Chronic allograft injury (CAI) is the leading cause of kidney transplant failure, manifesting as progressive tubulointerstitial fibrosis and atrophy (IFTA): inevitably increasing during the life of the allograft, adversely influencing graft function and survival (1, 2). It has been hypothesized that IFTA is the result of a natural equilibrium imbalance in collagen deposition and remodeling/removal, toward increased collagen deposition by activation of myofibroblasts in response to tissue injury. The cause of CAI is yet to be elucidated, but it is generally accepted to be a reflection of the nonspecific end pathway of tubular cell injury in response to various immune and nonimmune insults (2, 3). A postulated mechanism of CAI in the renal allograft is through epithelial-to-mesenchymal transition (EMT)—where the tubular epithelial cells transition into interstitial myofibroblasts. Because myofibroblasts are central to renal fibrosis, their origin and activation forms a crucial component of the axis of progressive IFTA.
In the kidney, the postulated events of EMT begins with the loss of epithelial markers (E-cadherin) and function, migration into the extracellular space and morphosis into a myofibroblast (α-smooth muscle actin [SMA] positive), ending with secretion of matrix proteins (collagens) and initiation of renal fibrosis (4–8). The role of EMT in renal fibrosis is contentious with a number of in vivo studies proposing it as a major mechanism of damage (8–10), whereas others as minor or unrelated to fibrosis (11–14). Most agree on the occurrence of EMT in the kidney, with in vitro studies proving the ability of a tubular epithelial cell to transition into a fibroblast under ideal conditions (6, 11, 15, 16), thus fueling the EMT-renal fibrosis association hypothesis, with key issues being mechanism driving EMT and its degree of overall fibrotic effect in vivo. Unfortunately, most studies into renal EMT have been performed in animal models, cells lines, or adult in vivo clinical investigations, thus neglecting pediatric transplantation. This study is the first of its kind, in attempt to shed light on the role of EMT in pediatric transplantation, specifically looking at the molecular responses during the adaptation of an adult-sized renal allograft in size discrepant, young pediatric recipients. This study combines focused microarray analysis and histology of 126 posttransplant renal biopsies at 3, 6, 12, and 24 months from 83 thoroughly clinically categorized pediatric renal transplant recipients.
Study recipients were 10.8±6.4 years old, of which 64% were men with a body surface area (BSA) of 1.1±0.5 m2 at implantation. The donor age was 32.8±10.7 years, of which 47% were men with 28% of allografts originating from deceased donors. The average human leukocyte antigen mismatch score was 4.1±1.4 with a cold ischemic time of 198±252 min. This was an immunologically pristine dataset that excluded patients with cases of delayed graft function and interval acute rejection. Immunosuppression consisted of induction with daclizumab, followed by maintenance on mycophenolate mofetil and tacrolimus in combination with steroids (n=16) or without (n=67). Finally, none of the allograft recipients developed recurrent primary renal disease, BK tubulointerstitial nephropathy, or infections.
Progressive IFTA in Pediatric Recipients of Adult Renal Allografts Is Driven by Allograft Size Mismatch, Ongoing Calcineurin Inhibitor Toxicity, and Immune Injury
This data set has minimal impact on donor-derived morbidity in the development of CAI because all allograft recipients had a Remuzzi score less than 1 on their implantation biopsy. As expected and as shown previously by our group (17), time posttransplantation was the greatest correlate (r=0.78, P<0.001) of increasing chronic allograft damage index (CADI) and calcineurin inhibitor toxicity (CNIT) scores over the 24-month period of study (mean: 24 months, CADI score: 4.4±2.7; mean: 24 months, CNIT score: 5.1±6.7). In addition, we observed a steady increase in the total i score over time, from 18% of biopsies at 3 months with a total i more than 1 to 37% at 6 months, 62% at 12 months, and 71% at 24 months, which correlated with increasing tubulointerstitial damage (TID; r=0.66, P<0.001). Stratifying patients by BSA, we found, similar to our previous publication (17), that children with the lowest BSA (<0.75 m2) had progressive and significant IFTA (r=0.52, P=0.001) and CADI (r=0.37, P=0.05) injury over the first 2 years. Using logistic regression (odds probability), we found that the total i score, CNIT score, and low BSA at implantation were the main risk factors (Fig. 1) for chronic and progressive no-immune graft injury.
