Messenger RNA for FOXP3 in the Urine of Renal-Allograft Recipients
http://www.100md.com
《新英格兰医药杂志》
ABSTRACT
Background The outcome of renal transplantation after an episode of acute rejection is difficult to predict, even with an allograft biopsy.
Methods We studied urine specimens from 36 subjects with acute rejection, 18 subjects with chronic allograft nephropathy, and 29 subjects with normal biopsy results. Levels of messenger RNA (mRNA) for FOXP3, a specification and functional factor for regulatory T lymphocytes, and mRNA for CD25, CD3, perforin, and 18S ribosomal RNA (rRNA) were measured with a kinetic, quantitative polymerase-chain-reaction assay. We examined associations of mRNA levels with acute rejection, rejection reversal, and graft failure.
Results The log-transformed mean (±SE) ratio of FOXP3 mRNA copies to 18S ribosomal RNA copies was higher in urine from the group with acute rejection (3.8±0.5) than in the group with chronic allograft nephropathy (1.3±0.7) or the group with normal biopsy results (1.6±0.4) (P<0.001 by the Kruskal–Wallis test). FOXP3 mRNA levels were inversely correlated with serum creatinine levels measured at the time of biopsy in the acute-rejection group (Spearman's correlation coefficient = –0.38, P=0.02) but not in the group with chronic allograft nephropathy or the group with normal biopsy results. Analyses of receiver-operating-characteristic curves demonstrated that reversal of acute rejection can be predicted with 90 percent sensitivity and 73 percent specificity with use of the optimal identified cutoff for FOXP3 mRNA of 3.46 (P=0.001). FOXP3 mRNA levels identified subjects at risk for graft failure within six months after the incident episode of acute rejection (relative risk for the lowest third of FOXP3 mRNA levels, 6; P=0.02). None of the other mRNA levels were predictive of reversal of acute rejection or graft failure.
Conclusions Measurement of FOXP3 mRNA in urine may offer a noninvasive means of improving the prediction of outcome of acute rejection of renal transplants.
Kidney transplantation is the treatment of choice for most patients with end-stage renal disease (ESRD), but a shortage of organs limits its availability.1 Renal-allograft failure is the fourth most common cause of ESRD in the United States2 and contributes to the shortage of organs.
Acute rejection is an important risk factor for allograft failure.3,4,5,6 The current approach to treatment of acute rejection is uniform, although it is well recognized that some rejection episodes are not fully reversible and lead to long-term graft dysfunction and failure, whereas others are easily treatable and benign.6 The outcome of acute rejection is difficult to predict, and histologic features that are observed in allograft tissue obtained by core needle biopsy are currently the best predictors.7,8 The invasive procedure of allograft biopsy, however, is associated with complications such as bleeding, arteriovenous fistula, and even graft loss.9,10
We previously reported a method using a quantitative polymerase chain reaction (PCR) to measure messenger RNA (mRNA) levels of immune products within urinary cells of renal-transplant recipients.11,12,13 This noninvasive and nucleic acid–based technique allows for the early diagnosis of acute rejection by detection of increased levels of cytolytic T-cell products such as granzyme B and perforin,11 integrins such as CD103,12 and key chemokine and chemokine-receptor combinations such as gamma-inducible protein of 10kD (IP-10) and its receptor, CXCR3, which promote effector T-cell recruitment to a transplant.13
Recent studies have highlighted the role of a specialized subgroup of CD4+CD25+ T lymphocytes, termed regulatory T lymphocytes (Treg cells), in the suppression of autoimmunity.14,15 Treg cells specifically express the X-linked forkhead/winged helix transcription factor, FOXP3,16 and mutations in the human FOXP3 gene result in an autoimmune disease characterized by polyendocrinopathy and enteropathy that is fatal in infancy.17,18 In a similar manner, male mice with a loss-of-function mutation in the Foxp3 gene (scurfy mice) or with a deficiency of the Foxp3 gene generated by homologous recombination waste away and die within three to four weeks of life with multiorgan lymphocytic infiltrates; transfer of the Foxp3 gene reverses this process.16,19,20
In view of the dominant role of Treg cells in the maintenance of self-tolerance14,15 and in view of the cells' suppressive role in experimental models of transplantation tolerance,21,22,23 we reasoned that measurement of FOXP3 mRNA levels in urinary cells might provide insight into the immunologic events within a renal allograft undergoing acute rejection. We developed a kinetic, quantitative PCR assay and demonstrate that levels of FOXP3 mRNA in urinary cells predict the reversibility of acute rejection and identify patients at high risk for graft loss after an episode of acute rejection.
Methods
Study Cohorts
We examined urine samples from 83 kidney-transplant recipients. In this group were 36 subjects with graft dysfunction (mean creatinine level, 3.6±2.4 mg per deciliter ) and biopsy-confirmed acute rejection (mean age, 41±12 years; 15 men and 21 women; 13 white, 12 black, and 11 with other racial or ethnic backgrounds; with 20 living and 16 deceased donors), 29 subjects with stable allograft function (mean creatinine level, 1.4±0.4 mg per deciliter ) and normal allograft biopsy (mean age, 44±14 years; 15 men and 14 women; 12 white, 4 black, and 13 with other racial or ethnic backgrounds; with 26 living and 3 deceased donors), and 18 subjects with allograft dysfunction (mean creatinine level, 3.1±1.6 mg per deciliter ) and biopsies classified as indicating chronic allograft nephropathy (mean age, 52±12 years; 9 men and 9 women; 9 white, 2 black, and 7 with other racial or ethnic backgrounds; with 5 living and 13 deceased donors).
Seventy-five of the 83 urine specimens were collected before the biopsy procedure, and 8 samples were obtained after the procedure. Formalin-fixed, paraffin-embedded renal-biopsy specimens were stained with hematoxylin and eosin, periodic acid–Schiff, and Masson's trichrome stains and were scored with the use of the Banff 97 classification24 by a pathologist who was blinded to the results of molecular studies. Immunosuppression consisted of a calcineurin inhibitor–based regimen (cyclosporine or tacrolimus), with the administration of glucocorticoids, antilymphocyte antibodies (muromonab-CD3 or antithymocyte globulin), or both for the treatment of acute rejection.11 The study was approved by the institutional review board at the Weill Medical College of Cornell University in New York, and each patient gave written informed consent.
Quantitation of mRNA by Kinetic, Quantitative PCR
Total RNA was isolated from urine-cell pellets, quantified and reverse transcribed to complementary DNA (cDNA).11 We designed and synthesized oligonucleotide primers and fluorogenic probes for the measurement of mRNA levels of FOXP3, CD25, CD3, perforin, and 18S ribosomal RNA (rRNA) (Table 1 in the Supplementary Appendix, available with the full text of this article at www.nejm.org). PCR analysis was performed by a two-step process, a preamplification step followed by measurement of mRNA with an ABI Prism 7700 system (the PCR protocol is provided in the Supplementary Appendix). Transcript levels were calculated by a standard curve method,12 and mRNA copy numbers were normalized with the use of 18S rRNA copy numbers (the number of mRNA copies in 1 μg of RNA divided by the number of 18S rRNA copies in 1 fg of RNA). When no detectable level of a transcript was found, a value equal to half the minimum observed 18S-normalized level was assigned. For an estimation of group means, this method is considered a reasonable substitute for the value of zero or the minimum detected value; moreover, the nonparametric statistical tests of group differences reported below are not affected by the choice of value.25
Statistical Analysis
The levels of mRNA for FOXP3, CD25, CD3, perforin, and 18S rRNA deviated from a normal distribution (P<0.001), but a log transformation substantially reduced the positive skew. We used the 18S-normalized level as the dependent variable in a Kruskal–Wallis test to identify any differences among the group with acute rejection, the group with chronic allograft nephropathy, and the group with normal biopsy results and then used Dunn's test for multiple comparisons. The Mann–Whitney test, equivalent to the Kruskal–Wallis test when applied to two groups, was used when mRNA levels were compared between two groups. Spearman's rank-order correlations were used to test for a monotonic association of the 18S-adjusted mRNA transcript levels with serum creatinine levels and time (in days) from kidney transplantation to biopsy. An episode of acute rejection was classified as reversible if the serum creatinine level returned to within 15 percent of the prerejection level within four weeks after the initiation of antirejection treatment. A second end point was the loss of the graft during the first six months after the diagnosis of acute rejection. We used receiver-operating-characteristic (ROC) curves to analyze mRNA levels in order to determine the cutoff points that yielded the highest combined sensitivity and specificity for predicting the outcome of an episode of acute rejection.
