Secondary Logo

Journal Logo

Noninvasive Tim-3 Messenger RNA Evaluation in Renal Transplant Recipients With Graft Dysfunction

Manfro, Roberto C.1,2,3; Aquino-Dias, Esther C.2; Joelsons, Gabriel2; Nogare, Aline L.2; Carpio, Virna N.2; Gonçalves, Luiz Felipe S.1,2

doi: 10.1097/TP.0b013e3181914246
Original Articles: Immunobiology and Genomics

Background. Renal biopsies are usually needed to elucidate graft dysfunction. In this study, T-cell immunoglobulin domain, mucin domain mRNA expression in the peripheral blood leukocytes (PBL) and urinary cells (UC) were studied as a noninvasive method for the diagnosis of acute rejection (AR) of kidney transplant patients with dysfunction.

Methods. One hundred sixty biopsies were obtained from 115 patients. Blood and urine samples were collected immediately before the biopsies. Histopathologic diagnoses were acute tubular necrosis with superimposed AR or acute tubular necrosis in patients with delayed graft function (DGF), and (AR), or calcineurin inhibitor nephrotoxicity (CIN), or interstitial fibrosis and tubular atrophy in patients with acute graft dysfunction (AGD). Fifteen protocol biopsies of stable grafts were used as controls. mRNA relative quantification was performed by real-time polymerase chain reaction.

Results. Gene expression in tissue, PBL, and UC was always higher in patients with AR than in patients with the other causes of graft dysfunction (P<0.001). Significant correlations of gene expression in different compartments were observed (P<0.001). The obtained diagnostic parameters were 100% accurate in the DGF group and, respectively, for blood and urine: sensitivity (87% and 84%); specificity (95% and 96%); positive predictive value (87% and 89%); negative predictive value (93% and 94%); and accuracy (91% and 93%) for the group of patients with AGD.

Conclusion. T-cell immunoglobulin domain, mucin domain mRNA quantification by real-time polymerase chain reaction in PBL and UC of renal transplant patients undergoing DGF or AGD may become a useful tool for an accurate noninvasive diagnosis of AR.

1 Department of Internal Medicine, School of Medicine, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil.

2 Division of Nephrology, Hospital de Clínicas de Porto Alegre, Porto Alegre, RS, Brazil.

This work was supported by Comissão de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Comissão Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Ministery of Education, Brazil and Research Incentive Fund from Hospital de Clínicas de Porto Alegre.

The authors declare no conflict of interests.

3 Address correspondence to: Roberto C. Manfro, M.D., Ph.D., Division of Nephrology, Renal Transplant Unit, Hospital de Clínicas de Porto Alegre, 2350 Ramiro Barcelos Street, Room 2030, Porto Alegre, RS, Brazil 90035-003.


Received 8 August 2008. Revision requested 9 September 2008.

Accepted 13 October 2008.

The practice of kidney transplantation has presented significant progress over the last decade. Patient and graft survival had achieved excellent short-term rates, and the use of more efficient immunosuppressive drugs has significantly decreased the incidence of acute rejection (AR). However, episodes of acute graft dysfunction (AGD) are still common and can be due to a variety of causes including a considerable incidence of AR that is associated with lower graft survival (1). Also, AR is even more frequent in patients with delayed graft function (DGF) leading to additional detrimental effects on patient and graft survival (2–6). The elucidation of graft dysfunction episodes many times requires a biopsy. Furthermore, during the course of DGF, in the absence of a functional parameter, the diagnosis of AR is based on the histologic analysis of graft samples obtained by surveillance biopsies. Graft biopsy, although considered a routine procedure, is uncomfortable and expensive, with a low incidence of complications that can, however, be significant. Also, it is subjected to sampling error, mainly due to the focal nature of the rejection process (7, 8).

Transcriptional profiling initially applied to graft tissue, latter provided the opportunity to the development of noninvasive diagnostic tools by analyzing in the peripheral blood and in urinary cells (UC) mRNA from genes associated with the immune response to the graft tissue (9–16). A variety of genes evaluated as noninvasive diagnostic markers of AR also led to a better understanding of the clinical and sub clinical AR process (17, 18).

