Experimental design
Mice were divided into three groups: (1) normal control; (2) adriamycin-alone; and (3) MR1 + adriamycin. A model of chronic glomerular disease was induced as previously described. Murine adriamycin nephropathy was induced as previously described10 by a single intravenous injection of adriamycin (9.6 mg/kg). Control animals were treated with an equivalent intravenous volume of saline vehicle.
The hamster anti-CD40L antibody (clone MR1, 0.4 mg per mouse) was purified by protein G chromatography from the culture supernatant of an immunoglobulin G (IgG)-producing hybridoma11. In the MR1 + adriamycin group, mice were given 0.4 mg of MR1 antibody at days 5, 7, 9, and 11 after adriamycin treatment. The timing of MR1 treatment was designed to occur after the induction of proteinuria in this model10. Adriamycin-alone animals were given an equivalent amount of murine IgG.
Animal subgroups were anaesthetized and sacrificed on days 14 and 42 after adriamycin. Blood samples were collected by cardiac puncture and both kidneys were removed for various analyses. A group of four normal mice was injected four times with same amount of MR1 antibody for examination of the possibility that antibody per se causes renal injury, and none of them had any histologic injury.
Another group in our department tested efficacy of the anti-CD40L antibody. A dose of 500 
g of MR1 antibody given intraperitoneally on days 0, 2, 4 and 6 prolonged survival of islet xenografts to 66 days and with cytotoxic T-lymphocyte associated antigen-4 (CTLA4)-Fc treatment produced indefinite graft survival12. These data demonstrated that this dose of MR1 antibody effectively blocked CD40-CD40L interaction. The same batch and dose of MR1 antibody was used in the present studies.
Renal functional assay
Fasting animals were placed in metabolic cages for 12 hours to collect urine for determination of urinary protein and creatinine. Creatinine clearance was calculated as creatinine excretion divided by plasma creatinine concentration. Venous blood for serum creatinine and albumin was collected at the time of sacrifice.
Histologic assessment
At week 6, the kidney was removed and weighed. A coronal slice was fixed in 10% neutral-buffered formalin for 24 hours and then dehydrated in graded alcohols and embedded in paraffin. Tissues were cut at 5 m and stained with hematoxylin and eosin, periodic acid-Schiff (PAS), and Masson's trichrome. The degree of glomerulosclerosis was measured using a quantitative method13.
Images viewed under a light microscope were digitalized using a video camera and then transferred onto a computer screen (Tang Computer, Sydney, Australia) using image analysis software (Optimas, version 5, Media Cybernetics, Seattle, WA, USA). The outline of the glomerular capillary tuft was traced, and the computed area was used as a measure of total glomerular area. The area covered by PAS-positive staining in the same glomerulus was then determined. The percentage of glomerulosclerosis was calculated by dividing the total PAS-positive area by the total glomerular area. The mean value of 20 randomly selected glomeruli was determined for each section. The degree of interstitial expansion was determined by quantitation of the relative interstitial volume. Random cortical fields were viewed at a magnification of 
200. The percentage relative interstitial volume was calculated from the area occupied by trichrome-stained interstitium divided by the total area. The mean value of five cortical fields was determined for each section. To avoid selection bias13, the areas to be viewed for morphometric analysis were anatomically identical for each section and positioned prior to microscopic visualization. Tubular atrophy was defined by the finding of a tubule with dilated tubular lumen and low cell height with absence of brush border.
Immunohistochemistry
Coronal slices of kidneys were embedded in 22-oxycalcitriol (OCT) compound (Tissue-Tek; Sakura Finetek, Torrance, CA, USA), frozen in liquid nitrogen, and stored at –70°C. Frozen sections were cut at 5 m, fixed with acetone at 4°C for 15 minutes, and blocked with 5% normal rabbit serum to minimize nonspecific antibody binding. Sections were incubated with antibodies against murine macrophages (1:100, clone F4/80; Serotec, Oxford, UK), murine CD4 (1:100, clone RM4-5; BD Biosciences, San Diego, CA, USA), murine CD8 (1:100, clone 53-6.7; BD Biosciences), and murine CD40 (1:10, clone 3/23; BD Biosciences) for 60 minutes at room temperature. After washing in phosphate-buffered saline (PBS), the section was blocked with 3% H2O2 for 3 minutes to eliminate endogenous peroxidase. Sections were incubated in biotinylated rabbit anti-immunoglobulin (Ig) (1:200) and then in avidin-biotin-horseradish peroxidase complex for 30 minutes each at room temperature. After PBS washing, the reaction was visualized by the addition of freshly prepared 3, 3-diaminobenzidine tetrahydrochloride (7.5 mg in 15 mL PBS with 15 L 30% H2O2) and examined by light microscopy. Positive control tissue for all antibodies was splenic tissue from mice injected intravenously with lipopolysaccharide. Negative controls were incubated with irrelevant antibodies of the same strain and isotype as the primary antibody in use.
