CD4⁺CD25⁺ Regulatory T Cells Protect against Injury in an Innate Murine Model of Chronic Kidney Disease

Mice

Male SCID mice that were approximately 6 to 8 wk of age and weighed 22 to 24 g were purchased from Australian Research Council (Perth, Australia). All mice were housed in the Department of Animal Care at Westmead Hospital under standard conditions and allowed free access to standard food and water. Experiments were carried out in accordance with protocols approved by the Animal Ethics Committee of Sydney West Area Health Service. AN was induced in SCID mice by a single tail-vein injection of Adriamycin (5.2 mg/kg; David Bull Laboratories, Victoria, Australia). This model has been described in detail in BALB/c mice previously (17). AN in SCID mice is similar to that in BALB/c mice. In SCID AN, proteinuria was observed from days 3 to 5 and remained elevated throughout the 4 wk of observation. A focal increase in reabsorption droplets in tubular cells appeared at week 1, and tubular and glomerular injury appeared after week 2 and progressed to become severe by week 4. Macrophage infiltration was observed from days 7 to 9 within both interstitium and glomeruli. CD4⁺CD25⁺ T cells were reconstituted at a time when proteinuria and renal injury with macrophage infiltration was already apparent.

Cell Isolation and Purification

CD4⁺ T cells were isolated from blood, spleen, and lymph nodes of BALB/c mice. Single-cell suspensions were prepared and fractionated by Lymphoprep, and CD4⁺ T cells were isolated from leukocytes by positive selection with DYNAL separation system (Dynal Biotech, Lake Success, NY). CD4⁺CD25⁺ T cells and CD4⁺CD25⁻ T cells were isolated from blood, spleen, and lymph nodes of BALB/c mice using Miltenyi Biotech isolation kit (Bergisch Gladbach, Germany). Briefly, non-CD4⁺ T cells were depleted using a biotinylated antibody cocktail and anti-biotin microbeads. CD4⁺CD25⁺ T cells then were positively selected using phycoerythrin (PE)-labeled anti-CD25 mAb and anti-PE microbeads. The purity of CD4⁺CD25⁺ Treg and CD4⁺CD25⁻ T cells was >96 and 95%, respectively. Macrophages were isolated by positive selection using anti-CD11b mAb (Miltenyi Biotech), followed by incubation at 37°C in a CO₂ incubator for 2 h and collection of only the adherent cells (average purity >95%).

T Cell Reconstitution

SCID mice were reconstituted by intravenous injections of 3 × 10⁶ CD4⁺ or non-CD4⁺ T cells or of 1 × 10⁶ CD4⁺CD25⁺ or CD4⁺CD25⁻ T cells at day 8 after Adriamycin administration. Antibody treatment commenced the day after T cell reconstitution. Where indicated, mice were treated with 250 μg of anti–TGF-β (1D11.16.8; Bioexpress, West Lebanon) every fourth day after reconstitution for the duration of the experiment or an isotype control by intraperitoneal injection. All mice were killed 4 wk after Adriamycin administration.

Assessment of Renal Injury

For assessment of renal function, fasting mice were placed in metabolic cages for 12 h to collect urine for determination of urinary protein and creatinine. Creatinine clearance (Ccr) was calculated as urinary creatinine excretion divided by plasma creatinine concentration. Venous blood for serum creatinine and albumin was collected when the mice were killed. All urine and blood specimens were analyzed by the Institute of Clinical Pathology and Medical Research, Westmead Hospital, using a BM/Hitachi 747 analyser (Tokyo, Japan).

Histology and Immunohistochemistry

Formalin-fixed coronal sections of renal tissue were embedded in paraffin, cut in 5-μm sections, and stained with periodic acid-Schiff for histologic analysis. The degree of renal injury was estimated by evaluating the percentage of renal injury per field and was graded on a scale of 0 to 3. A semiquantitative score from two blinded trained observers was used to evaluate the degree of renal injury, and a minimum of eight consecutive fields at a magnification of ×400 were assessed and scored in each section. For the CD4⁺ reconstitution experiment, a computer-assisted image analysis system was used to quantify glomerular and tubulointerstitial structure in the renal cortex, as described previously (16).

