Dr Mi-La Cho, Department of Internal Medicine, Catholic Institute of Medical Science, 505 Banpo-Dong, Seocho-Ku, 137-040, Seoul, Korea. Email: firstname.lastname@example.org and Dr Chul Woo Yang, Department of Internal Medicine, Seoul St Mary’s Hospital, 505 Banpo-Dong, Seocho-Ku, 137-040, Seoul, Korea. Email: email@example.com Senior author: Chul Woo Yang, email: firstname.lastname@example.org
The aim of this study is to investigate the clinical significance of the ratio between interleukin-17 (IL-17) secreting cell and FOXP3-positive regulatory T cell (FOXP3+ Treg) infiltration in renal allograft tissues with acute T-cell-mediated rejection (ATCMR). Fifty-six patients with biopsy-proven ATCMR were included. Infiltration of FOXP3+ Treg and IL-17-secreting cells was evaluated with immunostaining for FOXP3 or IL-17 on the biopsy specimens, and the patients were divided into the FOXP3 high group (Log FOXP3/IL-17 > 0·45) or the IL-17 high group (Log FOXP3/IL-17 < 0·45). We compared the allograft function, severity of tissue injury, and clinical outcome between the two groups. In the IL-17 high group, allograft function was significantly decreased compared with the FOXP3 high group (P <0·05). The severity of interstitial and tubular injury in the IL-17 high group was higher than the FOXP3 high group (P <0·05). The proportions of steroid-resistant rejection, incomplete recovery and recurrent ATCMR were higher in the IL-17 high group than in the FOXP3 high group (all indicators, P <0·05). The IL-17 high group showed lower 1-year (54% versus 90%, P <0·05) and 5-year (38% versus 85%, P <0·05) allograft survival rates compared with the FOXP3 high group. Multivariate analysis revealed that the FOXP3/IL-17 ratio was a significant predictor for allograft outcome. The FOXP3/IL-17 ratio is a useful indicator for representing the severity of tissue injury, allograft dysfunction and for predicting the clinical outcome of ATCMR.
FOXP3+ regulatory T cells (Treg) play a critical role in suppressing the immune responses of recipients to allografts.1,2 Therefore, high infiltration of FOXP3+ Treg in allograft tissue is expected to have significant associations with a favourable allograft outcome. Indeed, the higher numbers of FOXP3+ Treg in a protocol biopsy are associated with the donor-specific hyporesponsiveness.3 In other studies, they were associated with favourable outcomes in subclinical rejection or chronic inflamed fibrotic tissue.4,5 In contrast, the detection of FOXP3+ Treg in acute T-cell-mediated rejection (ATCMR) did not suggest a favourable outcome. Veronese et al.6 observed that the presence of Treg had no significant association with the allograft outcome in patients undergoing biopsy-proven ATCMR. In another study, FOXP3 expression in allograft tissue with ATCMR did not correlate with a favourable outcome, and they concluded that the effect of inflammation could mask the benefits of FOXP3+ Treg in biopsies with ATCMR.7
Interleukin-17 (IL-17) is pro-inflammatory cytokine that has an important role in both autoimmune disorders and alloimmune reactions in solid organ transplantation.8 Even though it is a pro-inflammatory mediator, it has close connections to FOXP3+ Treg.9,10 For example, T helper type 17 (Th17) cells, the major source of IL-17, developed from a common precursor with FOXP3+ Treg and it can interconvert with Treg according to the microenvironment.11–13 In addition, FOXP3+ Treg can differentiate into IL-17-producing cells under certain circumstances.14,15 Therefore, the interplay between IL-17 and Treg is an important mechanism for modulating the immune responses in various immunological disorders.16–19 In previous reports, the ratio between FOXP3+ Treg and IL-17-secreting T cells was associated significantly with the disease activity in autoimmune disease, graft-versus-host disease after haematopoietic stem cell transplantation, and the atherosclerotic inflammatory condition.19–22
Based on previous reports about the interplay between IL-17-secreting cells and FOXP3+ Treg, we hypothesized that the ratio between FOXP3+ Treg and IL-17-secreting cell infiltration is a prognostic marker for determining allograft outcome in ATCMR. In this report, we investigated the cell infiltration that expresses FOXP3 or IL-17 in allograft tissue with biopsy-proven ATCMR, and we intended to appraise whether the ratio between them is associated with allograft outcome after ATCMR.