Enrichment of EMT-Associated Genes
To assess the influence of EMT on CAI, we performed an intersection comparison of the differentially expressed genes (DEGs) in relation to CAI on biopsy; with the 243 EMT-associated gene set (Fig. 2; see Table, SDC 1,http://links.lww.com/TP/A483 ). From these CAI DEGs (details in Ref. 18), there was enrichment (P=0.05) of EMT-associated genes (n=13), such as TGFβR2, NOTCH2, MMP7, VIM, indicating that EMT may play a role in progressive IFTA (data shown in SDC 2,http://links.lww.com/TP/A484).
To further investigate the specific influence of EMT on CAI, we filtered the microarrays using the 243 EMT- associated gene set and compared CAI versus no-CAI on biopsy (Fig. 3, Table 1; see SDC 3,http://links.lww.com/TP/A485, SDC 4, http://links.lww.com/TP/A486, and SDC 5,http://links.lww.com/TP/A487) at the various time points. The first analysis at 3 months identified 48 DEGs (see Table, SDC 3,http://links.lww.com/TP/A485), with 36 upregulated and 12 downregulated genes. A number of these DEGs represent EMT regulators, such as SMURF1 and 2, NOTCH2, SMAD2 and 6, TGFβ3, BMP1 and 2, FGFR1 and 2, and SKI (Table 1). The EMT profile of gene expression seems to suggest a pro-EMT pattern, with the majority of above- mentioned genes upregulated in the absence of pro– mesenchymal-to-epithelial transition (MET). This EMT association with early CAI is further evident from its separation from no-CAI at 3 months (Fig. 3B). One of the differentially expressed EMT-related genes were validated (MMP2) by quantitative polymerase chain reaction (QPCR) in an independent cohort of 3-month protocol biopsies that had presence of CAI (n=6) or absence of CAI (n=6). We selected MMP2 as a DEG in CAI at 3 months because it has been recognized to play a role in EMT and has a q score in the microarray data set of 5.39 and a false discovery rate of less than 5%. MMP2 was observed to be similarly upregulated through QPCR with significant increase in transcription for CAI (P=0.03) (see SDC 2,http://links.lww.com/TP/A484).
The EMT pattern at 6 months differed, with 10 of the 47 DEGs (all upregulated) in CAI were common to 3 months (Fig. 3A, Table 1; see Table, SDC 4,http://links.lww.com/TP/A486). The EMT expression in CAI at 6 months continues to be pro-EMT with the upregulation of AKT1, SMURF1, BMP1 and 2, SMAD3, FGFR1, and TIMP1. Unlike at 3 months, hierarchical clustering was unable to distinctly separate CAI from no-CAI (Fig. 3C), indicating additional mechanisms influencing the development of CAI at 6 months posttransplant.
There seems to be a time course to EMT-related genes expression, with majority and strongest EMT molecular changes occurring early posttransplant time course (3 and 6 months) with the absence of any EMT-related DEGs at 12 months. There was late resurgence of EMT gene expression at 24 months with 24 DEGs EMT genes, 10 upregulated and 14 downregulated (Table 1; see Table, SDC 5,http://links.lww.com/TP/A487). In addition, a pro-MET pattern appeared with an upregulation of hepatocyte growth factor (HGF) and downregulation of epidermal growth factor and insulin-like growth factor 1 receptor. A number of the pro-EMT genes (TCF4, WNT5A, VCAN, and IL1RN) also play a role in B-cell signaling and differentiation, supporting our previous findings of a subtle immune injury in the absence of overt rejection episodes. Hierarchical clustering was unable to separate CAI from no-CAI (Fig. 3D).
CNIT and Overall Immune Burden Promotes Early EMT in CAI, Pro-MET Late to Combat Immune Injury
To elucidate the mechanisms and role of EMT-related genes at each defined posttransplant time point, Spearman's rank correlations were performed against the main histopathologic (total i and CNIT) and demographical (BSA and recipient age) drivers of CAI (Table 2; for DEGs, see SDC 3,http://links.lww.com/TP/A485, SDC 4,http://links.lww.com/TP/A486, and SDC 5,http://links.lww.com/TP/A487).