The number of macrophages, CD4- and CD8-positive cells was quantitated in 10 nonoverlapping cortical fields (400, measuring 0.075 mm2 each). Using image analysis, the mean number of positive cells per interstitial field was calculated for each section and expressed as cells per mm2. For CD40, expression at 14 and 42 days after adriamycin was compared between the three animal subgroups.
Staining for CD40L was attempted using the MR1 antibody, with a biotin-conjugated mouse antihamster antibody as the secondary reagent (1:100, BD Biosciences). Splenic tissue from lipopolysaccharide-treated mice was used as a positive control tissue. Immunofluorescent and enzymatic methodology were trialed in an attempt to detect primary antibody binding.
Comparative expression of chemokine mRNA
Relative expression of the chemokines MCP-1 and RANTES were compared by reverse-transcription polymerase chain reaction (RT-PCR). Cortical tissue from the pole of each kidney was obtained at the time of animal sacrifice and snap-frozen in liquid nitrogen prior to storage at -70°C. Total cellular RNA was extracted using TRIreagent (Sigma Chemical Co., St. Louis, MO, USA), according to the guanidine thiocyanate/phenol/chloroform method14. One microgram of isolated RNA was reverse transcribed into complementary DNA (cDNA) in a standardized reaction using the MuLV enzyme (2.5U/L) (Roche Molecular Systems, Branchburg, NJ, USA). For MCP-1 and RANTES, 50 ng of cDNA was then amplified in a PCR reaction using the thermostable DNA polymerase Taq (Roche Molecular Systems). Twenty nanograms of cDNA was used for the chosen housekeeping gene (18S RNA). The sequence of the primers used for these experiments is shown in Table 1. For each PCR reaction, the linear phase of PCR amplification was determined prior to the final PCR of the experimental samples. PCR products were measured by densitometric analysis after ethidium bromide-agarose gel electrophoresis using a gel documentation system (Fluor-S; Bio-Rad, Hercules, CA, USA). The relative RNA expression for each sample was then calculated by dividing the density of the chemokine RNA by the amount of 18S RNA for each sample.
Table 1 - Sequence of upstream and downstream murine primers used for reverse transcription-polymerase chain reaction (RT-PCR) analysis.
Full table Statistical analysis
The results are expressed as mean 
SD, and statistical significance of differences among groups was determined by one-way analysis of variance (ANOVA). A P value of less than 0.05 was considered statistically significant.
RESULTS
Functional and pathologic changes
All adriamycin-treated mice developed renal injury. Body weight was reduced in both the adriamycin and MR1 + adriamycin groups. Kidneys of adriamycin-treated mice were pale and atrophic, whereas those of the MR1 + adriamycin group were hypertrophied compared to those of normal control. Proteinuria appeared 5 days after adriamycin administration, and remained at similar levels throughout the study period, as evaluated by urinalysis. Renal function at the end of experiment is summarized in Table 2. Serum creatinine and creatinine clearance were severely impaired in the adriamycin-alone group, but with MR1 treatment were close to the levels of normal mice. Proteinuria was improved with MR1 treatment.
Table 2 - Comparison of functional markers of disease severity at 42 days after adriamycin.
Full table Histologic analysis showed that MR1 treatment significantly reduced structural damage in adriamycin nephropathy (Table 3 and Figure 1). Glomeruli and tubules were only mildly damaged at week 6 in the MR1 + adriamycin mice when compared with the adriamycin-alone group. Glomerular surface area was significantly reduced in the adriamycin-alone mice as a result of shrinkage due to glomerulosclerosis, but there was no difference in glomerular size between MR1 + adriamycin and normal mice. There was much less glomerulosclerosis in the MR1 + adriamycin group when compared to the adriamycin-alone group (P < 0.01), as well as more nuclei per glomerular cross-section (P < 0.001). Interstitial expansion was severe in the adriamycin-alone mice, but much less severe in the MR1 + adriamycin group (P < 0.01). Tubular atrophy was significantly reduced by MR1 treatment in comparison with the adriamycin-alone mice Figure 1.
Figure 1.