Subjects and Methods
The study population consisted of 71 clinically indicated renal allograft biopsies performed on 56 renal transplant recipients in our transplant centre from August 1999 to August 2008. Of the 71 biopsy samples, 56 biopsies were a first-time ATCMR and the other 15 specimens were repeat ATCMR biopsy samples (13 specimens were the second ATCMR and two specimens were third ATCMR).
The indication for the allograft biopsy was graft dysfunction defined as a serum creatinine increment of greater than or equal to 10% from the baseline value. These cases were selected only for the diagnosis of ATCMR type I or II according to Banff’s working classification and the availability of sufficient paraffin-embedded tissue.23,24 BK virus or cytomegalovirus nephropathy, lymphoproliferative disorder, interstitial fibrosis/tubular atrophy (IF/TA) grade III was not present in these patients or biopsies.
Out of 56 patients, 33 patients (59%) were a living related donor, 13 cases (23%) were a living unrelated donor, and 10 cases (17·9%) were deceased donor transplantation. The HLA mismatch number was 3·7 ± 1·3 and four cases (7%) were a second transplantation. The flow-cross-match test before transplantation was negative and the Panel reactive antibody was less than 20% in all patients. Our centre’s protocol for immune suppression is described in a previous publication.25 Briefly, the main immunosuppressive agents used were cyclosporine (n = 31, 55%) or tacrolimus (n = 25, 45%). Mycophenolate mofetil was added as a primary immunosuppressant in 42 patients (75%). Basiliximab was used as an additional induction therapy in 22 patients (39%). Patients were followed from the date of transplantation to the date of nephrectomy, permanent dialysis, re-transplantation, or death.
During the study period, ATCMR was treated with three to five daily boluses of intravenous methylprednisolone (500 mg/day), followed by a 5–7-day oral steroid taper. When the serum creatinine level failed to decrease within 5 days, muromonab-CD3 (OKT3) or anti-thymocyte globulin (ATG) was applied. The Institutional Review Board of Seoul St Mary’s Hospital approved the study.
Immunohistochemistry for FOXP3 and IL-17
All biopsies were examined for FOXP3+ cell and IL-17+ cell infiltration. Paraffin sections were immersed in three changes of xylene and hydrated using a graded series of alcohols. Antigen retrieval was performed routinely by immersing the sections in sodium citrate buffer (pH 6·0) in a microwave for 15 min. The sections were depleted of endogenous peroxidase activity by adding methanolic hydrogen peroxide and were blocked with normal serum for 30 min. After overnight incubation with polyclonal antibodies against FOXP3 (sc-21072; Santa Cruz Biotechnology, Santa Cruz, CA) and anti-human IL-17 monoclonal antibodies (R&D Systems Inc., Minneapolis, MN), the samples were incubated with the secondary antibodies, biotinylated with anti-IgG for 20 min and then incubated with a streptavidin–peroxidase complex (Vector, Peterborough, UK) for 1 hr. This was followed by incubation with 3,3’-diaminobenzidine (Dako, Glostrup, Denmark). The sections were counterstained with haematoxylin, and samples were photographed with an Olympus photomicroscope (Tokyo, Japan). The positivity for each immunohistochemistry stain was examined in a blind fashion relative to the clinical information.
Analysis of imunohistochemistry results
Analysis was performed by counting the total number of infiltrating cells that express FOXP3 or IL-17 in the cortex. The area of cortex was measured with a loupe and the data were expressed as the number of cells/mm2. The counting of the FOXP3+ and IL-17+ cells was performed by HistoQuest Experiment (TissueQuest Software, TissueGenostics, Vienna, Austria). A pathologist blinded to the results of the HistoQuest Experiment, manually counted the cell number. The FOXP3+ cell and IL-17+ cell numbers counted by pathologist and HistoQuest Experiment were highly correlated (r = 0·901, P = 0·00) and the result did not change the classification of the patient.
Indirect immunofluorescence staining was performed using monoclonal antibodies against complement protein C4d (Biogenesis, Poole, UK) in 48 (68%) biopsies. In 23 (32%) biopsies where no C4d staining was performed on frozen sections, sections were obtained from paraffin blocks and stained for immunohistochemistry with C4d using a rabbit polyclonal antibody (Biogenesis, Poole, UK). C4d positivity was defined as diffuse (> 50%) and linear staining of peritubular capillaries.