We observed that the majority (96%) of the EMT DEGs significantly correlated with the degree of interstitial fibrosis, as indicated by the Banff ci score, early posttransplant (3 months) with a pro-EMT pattern (Table 2; see Table, SDC 6,http://links.lww.com/TP/A488). In addition, it seems that both the total interstitial lymphocytic infiltrate (total i) and CNIT are contributing to the pro-EMT signal at 3 months with 23% and 19% of EMT DEGs significantly correlated, respectively (Table 2). Each pathologic entity seems to influence different EMT genes with the total i positively correlating with such pro-EMT genes as SMAD2, TGFβ3, MMP2, FGF18, and FGFR1. CNIT correlated positively with pro-EMT genes such as insulin-like growth factor 1 receptor, MITF, and SMAD9. Interestingly, it seems that small (BSA) young recipients are somewhat protected against the pro-EMT gene expression as seen by their significant correlations contrary to what one would expect. This is exhibited by the significant negative correlations against some major pro-EMT such as BMP1, JAG1, NOTCH3, and MITF and similar trends in SMAD2, SMAD9, SKI, RAC1, and MMP2 with young recipients and positive correlations of the same genes with larger recipients (see Table, SDC 6,http://links.lww.com/TP/A488).
Unlike at 3 months, there were no significantly correlated EMT genes at 6 and 24 months with Banff ci, BSA or recipient age, although the Banff total i score was the main group that correlated with EMT gene expression (17%), with positive correlations toward a number of mitogen-activated protein kinases (MAPKs), integrins, and TIMP1, which may represent local immune-related mechanisms in an injured graft (Table 2; see Table, SDC 6,http://links.lww.com/TP/A488).
EMT genes that were significantly correlated with the total infiltrate at 24 months were IL1B, IL1RN, VCAN, and ITGAX (Table 2; see Table, SDC 6,http://links.lww.com/TP/A488). In addition, the major pro-MET gene HGF was positively correlated with the total infiltrate, suggesting an active attempt to reverse or counteract the immune-directed injury. This is also evident from its significant correlation with increasing interstitial fibrosis, further suggesting an attempt of the allograft to respond to the injury burden.
Decreased E-Cadherin Staining (CDH1) in Patients With CAI
We performed IHC staining for E-cadherin (CDH1; see SDC 2, http://links.lww.com/TP/A484) to assess for loss of E-cadherin staining in CAI. We observed a significant decrease in the number of CDH1-positive-only cells (see Figure, SDC 2,http://links.lww.com/TP/A484) in the tubules of biopsies with CAI compared with no-CAI biopsies (P=0.04, 132.9±38.6 vs. 161.4±16.4 per high-power field, respectively). The decrease in CDH1 staining can be considered a reflection of tubular injury and tubular epithelial cell loss, which is the first stage of the EMT mechanism and a measure of EMT.
This is the first and largest study to date to investigate the role of EMT in the evolution of CAI in pediatric recipients of adult-sized, stable renal allografts, without the influence of acute renal allograft rejection, which in itself is a major player in the progression of CAI. Not unlike the adult kidney transplant population, CAI is one of the largest causes of allograft failure, manifesting itself as indiscriminate IFTA (1, 2). This study demonstrates that the EMT-MET equilibrium is a potential key mechanism used by the allograft in the development or response to the evolution of chronic injury. We observed an enrichment of EMT-associated genes in protocol transplant kidney biopsies exhibiting phenotypic signs of TID using whole transcriptome analysis with microarrays. To elucidate the role of EMT, we interrogated the performance of a selected panel of 243 EMT-related genes using whole genome biopsy transcriptome analysis, semiquantitative histopathologic analysis of Banff, CADI and CNIT biopsy scores, and patient clinical and demographic data. From this focused analysis, we observed a time-dependent expression of EMT-related genes in CAI, with a prominent early (3–6 months) stage and resurgent late (24 months) stage, relating to tissue remodeling in response to allograft injury. The major confounders associated with the early pro-EMT response in the allograft are molecular markers of immune injury and CNIT biopsy scores. A counter-EMT response is enriched in young recipients of an adult-sized kidney, with low BSA (<0.75 m2), suggesting a confounding effect of natural developmental biology in young recipients (<6 years). A counter-EMT pattern of expression was observed late posttransplant (24 months) as indicated by increased expression of HGF, a known inducer of MET (19, 20), which is associated with interstitial fibrosis and infiltrating immune cells. Our study using a combination of whole genome transcription analysis, histopathology, and clinical data was able to determine that EMT has a role in early allograft injury and fibrosis, with the allograft appearing to use the reciprocal pathway of MET to counteract and repair damage caused by chronic immune and CNIT injury.