Results
Levels of FOXP3 mRNA in Urinary Cells
The log-transformed mean (±SE) ratio of FOXP3 mRNA copies to 18S-rRNA copies in urinary cells was 3.8±0.5 in the 36 subjects with acute rejection and was higher than the levels in both the 18 subjects with chronic allograft nephropathy (1.3±0.7) and the 29 subjects with normal biopsy results (1.6±0.4, P<0.001 by the Kruskal–Wallis test) (Figure 1A). Among the three groups, the 18S-normalized, log-transformed mRNA levels of CD25 (6.9±0.4, 4.0±0.5, and 2.8±0.6, respectively; P<0.001), CD3 (8.2±0.4, 4.3±0.5, and 1.6±0.5; P<0.001), and perforin (7.6±0.4, 4.5±0.4, and 2.8±0.4; P<0.001) were also highest in the acute-rejection cohort (Figure 1B, Figure 1C, and Figure 1D).
Figure 1. Levels of FOXP3 mRNA in Urinary Cells.
Box plots show the 10th, 25th, 50th (median), 75th, and 90th percentile values for log-transformed ratios of mRNA copies to 18S rRNA copies for FOXP3, CD25, CD3, and perforin in urine samples obtained from 36 subjects with biopsy-confirmed acute rejection, 18 subjects with biopsy-confirmed chronic allograft nephropathy, and 29 subjects with stable graft function and normal biopsy results. The levels of mRNA for FOXP3, CD25, CD3, and perforin were higher in the urinary cells from subjects with acute rejection than in the subjects with chronic allograft nephropathy or normal biopsy results. P values are based on the Kruskal–Wallis test, with the log-transformed mRNA levels treated as the dependent variable. Dunn's multiple-comparison test showed that levels of FOXP3 mRNA in the acute-rejection group were higher than those in both the group with chronic allograft nephropathy (P<0.05) and the group with normal biopsy results (P<0.01) (Panel A). CD25 mRNA levels were higher in the acute-rejection group than in both the group with chronic allograft nephropathy (P<0.001) and the group with normal biopsy results (P<0.001) (Panel B). CD3 mRNA levels were higher in the acute-rejection group than in both the group with chronic allograft nephropathy (P<0.01) and the group with normal biopsy results (P<0.001); CD3 mRNA levels were also higher in the group with chronic allograft nephropathy than in the group with normal biopsy results (P<0.05) (Panel C). Perforin mRNA levels were higher in the acute-rejection group than in both the group with chronic allograft nephropathy (P<0.001) and the group with normal biopsy results (P<0.001) (Panel D). In all cases, log-transformed levels, normalized for 18S rRNA, are shown.
FOXP3 mRNA Levels and Disease Severity
We observed a significant inverse relationship between the levels of FOXP3 mRNA and serum creatinine measured during an episode of acute rejection (Spearman's correlation coefficient =–0.38, P=0.02). By contrast, serum creatinine levels were not significantly related to mRNA levels of CD25 (rs=–0.01, P=0.93), CD3 (rs=–0.11, P=0.54), or perforin (rs=–0.23, P=0.18) in the acute-rejection group. Also, the mean (±SE) serum creatinine level in the 16 subjects with acute rejection of Banff grade IA (moderate tubulitis) did not differ significantly from that of the 20 subjects with grade IB (severe tubulitis) or more (3.3±0.6 mg per deciliter as compared with 3.8±0.6 mg per deciliter , P=0.57).
There was no correlation between the levels of FOXP3 mRNA and serum creatinine that were measured in the group with chronic allograft nephropathy (rs=0.02, P=0.93) or the group with normal biopsy results (rs=–0.08, P=0.67).
FOXP3 mRNA Levels and Reversal of Acute Rejection
Twenty-six of the 36 episodes of acute rejection qualified as successfully reversed; the remaining 10 did not. Levels of FOXP3 mRNA in urinary cells were significantly higher in the group with successful reversal than in the group without reversal (mean level, 4.7±0.5 and 1.5±0.7, respectively; P=0.001) (Figure 2A). In the two groups, the levels of mRNA for CD25 (7.3±0.4 and 6.0±0.9, P=0.22), CD3 (8.5±0.5 and 7.4±0.8, P=0.35), and perforin (7.8±0.5 and 7.3±0.7, P=0.43) were not informative of outcome (Figure 2B, Figure 2C, and Figure 2D).
Figure 2. Correlation between Levels of FOXP3 mRNA in Urinary Cells and Reversal of an Episode of Acute Rejection.
Box plots show the 10th, 25th, 50th (median), 75th, and 90th percentiles for levels of mRNA for FOXP3, CD25, CD3, and perforin in urine samples obtained from 26 subjects with successful reversal of acute rejection (classified as reversible and defined by the return of serum creatinine levels to within 15 percent of prerejection levels within four weeks after the initiation of antirejection treatment) and 10 patients without reversal of acute rejection (nonreversible). The levels of mRNA for FOXP3 but not for CD25, CD3, and perforin were significantly higher in subjects with reversible acute rejection than in subjects with nonreversible acute rejection. Two-tailed P values are based on the Mann–Whitney test. In all cases, log-transformed levels, normalized for 18S rRNA, are shown.
The ROC curves (Figure 3) show the fraction of true positive results (sensitivity) and false positive results (1 – specificity) for various cutoff levels of mRNA for FOXP3, CD25, CD3, and perforin. The log-transformed threshold that gave the maximal sensitivity and specificity for FOXP3 mRNA was 3.46; using the cutoff value of 3.46 derived from the data, the FOXP3 mRNA level predicted rejection reversal with a sensitivity of 90 percent and a specificity of 73 percent (P=0.001) (Figure 3A). The levels of mRNA for CD25, CD3, and perforin were not predictive of reversal of acute rejection (Figure 3B, Figure 3C, and Figure 3D).
Figure 3. Receiver-Operating-Characteristic (ROC) Curves for mRNA Levels.
The fraction of true positive results (sensitivity) and false positive results (1–specificity) for levels of mRNA for FOXP3, CD25, CD3, and perforin, each normalized for 18S rRNA, as predictors of reversal of acute rejection are shown. The calculated area under the curve was 0.85 (95 percent confidence interval, 0.71 to 0.99) for FOXP3 mRNA levels (Panel A) but only 0.63 (95 percent confidence interval, 0.42 to 0.84) for CD25 mRNA levels (Panel B), 0.60 (95 percent confidence interval, 0.39 to 0.81) for CD3 mRNA levels (Panel C), and 0.58 (95 percent confidence interval, 0.38 to 0.79) for perforin mRNA levels (Panel D). A value of 0.5 is no better than that expected by chance (the null hypothesis), and a value of 1.0 reflects a perfect indicator. Of the four mRNA measures, only FOXP3 predicts successful reversal significantly better than chance (P=0.001).
Successful reversal of acute rejection, as compared with unsuccessful reversal, was not predicted by the subjects' age (mean , 41±2.2 years and 40±4.6 years, respectively; P=0.68), sex (10 men and 16 women vs. 5 men and 5 women, P=0.71), race (10 white, 6 black, and 10 with other race or ethnic background vs. 3 white, 6 black, and 1 with other race or ethnic background; P=0.08), graft-donor source (15 living and 11 deceased vs. 5 living and 5 deceased, P=0.68), Banff histologic grade (11 with IA and 15 with >IA vs. 5 with IA and 5 with >IA, P=0.68), or initial antirejection treatment (24 with glucocorticoids and 2 with antilymphocyte antibodies vs. 7 with glucocorticoids and 3 with antilymphocyte antibodies, P=0.12). Among subjects with successful reversal, as compared with those with unsuccessful reversal, serum creatinine levels (median levels, 2.3 mg per deciliter and 6.5 mg per deciliter, respectively; P<0.001) and the time from kidney transplantation to the development of acute rejection (median time, 82 days and 523 days, respectively; P=0.008) were lower. In logistic-regression analyses predicting nonresponse, levels of FOXP3 mRNA in urinary cells remained significant after statistical control for serum creatinine level (P=0.04) and the time from transplantation to rejection (P=0.02).
A linear combination of levels of FOXP3 mRNA and creatinine was a better predictor of rejection reversal (90 percent sensitivity and 96 percent specificity) than FOXP3 mRNA levels alone (90 percent sensitivity and 73 percent specificity) or serum creatinine levels alone (85 percent sensitivity and 90 percent specificity).
FOXP3 mRNA Levels and Allograft Failure
Ten of the 36 subjects with acute rejection lost their grafts within six months after the incident episode of acute rejection, and 9 of those 10 subjects did not respond to the initial antirejection therapy. Renal-allograft recipients with a failed allograft within six months after the episode of acute rejection had significantly lower FOXP3 mRNA levels in their urinary cells than the 26 subjects who had a functioning allograft (2.0±0.8 and 4.5±0.5, respectively; P=0.01). In the two groups, the levels of mRNA for CD25 (6.6±0.7 and 7.1±0.5, P=0.33), CD3 (7.9±0.7 and 8.3±0.5, P=0.76), and perforin (7.8±0.6 and 7.6±0.5, P=0.90) did not predict allograft loss.