Sections that were cut from frozen tissue (5 μm) were used for immunohistochemistry using standard techniques. Sections were incubated with antibodies (Becton Dickinson, Mountain View, CA) against macrophages (ED1) and CD4 T cells for 60 min at room temperature. After blocking with 3% H₂O₂ to eliminate endogenous peroxidases, sections were incubated in biotinylated rabbit anti-IgG (1:200) and then in an avidin-biotin-horseradish peroxidase complex. Reaction was visualized by the addition of freshly prepared 3,3-diaminobenzidine tetrahydrochloride (7.5 mg in 15 ml of PBS with 15 μl of 30% H₂O₂) and examined by light microscopy.

Reverse Transcription–PCR

Total RNA was extracted from renal cortical cells using TRIZOL Reagent (Invitrogen, Life Technologies, Scotland, UK) and reverse-transcribed using Superscript 11 reverse-transcriptase (Invitrogen Japan K.K., Tokyo, Japan). Foxp3 mRNA levels were quantified by real-time PCR using the ABI PRISM 7700 sequence detection system (PE Applied Biosystems, Foster City, CA). The primers and TaqMan probe sequences for real-time PCR were as follows: Foxp3 primers 5′-TTG GCC AGC GCC ATC TT-3′ and 5′-TGC CTC CTC CAG AGA GAA GTG-3′; Foxp3 probe 5′-6FAMCAG CTG CTG CTC CAGMGBNFQ-3′; glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers: 5′-TGC ACC ACC AAC TGC TTA GC −3′ and 5′-GGA AGG CCA TGC CAG TGA-3′; and GAPDH probe 5′-VICCCT GGC CAA GGT CAT CCA TGA CAA CTTMGBNFQ-3′. A normalized value for Foxp3 mRNA expression in each sample was calculated as the relative quantity of Foxp3 divided by the relative quantity of GAPDH from five animals that were picked randomly from each group. All samples were run in triplicate.

Flow Cytometric Analysis

FACS analysis was performed using FITC-conjugated anti-mouse CD25 (IL-2 receptor α chain), PE-conjugated anti-mouse CD3 (145 2C11), and PE-conjugated anti-mouse CD4 (L3T4; BD Bioscience, New South Wales, Australia). For detection and quantification of cell surface markers, cells were treated with mAb that were directed against these molecules. Cells were stained with adequate concentration of FITC- or PE-labeled antibodies and allowed to stand at room temperature for 30 min. The cells were washed and resuspended in PBS for analysis. All samples were analyzed on a FACScan analyzer (Becton Dickinson). CellQuest software was used for acquisition and analysis (Becton Dickinson).

In Vitro Experiments

Purified macrophages as well as CD4⁺CD25⁺ and CD4⁺CD25⁻ T cells were stimulated with LPS (2 μg/μl) and anti-CD3 antibody, respectively, for 2 h in flat-bottom plates (Nunc, Roskilde, Denmark). Cells were cultured in RPMI 1640 medium (Life Technologies BRL Life Technologies, Merelbeke, Belgium) supplemented with 1% streptomycin and 10% heat-inactivated FCS. All cells were washed three times after stimulation and co-cultured in varying responder:suppressor ratios for 16 h, 24 h, and 3 d. In another set of experiments, stimulated CD4⁺CD25⁺ T cells were co-cultured with stimulated macrophages in the presence of neutralizing anti–TGF-β (2 to 10 μg/ml) and anti–IL-10 (2 μg/ml) antibodies. Expression of TNF-α, IL-12, CCL3, CCL17, inducible nitric oxide synthetase (iNOS), IL-10, and TGF-β was assessed using reverse transcription–PCR. Because no difference was observed in the suppression using the three different time points above, cells were co-cultured for 24 h for all of the experiments.

