Transcriptional profiling to assess the clinical status of kidney transplants Terry B Strom* and Manikkam Suthanthiran
TB Strom is Professor of Medicine and Surgery, Co-Director of the Transplant Center and Director of the Transplant Research Center at Beth Israel Deaconess Medical Center, Boston, MA, USA. M Suthanthiran is Chief of Nephrology and Transplant Medicine at New York-Presbyterian Hospital, and Stanton Griffis Distinguished Professor of Medicine at Weill Medical College of Cornell University, New York, NY, USA.
Correspondence *Harvard Institutes of Medicine Room 1026 77 Avenue Louis Pasteur Boston MA 02115 USA [email protected]
Received 26 July 2005 Accepted 3 January 2006 www.nature.com/clinicalpractice doi:10.1038/ncpneph0115
Over the past decade, we have investigated the hypothesis that a molecular diagnostic approach based on transcriptional profiling can refine the classification of kidney transplant biopsies. More recently, we have tested the idea that noninvasive transcriptional analysis of circulating blood cells and urine sediment cells can function not only as a surrogate for the invasive biopsy procedure, but can also provide predictive, diagnostic, and prognostic information. Human renal1 and rodent cardiac allografts1,2 serve as magnets for activated recipient antidonor T cells.2 Activated host antidonor CD8+ T cells selectively kill donor cells, whereas naive T cells do not. These cytotoxic T lymphocytes (CTLs) home to and concentrate within human and rodent allografts during episodes of allograft rejection, and can therefore function as an indicator of rejection.2 Unfortunately however, the cumbersome cell-based methods used experimentally to detect activated host antidonor T cells do not enable routine clinical surveillance of renal allograft recipients. We have therefore begun investigating a molecular transcriptional profiling approach to clinical monitoring of renal allografts. The polymerase chain reaction (PCR) provides a sensitive and convenient means of rigorously quantifying the expression of numerous gene transcripts within a small sample of tissue. The development of DNA ARRAY technology provides another potent tool for transcriptional profiling. With the knowledge that host antidonor CTLs are concentrated within rejecting allografts, and that a carefully orchestrated succession of gene expression events facilitates the transition of naive T cells to proliferating effector T cells, we decided to investigate whether the intragraft expression of transcripts encoding the CTL cell-killing machinery and other components of T-cell activation could serve as an indicator of rejection. We found that these genes are expressed within rejecting, but not quiescent, mouse and human allografts.3–5 In fact, CTL-related genes are coordinately (simultaneously) expressed, thereby
116 NATURE CLINICAL PRACTICE NEPHROLOGY
allowing the results to be authenticated by crosschecking the expression patterns of multiple genes.5 Many other T-cell activation transcripts (e.g. CD154) can be readily detected in rejecting, but not quiescent, human renal allografts.6 Many of the activated T-cell gene transcripts present in rejecting grafts encode proteins that are critical for rejection-mediated damage of the allograft. Interestingly however, some transcripts—such as CTLA, found abundantly in rejecting allografts—encode molecules that are linked to suppressor-type immune responses.5 This finding illustrates that the immune response to foreign tissues possesses both cytoprotective and destructive components. Although the destructive component predominates during rejection, protective components are also present. Ongoing preclinical and clinical investigations emphasize that the expression of these protective molecules is profoundly relevant to clinical outcome.7 As mentioned above, T-cell-lineage-specific transcripts, such as those encoding the T-CELL RECEPTOR FOR ANTIGEN8 and CD,9 are abundant within rejecting but not quiescent allografts. In a recent study, the presence of abundant intragraft B-cell-specific transcripts, combined with T-cell activation, was found to correlate with corticosteroid-resistant cellular rejection.10 This study used DNA array technology, rather than REVERSE TRANSCRIPTASE-PCR, enabling analysis of gene expression for thousands of transcripts in a single step—albeit with slightly diminished sensitivity compared with PCR. Clearly, quantitative molecular analysis of transcripts in renal allograft biopsy specimens provides a sensitive and powerful means of evaluating kidney biopsy specimens. Should this approach provide, as we believe it will, data that aid clinical management, this would represent a significant advance. It is important to bear in mind that, at present, renal biopsies are only performed to determine whether deteriorating renal function is due to graft rejection. Surveillance biopsies are not routinely
MARCH 2006 VOL 2 NO 3 ©2006 Nature Publishing Group