Figure 1(a,b) shows representative stains of FOXP3 and IL-17. The cell numbers of the FOXP3+ cell and IL-17+ cell infiltrations were 11·6 ± 12·2 cells/mm2 and 5·6 ± 8·0 cells/mm2, respectively. The average value of the ratio between FOXP3+ cell and IL-17+ cell (FOXP3/IL-17) was 5·6 ± 8·2. We used log transformation to correct data skewness for the FOXP3/IL-17 ratio. When log transformation of the FOXP3/IL-17 ratio (Log FOXP3/IL-17) is 0·45, it conferred the highest sensitivity (0·713) and specificity (0·724) in the prediction of allograft failure by receiver operating characteristic analysis. Therefore, when Log FOXP3/IL-17 was > 0·45, the biopsy was considered as the FOXP3 high group (n = 30) and when it was < 0·45, the biopsy was considered as the IL-17 high group (n = 26). Only the first biopsy tissues were considered in the evaluation of the clinical outcome after ATCMR and the long-term allograft survival.
Comparison of clinical indicator according to immunohistochemistry results
Clinical information was collected by retrospective chart review. If the ATCMR episode developed later than 6 months after transplantation, it was defined as late-onset ATCMR. Clinical indicators, such as steroid-resistant ATCMR, incomplete functional recovery after anti-rejection treatment, recurrence of ATCMR within 6 months after a previous ATCMR episode, and allograft survival rate after ATCMR, were compared according to the FOXP3/IL-17 ratio. Steroid-resistant ATCMR was defined when serum creatinine levels did not return to within 20% of baseline within 5 days after the last steroid pulse, and incomplete functional recovery was defined when the anti-rejection treatment did not recover allograft function to within 10% of the baseline value.26 The baseline estimated glomerular filtration rate was calculated from the stable serum creatinine concentration at 2 to 4 weeks before the ATCMR episode by using the modified diet in the renal disease formula.27 Recurrence of ATCMR within 6 months was evaluated in 52 patients after exclusion of four patients who suffered allograft failure immediately after the first ATCMR.
Statistical analysis was performed using spss software version 16·0 (SPSS Inc., Chicago, IL). Data are presented as mean ± SD or counts and percentages, depending on the data type. For continuous variables, means were compared using Student’s t-test. For categorized variables, Pearson’s chi-square test and Fisher’s exact test were used. Allograft survival was analysed by the Kaplan–Meier method with a log-rank test, and it was censored in cases of patient death with a functioning allograft. Cox regression analysis was used for the multivariate analysis to evaluate risk factors for allograft failure. The results were considered significant when the P value was below 0·05.
Comparison of clinical characteristics between the FOXP3 high and the IL-17 high groups
Demographic and pre-transplant baseline characteristics did not differ significantly between the FOXP3 high and the IL-17 high groups (Table 1). However, in the FOXP3 high group, the proportion of patients who took basiliximab as an induction therapy was higher (P =0·03). Interval from transplantation to biopsy was 8·5 ± 14·7 months. Time from transplantation to biopsy and the proportion of late-onset ATCMR (> 6 months from transplant) did not differ significantly between the FOXP3 high and IL-17 high groups (Table 2).
Table 1. Comparison of demographic and clinical characteristics of patients according to the ratio between FOXP3 and interleukin-17 (IL-17) infiltration in allograft tissue with acute T-cell-mediated rejection
Table 2. Comparison of clinical characteristics and histological diagnosis of biopsy according to the ratio between FOXP3+ regulatory T cells (Treg) and interleukin-17 (IL-17) -secreting cell infiltration in allograft tissue with acute T-cell-mediated rejection
Comparison of allograft function between the FOXP3 high and the IL-17 high groups
Calculated estimated glomerular filtration rate at biopsy was significantly decreased in the IL-17 high group compared with the FOXP3 high group (31·4 ± 15·2 ml/min versus 41·6 ± 15·5 ml/min, P =0·04). Serum creatinine at biopsy was higher in the IL-17 group compared with the FOXP3 group, even though it did not reach statistical significance (2·9 ± 1·8 mg/dl versus 2·3 ± 1·3 mg/dl, P =0·08) (Table 2).
Comparison of the severity of tissue injury between the FOXP3 high and the IL-17 high groups
Based on the Banff classification, the distribution of the ATCMR stage did not differ significantly between two groups (P =0·39). However, the development of IF/TA was significantly higher in the IL-17 high group (P =0·04). Among the Banff scores, i score (interstitial inflammation), ci score (interstitial fibrosis), t score (tubulitis), and ct score (tubular atrophy) were significantly higher in the IL-17 high group. In contrast, scores for vascular injury (v, cv) or glomerular injury (g, cg) did not differ significantly between the two groups (Table 2).