The rate of and relative risk of graft failure within six months after an episode of acute rejection, for thirds of each mRNA measure, are shown in Figure 4. At the highest third of FOXP3 mRNA levels, the graft failure rate was 8 percent; at the middle third, the graft failure rate was 25 percent and the relative risk was 3; and at the lowest third, the graft failure rate was 50 percent and the relative risk was 6 (P=0.02 by the chi-square test for linear trend) (Figure 4A). In contrast, the rate of graft failure after an episode of acute rejection did not differ significantly across the thirds of mRNA levels for CD25, CD3, and perforin (Figure 4B, Figure 4C, and Figure 4D).
Figure 4. Relative Risk (RR) of Graft Failure after an Episode of Acute Rejection.
The rates and relative risks of graft failure within six months after an episode of acute rejection for thirds of FOXP3, CD25, CD3, and perforin mRNA levels are shown. The graft-failure rate was 50 percent and the relative risk was 6 for the lowest third of FOXP3 mRNA levels; 25 percent and 3, respectively, for the middle third; and 8 percent for the highest third (P=0.02 by the chi-square test for linear trend) (Panel A). There was no relationship between graft failure after an episode of acute rejection and the thirds of mRNA levels for CD25, CD3, and perforin (Panels B, C, and D).
Graft failure as compared with graft success was not predicted by the subjects' age (mean ages, 39±4.2 years and 42±2.4 years, respectively; P=0.52), sex (4 men and 6 women vs. 11 men and 15 women, P=0.90), race (3 white, 6 black, and 1 with other race or ethnic background vs. 10 white, 6 black, and 10 with other race or ethnic background; P=0.08), graft-donor source (5 living and 5 deceased vs. 15 living and 11 deceased, P=0.68), Banff histologic grade (5 with IA and 5 with >IA vs. 11 with IA and 15 with >IA, P=0.68), or initial antirejection treatment (24 with glucocorticoids and 2 with antilymphocyte antibodies vs. 7 with glucocorticoids and 3 with antilymphocyte antibodies, P=0.12). In subjects with graft failure, as compared with subjects with graft success, serum creatinine levels (median levels, 6.5 mg per deciliter and 2.3 mg per deciliter , respectively; P<0.001) and the time from kidney transplantation to the development of acute rejection (median time, 562 days and 82 days; P=0.003) were significantly greater. In a logistic-regression analysis, FOXP3 mRNA levels became nonsignificant after control for serum creatinine levels (P=0.13) or time between transplantation and rejection (P=0.09).
A linear combination of levels of FOXP3 mRNA and creatinine was a better predictor of graft failure (90 percent sensitivity and 92 percent specificity) than were either FOXP3 mRNA levels alone (80 percent sensitivity and 69 percent specificity) or serum creatinine levels alone (85 percent sensitivity and 90 percent specificity).
FOXP3 mRNA Levels and Time to Acute Rejection
Late acute rejection (acute rejection occurring at least three months after transplantation) results in an outcome that is inferior to that of early acute rejection.26,27 We found a strong inverse relationship between levels of FOXP3 mRNA in urinary cells and the time from kidney transplantation to the development of acute rejection (rs=–0.42, P=0.01) (Figure 5A). Levels of FOXP3 mRNA in urinary cells were lower in 11 urine specimens from patients with late acute rejection than in 25 specimens from patients with early acute rejection (mean level, 2.5±0.6 and 4.7±0.5; P=0.009). CD25 mRNA levels also showed an inverse relation (rs=–0.45, P=0.006) (Figure 5B), and the levels were lower during late rejection than during early acute rejection (5.8±0.8 and 7.4±0.4, P=0.07). There was no inverse relation between the time from kidney transplantation to the development of acute rejection and the mRNA levels of CD3 (rs=–0.26, P=0.12) (Figure 5C) or perforin (rs=–0.02, P=0.91) (Figure 5D). There was also no correlation between the serum creatinine levels at the time of acute rejection and the time from kidney transplantation to the development of acute rejection (rs=0.23, P=0.17).
Figure 5. Levels of mRNA in Urinary Cells and the Time from Kidney Transplantation to the Development of Acute Rejection.
The relationship between levels of mRNA in urine samples and the time from kidney transplantation to the development of biopsy-confirmed acute rejection is shown, along with Spearman's rank-order correlation (rs). A significant inverse relationship between the levels of FOXP3 mRNA and CD25 mRNA in urinary cells and the time from kidney transplantation to the development of acute rejection was found (Panels A and B). There was no such relationship between levels of CD3 mRNA or perforin mRNA in urinary cells and the time from kidney transplantation to the development of acute rejection (Panels C and D). Log-transformed mRNA levels, normalized for 18S rRNA, are shown. The numbers of days from renal transplantation to the development of acute rejection are shown on a log scale.
Discussion
Previous studies have associated acute allograft rejection with cytotoxic T cells and have shown that monitoring the levels of these cells in blood, urine, or both is helpful in the treatment of renal-allograft recipients.11,28,29 Our study suggests that levels of FOXP3 mRNA in urinary cells may serve as a mechanistically informative biomarker of acute-rejection outcome.
Depletion or dysfunction of Treg cells can result in autoimmune disease; tolerance to experimental allografts, on the other hand, is associated with graft infiltration by Treg cells.14,15,16,21,22,23 An attractive hypothesis is that the Treg cells prevent the emergence of effector T cells, and that the absence of disease or tolerance is due to the lack of an immune response. An alternative hypothesis is that Treg cells play a "damage control" role rather than a preventive one.30 We suggest that the host antidonor immune repertoire during acute rejection includes the activation of graft-destructive effector cells as well as graft-protective Treg cells. We observed that levels of FOXP3 mRNA in urinary cells, a defining functional marker of Treg cells, and mRNA for perforin, a cytotoxic effector, are both expressed in a heightened fashion during acute rejection and that high levels of FOXP3 mRNA are associated with reversible acute rejection and a lower risk of graft failure. These findings are consistent with the hypothesis that Treg cells serve to limit antiallograft immunity and that the lack of counterregulation by Treg cells during an episode of acute rejection results in unrestrained effector-cell activity, impaired allograft function, and even graft failure.
Histologic analysis of renal allografts is considered to be the best predictor of acute rejection.7,8 However, it has long been recognized that cellular interstitial infiltration is not invariably associated with allograft dysfunction or failure.31 In the current study, Banff grades of acute rejection did not predict rejection outcome, and the serum creatinine levels did not vary across Banff grades. We suggest that graft-infiltrating cells comprise both graft-destructive cells such as cytotoxic T cells and graft-protective FOXP3-expressing Treg cells and that graft dysfunction and response to therapy may be predicted more accurately when the heterogeneous nature of the cellular components is better resolved.
Elevated levels of serum creatinine are an established risk factor for renal allograft failure32 and in our studies were shown to be a strong predictor of the outcome of acute rejection. Do levels of FOXP3 mRNA in urinary cells provide information above and beyond that provided by serum creatinine levels? In our study, levels of serum creatinine and FOXP3 mRNA in urinary cells were independent predictors of reversal of acute rejection. Moreover, rejection reversal and graft loss were predicted with a higher degree of accuracy with the use of levels of both FOXP3 mRNA and creatinine than with either one alone. However, since our estimates of sensitivity and specificity were calculated from the same sample that was used to select the cutoff points, the estimates are upwardly biased and need to be reevaluated in an independent sample.
In both adult and pediatric recipients of renal allografts, an episode of late acute rejection is associated with a lower rate of graft survival than an episode of early acute rejection.26,27 Our finding of an inverse relationship between levels of FOXP3 mRNA in urinary cells and the time to acute rejection suggests a cellular mechanism for the hitherto unexplained poor outcome associated with late acute rejection.
A mechanistic hypothesis engendered by our study is that drugs that enhance the generation of Treg cells, or the administration of Treg cells themselves, may improve the outcome of acute rejection. Cyclosporine and tacrolimus both inhibit the production of interleukin-2, an essential growth factor for Treg cells,33 but induce the production of transforming growth factor 1,34,35 an inducer of FOXP3 and a promoter of the development of CD4+CD25+ Treg cells.36,37 Sirolimus (rapamycin) has been shown to promote the expansion of murine Treg cells in vitro,38 and glucocorticoids have been reported to increase the expression of FOXP3 mRNA in human CD4+ cells.39 However, the in vivo effects of drugs on the induction, expansion, and function of Treg cells in allograft recipients remain to be fully characterized.
How might FOXP3-expressing Treg cells exert their salutary activity during an episode of acute rejection? Treg cells have been shown to dampen or suppress local host immune responses by acting on antigen-presenting cells, directly modulating effector-cell functions, or both.14,15 Mechanisms of immunosuppression by FOXP3-expressing Treg cells include direct cell contact, cytokine signaling, and inhibition of transcription of genes central to effector functions.40 The role of these mechanisms in mitigating the acute-rejection response remains undetermined.