Comparison of histologic changes in periodic acid-Schiff (PAS)-stained sections at day 42 after adriamycin. Compared to the normal kidney (A), the adriamycin-alone group (B) showed severe changes of glomerulosclerosis, tubular atrophy, and interstitial expansion. These changes were significantly less in the MR1+ adriamycin group (C).
Full figure and legend (417K)Table 3 - Comparison of pathologic markers of disease severity at 42 days after adriamycin.
Full table CD40 expression
There was minimal expression of CD40 within the normal kidney Figure 2a. Following adriamycin, there was a progressive increase in CD40 expression within the kidney Figure 2 b and c that was greatest in areas of glomerular and tubulointerstitial injury. Expression of CD40 was predominantly glomerular and interstitial at 14 days after adriamycin Figure 2b. CD40 expression was extensive 42 days after adriamycin and present in glomeruli, tubules, and the interstitium Figure 2c. Expression of CD40 in the kidney of the MR1 + adriamycin-treated animals occurred in a similar pattern but was reduced in extent, in parallel with the reduction in pathologic injury.
Figure 2.
Immunoperoxidase staining of cortical CD40 expression in murine adriamycin nephrosis. In the normal kidney (A), there was minimal CD40 expression of CD40 within tubules and glomeruli (G). CD40 expression was significantly increased at 14 days and 42 days after adriamycin (B and C). At 14 days (B), this expression appeared largely confined to glomeruli (G) and the interstitium. At 42 days after adriamycin (C), there was widespread expression of CD40 by glomerular cells, tubular cells, and cells of the expanded interstitium.
Full figure and legend (184K)Significant cell-associated CD40L was not detected in splenic or kidney tissue of any mice by enzymatic or fluorescent methods of detection (data not shown).
Number of interstitial mononuclear cells
Adriamycin nephrosis was associated with an interstitial mononuclear cell infiltrate consisting of numerous macrophages, CD4, and CD8 lymphocytes. The number of infiltrating macrophages in mice with adriamycin nephropathy was significantly reduced by MR1 treatment Figures 3 and 4 at both 14 and 42 days. The number of infiltrating CD4+ and CD8+ lymphocytes was reduced at 14 days in the MR1 + adriamycin group Figure 4a; however, this result did not reach statistical significance (P = 0.06 for CD4+ lymphocytes, P = 0.3 for CD4+ lymphocytes). At 42 days after adriamycin, there was no significant difference in the number of CD4+ or CD8+ lymphocytes between the adriamycin-alone or MR1 + adriamycin groups Figure 4b.
Figure 3.
Immunostaining for renal macrophages at day 42 after adriamycin. There was an extensive interstitial macrophage infiltrate in the adriamycin-alone group (A). The number of interstitial macrophages was reduced in the MR1 + adriamycin group (B).
Full figure and legend (184K)Figure 4.
Comparison of the number of infiltrating mononuclear cells at 14 (A) and 42 days (B) after adriamycin. MR1 treatment was associated with a significant reduction in the number of infiltrating macrophages. The number of infiltrating CD4+ and CD8+ cells was not statistically different from the adriamycin-alone group. Bars represent mean values + standard deviation. **P < 0.01 vs. adriamycin-alone group.
Full figure and legend (24K)Cortical chemokine expression
Adriamycin nephrosis was associated with significantly increased relative cortical expression of MCP-1 and RANTES mRNA at the 42-day time point Figure 5. For both chemokines, MR1 treatment was associated with significantly reduced expression at the 42-day time point (P < 0.01 for MCP-1 and P < 0.05 for RANTES).
Figure 5.
Relative cortical mRNA expression of the chemokines monocyte chemoattractant protein (MCP-1) (A) and regulated on activation normal T cell expressed and secreted (RANTES) (B). Adriamycin nephrosis was associated with significantly increased cortical expression of MCP-1 and RANTES at the 42-day time point. MR1 treatment was associated with significantly reduced expression of these chemokines at this time point. Bars represent mean values + standard deviation. *P < 0.05; **P < 0.01 vs. control group.
Full figure and legend (22K)DISCUSSION
In this study, adriamycin nephropathy was associated with progressive inducible expression of CD40 by both parenchymal and infiltrating cells. Blockade of CD40-CD40L interactions protected against structural and functional injury in mice with established disease. MR1 treatment significantly reduced glomerular sclerosis, tubular atrophy, and interstitial expansion and protected against functional loss. In addition, MR1 treatment was associated with a reduction of cortical proinflammatory signals and reduced macrophage infiltration in the cortex.