As we have shown in previous studies, allograft size mismatch in the form of adult size allografts transplanted into small pediatric recipients is a major risk factor and driving force for accelerated CAI and allograft dysfunction (17, 21–24). The rationale behind this phenomenon is of chronic ischemic injury to the allograft as a result of reduced perfusion by the recipients smaller cardiac output compared with that of its healthy adult donor. This ischemic injury leads to the release of the full plethora of injury mechanisms resulting in TID, which combined with the other standard immune and nonimmune drivers of CAI results in an accelerated process and further reduced allograft survival (17). This study supports our previous findings where low BSA of the recipient presented as the major demographic risk factor (hazard odds ratio of 2.0) for the presence of CAI on biopsy. In addition to the ischemic injury, this study is consistent with our previously published work and others in the adult setting, where both the total immune burden and CNIT are also critical driving forces to CAI with an increased hazard ratio of 11.5 and 2.5, respectively (3, 25–28).
EMT is a critical process of embryogenesis during specific phases throughout metazoan development and formation of organs, limbs, and other specialized structures. The formation of these various structures is based on the creation of mesenchymal cells from an epithelia progenitor cell type, hence EMT. These de novo mesenchymal cells allow migration into and through the extracellular environment, settling in areas where organ development is to occur. This process is reversed once these mesenchymal cells have settled; forming epithelial-based organs through MET. Both EMT and MET are used to form the kidney during embryogenesis and development (29, 30) and are in equilibrium. For EMT to commence and be maintained to completion, a number of major and profound changes in the surrounding microenvironment and epithelial cell organization must occur. These changes occur in six rate-limiting steps, as outlined by Baum et al. (31), where malfunction at any stage leads to EMT unable to proceed.
We observed two distinct phases of EMT/MET-related expression following the pattern of allograft injury (17, 18), one early pro-EMT (0–6 months) and one late mixed pro-EMT/pro-MET (12–24 months). Early posttransplant (0–6 months), saw the upregulation of pro-EMT genes (Table 1), with the majority of these EMT-related genes correlated strongly with early IFTA, CNIT, and overall immune burden (Table 2), distinguishing CAI from no-CAI with supervised hierarchical clustering at 3 months, abet four miss classifiers (Fig. 3B). Contrary to expectation, pro-EMT gene expression did not associate with prolonged ischemic injury as represented by organ-recipient size mismatch disparity. One possible explanation is the general biology of the younger recipients may be causing a dampening or counteraction of EMT mechanisms because of normal renal development. This would require further investigation to elucidate the influence of recipient age on EMT mechanisms, including a young to young control group, which unfortunately was not possible in the current study.
The phenotypical transition from an epithelial cell to a mesenchymal cell through the process of EMT is a complex and dramatic cascade of events that requires a great deal of both transcriptomic and structural changes to occur successfully. With this being the case, there are a number of key pathways and genes that are top-end regulators of EMT: some of which are TGF-β, E-cadherin, Snail, and Twist as shown in studies of organ development, cancer, and fibrosis (4, 32–34). As with many of EMT-associated genes, they have many different roles such as TIMP1, MMPs, and collagen in breakdown and extracellular matrix remodeling. Others are involved in proliferation and dedifferentiation, such as SMAD2 and HGF, respectively, with some playing dual roles depending on the situation, such as TGFβ3. This is probably the main cause of contrasting opinions with the role of EMT fibrosis (8–14); however, most agree that EMT is present posttransplant. In a previous adult-based study (12), EMT may play a minor role in CAI, with this current study showing that EMT potentially playing a larger role in early pediatric transplantation. This discrepancy between age-confounded studies is not uncommon as we have seen previously in transcriptome analysis of acute rejection in renal allografts, with only 16% of the pediatric-derived DEGs common with adults (35), likely because of differences in immunobiology, population, array platform, and other demographic variables.
The second stage of EMT/MET was observed through the disappearance of EMT-related mechanisms between the 6- and 24-month posttransplant period and low resurgence at 24 months. This indicates a potential shift in the mechanisms influencing EMT or a transitional point in EMT signaling, thus suggesting a late phase of MET-related mechanisms. This absence of EMT-related expression was observed despite continuous increase in allograft injury, potentially reflecting the overall net effect of mRNA message in tissue samples in an attempt to maintain the EMT-MET equilibrium. This can be postulated as evidence of epithelial-mesenchymal-epithelial (EME) cycling of the allograft to counteract the rampant uncontrolled healing response to early injury.