Comparison of clinical outcomes after ACR between the FOXP3 high and the IL-17 high groups
The proportion of steroid-resistant ATCMR was significantly higher in the IL-17 high group (P =0·00). In the FOXP3 high group, only 7% (2/30) did not respond to steroid pulse therapy. In contrast, 46% (12/26) were resistant to steroid pulse therapy in the IL-17 group (Fig. 2a). Out of two steroid-resistant ATCMR cases in the FOXP3 high group, one did not recover completely after ATG therapy; hence the overall incomplete recovery rate was 4% (1/30). In the IL-17 high group, eight patients did not recover completely after OKT3 (n = 2) or ATG (n = 10), hence the overall incomplete recovery rate was 31% (P =0·01) (Fig. 2b). Recurrence of ATCMR within 6 months after first ATCMR episode was also more frequent in the IL-17 high group (57% (13/23) versus 28% (8/29), P =0·03) (Fig. 2c).
Comparison of long-term allograft survival between the FOXP3 high and the IL-17 high groups
In the comparison of long-term allograft outcomes after ATCMR episode, the FOXP3 high group was significantly superior to the IL-17 high group (P =0·00). The 1-year and 5-year graft survival rates were 90% and 85%, respectively, in the FOXP3 high group, but they were only 54% and 38%, respectively, in the IL-17 high group (Fig. 2d).
The FOXP3/IL-17 ratio as a prognostic factor for allograft failure
To evaluate whether the FOXP3/IL-17 ratio is a significant prognostic factor for allograft outcome, we performed univariate and multivariate analysis. Univariate analysis revealed that late-onset ATCMR, development of IF/TA, elevated serum creatinine at biopsy, positive C4d, and low Log (FOXP3/IL-17) were significant risk factors for allograft failure. Multivariate analysis using the Cox regression hazard model showed that elevated serum creatinine at biopsy, development of IF/TA, and low Log (FOXP3/IL-17) were independent risk factors for allograft failure (Table 3).
Table 3. Multivariate analysis for allograft outcome after acute T-cell-mediated rejection
The change of the FOXP3/IL-17 ratio in the repeat rejection
Twenty-seven repeat ATCMR developed in 21 patients. The interval between the first rejection and the second rejection was 8·2 ± 10·4 months. Out of them, 15 allograft tissues from 13 patients were available for immunohistochemistry evaluation. We compared the FOXP3/IL-17 ratio, allograft function at biopsy, and the severity of tissue injury between the first rejection and the repeat rejection in those 13 patients. The FOXP3/IL-17 ratio significantly decreased in the repeat rejection compared with the first rejection (Log FOXP3/IL-17, 0·50 ± 0·41 versus 0·12 ± 0·58, P =0·04) (Fig. 3). The severity of interstitial fibrosis (ci score, 0·38 ± 0·50 versus 1·07 ± 0·88, P =0·04) and tubular atrophy (ct score, 0·38 ± 0·51 versus 1·07 ± 0·88, P =0·02) significantly increased in the repeat ATCMR. In contrast, allograft function (serum creatinine, 2·5 ± 1·2 mg/dl versus 2·9 ± 1·8 mg/dl, P = 0·47), the severity of interstitial infiltration (i score, 1·62 ± 0·96 versus 1·92 ± 0·64, P = 0·34) and tubulitis (t score, 1·92 ± 0·76 versus 2·15 ± 0·99, P = 0·50) did not change significantly.
This study was performed to evaluate the role of FOXP3/IL-17 ratio in ATCMR, and the results demonstrate clearly that in allograft tissue, the IL-17-secreting cell dominant state compared with FOXP3+ Treg is associated with more severe allograft dysfunction, and tissue injury. Furthermore, it is an important risk factor for poor clinical outcome with ATCMR. This finding suggests that it could be a useful marker for predicting the prognosis of an allograft after ATCMR.