Previous investigators have shown that CD40L blockade protects against the initiation and progression of renal disease in murine models of membranous and lupus nephritis8,9,15. In both of these models of injury, the effect of CD40L blockade was thought to be due to impaired production of pathogenic antibodies by B lymphocytes. In this study, we found that CD40L blockade significantly protected against renal inflammation and injury in adriamycin nephrosis, a model in which the initiation of injury is not antibody-dependent. Protection in adriamycin nephrosis was also seen when CD40L blockade was started after glomerular injury was initiated, suggesting a broader role for CD40-CD40L in the pathogenesis of renal inflammation in this model.
Several possibilities exist to explain the reduction in glomerular damage and tubulointerstitial inflammation and injury in this model. One explanation is that blockade may interfere with CD40L-dependent effects on T-lymphocyte antigen recognition. Interaction between CD40 expressed by an antigen-presenting cell with membrane-associated CD40L on T lymphocytes is an important costimulatory signal necessary for antigen recognition16. A number of experimental findings support the possibility that cognate T-cell responses contribute to inflammation in chronic renal diseases. In vitro stimulation studies have demonstrated that tubular epithelial cells are capable of inducible expression of all the molecules necessary for them to act as antigen-presenting cells17,18, although the nature of potential antigen(s) involved in this scenario remains unclear. We were unable to demonstrate significant membrane-associated CD40L within the kidney of mice with adriamycin nephrosis. Although membrane-associated CD40L has been demonstrated on infiltrating lymphocytes of rejecting renal allografts19, it has been difficult to demonstrate cells bearing CD40L in chronic renal injury models20. Several possible explanations exist for this finding. In vitro studies suggest that surface expression of CD40L by T lymphocytes is rapidly and transiently induced after activation21,22 and declines to baseline levels after 48 hours, despite ongoing activation. A further explanation is for the apparent absence of membrane-associated CD40L is the possibility that the soluble cytokine-like form of CD40L is the predominant form in this model and exerts its effect remotely by binding to CD40 in this model.
The protective effect of MR1 could also be explained by its interference with CD40-dependent effects on inflammation. In this experiment, we found that MR1 treatment in adriamycin nephropathy was associated with a significant reduction in the number of infiltrating macrophages. Treatment was also associated with a trend toward a reduction in the numbers of infiltrating CD4+ and CD8+ lymphocytes at the day 14 time point. The lack of statistical significance for the latter result may partially be explained by a 
effect. A possible explanation for this powerful anti-inflammatory effect may be an attenuated production of the chemokines and adhesion molecules necessary for interstitial inflammation to occur. In vitro ligation of CD40 on renal tubular epithelial cells and monocytes induces the production of multiple chemokines and adhesion molecules1,5,6. It is therefore possible that the reduction in inflammation seen with MR1 treatment in adriamycin nephropathy is due to an inhibition of interaction between CD40L (expressed on or secreted by mononuclear cells) and CD40 (abundantly expressed on parenchymal and interstitial cells in this model). In support of this hypothesis is the finding in the current experiments that MR1 treatment was associated with significant reductions in the expression of MCP-1 and RANTES mRNA at the 42-day time point Figure 5. Both of these chemokines have previously been shown to be produced by renal tubular cells in response to CD40 ligation5.
CONCLUSION
Blockade of CD40-CD40L interactions in established murine adriamycin nephrosis attenuates interstitial inflammation and protects against parenchymal injury. These results have potential therapeutic relevance for the treatment of renal inflammation in human chronic renal diseases if agents can be developed that safely disrupt CD40-CD40L interactions.