In sum, our study suggests that levels of FOXP3 mRNA in urinary cells may serve as a mechanistically informative biomarker of acute-rejection outcome, with lower levels associated with irreversible acute rejection and even graft failure. In addition to suggesting a robust cellular mechanism for the clinically important differences in the outcome of acute-rejection episodes, the strategy we present here may ultimately lead to individualized treatment of renal-allograft recipients and inform antirejection therapy, including the consideration of infusion of Treg cells to treat acute rejection of allografts.
Supported in part by awards (AI51652 and AI60706, to Dr. Suthanthiran, and AI54720, to Dr. Hancock) from the National Institutes of Health, from the International Society of Nephrology (to Dr. Naqvi), and from the American Society of Transplantation (to Dr. Lee).
A U.S. patent entitled "Methods of Evaluating Transplant Rejection" (6187534) was issued on February 13, 2001; Dr. Suthanthiran is one of the inventors. The patent is owned jointly by Harvard Medical School, Cornell University, and the Beth Israel–Deaconess Medical Center, Boston.
We are indebted to Dr. Phyllis August for her careful review of this article and to Linda Stackhouse for her assistance in the preparation of the manuscript.
Source Information
From the Division of Nephrology, Departments of Medicine (T.M., D.D., R.D., C.S., R.N., J.B.L., C.H., B.L., V.K.S., M.S.), Pathology (S.V.S.), and Surgery (S.K.), Weill Medical College of Cornell University; the Department of Transplantation Medicine, New York Presbyterian Hospital–Weill Cornell Medical Center (D.D., C.H., S.K., M.S.); and the Rogosin Institute (D.D., C.H., V.K.S.) — all in New York; the Department of Pathology and Laboratory Medicine, Children's Hospital of Philadelphia and the University of Pennsylvania School of Medicine, Philadelphia (W.W.H.); and the Department of Psychiatry, State University of New York at Stony Brook, Stony Brook (J.E.S.).
Address reprint requests to Dr. Suthanthiran at the Division of Nephrology and Department of Transplantation Medicine, 525 E. 68th St., Box 3, New York, NY 10021, or at msuthan@med.cornell.edu.
References
Langone AJ, Helderman JH. Disparity between solid-organ supply and demand. N Engl J Med 2003;349:704-706.
Agodoa L, Eknoyan G, Ingelfinger J, et al. Assessment of structure and function in progressive renal disease. Kidney Int 1997;63:Suppl:S144-S150.
Cheigh JS, Saal SD, Suthanthiran M, et al. Natural history of cadaveric kidney transplants in the absence of early acute rejection. Nephron 1983;35:6-10.
Gulanikar AC, MacDonald AS, Sungurtekin U, Belitsky P. The incidence and impact of early rejection episodes on graft outcome in recipients of first cadaver kidney transplants. Transplantation 1992;53:323-328.
Hariharan S, Johnson CP, Bresnahan BA, Taranto SE, McIntosh MJ, Stablein D. Improved graft survival after renal transplantation in the United States, 1988 to 1996. N Engl J Med 2000;342:605-612.
Opelz G. Critical evaluation of the association of acute with chronic graft rejection in kidney and heart transplant recipients: the Collaborative Transplant Study. Transplant Proc 1997;29:73-76.
Matas AJ, Sibley R, Mauer M, Sutherland DE, Simmons RL, Najarian JS. The value of needle renal allograft biopsy. I. A retrospective study of biopsies performed during putative rejection episodes. Ann Surg 1983;197:226-237.
Hayry P. Pathology makes a difference. Transplantation 2000;69:1038-1040.
Huraib S, Goldberg H, Katz A, et al. Percutaneous needle biopsy of the transplanted kidney: technique and complications. Am J Kidney Dis 1989;14:13-17.
Beckingham IJ, Nicholson ML, Bell PR. Analysis of factors associated with complications following renal transplant needle core biopsy. Br J Urol 1994;73:13-15.
Li B, Hartono C, Ding RC, et al. Noninvasive diagnosis of renal-allograft rejection by measurement of messenger RNA for perforin and granzyme B in urine. N Engl J Med 2001;344:947-954.
Ding R, Li B, Muthukumar T, et al. CD103 mRNA levels in urinary cells predict acute rejection of renal allografts. Transplantation 2003;75:1307-1312.
Tatapudi RR, Muthukumar T, Dadhania D, et al. Noninvasive detection of renal allograft inflammation by measurements of mRNA for IP-10 and CXCR3 in urine. Kidney Int 2004;65:2390-2397.
Sakaguchi S. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in immunological tolerance to self and non-self. Nat Immunol 2005;6:345-352.
Schwartz RH. Natural regulatory T cells and self-tolerance. Nat Immunol 2005;6:327-330.
Fontenot JD, Gavin MA, Rudensky AY. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 2003;4:330-336.
Wildin RS, Ramsdell F, Peake J, et al. X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nat Genet 2001;27:18-20.
Bennett CL, Christie J, Ramsdell F, et al. The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat Genet 2001;27:20-21.
Brunkow ME, Jeffery EW, Hjerrild KA, et al. Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat Genet 2001;27:68-73.
Khattri R, Cox T, Yasayko SA, Ramsdell F. An essential role for Scurfin in CD4+CD25+ T regulatory cells. Nat Immunol 2003;4:337-342.
Graca L, Cobbold SP, Waldmann H. Identification of regulatory T cells in tolerated allografts. J Exp Med 2002;195:1641-1646.
Zheng XX, Sanchez-Fueyo A, Sho M, Domenig C, Sayegh MH, Strom TB. Favorably tipping the balance between cytopathic and regulatory T cells to create transplantation tolerance. Immunity 2003;19:503-514.
Lee I, Wang L, Wells AD, Dorf ME, Ozkaynak E, Hancock WW. Recruitment of Foxp3+ T regulatory cells mediating allograft tolerance depends upon the CCR4 chemokine receptor. J Exp Med 2005;201:1037-1044.
Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999;55:713-723.
Helsel DR. Less than obvious -- statistical treatment of data below the detection limit. Environ Sci Technol 1990;24:1766-1774.
Tejani AH, Stablein DM, Sullivan EK, et al. The impact of donor source, recipient age, pre-operative immunotherapy and induction therapy on early and late acute rejections in children: a report of the North American Pediatric Renal Transplant Cooperative Study (NAPRTCS). Pediatr Transplant 1998;2:318-324.
Sijpkens YW, Doxiadis H, Mallat MJ, et al. Early versus late acute rejection episodes in renal transplantation. Transplantation 2003;75:204-208.
Strom TB, Tilney NL, Carpenter CB, Busch GJ. Identity and cytotoxic capacity of cells infiltrating renal allografts. N Engl J Med 1975;292:1257-1263.
Vasconcellos LM, Schachter AD, Zheng XX, et al. Cytotoxic lymphocyte gene expression in peripheral blood leukocytes correlates with rejecting renal allografts. Transplantation 1998;66:562-566.
Gonzalez A, Andre-Schmutz I, Carnaud C, Mathis D, Benoist C. Damage control, rather than unresponsiveness, effected by protective DX5+ T cells in autoimmune diabetes. Nat Immunol 2001;2:1117-1125.
Burdick JF, Berschorner WE, Smith WJ, et al. Characteristics of early routine renal allograft biopsies. Transplantation 1984;38:679-684.
Siddiqi N, McBride MA, Hariharan S. Similar risk profiles for post-transplant renal dysfunction and long-term graft failure: UNOS/OPTN database analysis. Kidney Int 2004;65:1906-1913.
Almeida AR, Legrand N, Papiernik M, Freitas AA. Homeostasis of peripheral CD4+ T cells: IL-2R and IL-2 shape a population of regulatory cells that controls CD4+ T cell numbers. J Immunol 2002;169:4850-4860.
Li B, Sehajpal PK, Khanna A, et al. Differential regulation of transforming growth factor beta and interleukin 2 genes in human T cells: demonstration by usage of novel competitor DNA constructs in the quantitative polymerase chain reaction. J Exp Med 1991;174:1259-1262.
Maluccio M, Sharma V, Lagman M, et al. Tacrolimus enhances transforming growth factor-beta1 expression and tumor progression in a dose-dependent fashion. Transplantation 2003;76:597-602.
Chen W, Jin W, Hardegen N, et al. Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J Exp Med 2003;198:1875-1886.
Luo X, Yang H, Kim IS, et al. Systemic transforming growth factor-beta1 gene therapy induces Foxp3+ regulatory cells, restores self-tolerance, and facilitates regeneration of beta cell function in overtly diabetic nonobese diabetic mice. Transplantation 2005;79:1091-1096.
Battaglia M, Stabilini A, Roncarolo MG. Rapamycin selectively expands CD4+ CD25+FoxP3+ regulatory T cells. Blood 2005;105:4743-4748.
Karagiannidis C, Akdis M, Holopainen P, et al. Glucocorticoids upregulate FOXP3 expression and regulatory T cells in asthma. J Allergy Clin Immunol 2004;114:1425-1433.
von Boehmer H. Mechanisms of suppression by suppressor T cells. Nat Immunol 2005;6:338-344.(Thangamani Muthukumar, M.)