We evaluated the severity of allograft dysfunction and tissue injury between the FOXP3 high and the IL-17 high groups, and our results showed that more severe allograft dysfunction and tissue injury were observed in the IL-17 high group compared with the FOXP3 high group. In the IL-17 high group, the tissue injury score for acute and chronic inflammation of the interstitial area and tubule was higher than in the FOXP3 high group. This finding suggests that the IL-17-dominant state is associated with both acute and chronic injuries, and previous reports may support this presumption in that acute inflammation induces the IL-17-dominant condition and, in turn hastens chronic changes in the allograft tissue in turn.28
We also evaluated the clinical indicators of ATCMR, which represent poor prognosis (steroid-resistant ATCMR, incomplete recovery, and recurrence of ATCMR) between the FOXP3 high and the IL-17 high groups. The results showed that all indicators in the IL-17 high group were higher than in the FOXP3 high group. The reason for this result is still unclear but we speculate several possibilities. First, renal epithelial cells exposed to IL-17 can produce inflammatory mediators with the potential to stimulate early alloimmune responses.29 Second, IL-17 could rapidly recruit neutrophils, which are observed frequently in biopsies with more severe rejection.30 Third, IL-17 could drive alloimmune responses by promoting lymphoid neogenesis.28 Therefore, it is possible that exposure to relatively higher levels of IL-17 during ATCMR induces stronger alloimmune responses and results in a poor clinical outcome in ATCMR.
As observed, with poor clinical outcome in the IL-17 high group, the FOXP3/IL-17 ratio also affected significantly the long-term allograft survival after ATCMR. The allograft survival rate at 1 year (90% versus 54%) and 5 years (85% versus 38%) in the FOXP3 high group was higher than in the IL-17 high group (P =0·00) (Fig. 2d). Furthermore, multivariate analysis revealed that the FOXP3/IL-17 ratio is a significant prognostic factor independent of other important confounding factors, such as chronic tissue injury and allograft dysfunction. This suggests that the IL-17-dominant state is not secondary to the outcome of allograft dysfunction or chronic tissue injury.
In patients who suffered from multiple episodes of ATCMR, the FOXP3/IL-17 ratio decreased in the repeat ATCMR compared with the first ATCMR in all patients (Fig. 3). In two patients who underwent three biopsies, the FOXP3/IL-17 ratio decreased sequentially (data not shown). Repetition of ATCMR promotes chronic change of allograft tissue, which results in the poor allograft outcome. Therefore, our results suggest that the IL-17-dominant state may involve in the development of chronic change by repeat ATCMR.
We investigated C4d positivity to evaluate whether the FOXP3/IL-17 ratio is associated with humoral immunity. Our results showed that C4d positivity and the coexistence of acute antibody-mediated rejection did not differ significantly between the two groups. In addition, glomerulopathy or vasculopathy, which is associated with humoral immunity, was not different between the two groups.31–33 These findings suggest that the impact of the Th17–Treg axis on humoral immunity is not as strong as its effect on T-cell-mediated immunity.
The results of our study demonstrated that the ratio between Treg and IL-17-secreting cell infiltration in the renal allograft represents the severity of ATCMR. But it is uncertain whether a similar ratio between these two cells is observed in peripheral blood mononuclear cells (PBMCs). In a previous report, significantly higher Treg infiltration in allograft tissue was observed even though its proportion in PBMCs was not elevated.34 It may be because the allograft is a more active site of immune stimulation than PBMCs. Therefore, it is possible that the ratio between Treg and IL-17-secreting cells in PBMCs is different from that in allograft.
Our study has some limitations. First, this study is retrospective and non-randomized. For example, the proportion of basiliximab induction therapy was significantly higher in the FOXP3 high group. However, basiliximab induction was not a significant prognostic factor for allograft outcome in this study. In addition, the FOXP3/IL-17 ratio did not differ significantly between the patients who took basiliximab induction and the patients who did not (data not shown). The above findings suggest that basiliximab induction did not have a significant effect on the development of an IL-17-secreting cell or FOXP3+ Treg dominant condition, and allograft outcome after ATCMR. Second, the microenvironment, which is associated with the IL-17-driven or the FOXP3+ Treg-driven condition, was not assessed. Therefore, randomized controlled trials investigating the inflammatory cytokines associated with IL-17-producing cell development, such as IL-6, IL-21 and tumour necrosis factor-α, may help to understand clearly the underlying mechanisms that drive the IL-17 high or FOXP3 high condition.35
In summary, it is helpful to assess IL-17-secreting cell infiltration combined with FOXP3+ Treg in predicting the clinical outcome after ATCMR. The ratio between FOXP3 and IL-17 was closely associated with allograft function and the severity of tissue injury. Their ratio was also associated with the clinical outcome of ATCMR and long-term allograft survival. Furthermore, their ratio predicts allograft outcome, independent of other risk factors. Therefore, the FOXP3/IL-17 ratio is a good marker for predicting graft survival in patients with ATCMR.
Conflict of interest
This study was supported by a grant (A092258) from the Korea Healthcare Technology R&D Project, Ministry for Health, Welfare & Family Affairs, Republic of Korea.