EME cycling is a process where an injured epithelial cell transitions into a mesenchymal cell by means of EMT, then reverting back to an epithelial cell once the damage is repaired. This process is postulated to occur naturally in the kidney, with evidence deriving from various studies in vivo of acute tubular necrosis (36, 37) and the presence of EMT in healthy kidney donor biopsies (12). The process of EME allows epithelial cells to migrate to damaged tubules, where they back-differentiate, proliferate, and repopulate it (38–40). Many of the same markers implicated in EMT are associated with the dedifferentiation and migration of the epithelial cells, including TGF-β1, vascular endothelial growth factor, HGF, epidermal growth factor, and various others (41–43). Redifferentiation and back conversion of these cells to form a tubule requires the absence of the EMT-inducing markers and cell-cell contact to be re-established, one such maker being HGF (40) with MAPKs playing a major role (44, 45).
We observed a gene expression pattern which suggests EME cycling is ongoing in retaliation to chronic immune injury and initial ischemic insult of the allograft, as seen by the DEG and positive correlations of HGF in late biopsies with IFTA. This hypothesis is supported by the absence of the major pro-EMT genes that were observed earlier during the combination of injury mechanisms CNIT, ischemia, and immune-related processes. The late pro-MET–related gene expression may be an indication of the allografts response to this ongoing allograft injury. One may go further and postulate that the allograft uses some of the EMT observed at this late stage in the process of EME cycling to counteract the immune and ischemic injury, the two significant drivers of fibrosis (17) and EMT (6, 11), as supported by the correlation HGF and MAPKs and minor EMT-related genes late. With this said, a more comprehensive focused study into the role and mechanisms of EME cycling in renal transplantation is required to elucidate and validate this working hypothesis on the allografts response to chronic injury.
In conclusion, we observed an enrichment of EMT-related gene expression in our pediatric recipients of an adult allograft with consistent and gradual CAI. We were able to elucidate somewhat the role of EMT in relation to the progressive immune, CNIT, and ischemic-based injury as the allografts response to such, in an attempt to counteract and repair itself. We postulate that this is mediated through the natural EME cycling equilibrium, where the allograft attempts to shift this equilibrium in favor of healing, over fibrosis.
MATERIALS AND METHODS
Study Population and Patient Samples
The study consisted of longitudinal surveillance biopsies from pediatric recipients of an adult kidney transplant from 2004 to 2006 at Lucile Packard Children's Hospital at Stanford. All patients received induction with daclizumab and maintenance immunosuppression with tacrolimus, mycophenolate mofetil, with or without steroid avoidance (46, 47); all included patients in the study had excellent and immediate graft function, and there were no patients with delayed graft function in this study. All the samples were taken at immunologic quiescent phase with no Banff gradable acute rejection. With this study design, any confounding influence of DGF and interval acute rejection were controlled for, allowing for a clean analysis of EMT-related transcriptional changes in the development of CAI. The study was approved by the Ethics Committee of Stanford University Medical School, and all patients/guardians provided informed consent to participate in the research, in full adherence to the Declaration of Helsinki. As per Stanford institutional review board–approved protocol, biopsies were collected (see Methods, SDC 2,http://links.lww.com/TP/A484) at 3 (n=20), 6 (n=45), 12 (n=19), and 24 (n=42) months posttransplant. Full clinical details matched at the time of biopsy were collected on all participating patients.
Histopathology and CAI Definition
All biopsies collected were blindly scored by a single pathologist at Stanford (Neeraja Kambham) according to Banff criteria (48–50), with all biopsies given a CADI (51), CNIT (52), and IFTA/TID score additional to Banff diagnosis. Biopsies were scored semiquantitatively for both chronic and acute injury, as per standard Banff schema (48–50) and a CADI score of more than 5 defined a biopsy as exhibiting signs of significant CAI (previously CAN), as described in our previous publication (18). For comparative analysis, the non-CAI group was chosen to include subjects with biopsies not showing any significant chronic injury (CADI score <1) over the first 2 years posttransplant.
RNA Extraction and Microarrays
From each tissue biopsy, total RNA was extracted, quantitatively and qualitatively measured, and hybridized to Affymetrix GeneChip Human Genome U133 Plus 2.0 Arrays as previously described in Naesens et al. (18) and summarized in SDC 2, Methods and Results,http://links.lww.com/TP/A484.