Background The outcome of renal transplantation after an episode of acute rejection is difficult to predict, even with an allograft biopsy.
Methods We studied urine specimens from 36 subjects with acute rejection, 18 subjects with chronic allograft nephropathy, and 29 subjects with normal biopsy results. Levels of messenger RNA (mRNA) for FOXP3, a specification and functional factor for regulatory T lymphocytes, and mRNA for CD25, CD3, perforin, and 18S ribosomal RNA (rRNA) were measured with a kinetic, quantitative polymerase-chain-reaction assay. We examined associations of mRNA levels with acute rejection, rejection reversal, and graft failure.
Results The log-transformed mean (±SE) ratio of FOXP3 mRNA copies to 18S ribosomal RNA copies was higher in urine from the group with acute rejection (3.8±0.5) than in the group with chronic allograft nephropathy (1.3±0.7) or the group with normal biopsy results (1.6±0.4) (P<0.001 by the Kruskal–Wallis test). FOXP3 mRNA levels were inversely correlated with serum creatinine levels measured at the time of biopsy in the acute-rejection group (Spearman's correlation coefficient = –0.38, P=0.02) but not in the group with chronic allograft nephropathy or the group with normal biopsy results. Analyses of receiver-operating-characteristic curves demonstrated that reversal of acute rejection can be predicted with 90 percent sensitivity and 73 percent specificity with use of the optimal identified cutoff for FOXP3 mRNA of 3.46 (P=0.001). FOXP3 mRNA levels identified subjects at risk for graft failure within six months after the incident episode of acute rejection (relative risk for the lowest third of FOXP3 mRNA levels, 6; P=0.02). None of the other mRNA levels were predictive of reversal of acute rejection or graft failure.
Conclusions Measurement of FOXP3 mRNA in urine may offer a noninvasive means of improving the prediction of outcome of acute rejection of renal transplants.
Kidney transplantation is the treatment of choice for most patients with end-stage renal disease (ESRD), but a shortage of organs limits its availability.1 Renal-allograft failure is the fourth most common cause of ESRD in the United States2 and contributes to the shortage of organs.
Acute rejection is an important risk factor for allograft failure.3,4,5,6 The current approach to treatment of acute rejection is uniform, although it is well recognized that some rejection episodes are not fully reversible and lead to long-term graft dysfunction and failure, whereas others are easily treatable and benign.6 The outcome of acute rejection is difficult to predict, and histologic features that are observed in allograft tissue obtained by core needle biopsy are currently the best predictors.7,8 The invasive procedure of allograft biopsy, however, is associated with complications such as bleeding, arteriovenous fistula, and even graft loss.9,10
We previously reported a method using a quantitative polymerase chain reaction (PCR) to measure messenger RNA (mRNA) levels of immune products within urinary cells of renal-transplant recipients.11,12,13 This noninvasive and nucleic acid–based technique allows for the early diagnosis of acute rejection by detection of increased levels of cytolytic T-cell products such as granzyme B and perforin,11 integrins such as CD103,12 and key chemokine and chemokine-receptor combinations such as gamma-inducible protein of 10kD (IP-10) and its receptor, CXCR3, which promote effector T-cell recruitment to a transplant.13
Recent studies have highlighted the role of a specialized subgroup of CD4+CD25+ T lymphocytes, termed regulatory T lymphocytes (Treg cells), in the suppression of autoimmunity.14,15 Treg cells specifically express the X-linked forkhead/winged helix transcription factor, FOXP3,16 and mutations in the human FOXP3 gene result in an autoimmune disease characterized by polyendocrinopathy and enteropathy that is fatal in infancy.17,18 In a similar manner, male mice with a loss-of-function mutation in the Foxp3 gene (scurfy mice) or with a deficiency of the Foxp3 gene generated by homologous recombination waste away and die within three to four weeks of life with multiorgan lymphocytic infiltrates; transfer of the Foxp3 gene reverses this process.16,19,20
In view of the dominant role of Treg cells in the maintenance of self-tolerance14,15 and in view of the cells' suppressive role in experimental models of transplantation tolerance,21,22,23 we reasoned that measurement of FOXP3 mRNA levels in urinary cells might provide insight into the immunologic events within a renal allograft undergoing acute rejection. We developed a kinetic, quantitative PCR assay and demonstrate that levels of FOXP3 mRNA in urinary cells predict the reversibility of acute rejection and identify patients at high risk for graft loss after an episode of acute rejection.
Methods
Study Cohorts
We examined urine samples from 83 kidney-transplant recipients. In this group were 36 subjects with graft dysfunction (mean creatinine level, 3.6±2.4 mg per deciliter ) and biopsy-confirmed acute rejection (mean age, 41±12 years; 15 men and 21 women; 13 white, 12 black, and 11 with other racial or ethnic backgrounds; with 20 living and 16 deceased donors), 29 subjects with stable allograft function (mean creatinine level, 1.4±0.4 mg per deciliter ) and normal allograft biopsy (mean age, 44±14 years; 15 men and 14 women; 12 white, 4 black, and 13 with other racial or ethnic backgrounds; with 26 living and 3 deceased donors), and 18 subjects with allograft dysfunction (mean creatinine level, 3.1±1.6 mg per deciliter ) and biopsies classified as indicating chronic allograft nephropathy (mean age, 52±12 years; 9 men and 9 women; 9 white, 2 black, and 7 with other racial or ethnic backgrounds; with 5 living and 13 deceased donors).
Seventy-five of the 83 urine specimens were collected before the biopsy procedure, and 8 samples were obtained after the procedure. Formalin-fixed, paraffin-embedded renal-biopsy specimens were stained with hematoxylin and eosin, periodic acid–Schiff, and Masson's trichrome stains and were scored with the use of the Banff 97 classification24 by a pathologist who was blinded to the results of molecular studies. Immunosuppression consisted of a calcineurin inhibitor–based regimen (cyclosporine or tacrolimus), with the administration of glucocorticoids, antilymphocyte antibodies (muromonab-CD3 or antithymocyte globulin), or both for the treatment of acute rejection.11 The study was approved by the institutional review board at the Weill Medical College of Cornell University in New York, and each patient gave written informed consent.
Quantitation of mRNA by Kinetic, Quantitative PCR
Total RNA was isolated from urine-cell pellets, quantified and reverse transcribed to complementary DNA (cDNA).11 We designed and synthesized oligonucleotide primers and fluorogenic probes for the measurement of mRNA levels of FOXP3, CD25, CD3, perforin, and 18S ribosomal RNA (rRNA) (Table 1 in the Supplementary Appendix, available with the full text of this article at www.nejm.org). PCR analysis was performed by a two-step process, a preamplification step followed by measurement of mRNA with an ABI Prism 7700 system (the PCR protocol is provided in the Supplementary Appendix). Transcript levels were calculated by a standard curve method,12 and mRNA copy numbers were normalized with the use of 18S rRNA copy numbers (the number of mRNA copies in 1 μg of RNA divided by the number of 18S rRNA copies in 1 fg of RNA). When no detectable level of a transcript was found, a value equal to half the minimum observed 18S-normalized level was assigned. For an estimation of group means, this method is considered a reasonable substitute for the value of zero or the minimum detected value; moreover, the nonparametric statistical tests of group differences reported below are not affected by the choice of value.25
Statistical Analysis
The levels of mRNA for FOXP3, CD25, CD3, perforin, and 18S rRNA deviated from a normal distribution (P<0.001), but a log transformation substantially reduced the positive skew. We used the 18S-normalized level as the dependent variable in a Kruskal–Wallis test to identify any differences among the group with acute rejection, the group with chronic allograft nephropathy, and the group with normal biopsy results and then used Dunn's test for multiple comparisons. The Mann–Whitney test, equivalent to the Kruskal–Wallis test when applied to two groups, was used when mRNA levels were compared between two groups. Spearman's rank-order correlations were used to test for a monotonic association of the 18S-adjusted mRNA transcript levels with serum creatinine levels and time (in days) from kidney transplantation to biopsy. An episode of acute rejection was classified as reversible if the serum creatinine level returned to within 15 percent of the prerejection level within four weeks after the initiation of antirejection treatment. A second end point was the loss of the graft during the first six months after the diagnosis of acute rejection. We used receiver-operating-characteristic (ROC) curves to analyze mRNA levels in order to determine the cutoff points that yielded the highest combined sensitivity and specificity for predicting the outcome of an episode of acute rejection.
