|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
Background Nephrogenic adenomas are benign, tumor-like lesions within the urothelial mucosa of the urinary tract that are not uncommon in renal-transplant recipients. We investigated the origin of nephrogenic adenomas in renal-transplant recipients.
Methods Tissue sections were analyzed by fluorescence in situ hybridization with the use of probes for the X and Y chromosomes, by immunohistochemical methods with the use of antibodies to renal tubular antigens, and by lectin histochemical methods. Forty-six nephrogenic adenomas from 29 patients were analyzed.
Results All nephrogenic adenomas in 14 female recipients of transplants from male donors and 10 male recipients of transplants from female donors showed the same sex-chromosome status as the donor kidney, but not the same sex-chromosome status as the recipient's surrounding bladder tissue. The nephrogenic adenomas from all 6 female recipients of transplants from female donors showed female chromosomes, and those from the 16 male recipients of transplants from male donors showed male chromosomes. The presence of aquaporin 1, PAX2, and lectin-binding capacity for peanut agglutinin, Lotus tetragonolobus agglutinin, and Sophora japonica agglutinin in nephrogenic adenomas indicated an origin from renal tubular cells.
Methods
Patients
From March 1994 through February 2001, 29 renal-transplant recipients underwent cystoscopy because of hematuria or dysuria, followed by transurethral resection of a nephrogenic adenoma. The patients' ages ranged from 16 to 78 years (median, 49.6). Forty-six nephrogenic adenomas were resected, including 17 recurrent lesions. The interval between renal transplantation and the diagnosis of nephrogenic adenoma was 6 to 120 months (median, 44.6). Table 1 shows the sexes of the donors and recipients of the renal transplants. Post-transplantation renal-biopsy specimens were available from all patients and were obtained before any manifestation of a nephrogenic adenoma. Quantification of the grade of renal rejection in post-transplantation biopsies was performed with the Banff 97 classification.18
|
Specimens from transurethral resections were completely embedded in paraffin and cut into serial sections 4 to 6 µm in thickness. One section was stained with hematoxylin and eosin, and adjacent sections were used for fluorescence in situ hybridization, immunohistochemical analysis, and lectin staining. In selected cases, sections that were analyzed by fluorescence in situ hybridization were restained with hematoxylin and eosin to compare sex-chromosome analysis and histologic details in the same tissue section. Nephrogenic adenoma was diagnosed when typical histologic characteristics (tubular or cystic proliferations, or both, with or without papillary formations)5 were seen.
Fluorescence in Situ Hybridization
Food and Drug Administrationapproved X and Y centromeric probes, directly labeled with Spectrum Orange and Spectrum Green (Vysis), were used according to the protocols of the manufacturer. Because tissue sections and not cytologic preparations were analyzed, some nuclei were cut through, and single cells did not show all sex chromosomes on the sections. We therefore assigned male chromosomal status to nephrogenic-adenoma tissue in a female patient if at least 75 percent of all nuclei of the nephrogenic-adenoma tissue showed one clear-cut Y chromosome, identified as a green, dotlike intranuclear hybridization signal, and no nuclei of the adjacent bladder tissue showed Y chromosomes. Similarly, we assigned female chromosomal status to nephrogenic adenoma tissue in a male patient if the nuclei of the nephrogenic-adenoma tissue did not show Y chromosomes and at least 75 percent of the nuclei of the adjacent bladder tissue showed Y chromosomes. Evaluation of fluorescence signals was performed on transparencies. Three female and three male patients who had nephrogenic adenomas without preceding renal transplantation (five patients with bladder endometriosis and one with prostatic hyperplasia) served as controls for fluorescence in situ hybridization experiments.