RNA was isolated using TRIZOL Reagent (Invitrogen, Life Technologies) according to the manufacturer’s instructions. The total amount of RNA was reverse transcribed using Superscript 11 reverse-transcriptase and oligo (dT) primer in a final volume of 20 μl. Semiquantitative PCR to assess cytokines, chemokines, and iNOS was carried out after reverse transcription of equal amounts of RNA (2 μg) from each sample. PCR amplification was tested on each primer pair at different cycles to ensure amplification in the linear range.

Statistical Analyses

Values are expressed as mean ± SD. Statistical analysis was performed using SPSS 11.0 (SPSS Inc., Chicago, IL). Statistical significance of differences among groups was determined by one-way ANOVA. The post test of ANOVA was the post hoc least significant difference test. P < 0.05 was considered statistically significant.

CD4⁺CD25⁺ Treg Suppress Macrophage Phenotype and Proinflammatory Cytokine Production

Because macrophages secrete and synthesize TNF-α and IL-12, proinflammatory cytokines that are known to induce glomerular damage (18), we examined in vitro the effect of stimulated CD4⁺CD25⁺ T cells on macrophage TNF-α and IL-12 expression. Macrophages that were cultured alone, without any stimulus, did not produce any cytokines but when stimulated with LPS showed increased expression of TNF-α and IL-12. CD4⁺CD25⁺ T cells showed a suppressive effect on the expression of both of these cytokines (Figure 1, A and B) when co-cultured with stimulated macrophages irrespective of responder:suppressor ratio of cells (data not shown). CD4⁺CD25⁺ T cells, being more potent, exhibited complete suppression even at a ratio of 1:0.5.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

(A) Reverse transcription–PCR analysis of purified macrophages and CD4⁺CD25⁺ T cells alone and in coculture (1:1) without inserts, showing that the effects were contact-dependent. After 24 h, purified RNA was reverse transcribed, and primers for the indicated genes were used to amplify the cDNA. IL-10 and TGF-β were expressed only by CD4⁺CD25⁺ T cells; other genes were expressed only by macrophages. Representative of three independent experiments, each performed in triplicate. (B) Relative intensity of gene expression by macrophages alone or in coculture (1:1) with CD4⁺CD25⁺ T cells, with respect to housekeeping gene (18S). Data are mean ± SD from three independent experiments, each performed in triplicate. *P < 0.001 versus unstimulated macrophages, CD4⁺CD25⁺ T cells + stimulated (s) macrophages, and CD4⁺CD25⁺ T cells + s macrophages + anti–IL-10; ^†P < 0.001 versus all other groups. (C) Inhibitory effect of CD4⁺CD25⁻ T cells on the expression of TNF-α, IL-12, CCL3, and inducible nitric oxide synthase (iNOS) depended on the responder:suppressor ratio of macrophages and CD4⁺CD25⁻ T cells. CD4⁺CD25⁺ T cells were fully suppressive at each ratio (data not shown), whereas no effect was seen when macrophages were cultured in excess of CD4⁺CD25⁻ T cells. Data are mean ± SD from three independent experiments, each performed in triplicate. *P < 0.05, **P < 0.001 versus the other two ratios. (D) Dose-response effect of anti–TGF-β, showing an inhibition of the suppressive effect of CD4⁺CD25⁺ T cells on macrophage expression of TNF-α and IL-12 (from 2 μg/ml) and of CCL3 and iNOS (from 5 μg/ml). Data are mean ± SD from three independent experiments, each performed in triplicate. P < 0.001 versus 0 anti–TGF-β for all comparisons (except CCL3 and iNOS at 2 μg/ml).

CD4⁺CD25⁻ T cells also demonstrated inhibition of these cytokines, but the inhibitory effect depended on the responder:suppressor ratio. CD4⁺CD25⁻ T cells did not show any suppressive effect when macrophages were cultured in excess of CD4⁺CD25⁻ T cells (Figure 1C).

The phenotype of LPS-induced macrophages was investigated on the basis of expression of chemokines CCL3 and CCL17 and iNOS. LPS-induced macrophages showed increased expression of CCL3 and iNOS, which was absent in the unstimulated macrophages, but there was no expression of CCL17 or IL-10 in either of the macrophage populations (Figure 1, A and B), suggesting that LPS-stimulated macrophages are type I or classically activated macrophages (19,20). IL-10 and TGF-β expression was observed in the CD4⁺CD25⁺ T cells alone and when co-cultured with macrophages. Macrophages alone and CD4⁺CD25⁻ T cells, however, did not produce any TGF-β in vitro under the conditions described in our study. In addition, CD4⁺CD25⁺ T cells inhibited the expression of CCL3 and iNOS by stimulated macrophages when co-cultured together (Figure 1A). CD4⁺CD25⁻ T cells also showed an inhibitory effect, which depended on the responder:suppressor ratio (Figure 1C).

Neutralizing anti–TGF-β and anti–IL-10 antibodies were used to define the role of these cytokines in the suppressive effect shown by CD4⁺CD25⁺ Treg. When macrophages were co-cultured with CD4⁺CD25⁺ T cells in the presence of neutralizing antibodies, anti–TGF-β reversed the suppressive effect of CD4⁺CD25⁺ Treg in a dose-dependent manner (Figure 1, A, lane 6, and D). Anti–IL-10 antibodies, however, did not show any effect on the suppressive effect of CD4⁺CD25⁺ Treg (Figure 1A, lane 7).

SCID Mice Are More Sensitive to AN

All SCID mice that received an injection of Adriamycin developed severe AN, characterized by proteinuria, hypoalbuminemia, hypercreatininemia, extensive interstitial inflammatory infiltrates, and progressive renal injury. The histologic changes were dominated by tubulointerstitial injury, including tubule cell atrophy with brush border loss and interstitial expansion with monocytic infiltration, and glomerular sclerosis. The mean body weight of mice with AN fell >15% at the end of week 1, which was greater than shown previously in immunocompetent BALB/c mice (17) despite use of a much lower dose (5.2 mg/kg body wt in SCID and 9.8 mg/kg body wt in BALB/c mice).

Adoptive Transfer of CD4⁺ T Cells Protects against T Cell–Independent Injury

Mice that were reconstituted with CD4⁺ T cells had significantly reduced glomerular sclerosis and mesangial matrix expansion than mice that were treated with Adriamycin alone and mice that were reconstituted with non-CD4⁺ leukocytes (Table 1). Tubular atrophy and interstitial volume were significantly decreased in mice that were reconstituted with CD4⁺ T cells but increased in mice that were reconstituted with non-CD4⁺ cells (Figure 2). Urine protein levels were significantly lower in mice that were reconstituted with CD4⁺ T cells as compared with other groups (Table 2), consistent with the histologic changes.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Representative renal cortical sections stained by periodic acid-Schiff (PAS) from normal SCID mice (A), SCID mice with Adriamycin nephropathy (AN; 4 wk after Adriamycin; B), SCID mice with AN (4 wk after Adriamycin and 3 wk after reconstitution with CD4⁺ T cells; C), and SCID mice with AN (4 wk after Adriamycin and 3 wk after reconstitution with non-CD4⁺ cells; D). The results of morphometric analysis of the histologic changes are summarized in Table 1.

CD4⁺CD25⁺ Treg Protect against T Cell–Independent Injury

To examine whether Treg could prevent renal injury that was induced by Adriamycin in a lymphocyte-independent model, we isolated CD4⁺CD25⁺ T cells and CD4⁺CD25⁻ T cells from normal BALB/c mice and injected these subpopulations into SCID recipients with established AN. All mice that received an injection of Adriamycin alone developed severe nephropathy (Figures 3 and 4). Transfer of CD4⁺CD25⁺ T cells significantly inhibited functional and structural injury. Urinary protein excretion and Ccr were significantly improved by reconstitution with CD4⁺CD25⁺ T cells (Figure 3). There was a minor improvement in Ccr but not proteinuria with CD4⁺CD25⁻ T cell reconstitution. By light microscopy, there was only mild damage in glomeruli and tubules in the mice that were reconstituted with CD4⁺CD25⁺ T cells as compared with unreconstituted mice with AN (Figures 4 and 5). There was no histologic protection from reconstitution with CD4⁺CD25⁻ cells. There was a significant decrease in the number of macrophages in interstitium of mice that were reconstituted with Treg as compared with unreconstituted mice with AN (Figure 6). There was a similar increase in the number of cortical CD4⁺ T cells in the mice that were reconstituted with either CD4⁺CD25⁺ or CD4⁺CD25⁻ T cells (Figure 7).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Urinary protein (A) and creatinine clearance (Ccr; B) at 4 wk in SCID mice with AN, untreated except for Adriamycin (n = 12); reconstituted with CD4⁺CD25⁺ T cells (n = 12) or CD4⁺CD25⁻ T cells (n = 12); reconstituted with CD4⁺CD25⁺ T cells (n = 7) or CD4⁺CD25⁻ T cells (n = 7) and treated with 250 μg of anti–TGF-β (1D11.16.8; Bioexpress) intraperitoneally every fourth day after reconstitution for the duration of the experiment; or treated with anti–TGF-β alone (n = 7). Reconstituted mice that were treated with an isotype control antibody (n = 5; data not shown) were no different from reconstituted mice that received no antibody. Urinary protein was significantly lower and Ccr significantly higher in AN mice with CD4⁺CD25⁺ T cells than in each of the other groups. *P < 0.05 versus AN alone and AN/anti–TGF-β; **P < 0.001 versus all other groups. Data are mean ± SD.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Semiquantification of glomerular sclerosis (A) tubular atrophy (B), and interstitial injury (C) at 4 wk in SCID mice with AN, untreated except for Adriamycin (n = 12); reconstituted with CD4⁺CD25⁺ T cells (n = 12) or CD4⁺CD25⁻ T cells (n = 12); reconstituted with CD4⁺CD25⁺ T cells (n = 7) or CD4⁺CD25⁻ T cells (n = 7) and treated with 250 μg of anti–TGF-β intraperitoneally every fourth day after reconstitution for the duration of the experiment; or treated with anti–TGF-β alone (n = 7). Reconstituted mice that were treated with an isotype control antibody (n = 5; data not shown) were no different from reconstituted mice that received no antibody. SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells showed significantly less glomerular sclerosis, tubular atrophy, and interstitial injury than AN mice without reconstitution. *P < 0.001 versus all other groups.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Representative renal cortical sections stained by PAS. (A) Normal SCID mice showed morphologically normal glomeruli and tubules. (B) SCID mice with AN (4 wk after Adriamycin) showed glomerulosclerosis, tubular atrophy, interstitial expansion, and hyaline casts. (C) SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells from day 7 had less renal injury at 4 wk than mice with AN alone. SCID mice that had AN and were reconstituted with CD4⁺CD25⁻ T cells (D), reconstituted with CD4⁺CD25⁺ T cells and treated with 250 μg of anti–TGF-β (ID11) intraperitoneally every fourth day after reconstitution for the duration of the experiment (E), or reconstituted with CD4⁺CD25⁻ T cells and treated with anti–TGF-β (F) each had injury that was no different from SCID mice with AN alone.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Representative immunostaining for renal macrophages (A) and quantification of interstitial macrophages (B) in normal SCID mice (n = 5; A), SCID mice with AN (4 wk after Adriamycin; n = 12; B), SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells (n = 12; C), SCID mice that had AN and were reconstituted with CD4⁺CD25⁻ T cells (n = 12; D), and SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells (n = 7; E) or CD4⁺CD25⁻ T cells (n = 7) and treated with 250 μg of anti–TGF-β (ID11; F). Interstitial macrophage infiltration was significantly reduced in SCID mice with AN after reconstitution with CD4⁺CD25⁺ T cells. *P < 0.001 versus all other groups. Data are mean ± SD.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