Results
Levels of FOXP3 mRNA in Urinary Cells
The log-transformed mean (±SE) ratio of FOXP3 mRNA copies to 18S-rRNA copies in urinary cells was 3.8±0.5 in the 36 subjects with acute rejection and was higher than the levels in both the 18 subjects with chronic allograft nephropathy (1.3±0.7) and the 29 subjects with normal biopsy results (1.6±0.4, P<0.001 by the Kruskal–Wallis test) (Figure 1A). Among the three groups, the 18S-normalized, log-transformed mRNA levels of CD25 (6.9±0.4, 4.0±0.5, and 2.8±0.6, respectively; P<0.001), CD3 (8.2±0.4, 4.3±0.5, and 1.6±0.5; P<0.001), and perforin (7.6±0.4, 4.5±0.4, and 2.8±0.4; P<0.001) were also highest in the acute-rejection cohort (Figure 1B, Figure 1C, and Figure 1D).
Figure 1. Levels of FOXP3 mRNA in Urinary Cells.
Box plots show the 10th, 25th, 50th (median), 75th, and 90th percentile values for log-transformed ratios of mRNA copies to 18S rRNA copies for FOXP3, CD25, CD3, and perforin in urine samples obtained from 36 subjects with biopsy-confirmed acute rejection, 18 subjects with biopsy-confirmed chronic allograft nephropathy, and 29 subjects with stable graft function and normal biopsy results. The levels of mRNA for FOXP3, CD25, CD3, and perforin were higher in the urinary cells from subjects with acute rejection than in the subjects with chronic allograft nephropathy or normal biopsy results. P values are based on the Kruskal–Wallis test, with the log-transformed mRNA levels treated as the dependent variable. Dunn's multiple-comparison test showed that levels of FOXP3 mRNA in the acute-rejection group were higher than those in both the group with chronic allograft nephropathy (P<0.05) and the group with normal biopsy results (P<0.01) (Panel A). CD25 mRNA levels were higher in the acute-rejection group than in both the group with chronic allograft nephropathy (P<0.001) and the group with normal biopsy results (P<0.001) (Panel B). CD3 mRNA levels were higher in the acute-rejection group than in both the group with chronic allograft nephropathy (P<0.01) and the group with normal biopsy results (P<0.001); CD3 mRNA levels were also higher in the group with chronic allograft nephropathy than in the group with normal biopsy results (P<0.05) (Panel C). Perforin mRNA levels were higher in the acute-rejection group than in both the group with chronic allograft nephropathy (P<0.001) and the group with normal biopsy results (P<0.001) (Panel D). In all cases, log-transformed levels, normalized for 18S rRNA, are shown.
FOXP3 mRNA Levels and Disease Severity
We observed a significant inverse relationship between the levels of FOXP3 mRNA and serum creatinine measured during an episode of acute rejection (Spearman's correlation coefficient =–0.38, P=0.02). By contrast, serum creatinine levels were not significantly related to mRNA levels of CD25 (rs=–0.01, P=0.93), CD3 (rs=–0.11, P=0.54), or perforin (rs=–0.23, P=0.18) in the acute-rejection group. Also, the mean (±SE) serum creatinine level in the 16 subjects with acute rejection of Banff grade IA (moderate tubulitis) did not differ significantly from that of the 20 subjects with grade IB (severe tubulitis) or more (3.3±0.6 mg per deciliter as compared with 3.8±0.6 mg per deciliter , P=0.57).
There was no correlation between the levels of FOXP3 mRNA and serum creatinine that were measured in the group with chronic allograft nephropathy (rs=0.02, P=0.93) or the group with normal biopsy results (rs=–0.08, P=0.67).
FOXP3 mRNA Levels and Reversal of Acute Rejection
Twenty-six of the 36 episodes of acute rejection qualified as successfully reversed; the remaining 10 did not. Levels of FOXP3 mRNA in urinary cells were significantly higher in the group with successful reversal than in the group without reversal (mean level, 4.7±0.5 and 1.5±0.7, respectively; P=0.001) (Figure 2A). In the two groups, the levels of mRNA for CD25 (7.3±0.4 and 6.0±0.9, P=0.22), CD3 (8.5±0.5 and 7.4±0.8, P=0.35), and perforin (7.8±0.5 and 7.3±0.7, P=0.43) were not informative of outcome (Figure 2B, Figure 2C, and Figure 2D).
Figure 2. Correlation between Levels of FOXP3 mRNA in Urinary Cells and Reversal of an Episode of Acute Rejection.
Box plots show the 10th, 25th, 50th (median), 75th, and 90th percentiles for levels of mRNA for FOXP3, CD25, CD3, and perforin in urine samples obtained from 26 subjects with successful reversal of acute rejection (classified as reversible and defined by the return of serum creatinine levels to within 15 percent of prerejection levels within four weeks after the initiation of antirejection treatment) and 10 patients without reversal of acute rejection (nonreversible). The levels of mRNA for FOXP3 but not for CD25, CD3, and perforin were significantly higher in subjects with reversible acute rejection than in subjects with nonreversible acute rejection. Two-tailed P values are based on the Mann–Whitney test. In all cases, log-transformed levels, normalized for 18S rRNA, are shown.
The ROC curves (Figure 3) show the fraction of true positive results (sensitivity) and false positive results (1 – specificity) for various cutoff levels of mRNA for FOXP3, CD25, CD3, and perforin. The log-transformed threshold that gave the maximal sensitivity and specificity for FOXP3 mRNA was 3.46; using the cutoff value of 3.46 derived from the data, the FOXP3 mRNA level predicted rejection reversal with a sensitivity of 90 percent and a specificity of 73 percent (P=0.001) (Figure 3A). The levels of mRNA for CD25, CD3, and perforin were not predictive of reversal of acute rejection (Figure 3B, Figure 3C, and Figure 3D).
Figure 3. Receiver-Operating-Characteristic (ROC) Curves for mRNA Levels.
The fraction of true positive results (sensitivity) and false positive results (1–specificity) for levels of mRNA for FOXP3, CD25, CD3, and perforin, each normalized for 18S rRNA, as predictors of reversal of acute rejection are shown. The calculated area under the curve was 0.85 (95 percent confidence interval, 0.71 to 0.99) for FOXP3 mRNA levels (Panel A) but only 0.63 (95 percent confidence interval, 0.42 to 0.84) for CD25 mRNA levels (Panel B), 0.60 (95 percent confidence interval, 0.39 to 0.81) for CD3 mRNA levels (Panel C), and 0.58 (95 percent confidence interval, 0.38 to 0.79) for perforin mRNA levels (Panel D). A value of 0.5 is no better than that expected by chance (the null hypothesis), and a value of 1.0 reflects a perfect indicator. Of the four mRNA measures, only FOXP3 predicts successful reversal significantly better than chance (P=0.001).
Successful reversal of acute rejection, as compared with unsuccessful reversal, was not predicted by the subjects' age (mean , 41±2.2 years and 40±4.6 years, respectively; P=0.68), sex (10 men and 16 women vs. 5 men and 5 women, P=0.71), race (10 white, 6 black, and 10 with other race or ethnic background vs. 3 white, 6 black, and 1 with other race or ethnic background; P=0.08), graft-donor source (15 living and 11 deceased vs. 5 living and 5 deceased, P=0.68), Banff histologic grade (11 with IA and 15 with >IA vs. 5 with IA and 5 with >IA, P=0.68), or initial antirejection treatment (24 with glucocorticoids and 2 with antilymphocyte antibodies vs. 7 with glucocorticoids and 3 with antilymphocyte antibodies, P=0.12). Among subjects with successful reversal, as compared with those with unsuccessful reversal, serum creatinine levels (median levels, 2.3 mg per deciliter and 6.5 mg per deciliter, respectively; P<0.001) and the time from kidney transplantation to the development of acute rejection (median time, 82 days and 523 days, respectively; P=0.008) were lower. In logistic-regression analyses predicting nonresponse, levels of FOXP3 mRNA in urinary cells remained significant after statistical control for serum creatinine level (P=0.04) and the time from transplantation to rejection (P=0.02).
A linear combination of levels of FOXP3 mRNA and creatinine was a better predictor of rejection reversal (90 percent sensitivity and 96 percent specificity) than FOXP3 mRNA levels alone (90 percent sensitivity and 73 percent specificity) or serum creatinine levels alone (85 percent sensitivity and 90 percent specificity).
FOXP3 mRNA Levels and Allograft Failure
Ten of the 36 subjects with acute rejection lost their grafts within six months after the incident episode of acute rejection, and 9 of those 10 subjects did not respond to the initial antirejection therapy. Renal-allograft recipients with a failed allograft within six months after the episode of acute rejection had significantly lower FOXP3 mRNA levels in their urinary cells than the 26 subjects who had a functioning allograft (2.0±0.8 and 4.5±0.5, respectively; P=0.01). In the two groups, the levels of mRNA for CD25 (6.6±0.7 and 7.1±0.5, P=0.33), CD3 (7.9±0.7 and 8.3±0.5, P=0.76), and perforin (7.8±0.6 and 7.6±0.5, P=0.90) did not predict allograft loss.