T cell immunoglobulin domain, mucin domain (Tim-3), a type 1 membrane protein with a extracellular domain consisting of a immunoglobulin variable region-like domain and a mucin-like region was recently described as molecule selectively expressed on the surface of terminally differentiated T-helper (Th)1 cells and that seems to act in the transport or effector functions of these cells or both (19).

In this study, we tested the hypothesis by measuring Tim-3 mRNA in urinary or peripheral blood cells or both, it is possible to diagnose AR in kidney grafts with AGD episodes as well as in grafts with DGF.

Back to Top | Article Outline


Patients and Sample Collection

Kidney transplant recipients with DGF episodes or with AGD who underwent surveillance or per cause graft biopsies were included in this study. Before the biopsy procedure, vascular, urological, nephrotoxic, or infectious causes of graft dysfunction were ruled out by Doppler ultrasound, nuclear scans, immunosuppressive drug levels, and urine and blood cultures. Acute dysfunction episodes were suspected in the presence of a significantly confirmed increment of the serum creatinine level. Patients with DGF, defined by the need for dialysis during the first week after transplantation, were submitted to surveillance biopsies every 7 to 10 days until kidney function was recovered or graft was lost. Immediately before each biopsy, peripheral blood and sterile urine samples were collected from patients with diuresis. The kidney biopsies were performed through ultrasound guidance, using a semi-automatic gun with 16-G needle. One and a half fragments were used for histologic analyses and half fragment was immediately frozen in liquid nitrogen and maintained at −70°C. Right after the collection, the blood and UC were isolated and frozen until the RNA extraction. Slide evaluation was performed by a renal pathologist unaware of the clinical data. The Banff ’97 classification was used for the histopathologic diagnoses (20).

Seventy-nine samples were obtained from 50 patients with DGF who were subsequently classified, according to the histologic findings, as having pure acute tubular necrosis (ATN; n=38) or ATN with superimposed AR (ATN-AR; n=41). Sixty-six samples from 50 patients with AGD, who based on the biopsy results, were subsequently classified as having AR (n=24), calcineurin inhibitor nephrotoxicity (CIN; n=13) or interstitial fibrosis, and tubular atrophy (IFTA; n=29). As a control for gene expression, we also examined mRNA expression in graft tissue, peripheral blood monocyte cells, and UC from 15 kidney transplant recipients with stable graft function who had protocol biopsies, interpreted as normal (NOR), within the first year posttransplantation (NOR; n=15).

All patients received corticosteroids combined with cyclosporine or tacrolimus and mycophenolate (mofetil or sodium) as immunosuppressive therapy. Interleukin-2 antireceptor antibodies (Basiliximab) were given within 24 hr of the transplantation to most patients with DGF. Antilymphocytic antibodies (Thymoglobulin) were given preoperatively to all patients considered as having elevated immunologic risk.

The study was approved by the Research Ethics Committee of the Hospital de Clínicas de Porto Alegre, accredited by the National Research Council of Brazilian Department of Health, and registered at the Office for Human Research Protections-OHRP-USDHHS (Institutional Review Board-IRB 00000921). Patients gave written informed consent to the procedures and research.

Back to Top | Article Outline

RNA Extraction

Graft fragments were defrosted, macerated, and processed for the RNA extraction, using QIAamp RNA Blood mini kit (QIAGEN Inc., Chatsworth, CA) method, according to the manufacturer’s instructions. Peripheral blood was collected in tubes containing EDTA and the cells were separated with an erythrocyte lysis buffer. UC were obtained by centrifugation at 500g for 20 min. The floating material was discarded and cell pellets were resuspended in phosphate- buffered saline solution; centrifuged again for 10 min and stored at −70°C. RNA was extracted from cell pellets isolated from peripheral blood and urine sediment cells by the same method as for the renal tissue. RNA quality was assessed by evaluating the optical density 260-to-280 ratio, and only RNA samples with optical density ratio higher than 1.7 were analyzed, provided that a sufficient amount was available.