EMT Gene Set Analysis
A selected group of EMT-associated genes were chosen on two criteria: (1) genes identified in the published literature to play a role in EMT and renal injury and EMT signaling cascades (see Table, SDC 1,http://links.lww.com/TP/A483); and (2) genes ascertained to play a role in EMT in chronic injury in adult kidney transplantation (12). This resulted in 247 individual discrete annotated genes (643 Affymetrix-designed probe sets), through AILUN (53).
Microarray and Pathway Analysis
The microarrays were normalized and analyzed using standard statistical and bioinformatic tools, such as R, Bioconductor, significance analysis of microarrays, ingenuity pathway analysis, and hierarchical clustering as described by Naesens et al. (18) and summarized in the Methods, SDC 2,http://links.lww.com/TP/A484.
QPCR was performed on an independent cohort of 12 patients on their 3-month protocol biopsy to validate MMP2 (important deferentially expressed EMT gene) using one house keeping gene (GAPD) as an endogenous control. Reverse transcription and cDNA synthesis was performed using the Superscript III following the manufactures protocol (Invitrogen, Carlsbad, CA). The quantitative real-time polymerase chain reaction was performed using SYBR Green chemistry as previously described (54), comparing the ratio of expression of CAI over no-CAI, as calculated in the microarrays (see Methods and Results, SDC 2,http://links.lww.com/TP/A484).
Immunohistochemistry Staining for E-cadherin (CDH1)
Staining for EMT-related tubular injury was performed using the standard epithelial marker CDH1 on an independent cohort of 12 patients on 3-month protocol biopsy. The immunohistochemistry was performed using standard techniques as previously described (12). In brief, 3-μm-thick sections from formalin-fixed paraffin-embedded tissue renal biopsies were dewaxed and treated for antigen retrieval, and endogenous peroxidases were blocked. CDH1 (clone ECH-6, Ventana, Tucson, AZ) was stained and visualized, all on an automated immunostainer (Ventana). Single stained CDH1-positive tubular cells were counted and averaged over three randomly chosen high-power fields (600×) per biopsy by a blinded observer to clinical phenotype.
Normalized gene expression values from biopsy microarrays were assessed for various statistical correlations and assessment with transplant variables, with a P value of less than 0.05 considered significant for all analyses (see Methods, SDC 2,http://links.lww.com/TP/A484).
The Sarwal Lab thanks all of the patients and families for their support in this research. The authors thank Neeraja Kambham for her pathologic assessment and scoring of the transplant biopsies for Banff criteria. Matthew Vitalone specifically thanks the Lucile Packard Foundation for Children's Health and the Stanford CTSA (UL1RR025744) for providing fellowship funds to complete this research and his wife Laurice for her love and support.
1.Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronic allograft nephropathy. N Engl J Med
2003; 349: 2326.
2.Nankivell BJ, Fenton-Lee CA, Kuypers DR, et al. Effect of histological damage on long-term kidney transplant outcome. Transplantation
2001; 71: 515.
3.Nankivell BJ, Borrows RJ, Fung CL, et al. Natural history, risk factors, and impact of subclinical rejection in kidney transplantation. Transplantation
2004; 78: 242.
4.Neilson EG. Setting a trap for tissue fibrosis. Nat Med
2005; 11: 373.
5.Liu Y. Epithelial to mesenchymal transition in renal fibrogenesis: Pathologic significance, molecular mechanism, and therapeutic intervention. J Am Soc Nephrol
2004; 15: 1.
6.Yang J, Liu Y. Dissection of key events in tubular epithelial to myofibroblast transition and its implications in renal interstitial fibrosis. Am J Pathol
2001; 159: 1465.
7.El-Nahas AM. Plasticity of kidney cells: Role in kidney remodeling and scarring. Kidney Int
2003; 64: 1553.
8.Iwano M, Plieth D, Danoff TM, et al. Evidence that fibroblasts derive from epithelium during tissue fibrosis. J Clin Invest
2002; 110: 341.
9.Vongwiwatana A, Tasanarong A, Rayner DC, et al. Epithelial to mesenchymal transition during late deterioration of human kidney transplants: The role of tubular cells in fibrogenesis. Am J Transplant
2005; 5: 1367.