The rate of and relative risk of graft failure within six months after an episode of acute rejection, for thirds of each mRNA measure, are shown in Figure 4. At the highest third of FOXP3 mRNA levels, the graft failure rate was 8 percent; at the middle third, the graft failure rate was 25 percent and the relative risk was 3; and at the lowest third, the graft failure rate was 50 percent and the relative risk was 6 (P=0.02 by the chi-square test for linear trend) (Figure 4A). In contrast, the rate of graft failure after an episode of acute rejection did not differ significantly across the thirds of mRNA levels for CD25, CD3, and perforin (Figure 4B, Figure 4C, and Figure 4D).
Figure 4. Relative Risk (RR) of Graft Failure after an Episode of Acute Rejection.
The rates and relative risks of graft failure within six months after an episode of acute rejection for thirds of FOXP3, CD25, CD3, and perforin mRNA levels are shown. The graft-failure rate was 50 percent and the relative risk was 6 for the lowest third of FOXP3 mRNA levels; 25 percent and 3, respectively, for the middle third; and 8 percent for the highest third (P=0.02 by the chi-square test for linear trend) (Panel A). There was no relationship between graft failure after an episode of acute rejection and the thirds of mRNA levels for CD25, CD3, and perforin (Panels B, C, and D).
Graft failure as compared with graft success was not predicted by the subjects' age (mean ages, 39±4.2 years and 42±2.4 years, respectively; P=0.52), sex (4 men and 6 women vs. 11 men and 15 women, P=0.90), race (3 white, 6 black, and 1 with other race or ethnic background vs. 10 white, 6 black, and 10 with other race or ethnic background; P=0.08), graft-donor source (5 living and 5 deceased vs. 15 living and 11 deceased, P=0.68), Banff histologic grade (5 with IA and 5 with >IA vs. 11 with IA and 15 with >IA, P=0.68), or initial antirejection treatment (24 with glucocorticoids and 2 with antilymphocyte antibodies vs. 7 with glucocorticoids and 3 with antilymphocyte antibodies, P=0.12). In subjects with graft failure, as compared with subjects with graft success, serum creatinine levels (median levels, 6.5 mg per deciliter and 2.3 mg per deciliter , respectively; P<0.001) and the time from kidney transplantation to the development of acute rejection (median time, 562 days and 82 days; P=0.003) were significantly greater. In a logistic-regression analysis, FOXP3 mRNA levels became nonsignificant after control for serum creatinine levels (P=0.13) or time between transplantation and rejection (P=0.09).
A linear combination of levels of FOXP3 mRNA and creatinine was a better predictor of graft failure (90 percent sensitivity and 92 percent specificity) than were either FOXP3 mRNA levels alone (80 percent sensitivity and 69 percent specificity) or serum creatinine levels alone (85 percent sensitivity and 90 percent specificity).
FOXP3 mRNA Levels and Time to Acute Rejection
Late acute rejection (acute rejection occurring at least three months after transplantation) results in an outcome that is inferior to that of early acute rejection.26,27 We found a strong inverse relationship between levels of FOXP3 mRNA in urinary cells and the time from kidney transplantation to the development of acute rejection (rs=–0.42, P=0.01) (Figure 5A). Levels of FOXP3 mRNA in urinary cells were lower in 11 urine specimens from patients with late acute rejection than in 25 specimens from patients with early acute rejection (mean level, 2.5±0.6 and 4.7±0.5; P=0.009). CD25 mRNA levels also showed an inverse relation (rs=–0.45, P=0.006) (Figure 5B), and the levels were lower during late rejection than during early acute rejection (5.8±0.8 and 7.4±0.4, P=0.07). There was no inverse relation between the time from kidney transplantation to the development of acute rejection and the mRNA levels of CD3 (rs=–0.26, P=0.12) (Figure 5C) or perforin (rs=–0.02, P=0.91) (Figure 5D). There was also no correlation between the serum creatinine levels at the time of acute rejection and the time from kidney transplantation to the development of acute rejection (rs=0.23, P=0.17).
Figure 5. Levels of mRNA in Urinary Cells and the Time from Kidney Transplantation to the Development of Acute Rejection.
The relationship between levels of mRNA in urine samples and the time from kidney transplantation to the development of biopsy-confirmed acute rejection is shown, along with Spearman's rank-order correlation (rs). A significant inverse relationship between the levels of FOXP3 mRNA and CD25 mRNA in urinary cells and the time from kidney transplantation to the development of acute rejection was found (Panels A and B). There was no such relationship between levels of CD3 mRNA or perforin mRNA in urinary cells and the time from kidney transplantation to the development of acute rejection (Panels C and D). Log-transformed mRNA levels, normalized for 18S rRNA, are shown. The numbers of days from renal transplantation to the development of acute rejection are shown on a log scale.
Discussion
Previous studies have associated acute allograft rejection with cytotoxic T cells and have shown that monitoring the levels of these cells in blood, urine, or both is helpful in the treatment of renal-allograft recipients.11,28,29 Our study suggests that levels of FOXP3 mRNA in urinary cells may serve as a mechanistically informative biomarker of acute-rejection outcome.
Depletion or dysfunction of Treg cells can result in autoimmune disease; tolerance to experimental allografts, on the other hand, is associated with graft infiltration by Treg cells.14,15,16,21,22,23 An attractive hypothesis is that the Treg cells prevent the emergence of effector T cells, and that the absence of disease or tolerance is due to the lack of an immune response. An alternative hypothesis is that Treg cells play a "damage control" role rather than a preventive one.30 We suggest that the host antidonor immune repertoire during acute rejection includes the activation of graft-destructive effector cells as well as graft-protective Treg cells. We observed that levels of FOXP3 mRNA in urinary cells, a defining functional marker of Treg cells, and mRNA for perforin, a cytotoxic effector, are both expressed in a heightened fashion during acute rejection and that high levels of FOXP3 mRNA are associated with reversible acute rejection and a lower risk of graft failure. These findings are consistent with the hypothesis that Treg cells serve to limit antiallograft immunity and that the lack of counterregulation by Treg cells during an episode of acute rejection results in unrestrained effector-cell activity, impaired allograft function, and even graft failure.
Histologic analysis of renal allografts is considered to be the best predictor of acute rejection.7,8 However, it has long been recognized that cellular interstitial infiltration is not invariably associated with allograft dysfunction or failure.31 In the current study, Banff grades of acute rejection did not predict rejection outcome, and the serum creatinine levels did not vary across Banff grades. We suggest that graft-infiltrating cells comprise both graft-destructive cells such as cytotoxic T cells and graft-protective FOXP3-expressing Treg cells and that graft dysfunction and response to therapy may be predicted more accurately when the heterogeneous nature of the cellular components is better resolved.
Elevated levels of serum creatinine are an established risk factor for renal allograft failure32 and in our studies were shown to be a strong predictor of the outcome of acute rejection. Do levels of FOXP3 mRNA in urinary cells provide information above and beyond that provided by serum creatinine levels? In our study, levels of serum creatinine and FOXP3 mRNA in urinary cells were independent predictors of reversal of acute rejection. Moreover, rejection reversal and graft loss were predicted with a higher degree of accuracy with the use of levels of both FOXP3 mRNA and creatinine than with either one alone. However, since our estimates of sensitivity and specificity were calculated from the same sample that was used to select the cutoff points, the estimates are upwardly biased and need to be reevaluated in an independent sample.
In both adult and pediatric recipients of renal allografts, an episode of late acute rejection is associated with a lower rate of graft survival than an episode of early acute rejection.26,27 Our finding of an inverse relationship between levels of FOXP3 mRNA in urinary cells and the time to acute rejection suggests a cellular mechanism for the hitherto unexplained poor outcome associated with late acute rejection.
A mechanistic hypothesis engendered by our study is that drugs that enhance the generation of Treg cells, or the administration of Treg cells themselves, may improve the outcome of acute rejection. Cyclosporine and tacrolimus both inhibit the production of interleukin-2, an essential growth factor for Treg cells,33 but induce the production of transforming growth factor 1,34,35 an inducer of FOXP3 and a promoter of the development of CD4+CD25+ Treg cells.36,37 Sirolimus (rapamycin) has been shown to promote the expansion of murine Treg cells in vitro,38 and glucocorticoids have been reported to increase the expression of FOXP3 mRNA in human CD4+ cells.39 However, the in vivo effects of drugs on the induction, expansion, and function of Treg cells in allograft recipients remain to be fully characterized.
How might FOXP3-expressing Treg cells exert their salutary activity during an episode of acute rejection? Treg cells have been shown to dampen or suppress local host immune responses by acting on antigen-presenting cells, directly modulating effector-cell functions, or both.14,15 Mechanisms of immunosuppression by FOXP3-expressing Treg cells include direct cell contact, cytokine signaling, and inhibition of transcription of genes central to effector functions.40 The role of these mechanisms in mitigating the acute-rejection response remains undetermined.
In sum, our study suggests that levels of FOXP3 mRNA in urinary cells may serve as a mechanistically informative biomarker of acute-rejection outcome, with lower levels associated with irreversible acute rejection and even graft failure. In addition to suggesting a robust cellular mechanism for the clinically important differences in the outcome of acute-rejection episodes, the strategy we present here may ultimately lead to individualized treatment of renal-allograft recipients and inform antirejection therapy, including the consideration of infusion of Treg cells to treat acute rejection of allografts.
Supported in part by awards (AI51652 and AI60706, to Dr. Suthanthiran, and AI54720, to Dr. Hancock) from the National Institutes of Health, from the International Society of Nephrology (to Dr. Naqvi), and from the American Society of Transplantation (to Dr. Lee).
A U.S. patent entitled "Methods of Evaluating Transplant Rejection" (6187534) was issued on February 13, 2001; Dr. Suthanthiran is one of the inventors. The patent is owned jointly by Harvard Medical School, Cornell University, and the Beth Israel–Deaconess Medical Center, Boston.
We are indebted to Dr. Phyllis August for her careful review of this article and to Linda Stackhouse for her assistance in the preparation of the manuscript.
Source Information
From the Division of Nephrology, Departments of Medicine (T.M., D.D., R.D., C.S., R.N., J.B.L., C.H., B.L., V.K.S., M.S.), Pathology (S.V.S.), and Surgery (S.K.), Weill Medical College of Cornell University; the Department of Transplantation Medicine, New York Presbyterian Hospital–Weill Cornell Medical Center (D.D., C.H., S.K., M.S.); and the Rogosin Institute (D.D., C.H., V.K.S.) — all in New York; the Department of Pathology and Laboratory Medicine, Children's Hospital of Philadelphia and the University of Pennsylvania School of Medicine, Philadelphia (W.W.H.); and the Department of Psychiatry, State University of New York at Stony Brook, Stony Brook (J.E.S.).
Address reprint requests to Dr. Suthanthiran at the Division of Nephrology and Department of Transplantation Medicine, 525 E. 68th St., Box 3, New York, NY 10021, or at msuthan@med.cornell.edu.
References
Langone AJ, Helderman JH. Disparity between solid-organ supply and demand. N Engl J Med 2003;349:704-706.
Agodoa L, Eknoyan G, Ingelfinger J, et al. Assessment of structure and function in progressive renal disease. Kidney Int 1997;63:Suppl:S144-S150.
Cheigh JS, Saal SD, Suthanthiran M, et al. Natural history of cadaveric kidney transplants in the absence of early acute rejection. Nephron 1983;35:6-10.
Gulanikar AC, MacDonald AS, Sungurtekin U, Belitsky P. The incidence and impact of early rejection episodes on graft outcome in recipients of first cadaver kidney transplants. Transplantation 1992;53:323-328.
Hariharan S, Johnson CP, Bresnahan BA, Taranto SE, McIntosh MJ, Stablein D. Improved graft survival after renal transplantation in the United States, 1988 to 1996. N Engl J Med 2000;342:605-612.
Opelz G. Critical evaluation of the association of acute with chronic graft rejection in kidney and heart transplant recipients: the Collaborative Transplant Study. Transplant Proc 1997;29:73-76.
Matas AJ, Sibley R, Mauer M, Sutherland DE, Simmons RL, Najarian JS. The value of needle renal allograft biopsy. I. A retrospective study of biopsies performed during putative rejection episodes. Ann Surg 1983;197:226-237.
Hayry P. Pathology makes a difference. Transplantation 2000;69:1038-1040.
Huraib S, Goldberg H, Katz A, et al. Percutaneous needle biopsy of the transplanted kidney: technique and complications. Am J Kidney Dis 1989;14:13-17.
Beckingham IJ, Nicholson ML, Bell PR. Analysis of factors associated with complications following renal transplant needle core biopsy. Br J Urol 1994;73:13-15.
Li B, Hartono C, Ding RC, et al. Noninvasive diagnosis of renal-allograft rejection by measurement of messenger RNA for perforin and granzyme B in urine. N Engl J Med 2001;344:947-954.
Ding R, Li B, Muthukumar T, et al. CD103 mRNA levels in urinary cells predict acute rejection of renal allografts. Transplantation 2003;75:1307-1312.
Tatapudi RR, Muthukumar T, Dadhania D, et al. Noninvasive detection of renal allograft inflammation by measurements of mRNA for IP-10 and CXCR3 in urine. Kidney Int 2004;65:2390-2397.
Sakaguchi S. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in immunological tolerance to self and non-self. Nat Immunol 2005;6:345-352.
Schwartz RH. Natural regulatory T cells and self-tolerance. Nat Immunol 2005;6:327-330.
Fontenot JD, Gavin MA, Rudensky AY. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 2003;4:330-336.
Wildin RS, Ramsdell F, Peake J, et al. X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nat Genet 2001;27:18-20.
Bennett CL, Christie J, Ramsdell F, et al. The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat Genet 2001;27:20-21.
Brunkow ME, Jeffery EW, Hjerrild KA, et al. Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat Genet 2001;27:68-73.
Khattri R, Cox T, Yasayko SA, Ramsdell F. An essential role for Scurfin in CD4+CD25+ T regulatory cells. Nat Immunol 2003;4:337-342.
Graca L, Cobbold SP, Waldmann H. Identification of regulatory T cells in tolerated allografts. J Exp Med 2002;195:1641-1646.
Zheng XX, Sanchez-Fueyo A, Sho M, Domenig C, Sayegh MH, Strom TB. Favorably tipping the balance between cytopathic and regulatory T cells to create transplantation tolerance. Immunity 2003;19:503-514.
Lee I, Wang L, Wells AD, Dorf ME, Ozkaynak E, Hancock WW. Recruitment of Foxp3+ T regulatory cells mediating allograft tolerance depends upon the CCR4 chemokine receptor. J Exp Med 2005;201:1037-1044.
Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999;55:713-723.
Helsel DR. Less than obvious -- statistical treatment of data below the detection limit. Environ Sci Technol 1990;24:1766-1774.
Tejani AH, Stablein DM, Sullivan EK, et al. The impact of donor source, recipient age, pre-operative immunotherapy and induction therapy on early and late acute rejections in children: a report of the North American Pediatric Renal Transplant Cooperative Study (NAPRTCS). Pediatr Transplant 1998;2:318-324.
Sijpkens YW, Doxiadis H, Mallat MJ, et al. Early versus late acute rejection episodes in renal transplantation. Transplantation 2003;75:204-208.
Strom TB, Tilney NL, Carpenter CB, Busch GJ. Identity and cytotoxic capacity of cells infiltrating renal allografts. N Engl J Med 1975;292:1257-1263.
Vasconcellos LM, Schachter AD, Zheng XX, et al. Cytotoxic lymphocyte gene expression in peripheral blood leukocytes correlates with rejecting renal allografts. Transplantation 1998;66:562-566.
Gonzalez A, Andre-Schmutz I, Carnaud C, Mathis D, Benoist C. Damage control, rather than unresponsiveness, effected by protective DX5+ T cells in autoimmune diabetes. Nat Immunol 2001;2:1117-1125.
Burdick JF, Berschorner WE, Smith WJ, et al. Characteristics of early routine renal allograft biopsies. Transplantation 1984;38:679-684.
Siddiqi N, McBride MA, Hariharan S. Similar risk profiles for post-transplant renal dysfunction and long-term graft failure: UNOS/OPTN database analysis. Kidney Int 2004;65:1906-1913.
Almeida AR, Legrand N, Papiernik M, Freitas AA. Homeostasis of peripheral CD4+ T cells: IL-2R and IL-2 shape a population of regulatory cells that controls CD4+ T cell numbers. J Immunol 2002;169:4850-4860.
Li B, Sehajpal PK, Khanna A, et al. Differential regulation of transforming growth factor beta and interleukin 2 genes in human T cells: demonstration by usage of novel competitor DNA constructs in the quantitative polymerase chain reaction. J Exp Med 1991;174:1259-1262.
Maluccio M, Sharma V, Lagman M, et al. Tacrolimus enhances transforming growth factor-beta1 expression and tumor progression in a dose-dependent fashion. Transplantation 2003;76:597-602.
Chen W, Jin W, Hardegen N, et al. Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J Exp Med 2003;198:1875-1886.
Luo X, Yang H, Kim IS, et al. Systemic transforming growth factor-beta1 gene therapy induces Foxp3+ regulatory cells, restores self-tolerance, and facilitates regeneration of beta cell function in overtly diabetic nonobese diabetic mice. Transplantation 2005;79:1091-1096.
Battaglia M, Stabilini A, Roncarolo MG. Rapamycin selectively expands CD4+ CD25+FoxP3+ regulatory T cells. Blood 2005;105:4743-4748.
Karagiannidis C, Akdis M, Holopainen P, et al. Glucocorticoids upregulate FOXP3 expression and regulatory T cells in asthma. J Allergy Clin Immunol 2004;114:1425-1433.
von Boehmer H. Mechanisms of suppression by suppressor T cells. Nat Immunol 2005;6:338-344.(Thangamani Muthukumar, M.)