Back to Top | Article Outline

Messenger RNA Quantification

The expression of each gene was analyzed, using the relative quantification technique, by real-time polymerase chain reaction (RT-PCR) using the ABI Prism 7000 (Applied Biosystems, Foster City, CA) detection system. Gene expression assay consists of a mixture of primers and TaqMan MGB (minor groove binding) probes at 360 μM 20× concentrated. Primers were designed based on the identification of target sequences at the gene bank. The sequences used had already been designed, tested, and validated previously by the manufacturer (Applied Biosystems. Gene expression assays/custom primers and probes). Fluorescent dies used as markers of the probes were 6-carboxy fluorescein (FAM) as reporter (at 5′) and 6- carboxytetramethyl rodamine (TAMRA) as quencher (at 3′). Tim-3 gene (ID: Hs 00262170_m1; gene bank reference 84868) was analyzed along with endogenous molecular controls, used for sample normalization of the mRNA amounts in each reaction. Β-actin (PN. 4310881E) was used as an endogenous molecular control, used for sample normalization, and for obtaining an equivalent mRNA amount in each reaction (TaqMan PDAR Endogenous Control).

The reactions were made in duplicates using the TaqMan EZ RT-PCR (PN. N808-0235) (Applied Biosystems) kit, according to the following protocol: 5.0 μL of 5× TaqMan EZ buffer, 3.0 μL of manganese acetate (25 mM), 0.75 μL of dATP (10 mM), 0.75 μL of dCTP (10 mM), 0.75 μL of dGTP (10 mM), 0.75 μL of dUTP (20 mM), 1.0 μL of rTth DNA polymerase (2.5 U/μL), 0.25 of AmpErase UNG (1 U/μL), and pure water to reach a volume of 23 μL. To this mixture, 1 μL of primers and probes (20×) and 1 μL of RNA were added to each reaction, to a final volume of 25 μL. The cycling program consisted of heating at 50°C for 2 min, 60°C for 30 min followed by heating to 95°C for 5 min and 40 cycles using the temperatures of 94°C for 20 sec and 62°C for 60 sec.

The analyses of amplified products were performed by the relative quantification method 2−ΔΔCT, which describes alterations to the target gene expression relative to a reference sample (21).

Back to Top | Article Outline

Statistical Analyses

Data are presented as absolute numbers, mean±standard deviations or percentages. Gene quantifications are shown as box-plot representations. Continuous variables with normal distribution were evaluated using analyses of variances followed by the Dunnett’s test. The mRNA levels were analyzed using Kruskal-Wallis test with all diagnostic groups. Tukey’s test was used for multiple comparisons among the various groups. Mann-Whitney’s test was used for comparisons between two groups. Fisher’s exact test was used for the analyses of categorical variables. The correlations between the mRNA levels of expression in the different compartments were calculated using the Spearman’s correlation test. Receiver operating characteristic curves were generated to find the best cutoff points for the diagnoses of AR (22). All analyses were performed using the SPSS (Statistical Package for the Social Sciences) program (version 14.0, Chicago, IL). The statistical significance level was established as P less than 0.05.

Back to Top | Article Outline


Table 1 shows the main demographic data of the groups, the number of biopsy samples per group within each histopathologic diagnosis, and relevant transplant information. Seventy percent of the patients received grafts from deceased donors, 60% were male recipients, and 7% received second grafts. Statistical analysis was performed within the major groups. No differences were found in the demographics of the group of patients with DGF. In the group of patients with AGD, differences were found in the time interval to obtain a graft biopsy and serum creatinine, both significantly lower in the group of patients with stable graft function (P<0.05), details are shown in Table 1.