References
| 1. | Schonbeck U & Libby P. The CD40/CD154 receptor/ligand dyad. Cell Mol Life Sci 2001; 58: 4−43. | PubMed | ISI | ChemPort | |
| 2. | Clark EA & Ledbetter JA. Activation of human B cells mediated through two distinct cell surface differentiation antigens, Bp35 and Bp50. Proc Natl Acad Sci USA 1986; 83: 4494−4498. | PubMed | ChemPort | |
| 3. | Foy TM, Shepherd DM & Durie FH et al. In vivo CD40-gp39 interactions are essential for thymus-dependent humoral immunity. II. Prolonged suppression of the humoral immune response by an antibody to the ligand for CD40, gp39. J Exp Med 1993; 178: 1567−1575. | Article | PubMed | ISI | ChemPort | |
| 4. | Weiler M, Kachko L & Chaimovitz C et al. CD40 ligation enhances IL-15 production by tubular epithelial cells. J Am Soc Nephrol 2001; 12: 80−87. | PubMed | ISI | ChemPort | |
| 5. | Woltman AM, de Haij S & Boonstra JG et al. Interleukin-17 and CD40-ligand synergistically enhance cytokine and chemokine production by renal epithelial cells. J Am Soc Nephrol 2000; 11: 2044−2055. | PubMed | ISI | ChemPort | |
| 6. | van Kooten C, Woltman AM & Daha MR. Immunological function of tubular epithelial cells: The functional implications of CD40 expression. Exp Nephrol 2000; 8: 203−207. | Article | PubMed | ISI | ChemPort | |
| 7. | Graf D, Muller S & Korthauer U et al. A soluble form of TRAP (CD40 ligand) is rapidly released after T cell activation. Eur J Immunol 1995; 25: 1749−1754. | PubMed | ISI | ChemPort | |
| 8. | Early GS, Zhao W & Burns CM. Anti-CD40 ligand antibody treatment prevents the development of lupus-like nephritis in a subset of New Zealand black x New Zealand white mice. Response correlates with the absence of an anti-antibody response. J Immunol 1996; 157: 3159−3164. | PubMed | ISI | ChemPort | |
| 9. | Biancone L, Andres G & Ahn H et al. Inhibition of the CD40-CD40 ligand pathway prevents murine membranous glomerulonephritis. Kidney Int 1995; 48: 458−468. | PubMed | ISI | ChemPort | |
| 10. | Wang Y, Wang YP, Tay YC & Harris DC. Progressive adriamycin nephropathy in mice: Sequence of histologic and immunohistochemical events. Kidney Int 2000; 58: 1797−1804. | Article | PubMed | ISI | ChemPort | |
| 11. | Noelle RJ, Roy M & Shepherd DM et al. A 39-kDa protein on activated helper T cells binds CD40 and transduces the signal for cognate activation of B cells. Proc Natl Acad Sci USA 1992; 89: 6550−6554. | PubMed | ChemPort | |
| 12. | Lehnert AM, Yi S, Burgess JS & O'Connell PJ. Pancreatic islet xenograft tolerance after short-term costimulation blockade is associated with increased CD4+ T cell apoptosis but not immune deviation. Transplantation 2000; 69: 1176−1185. | PubMed | ISI | ChemPort | |
| 13. | Vleming LJ, Baelde JJ & Westendorp RG et al. The glomerular deposition of PAS positive material correlates with renal function in human kidney diseases. Clin Nephrol 1997; 47: 158−167. | PubMed | ISI | ChemPort | |
| 14. | Chomczynski P & Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162: 156−159. | Article | PubMed | ISI | ChemPort | |
| 15. | Kalled SL, Cutler AH, Datta SK & Thomas DW. Anti-CD40 ligand antibody treatment of SNF1 mice with established nephritis: Preservation of kidney function. J Immunol 1998; 160: 2158−2165. | PubMed | ISI | ChemPort | |
| 16. | Yamada A, Salama AD & Sayegh MH. The role of novel T cell costimulatory pathways in autoimmunity and transplantation. J Am Soc Nephrol 2002; 13: 559−575. | PubMed | ISI | ChemPort | |
| 17. | Rubin-Kelley V & Singer G. The antigen presentation function of renal tubular epithelial cells. Exp Nephrol 1993; 1: 102−111. | PubMed | ISI | ChemPort | |
| 18. | Banu N & Meyers CM. IFN-gamma and LPS differentially modulate class II MHC and B7−1 expression on murine renal tubular epithelial cells. Kidney Int 1999; 55: 2250−2263. | Article | PubMed | ISI | ChemPort | |
| 19. | Biancone L, Segoloni G & Turello E et al. Expression of inducible lymphocyte costimulatory molecules in human renal allograft. Nephrol Dial Transplant 1998; 13: 716−722. | PubMed | ISI | ChemPort | |
| 20. | van Kooten C, Gerritsma JS & Paape ME et al. Possible role for CD40-CD40L in the regulation of interstitial infiltration in the kidney. Kidney Int 1997; 51: 711−721. | PubMed | ChemPort | |
| 21. | Roy M, Waldschmidt T & Aruffo A et al. The regulation of the expression of gp39, the CD40 ligand, on normal and cloned CD4+ T cells. J Immunol 1993; 151: 2497−2510. | PubMed | ISI | ChemPort | |
| 22. | Castle BE, Kishimoto K & Stearns C et al. Regulation of expression of the ligand for CD40 on T helper lymphocytes. J Immunol 1993; 151: 1777−1788. | PubMed | ISI | ChemPort | |
Acknowledgments
This study was supported by grants from the National Health and Medical Research Council (NHMRC) of Australia (grant # 107238) and from the Australia Kidney Foundation.