Immunohistochemical Analysis and Lectin Staining
Antibodies against aquaporins (water-channel membrane proteins) of the proximal tubules (aquaporin-1, dilution 1:400, Chemicon) and of the collecting ducts (aquaporin-2, 1:50, Chemicon) were used. In addition, Leu-M1 (1:10, Becton Dickinson); epithelial membrane antigen (1:50, Dako); vimentin (1:20, Immunotech); cytokeratins Cam 5.2 (undiluted, Becton Dickinson), CK7 (1:200, Dako), CK8 (1:25, Dako), and CK20 (1:20, Neomarkers); and PAX2, a transcription factor in renal development (1:100, Zymed), were used. Signal detection was performed with the peroxidaseantiperoxidase reaction. Antigen retrieval for aquaporin-1, aquaporin-2, and PAX2 was performed by microwave pretreatment in citrate buffer (pH 6.0) for 20 minutes at 120 W and three times for 5 minutes each at 450 W. Tumor-free renal tissue from five nephrectomy specimens (removed for renal tumor) and tumor-free bladder tissue from five cystoprostatectomy specimens served as controls. Evaluation was performed with use of a semiquantitative scale, with indicating negative, + moderate, ++ strong, and +++ very strong staining. The avidinbiotin method was used to analyze the lectin-binding properties for Lotus tetragonolobus agglutinin, which is specific for proximal tubules and the thin limb of Henle's loop19 (200 µg per milliliter, Sigma); Sophora japonica agglutinin, which is specific for the proximal tubules and collecting ducts19 (50 µg per milliliter, Vector Laboratories); and Arachis hypogaea agglutinin (peanut agglutinin), which is specific for the distal tubules and collecting ducts20 (50 µg per milliliter, Vector Laboratories). Negative controls were achieved by blocking the lectins with their corresponding sugars (
-L-fucose for L. tetragonolobus agglutinin, N-acetyl-
-D-galactosamine for S. japonica agglutinin, and
-D-galactose for peanut agglutinin) before the staining procedure.21
Both pretransplantation and post-transplantation graft-biopsy specimens were available for 13 patients. Immunohistochemical analysis for Cam 5.2 was performed in these 13 patients to demonstrate casts of detached tubular cells within the tubular lumina. The interval between transplantation and graft biopsy ranged from 5 to 721 days. The proliferation index was evaluated by the MIB1 antibody (1:50, Dako) to detect Ki-67 antigen in the nuclei of the tubular cells. In addition, the expression of PAX2 was investigated. The results of pretransplantation and post-transplantation graft biopsies were compared.
Statistical Analysis
Medians and standard deviations are reported. The WilcoxonMannWhitney U test and Fisher's exact test were used.
Results
Fluorescence in Situ Hybridization
In 14 of 14 nephrogenic adenomas from female recipients of renal transplants from male donors, a male chromosomal status was demonstrated within the epithelial component (Figure 1). In 10 of 10 nephrogenic adenomas from male recipients of renal transplants from female donors, a female chromosomal status was demonstrated (Figure 2). By contrast, the stromal component of all nephrogenic adenomas had the sex-chromosome status of the recipient. In 6 of 6 nephrogenic adenomas from women with renal transplants from female donors, a female chromosomal status was demonstrated, and in 16 of 16 nephrogenic adenomas from men with renal transplants from male donors, a male chromosomal status was demonstrated. In each nephrogenic adenoma from the renal-transplant recipients, the epithelial component had the same sex-chromosome status as the donor (Table 2). The control group of nephrogenic adenomas not associated with a kidney graft had the expected sex-chromosome status.
|
|
|
The results of analysis of the nephrogenic adenomas are summarized in Table 2. Epithelial components of the nephrogenic adenomas showed immunophenotypic characteristics of renal tubular cells: prominent staining for aquaporin-1 (Figure 3D), epithelial membrane antigen, and cytokeratins Cam 5.2 and CK7 and moderate reactivity for CK8, Leu-M1, and vimentin. CK20, which is expressed in urothelial cells, and aquaporin-2, which is confined to collecting ducts, were not detected. In addition, PAX2, which is usually found only during nephrogenesis, was found in epithelial components of nephrogenic adenomas, but not in the bladder urothelium (Figure 3E). Peanut agglutinin staining was found in all 46 nephrogenic adenomas, but not in the adjacent urothelium (Figure 3F). L. tetragonolobus agglutinin was found focally in 17 of the 46 nephrogenic adenomas (37 percent), but not in the adjacent urothelium. Focal staining was also found with S. japonica agglutinin in 26 of 46 (57 percent); however, the adjacent urothelium was positive.
|
Comparison between Patient Groups
No significant differences were found between female and male recipients in age, frequency of graft rejection, interval from transplantation to the diagnosis of nephrogenic adenoma, or frequency of recurrence. We also found no significant difference in these variables between patients who had received basic antirejection therapy and those also treated with antithymocyte globulin.
Discussion
Nephrogenic adenoma was initially described as a benign hamartomatous bladder lesion by Davis in 1949.17 The term "nephrogenic adenoma" was introduced in 1950 by Friedman and Kuhlenbeck22 because of the lesion's histologic similarities to renal tubules. Although nephrogenic adenomas are rare, a high incidence has been noted in recipients of renal transplants.7,8,9,10,11,12,13,14,15,16 Other possible predisposing factors are genitourinary trauma,23,24 mechanical irritation or genitourinary surgery,4,10,14,25,26,27,28,29,30,31 local chemical irritation,14,26,32,33,34 irradiation, and chronic inflammation. The idea that nephrogenic adenomas are metaplastic proliferations of resident urothelial mucosa is widely accepted3,11,25,35,36 but has never been proved. Redondo Martinez and Rey Lopez37 and Strand and Alfert38 reported nephrogenic adenomas after cystectomy that were located in a sigmoid neobladder and an ileal conduit, findings that argue against metaplasia of resident urothelial mucosa.