In the group of patients with DGF 79 surveillance biopsies were obtained. Forty-one had histopathologic features of ATN with superimposed AR and 38 had features of pure ATN. Serum creatinine levels did not differentiate ATN-AR from ATN (5.6±2.8 vs. 5.9±1.8 mg/dL, respectively; P=NS). Both in the graft, peripheral blood and urinary sediment cells the expression levels of Tim-3 mRNA were significantly higher in the group of patients with ATN-AR, as shown in the box-plot graphics in Figure 1 (A–C). The differences of expression reached a significant statistical difference (P<0.001). In this group, the diagnostic parameters to the diagnosis of AR were 100% accurate (Table 2). No significant differences in the levels of expression were found in patients with different antibody induction therapy regimens, ATG or Basiliximab, or between them and patients without induction.





In the group of patients with AGD 66 biopsies were obtained. Twenty-four had histopathologic features of AR, 13 of CIN, and 29 of IFTA. Fifteen protocol biopsies interpreted as NOR were used as comparators. Serum creatinine levels were not statistically different between the groups of patients with dysfunction but were significantly lower in the group of patients with NOR biopsies (P<0.05). In the graft tissue, peripheral blood and UC, compared with the CIN, IFTA, and NOR groups, Tim-3 mRNA expression was much higher in the group of patients with AR and the differences among the means reached highly significant statistical differences, as shown in Figure 1 (D–F) (P<0.001). The diagnostic parameters to the diagnosis of AR are shown in Table 2.

Table 3 shows Spearman’s correlation coefficients between Tim-3 mRNA quantifications in the different compartments. Strong and significant correlations were observed between the quantifications in tissue, and blood and urine cells.



In seven patients who had a histopathologic diagnosis of AR and were treated for rejection, a second set of samples was collected. After treatment Tim-3 levels of expression decreased significantly in the graft tissue (P=0.001), in the peripheral blood (P=0.003), and in the UC (P=0.029).

Back to Top | Article Outline


Tim-3 molecule ligates to galectin-9 and this pathway seems to be involved in the effective termination of effector TH1 cells (23). In experimental transplantation, the Tim-3 pathway has been suggested to be crucial to the development of peripheral tolerance because its blockage prevents the acquisition of tolerance possibly by dampening the immunosuppressive action of CD4(+)CD25(+) regulatory T-cell populations (24, 25). In the clinical setting, Tim-3 mRNA was demonstrated in increased amounts in renal biopsies and in UC of rejecting grafts and was considered a good marker for clinical AR episodes in renal transplant recipients (26, 27). These characteristics make Tim-3, a good candidate for a marker of immune mediated allograft injury in the clinical setting and worth testing as a molecular noninvasive test in the peripheral blood mononuclear cells and urinary sediment cells.

Graft dysfunction episodes are frequent in renal transplantation. In this study, we divided it into two settings. First, we evaluated kidney grafts with DGF, where we found that the quantitative analysis of Tim-3 mRNA was extremely accurate in differentiating pure ATN from ATN with superimposed AR. Transcriptional analyses of grafts undergoing DFG has not been described in the literature except for a few reports in a restricted number of patients. By using the competitive PCR method, Li et al., observed that, in 9 of 11 patients with DGF, the perforin and granzyme B expression levels were significantly higher in patients with AR (15). Yannaraki et al. (28) using the RT-PCR technique, evaluated five patients with DGF and reported that the detected amounts of perforin, granzyme B, and fas-ligand were also increased in patients with AR. Later, Renesto et al., analyzing the Tim-3 mRNA expression, pooled patients with DGF in a group along with another causes of dysfunction and also found higher expression in patients with AR (27). Finally, Aquino-Dias et al. recently reported that quantitative mRNA analyses of different genes, highlighting FOXP3, could be a useful diagnostic tool in this clinical situation in which a noninvasive method of adequate accuracy would be a cost-effective aid to clinical management (29).

Except for the latter work, gene expression analyses in patients with DGF have only been made in UC, which may represent a limitation because many patients might not produce urine for analyses or the urine might come from the native kidneys. In our sample, we were not able to collect urine for analysis for approximately 25% of the occasions. Despite, it was found that when urine is available for mRNA quantitative analyses accurate diagnostic parameters are achievable. However, the possible lack of urine reinforces the need for analyzing gene expression in the peripheral blood, which in this study produced diagnostic parameters identical to the urine analyses.

