Superagonistic CD28 protects against renal ischemia injury induced fibrosis through a regulatory T-cell expansion dependent mechanism

Six to eight-week-old male C57BL/6 J mice (weight, 20–25 g) were purchased from Silaike (China), 8–10 for each subgroup. Animals were housed in temperature- and humidity-controlled cages, with free access to water and rodent food on a 12-h light/dark cycle. Experiments were completed at Zhongshan Hospital Fudan University. The animal use protocols were approved by the Institutional Animal Care and Use Committee of Fudan University and Zhongshan Hospital. They also strictly adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All surgeries were performed under intraperitoneal 1% sodium pentobarbital anesthesia (3.0–4.5 ml/kg), intrarectal temperature of mice was maintained at 35.0 °C–36.0 °C with a heating pad during the surgeries. A midline abdominal incision was made, and the mice were subjected to bilateral renal pedicle clamping for 35 min using non-traumatic microvascular clips. After removal of the clamps, blood restoration in the kidneys was confirmed visually. To maintain body fluid balance, all mice were supplemented with 0.5 ml of saline administered subcutaneously. A sham operation was performed in a similar manner, except for renal pedicle clamping. After surgery, mice were transferred to the recovery cages, one cage for each. Parameters like vital signs, time to wake up from anesthesia were monitored. Food intake was daily monitored, and animals were weighed once every other day. Mice were sacrificed at 1, 3, 7, 14, and 28 days after the procedure, and serum and kidney tissues were collected for various analyses. One percent pentobarbital sodium (4 ml/kg) were injected intraperitoneally, 15 min later, mice were sacrificed by cervical dislocation. The early mortality rate (mice died before the indicated time point) of mice subject to IR, ranged from 10 to 30%.

Superagonistic CD28 was purchased from AbD Serotec (USA). LEAF-purified anti-mouse CD25 antibody was purchased from Biolegend (USA). For the CD28sa-IR group, a bolus injection of 4–4.5 mg/kg CD28sa in 1 ml phosphate buffer (PBS) was intraperitoneally administered to mice 6 days (day 1) before IR. The ischemia preconditioning (IPC) was induced 6 days before IR for 18-min pre-ischemia. The sham operation group and PBS group were used as controls. In order to deplete Tregs, anti-CD25 antibodies (PC61, Biolegend, USA) were administrated intraperitoneally to mice at a dose of 8–9 mg/kg on day 3 and day 5 for the CD28sa-PC61-IR group as reported in our previous study [14]. On day 7, mice were subjected to IRI injury. Blood, kidney, and spleen tissues were harvested at 24 h, 3 days, 7 days, 14 days, and 28 days after renal IRI. At indicated time point, 1% pentobarbital sodium 4 ml/kg were injected intraperitoneally, 15 min later, mice were sacrificed by cervical dislocation (Fig. 1).

The number of stain positive cells was analyzed using immunohistochemistry (IHC) as described previously [14]. Anti-Fibronectin rabbit antibody (Cell signaling technology, USA), anti-Collagen IV rabbit antibody (Abcam, USA), and anti-8-OHdg rabbit antibody (Abcam, USA) were used as primary antibodies according to the manufacturer’s instructions. Secondary antibodies (1:5000; Jackson ImmunoResearch) were also used according to the manufacturer’s instructions. All immunohistochemistry staining was analyzed as follows: under the microscope (200x or 400x), the sections were moved randomly to take 20–25 pictures per section, followed by analysis using Image Pro-Plus 6.0 Software (Media Cybernetics, USA) to identify the target cells.

For detection of CD4⁺Foxp3⁺ Tregs, spleens, blood, and kidneys of C57BL/6 mice were subjected to flow cytometry analysis as described previously [14]. Anti-CD4- fluorescein isothiocyanate (FITC), anti-CD25-Phycoerythrin (PE), anti-Foxp3-efluor450, anti-IL-17A-allophycocyanin (APC), anti-CD11c-PE, anti-MHCII-FITC antibodies were purchased from eBioscience (USA). FACS Attune Nxt (Life, Thermo) was used for the analysis.

Relative protein abundances were detected using western blotting as we previously described [6‐8, 14]. The antibodies we used in this study were from CST, Abcam, BioLegend and eBioscience.

Renal tubulointerstitial fibrosis was detected using Masson’s trichrome staining or sirius red staining, the extent of renal fibrosis was shown as the percentage of blue-stained area or sirius red-stained area in renal cortex and outer medulla.

The mouse cytokines interleukin-1α(IL-1α/β), IL-2, IL-3, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-17A, Eotaxin, granulocyte-macrophage colony stimulating factor (GM-CSF), granulocyte-colony stimulating factor (G-CSF), interferon-γ (IFN-γ), keratinocyte chemoattractant, monocyte chemotactic protein-1 (MCP-1, MCAF), macrophage inflammatory protein-1α (MIP-1α), macrophage inflammatory protein-1β (MIP-1β), RANTES, and tumor necrosis factor-α (TNF-α) were quantified using Mouse 23-plex Multi-Analyte Kit (Bio-Plex Suspension Array System; Bio-Rad, Hercules, CA, USA). The antibody array experiment was performed according to their established protocol by Wayen Biotechnology (Shanghai, China). The exact protocol was administered according to what had been reported before [15] .

When renal function was assessed after IRI, significant improvement was found in the CD28sa group compared to that in the IR group (Fig. 2a). Due to its ability to systemically neutralize CD28sa induced Treg expansion, PC61 abrogated the renoprotective effects of CD28sa on day 1 and day 7 post-IRI (Fig. 2a).

We measured the percentage of Tregs, dendritic cells, and Th17 cells in the kidney, peripheral blood, and spleens by flow cytometry at 24 h, 7 days, 14 days, and 28 days post-IRI in the mice model. Since we previously found Treg expansion reached a peak at 6 days after CD28sa treatment [14], we administered CD28sa or PBS at 6 days before IRI. CD28sa induced a significant increase in the percentage of Foxp3⁺CD4⁺ Tregs of CD4⁺ T cells from the spleen, blood, and kidney (Fig. 2b-d). When anti-CD25 antibody (PC61) was administered after injection of CD28sa on day 3 and 5, expansion of Tregs in the spleen, peripheral blood, and kidney was mostly abrogated in the short term (Fig. 2b-d).

The balance between Th17 and Tregs plays a key role in the maintenance of immune homeostasis in vivo. To illustrate whether the expansion effect of CD28sa could be connected to Th17 cells during IRI-induced inflammation and fibrosis, we checked the Th17 cell percentage at different time points after IRI. The percentage of IL-17A⁺CD4⁺ Th17 cells of the renal tissue indicated a remarkable decrease in the CD28sa-IR group compared with the IR group at 24 h after IRI (Fig. 2e-f). Tregs are capable of inhibiting Th17 cells and other effector T cells. The percentage of CD11c⁺MHCII⁺ dendritic cells in the kidney increased significantly in the CD28sa-ischemia reperfusion (IR) group (Fig. 2g-h). These results suggested that CD28sa treatment inhibited Th17 cell accumulation and promoted Tregs and CD11c⁺MHCII⁺ dendritic cell accumulation in the early stage of post-IRI inflammation.

To further verify the protective effects of CD28sa against IRI-induced renal fibrosis, we checked the histological changes of mouse kidneys. Tubular cell vacuolization, cast formation, and loss of brush border was predominant at the cortico-medullary junction by 28 days after IRI. In contrast, mice treated with CD28sa presented mild renal morphologic abnormalities (Fig. 3a).

Subsequently, we assessed kidney fibrosis at 28 days post-injury. Pathological examination showed no renal fibrotic lesions in the PBS-treated group. Tubulointerstitial fibrosis was prominent in the IRI group, with excess collagen deposition evidenced by masson staining and sirius red staining (Fig. 3b-c). Simultaneously, CD28sa-treated mice showed attenuated renal pathological damage and less collagen deposition (Fig. 3b-c). Western blot also showed that renal expression of collagen IV protein was reduced in the CD28sa-IR group compared with that in the IR group (Fig. 3d-e). As we previously reported, CD28sa mimicked the renoprotective effects of IPC on acute kidney ischemic injury. In this study, we have also observed that CD28sa mimicked the renoprotective effects of IPC on the long-term outcome. As shown in Fig. 3d-e, either IPC treatment or CD28sa treatment significantly attenuated renal protein expression of collagen IV at day 28 post IRI. Conversely, PC61 abolished all the beneficial effects conferred by CD28sa treatment.

Extracellular matrix (ECM) is a three-dimensional network of extracellular macromolecules such as collagen, enzymes, and glycoproteins that provide biochemical and structural support of surrounding cells. We examined the expression of fibronectin and collagen IV to fully indicate the ECM deposition of kidneys. Immunochemistry staining showed that fibronectin and collagen IV deposition induced by IR injury was significantly mitigated by CD28sa treatment (Fig. 4a-b).

Oxidative stress is known to have detrimental effects on T cells and natural killer (NK) cells both in chronic inflammatory conditions and in cancer [16]. Tregs have been reported to possess anti-oxidative capacity in the cancer microenvironment [17], therefore, we checked the expression of DNA associated oxidative stress marker, 8-Oxoguanine (8-OHdg) among different groups. Immunochemistry staining revealed that CD28sa treatment significantly downregulated 8-OHdg expression at day 28 post IRI (Fig. 4c), whereas depletion of Tregs by the anti-CD25 antibody PC61 significantly reversed the anti-oxidative effects secondary to CD28sa treatment. What’s more, CD28sa treatment showed similar anti-fibrotic and anti-oxidative effects with IPC treatment, as both CD28sa-IR group and IPC-IR group presented less fibronectin and Collagen IV deposition as well as less 8-OHdg expression, compared to the IR group (Fig. 4d-f).

Apoptosis and cell proliferation play a key role in tissue injury and repair [18]. At 28 days post injury, there was still an increased number of TUNEL positive cells in the kidneys of the IR group compared with those in the control group, which was attenuated by CD28sa treatment (Fig. 5a-b). Meanwhile, Ki67 staining of kidney specimens showed an increased number of proliferating cells in mice with CD28sa treatment (Fig. 5a,c). Similarly, these protective effects could be reversed by PC61 administration.

To further investigate the mechanism of CD28sa-induced anti-inflammation and anti-fibrosis effects, we examined the serum levels of 23 cytokines in mice over time after IRI. The expression of several cytokines differed between the IR group and CD28sa-IR. IL-10, the most well-known Treg effector cytokine, was increased in the CD28sa-IR group compared with the IR group at 24 h and 7 days after IRI (Fig. 6a). IL-6, a pro-inflammatory cytokine, which may associate with downregulated Th17 cells, presented a decrease in the CD28sa-IR group at 24 h and 7 days after IRI (Fig. 6b). IL-17A, the typical effector cytokine of Th17 cells was decreased significantly in the CD28sa-IR group compared with that in the IR-group at 24 h and 7 days (Fig. 6c).