10.Djamali A, Reese S, Oberley T, et al. Heat shock protein 27 in chronic allograft nephropathy: A local stress response. Transplantation
2005; 79: 1645.
11.Hertig A, Verine J, Mougenot B, et al. Risk factors for early epithelial to mesenchymal transition in renal grafts. Am J Transplant
2006; 6: 2937.
12.Vitalone MJ, O'Connell PJ, Jimenez-Vera E, et al. Epithelial-to- mesenchymal transition in early transplant tubulointerstitial damage
. J Am Soc Nephrol
2008; 19: 1571.
13.Duffield JS. Epithelial to mesenchymal transition in injury of solid organs: Fact or artifact? Gastroenterology
2010; 139: 1081, 1083 e1.
14.Callahan KL, Beck NK, Duffield EA, et al. Inactivation of methicillin-resistant Staphylococcus aureus
(MRSA) and vancomycin-resistant Enterococcus faecium
(VRE) on various environmental surfaces by mist application of a stabilized chlorine dioxide and quaternary ammonium compound-based disinfectant. J Occup Environ Hyg
2010; 7: 529.
15.McMorrow T, Gaffney MM, Slattery C, et al. Cyclosporine A induced epithelial-mesenchymal transition in human renal proximal tubular epithelial cells. Nephrol Dial Transplant
2005; 20: 2215.
16.Slattery C, Campbell E, McMorrow T, et al. Cyclosporine A-induced renal fibrosis: A role for epithelial-mesenchymal transition. Am J Pathol
2005; 167: 395.
17.Naesens M, Kambham N, Concepcion W, et al. The evolution of nonimmune histological injury and its clinical relevance in adult-sized kidney grafts in pediatric recipients. Am J Transplant
2007; 7: 2504.
18.Naesens M, et al. Microarray expression profiling associates progressive histological damage of renal allografts with innate and adaptive immunity. Kidney Int
2011; (in press).
19.Dai C, Liu Y. Hepatocyte growth factor antagonizes the profibrotic action of TGF-beta1 in mesangial cells by stabilizing Smad transcriptional corepressor TGIF. J Am Soc Nephrol
2004; 15: 1402.
20.Deuse T, Peter C, Fedak PW, et al. Hepatocyte growth factor or vascular endothelial growth factor gene transfer maximizes mesenchymal stem cell-based myocardial salvage after acute myocardial infarction. Circulation
2009; 120(11 suppl): S247.
21.Salvatierra O, Concepcion W, Sarwal M. A critical look at the immunologically favorable adult-sized kidney transplant in small children. Pediatr Transplant
2009; 13: 265.
22.Salvatierra O Jr, Sarwal M. Renal perfusion in infant recipients of adult-sized kidneys is a critical risk factor. Transplantation
2000; 70: 412.
23.Salvatierra O Jr, Singh T, Shifrin R, et al. Successful transplantation of adult-sized kidneys into infants requires maintenance of high aortic blood flow. Transplantation
1998; 66: 819.
24.Salvatierra O Jr, Singh T, Shifrin R, et al. Transplantation of adult-sized kidneys into infants induces major blood flow changes. Transplant Proc
1999; 31: 236.
25.Nankivell BJ, Borrows RJ, Fung CL, et al. Calcineurin inhibitor nephrotoxicity: Longitudinal assessment by protocol histology. Transplantation
2004; 78: 557.
26.Nankivell BJ, Chapman JR. The significance of subclinical rejection and the value of protocol biopsies. Am J Transplant
2006; 6: 2006.
27.Mengel M, Gwinner W, Schwarz A, et al. Infiltrates in protocol biopsies from renal allografts. Am J Transplant
2007; 7: 356.
28.Mengel M, Reeve J, Bunnag S, et al. Scoring total inflammation is superior to the current Banff inflammation score in predicting outcome and the degree of molecular disturbance in renal allografts. Am J Transplant
2009; 9: 1859.
29.Barasch J, Yang J, Ware CB, et al. Mesenchymal to epithelial conversion in rat metanephros is induced by LIF. Cell
1999; 99: 377.
30.Plisov SY, Yoshino K, Dove LF, et al. TGF beta 2, LIF and FGF2 cooperate to induce nephrogenesis. Development
2001; 128: 1045.
31.Baum B, Settleman J, Quinlan MP. Transitions between epithelial and mesenchymal states in development and disease. Semin Cell Dev Biol
2008; 19: 294.