The authors also analyzed Tim-3 mRNA expression in episodes of graft dysfunction that included AR, CIN, and IFTA. We also found that Tim-3 mRNA expression was an accurate method for the diagnosis of the immune-mediated graft injury present during AR episodes. In the peripheral blood cells and in the UC, the expression levels were much higher during AR and correlated with values obtained from graft tissue analyses.

The evaluation of mRNA amounts from peripheral blood leukocytes of kidney transplants patients with AGD was initially performed by Vasconcellos et al. (12) in studies in which mRNA was quantified by reverse transcription competitive PCR, who demonstrated that these evaluations could produce proper parameters to the diagnosis of AR.

The mRNA analyses of urine-sediment cells of kidney transplant patients are interesting because lymphocytes present in urine, under most circumstances, have trafficked through the kidney, so the status of these cells might accurately reflect the status of graft-infiltrating lymphocytes. Gene profiling in the urine have been performed by Muthukumar et al. who evaluated FOXP3 and perforin mRNA in the urine of kidney transplant recipients and found that the levels of these genes were significantly higher in patients with AR, as compared with groups with histologic diagnoses of chronic graft nephropathy and transplants with stable function (16). Also, Renesto et al. (27) have demonstrated increased amounts of Tim-3 mRNA in UC of rejecting grafts and considered the quantification of this gene, a good marker of AR in renal transplant recipients.

We acknowledge that in addition to AR other conditions, mainly acute CIN and pyelonephritis, might occur during DGF and play an important role in delaying functional graft recovery. However, at least regarding to CIN, the data obtained in patients with AGD suggest that increased expression would not occur.

The findings of this study allowed the conclusion that in patients with DGF or AGD quantification of Tim-3 mRNA, in peripheral blood leukocytes or in urine-sediment cells provides an accurate marker for the presence of AR. As previously suggested the validation of these results in multicenter longitudinal study and possible sophistications of used techniques might allow transcriptional profiling to replace or reduce the need for kidney biopsies for evaluation of graft dysfunction and during DGF, and evaluate its applicability to anticipate the diagnosis of AR (30).