Representative immunostaining (A) and quantification (B) for cortical CD4⁺ T cells in normal SCID mice (n = 5; A), SCID mice with AN (4 wk after Adriamycin; n = 12; B), SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells (n = 12; C), SCID mice that had AN and were reconstituted with CD4⁺CD25⁻ T cells (n = 12; D), and SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells (n = 7; E) or CD4⁺CD25⁻ T cells (n = 7) and treated with 250 μg of anti–TGF-β (ID11; F). *P < 0.001 versus AN/CD4⁺CD25⁺, AN/CD4⁺CD25⁻, AN/CD4⁺CD25⁺/anti–TGF-β and AN/CD4⁺CD25⁻/anti–TGF-β; ^†P < 0.001 versus all other groups. Data are mean ± SD.

Foxp3 Expression Correlates with Protection against Injury

The expression of Foxp3 mRNA was examined in cortical sections of all treated and control groups and in purified CD4⁺ and CD4⁺CD25⁺ T cells. Both total CD4⁺ and CD4⁺CD25⁺ T cell populations expressed Foxp3, and, as expected, the expression was greater in the CD4⁺CD25⁺ Treg (Figure 8A). There was a significant increase in renal cortical expression of Foxp3 in the mice that were reconstituted with CD4⁺CD25⁺ T cells as compared with unreconstituted mice with AN (Figure 8B), and the degree of protection against glomerulosclerosis, tubular atrophy, and interstitial injury that was provided by reconstitution with CD4⁺CD25⁺ T cells correlated with levels of cortical Foxp3 (Figure 9). The mice that were reconstituted with CD4⁺CD25⁻ T cells also showed higher expression of Foxp3 than unreconstituted mice but significantly less than the mice that were reconstituted with CD4⁺CD25⁺ T cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 8.

Foxp3 mRNA levels normalized against glyceraldehyde-3-phosphate dehydrogenase. (A) Levels of Foxp3 mRNA in freshly isolated cells. (B) SCID mice that had AN and were reconstituted with CD4⁺CD25⁺ T cells (n = 5) showed a significant increase in renal cortical Foxp3 mRNA in comparison with mice with AN alone (n = 5) as well as mice that had AN and were reconstituted with CD4⁺CD25⁻ T cells (n = 5). The results shown are mean ± SD of two independent measurements, each with triplicate wells. *P < 0.001 versus other groups.

• Download figure
• Open in new tab
• Download powerpoint

Figure 9.

Quantitative real-time PCR analysis for the detection of Foxp3 mRNA expression in renal cortex of SCID mice that had AN and were reconstituted or not with CD4⁺CD25⁺ or CD4⁺CD25⁻ T cells and treated or not with anti–TGF-β. The degree of protection, against glomerulosclerosis (A), tubular atrophy (B), and interstitial injury (C) correlated with the levels of cortical Foxp3 mRNA. P < 0.001 for each correlation. Each symbol represents the average value from three independent measurements in an individual mouse.

CD4⁺CD25⁺ Treg–Mediated Protection against T Cell–Independent Injury Is TGF-β–Dependent

For examination of whether TGF-β played a role in the inhibition of T cell–independent AN, SCID mice received an injection of Adriamycin, were reconstituted with CD4⁺CD25⁺ or CD4⁺CD25⁻ T cells, and were treated with mAb that were reactive with TGF-β. As shown in Figure 3, there was a significant increase in urinary protein excretion and decrease in Ccr in mice that were reconstituted with CD4⁺CD25⁺ T cells and treated with anti–TGF-β antibodies as compared with those with reconstitution alone. Isotype control antibody had no effect on the protective effect of CD4⁺CD25⁺ T cells. Treatment with anti–TGF-β completely abolished the ability of CD4⁺CD25⁺ T cells to protect against structural injury of AN (Figure 4). However, there was no effect on AN injury of unreconstituted mice that were treated with anti–TGF-β antibodies alone (Figures 3 and 4). Cortical Foxp3 expression of mice that were reconstituted with CD4⁺CD25⁺ T cells and treated with anti–TGF-β antibodies was significantly downregulated compared with the mice that were reconstituted with CD4⁺CD25⁺ T cells alone (Figure 9).