32.Thiery JP, Sleeman JP. Complex networks orchestrate epithelial- mesenchymal transitions. Nat Rev Mol Cell Biol
2006; 7: 131.
33.Moreno-Bueno G, Cubillo E, Sarrió D, et al. Genetic profiling of epithelial cells expressing E-cadherin repressors reveals a distinct role for Snail, Slug, and E47 factors in epithelial-mesenchymal transition. Cancer Res
2006; 66: 9543.
34.Barrallo-Gimeno A, Nieto MA. The Snail genes as inducers of cell movement and survival: Implications in development and cancer. Development
2005; 132: 3151.
35.Chen R, Sigdel TK, Li L, et al. Differentially expressed RNA from public microarray data identifies serum protein biomarkers for cross-organ transplant rejection and other conditions. PLoS Comput Biol
2010; 6: pii: e1000940.
36.Bonventre JV. Dedifferentiation and proliferation of surviving epithelial cells in acute renal failure. J Am Soc Nephrol
2003; 14(suppl 1): S55.
37.Gröne HJ, Weber K, Gröne E, et al. Coexpression of keratin and vimentin in damaged and regenerating tubular epithelia of the kidney. Am J Pathol
1987; 129: 1.
38.Ishibe S, Cantley LG. Epithelial-mesenchymal-epithelial cycling in kidney repair. Curr Opin Nephrol Hypertens
2008; 17: 379.
39.Montesano R, Matsumoto K, Nakamura T, et al. Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor. Cell
1991; 67: 901.
40.Ishibe S, Haydu JE, Togawa A, et al. Cell confluence regulates hepatocyte growth factor-stimulated cell morphogenesis in a beta-catenin-dependent manner. Mol Cell Biol
2006; 26: 9232.
41.Karihaloo A, O'Rourke DA, Nickel C, et al. Differential MAPK pathways utilized for HGF- and EGF-dependent renal epithelial morphogenesis. J Biol Chem
2001; 276: 9166.
42.Tilghman RW, Slack-Davis JK, Sergina N, et al. Focal adhesion kinase is required for the spatial organization of the leading edge in migrating cells. J Cell Sci
2005; 118(pt 12): 2613.
43.Zhuang S, Yan Y, Han J, et al. p38 kinase-mediated transactivation of the epidermal growth factor receptor is required for dedifferentiation of renal epithelial cells after oxidant injury. J Biol Chem
2005; 280: 21036.
44.O'Brien LE, Tang K, Kats ES, et al. ERK and MMPs sequentially regulate distinct stages of epithelial tubule development. Dev Cell
2004; 7: 21.
45.Hellman NE, Spector J, Robinson J, et al. Matrix metalloproteinase 13 (MMP13) and tissue inhibitor of matrix metalloproteinase 1 (TIMP1), regulated by the MAPK pathway, are both necessary for Madin-Darby canine kidney tubulogenesis. J Biol Chem
2008; 283: 4272.
46.Sarwal MM, Vidhun JR, Alexander SR, et al. Continued superior outcomes with modification and lengthened follow-up of a steroid-avoidance pilot with extended daclizumab induction in pediatric renal transplantation. Transplantation
2003; 76: 1331.
47.Li L, Chang A, Naesens M, et al. Steroid-free immunosuppression since 1999: 129 pediatric renal transplants with sustained graft and patient benefits. Am J Transplant
2009; 9: 1362.
48.Racusen LC, Halloran PF, Solez K. Banff 2003 meeting report: New diagnostic insights and standards. Am J Transplant
2004; 4: 1562.
49.Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int
1999; 55: 713.
50.Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: Updates and future directions. Am J Transplant
2008; 8: 753.
51.Isoniemi H, Taskinen E, Hayry P. Histological chronic allograft damage index accurately predicts chronic renal allograft rejection. Transplantation
1994; 58: 1195.
52.Nagarajan S, Shah S, Troxell M, et al. A novel drug toxicity pathology score in protocol biopsies can predict clinical graft and patient outcomes in pediatric renal transplantation. Boston, MA, American Transplant Congress 2004.
53.Chen R, Li L, Butte A. AILUN: Reannotating gene expression data automatically. Nat Methods
2007; 4: 879.
54.Vitalone MJ, O'Connell PJ, Wavamunno M. Transcriptome changes of chronic tubulointerstitial damage
in early kidney transplantation. Transplantation
2010; 89: 537.