performed because they are expensive and have the potential to cause complications. The capacity to detect rejection-associated transcripts noninvasively would therefore be of great benefit to clinicians. In our preclinical experiments, cellular assays indicated that host antidonor CTLs are concentrated within rejecting allografts, but that these cells are also present at other sites. We tested the hypothesis that sensitive PCR-based molecular techniques would detect expression of CTL effector-molecule transcripts in both the allograft and circulating blood of renal transplant recipients whose grafts are undergoing rejection. The results of our pilot trial indicate that this hypothesis was correct.11 Subsequently, a large study was undertaken to investigate the utility of transcriptional profiling of urinary sediment cells harvested from patients with graft dysfunction, as a noninvasive means of detecting rejection. Renal allograft biopsies were analyzed in parallel.9 Lymphocytes present in urine have undoubtedly trafficked through the kidney, so the status of these cells might accurately reflect the status of graft-infiltrating lymphocytes. This trial showed that the gene expression profile of CTL effector molecules in both renal biopsy specimens and urinary sediment cells predicted acute rejection with comparable and very high sensitivity and specificity. We have recently started to investigate whether subtle inflammation and immune activity detected in the intraoperative period are linked to adverse clinical outcomes after renal transplantation. Renal allografts were analyzed for expression of proinflammatory, inflammation-induced adhesion molecules, for immune activation and for the presence of ANTIAPOPTOTIC GENES expressed at zero hour (15 min after vascular reperfusion). We found that transcriptional profiling of the zero-hour kidney biopsy can predict post-transplantation clinical outcomes such as delayed graft function, acute rejection, and the quality of renal function 6 months post-transplantation.12 In conclusion, we hypothesize that molecular profiling will enable mechanistically driven
and individualized care of the renal transplant recipient in the near future. We posit that transcriptional profiles of allografts ascertained at the time of implantation, and noninvasive profiling of allograft recipients post-transplantation could augment, if not supplant, histopathologic examination and classification of allografts as a means of monitoring host anti-allograft responses. Clearly, validation of the predictive, diagnostic and prognostic utility of molecular diagnostic strategies in multicenter, prospective clinical trials will be required before this strategy can be recommended for routine use. Nevertheless, it is our belief that molecular profiling will prove a cost-effective aid to clinical management. References 1 Strom TB et al. (1975) Identity and cytotoxic capacity of cells infiltrating renal allografts. N Engl J Med 292: 1257–1263 2 Tilney NL et al. (1984) Mechanisms of rejection and prolongation of vascularized organ allografts. Immunol Rev 77: 185–216 3 Lipman ML et al. (1992) The strong correlation of cytotoxic T lymphocyte-specific serine protease gene transcripts with renal allograft rejection. Transplantation 53: 73–9 4 Lipman ML et al. (1994) Heightened intragraft CTL gene expression in acutely rejecting renal allografts. J Immunol 152: 5120–5127 5 Strehlau J et al. (1997) Quantitative detection of immune activation transcripts as a diagnostic tool in kidney transplantation. Proc Natl Acad Sci USA 94: 695–700 6 Zheng XX et al. (1998) Increased CD40 ligand gene expression during human renal and murine islet allograft rejection. Transplantation 65: 1512–1515 7 Muthukumar T et al. (2005) Messenger RNA for FOXP3 in the urine of renal-allograft recipients. N Engl J Med 353: 2342–2351 8 Pavlakis M et al. (1995) Intragraft T cell receptor transcript expression in human renal allografts. J Am Soc Nephrol 6: 281–285 9 Li B et al. (2001) Noninvasive diagnosis of renalallograft rejection by measurement of messenger RNA for perforin and granzyme B in urine. N Engl J Med 344: 947–954 10 Sarwal M et al. (2003) Molecular heterogeneity in acute renal allograft rejection identified by DNA microarray profiling. N Engl J Med 349: 125–138 11 Vasconcellos LM et al. (1998) Cytotoxic lymphocyte gene expression in peripheral blood leukocytes correlates with rejecting renal allografts. Transplantation 66: 562–566 12 Avihingsanon Y et al. (2005) On the intraoperative molecular status of renal allografts after vascular reperfusion and clinical outcomes. J Am Soc Nephrol 16: 1542–1548
MARCH 2006 VOL 2 NO 3
GLOSSARY DNA ARRAY Sets of oligonucleotides of known sequences synthesized in situ on a matrix in a predetermined order CTLA4 Cytotoxic T-lymphocyteassociated protein 4 T-CELL RECEPTOR FOR ANTIGEN A membrane-bound protein found on T cells, which recognizes antigen associated with the major histocompatibility complex CD3 A cell-surface protein that aids signal transduction via the T-cell receptor for antigen REVERSE TRANSCRIPTASEPOLYMERASE CHAIN REACTION (RT-PCR) A variation of the PCR technique in which cDNA is made from RNA via reverse transcription and the cDNA is then amplified using standard PCR protocols ANTIAPOPTOTIC GENES Genes coding for proteins that quench inflammation and cell death programs
Competing interests The authors declared competing interests; go to the article online for details.
NATURE CLINICAL PRACTICE NEPHROLOGY 117 ©2006 Nature Publishing Group