Back to Top | Article Outline


1. Opelz G, Döhler B; Collaborative Transplant Study Report. Influence of time of rejection on long-term graft survival in renal transplantation. Transplantation 2008; 85: 661.
2. Mikhalski D, Wissing KM, Ghisdal L, et al. Cold ischemia is a major determinant of acute rejection and renal graft survival in the modern era of immunosuppression. Transplantation 2008; 85(7 suppl): S3.
3. Quiroga I, McShane P, Koo DD, et al. Major effects of delayed graft function and cold ischaemia time on renal allograft survival. Nephrol Dial Transplant 2006; 21: 1689.
4. Perico N, Cattaneo D, Sayegh MH, et al. Delayed graft function in kidney transplantation. Lancet 2004; 364: 1814.
5. Szwarc I, Garrigue V, Delmas S, et al. [Delayed graft function: A frequent but still unsolved problem in renal transplantation]. Nephrol Ther 2005; 1: 325.
6. Johnston O, O’Kelly P, Spencer S, et al. Reduced graft function (with or without dialysis) vs. immediate graft function: A comparison of long-term renal allograft survival. Nephrol Dial Transplant 2006; 21: 2270.
7. Sorof JM, Vartanian RK, Olson JL, et al. Histopathological concordance of paired renal allograft biopsy cores. Effect on the diagnosis and management of acute rejection. Transplantation 1995; 60: 1215.
8. Colvin RB, Cohen AH, Saiontz C, et al. Evaluation of pathologic criteria for acute renal allograft rejection: Reproducibility, sensitivity, and clinical correlation. J Am Soc Nephrol 1997; 8: 1930.
9. Lipman ML, Stevens AC, Strom TB. Heightened intragraft CTL gene expression in acutely rejecting renal allografts. J Immunol 1994; 152: 5120.
10. Sharma VK, Bologa RM, Li B, et al. Molecular executors of cell death—Differential intrarenal expression of Fas ligand, Fas, granzyme B, and perforin during acute and/or chronic rejection of human renal allografts. Transplantation 1996; 62: 1860.
11. Strehlau J, Pavlakis M, Lipman MW, et al. Quantitative detection of immune activation transcripts as a diagnostic tool in kidney transplantation. Proc Natl Acad Sci USA 1997; 94: 695.
12. 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.
13. Simon T, Opelz G, Wiesel M, et al. Serial peripheral blood perforin and granzyme B gene expression measurements for prediction of acute rejection in kidney graft recipients. Am J Transplant 2003; 3: 1121.
14. Li B, Hartono C, Ding R, 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.
15. Kotsch K, Mashreghi MF, Bold G, et al. Enhanced granulysin mRNA expression in urinary sediment in early and delayed acute renal allograft rejection. Transplantation 2004; 77: 1866.
16. Muthukumar T, Dadhania D, Ding R, et al. Messenger RNA for FOXP3 in the urine of renal-allograft recipients. N Engl J Med 2005; 353: 2342.
17. Lipman ML, Shen Y, Jeffery JR, et al. Immune-activation gene expression in clinically stable renal allograft biopsies: Molecular evidence for subclinical rejection. Transplantation 1998; 66: 1673.
18. Aquino-Dias EC, Veronese FJ, Santos Gonçalves LF, et al. Molecular markers in subclinical acute rejection of renal transplants. Clin Transplant 2004; 18: 281.
19. Monney L, Sabatos CA, Gaglia JL, et al. Th1-specific cell surface protein Tim-3 regulates macrophage activation and severity of an autoimmune disease. Nature 2002; 415: 536.
20. Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: 713.
21. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2 (−ΔΔ C(T)) Method. Methods 2001; 25: 402.
22. Kirkwood BS, Sterne JAC. Measurement error: Assessment and implication. In: Kirkwood BS, Sterne JAC (ed). Essential medical statistics, [ed. 2]. London, Blackwell Publishing Company 2003, pp 432.
23. Zhu C, Anderson AC, Shubart A, et al. Tim-3 ligand galectin-9 negatively regulates T helper type 1 immunity. Nat Immunol 2005; 6: 1245.
24. Sánchez-Fueyo A, Tian J, Picarella D, et al. Tim-3 inhibits T helper type-1 mediated auto- and alloimmune responses and promotes immunological tolerance. Nat Immunol 2003; 4: 1093.
25. Sabatos CA, Chakravarti S, Cha E, et al. Interaction of Tim-3 and Tim-3 ligand regulates T helper type 1 responses and induction of peripheral tolerance. Nat Immunol 2003; 4: 1102.
26. Ponciano VC, Renesto PG, Nogueira E, et al. Tim-3 expression in human kidney allografts. Transplant Immunol 2007; 17: 215.
27. Renesto PG, Ponciano VC, Cenedeze MA, et al. High expression of Tim-3 mRNA in urinary cells from kidney transplant recipients with acute rejection. Am J Transplant 2007; 7: 1661.
28. Yannaraki M, Rebibou JM, Ducloux D, et al. Urinary cytotoxic molecular markers for a noninvasive diagnosis in acute renal transplant rejection. Transpl Int 2006; 19: 759.
29. Aquino-Dias E, Joelsons G, da Silva DM, et al. Non-invasive diagnosis of acute rejection in kidney transplants with delayed graft function. Kidney Int 2008; 73: 877.
30. Strom TB, Suthanthiran M. Transcriptional profiling to assess the clinical status of kidney transplants. Nat Clin Pract Nephrol 2006; 2: 116.

Kidney transplantation; Delayed graft function; Acute graft dysfunction; Acute rejection; Tim-3

© 2008 Lippincott Williams & Wilkins, Inc.