In our fluorescence in situ hybridization experiments, we found that nephrogenic adenomas from women who received renal transplants from male donors had a male sex-chromosome status (XY), whereas the surrounding bladder tissue had a female sex-chromosome status (XX). The reverse was found in nephrogenic adenomas from men with transplants from female donors. These results demonstrate that nephrogenic adenomas in patients with renal transplants originate from the graft.
Immunohistochemical and lectin staining was performed to characterize the cell type in the nephrogenic adenomas. Aquaporin-1, a membrane protein of water channels, is expressed in the proximal renal tubule and the descending thin limb of Henle's loop, but not in the other nephron segments, the collecting ducts, and the urothelial cells.39,40 Extrarenal aquaporin-1 has been reported in capillary endothelium and red cells.39,40 Aquaporin-2 is selectively expressed in renal collecting ducts.39,40 We found prominent staining for aquaporin-1 but not for aquaporin-2 in nephrogenic adenomas. We also investigated the epithelial cells of nephrogenic adenomas for antigens that are not specific to the kidney. Leu-M1 is expressed in the proximal tubular cells but not in the distal tubules, collecting ducts, and urothelial cells; cytokeratins CK7 and CK8 and epithelial membrane antigen occur in the renal tubules, collecting ducts, and urothelial cells.40,41,42 A moderate-to-strong immunoreactivity with antibodies against these antigens was shown. CK20, which is found in urothelial cells43 but not in nephrons and collecting ducts, was not detected in the nephrogenic adenomas of our patients.
Binding of L. tetragonolobus agglutinin and S. japonica agglutinin, which are typically found in mature proximal tubular cells, was demonstrated in many of the nephrogenic adenomas we studied. Binding of peanut agglutinin, which is usually found in mature distal tubules, has been reported in nephrogenic adenomas,44 and was found in all the adenomas we studied. This finding suggests that the adenomas arise from an early developmental stage of the renal tubules, since free peanut-agglutinin receptors occur in embryonal renal mesonephric and metanephric tubules.44 All our immunohistochemical and lectin-binding analyses suggest that the adenomas originate from renal tubular cells.
The PAX genes encode a family of transcription factors that act as key regulators during organogenesis.45 PAX2 and PAX8 seem to be required for the development and proliferation of the renal tubules. After renal differentiation, PAX2 expression is down-regulated in mature tissue, and usually no expression is seen in the adult kidney.45 Expression of PAX2 in some biopsy specimens from the transplanted kidneys of our patients with nephrogenic adenoma might indicate an activation of tubular precursor-cell antigens. The presence of aquaporin-1 in nephrogenic adenomas, together with PAX2, peanut agglutinin, and L. tetragonolobus agglutinin, argues against an origin in urothelial cells. Aquaporin-1 is strictly confined to renal tubules and has never been reported to be expressed in urothelium; furthermore, urothelium does not bind peanut agglutinin and L. tetragonolobus agglutinin.
Detachment of viable renal tubular cells probably occurs throughout life. A high incidence of detachment is possible in renal diseases and hypoxic conditions.46,47,48,49 Seeding and implantation of neoplastic or non-neoplastic cells in distant tissues are known in the literature, as in the case of endometriosis.50
It seems that in certain circumstances, urothelial mucosa is able to incorporate renal cells, but how implantation of tubular cells after seeding is mediated remains to be clarified in further investigations. Allogeneic renal tubular cells, already implanted, have to survive in the environment of the recipient's connective tissue. We looked at some nephrogenic adenomas for evidence of rejection but found none (data not shown).
Our results show that nephrogenic adenomas in renal-transplant recipients originate not from the recipient's bladder urothelial mucosa, but from renal tubular cells of the transplanted kidney. In our opinion, the terms "nephrogenic adenoma" and "nephrogenic metaplasia" are imprecise; we think these lesions are a kind of renal tubular satellite. Seeding, implantation, and growth of renal cells in urothelial mucosa may be enhanced by increased loss of tubular cells, trauma, infection, and immunosuppression.
Supported by the Fonds zur Förderung der Wissenschaft und Forschung (SFB 05, project 007).
Source Information
From the Department of Clinical Pathology and Center of Excellence in Clinical and Experimental Oncology (P.R.M., A.H., G.K., H.R., D.K., M.S.), the Department of Internal Medicine III (B.W.), and the Department of Urology (C.K.), University of Vienna General Hospital; and the Departments of Nephrology (R.S., F.T.M.) and Urology (R.A.-M., O.Z.), Wilhelminen Hospital all in Vienna, Austria.
Drs. Mazal and Schaufler contributed equally to the article.
Address reprint requests to Dr. Susani at the Department of Clinical Pathology, University of Vienna General Hospital, Währinger Gürtel 18-20, A-1090 Vienna, Austria, or at martin.susani{at}akh-wien.ac.at.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |