Adjuvant Role of p53 Immunostaining in Detecting BK Viral Infection in Renal Allograft Biopsies

Size: px

Start display at page:

Download "Adjuvant Role of p53 Immunostaining in Detecting BK Viral Infection in Renal Allograft Biopsies"

Annabelle Pearson
5 years ago
Views:

1 Available online at Annals of Clinical & Laboratory Science, vol. 40, no. 4, 2010 Adjuvant Role of p53 Immunostaining in Detecting BK Viral Infection in Renal Allograft Biopsies Wendy N. Wiesend, 1 Raviparasenna Parasuraman, 1 Wei Li, 2 Maryam A. Farinola, 1 Michele T. Rooney, 1 Sharon K. Hick, 1 Dilip Samarapungavan, 1 Steven R. Cohn, 1 Gampala H. Reddy, 1 Leslie L Rocher, 1 Francis Dumler, 1 Fan Lin, 3, and Ping L. Zhang 1 1 Departments of Anatomic Pathology, Nephrology, and Transplant Surgery, William Beaumont Hospital, Royal Oak, Michigan; 2 Department of Pathology and Laboratory Medicine, University of Texas Health Science Center, Houston, Texas; 3 Division of Laboratory Medicine, Geisinger Medical Center, Danville, Pennsylvania Abstract. BK virus infection is a significant threat to renal transplant outcome. Detecting viral infection in renal transplant biopsies using SV40 staining is less than ideal. SV40 antibody reacts with the large T-antigen of BK virus only at the early phases of infection and can miss cells in later stages of infection. As p53 is upregulated during both early and late phases of infection, this study set out to determine whether p53 staining could improve detection of BK virus infection in renal transplant patients. The control group consisted of 16 renal allograft biopsies without histologic evidence of BK virus infection, while the BK group consisted of 15 renal allograft biopsies with histologic evidence of BK virus infection. The biopsies from both groups were immunohistochemically stained with both SV40 and p53 antibodies. Dual staining with both markers was also performed to identify their nuclear co-localization. In the BK group, the percent of p53 staining (16.6 ± 4.8 %) was significantly higher than the percent of SV40 staining (5.4 ± 2.7%). BK virus infected cells revealed a unique p53 immunostaining pattern (strong nuclear staining with a central halo). Co-localization of SV40 and p53 was identified in cells that had characteristic nuclear features of BK virus infection by histology. The sensitivity and specificity for using p53 staining to identify BK infected cells was 92% and 86 %, respectively. In conclusion, p53 staining detects a higher percentage of BK virus infected cells than SV40 staining alone. Thus, for diagnosis of BK virus infection in renal allograft biopsies, p53 staining is a sensitive and specific method when used along with SV40 staining. Keywords: BK virus, p53, SV40, immunohistochemical staining, kidney transplants Introduction In recent years, cytomegalovirus (CMV) infection in renal allograft recipients has declined, partially due to prophylactic therapy, but BK virus nephropathy (BKVN) has increased in prevalence [1]. The incidence of BKVN ranges up to 10% in renal transplant biopsies [2]. This increase is associated, in part, with the introduction of new, potent Address correspondence to Ping L. Zhang, M.D., Ph.D., Dept. of Anatomic Pathology, William Beaumont Hospital, 3601 West 13 Mile Road, Royal Oak, MI , USA; tel ; fax ; ping.zhang@ beaumont.edu. immunosuppressive drugs to prevent acute cellular rejection. Clinical symptoms of BK virus infection can be insidious, or can resemble acute rejection. BKVN can have devastating effects on the renal allograft, causing renal allograft loss in 10 to >80% of cases [2]. Definitive diagnosis of BKVN requires demonstration of viral cytopathic changes in renal tissue and confirmation by immunohistochemical staining. The most widely used immunohistochemical stain is an antibody against the large T-antigen of Simian Virus 40 (SV40), which identifies BK virus infections due to cross-reaction [3]. Although renal biopsy is the gold standard in /10/ $ by the Association of Clinical Scientists, Inc.

2 p53 immunostaining in detecting BK infection in renal allograft biopsies 325 diagnosis of BKVN, diagnosis can be difficult in some cases because BK virus may be focal or patchy. Since BK virus infection is unevenly distributed in the renal core, there may a 27-35% chance to miss it if the biopsy is small or consists of only a single core [1]. Since SV40 is expressed only at early stages of BK virus infection, BK viral infection may be missed at later stages if one relies only on immunohistochemical staining against SV40 [1]. In vitro studies of BK virus infected cells have shown that the large T-antigen binds to p53 protein, causing p53 to accumulate in the nuclei of these cells [4,5]. In humans, p53 up-regulation is present at both the early and the late phases of BK virus infection [6]. It may therefore be helpful to use immunohistochemical staining for p53 along with SV40 to improve detection of BKVN in renal allograft biopsies. Fine granular expression of p53 has been seen in reactive nuclei of renal tubular epithelium during acute tubular injury [7]. However, the importance of nuclear p53 staining in assisting the diagnosis of BK infection in human renal allograft biopsies has not been addressed. This study was performed to characterize the immunohistochemical pattern of nuclear p53 staining in BK virus infected cells and to show how p53 staining improves the detection rate of BK virus infection in renal allograft biopsies. Methods Design. We conducted a retrospective study of 31 renal allograft core biopsies. Each biopsy was performed due to suspected graft dysfunction. The control group consisted of 16 renal allograft biopsies without histologic evidence of BK virus infection. Diagnoses in this group ranged from acute cellular rejection to acute tubular necrosis. The BK group consisted of 15 renal allograft biopsies with histologic evidence of BK virus infection. Stains. The renal core biopsies were routinely formalin-fixed and paraffin-embedded. The tissue blocks were cut into 4-µm sections and underwent hematoxylin and eosin (H&E) staining (3 sections), periodic acid-schiff (PAS) staining (3 sections), and Masson trichrome staining (3 sections). A human autopsy transplant kidney with BK virus infection and human brain tissue with JC virus infection were used as positive controls. For each block, one 4-μm section was dewaxed in xylene and rehydrated with graded ethanols to water. Antibody against BK virus (anti-sv40 T antigen) (monoclonal mouse SV40 antibody at 1:200 dilution) and p53 antibody (pre-diluted monoclonal p53 antibody) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) and DakoCytomation (Carpinteria, CA), respectively. The Table 1. Results of SV40 and p53 staining in the control group and BK group of renal allograft biopsies. Control SV40 P53 Final diagnosis BK group SV40 P53 Final diagnosis group staining staining staining staining intensity intensity intensity intensity score score score score Ia 1 3 (5%) 3 (10%) BK Ia 2 3 (5%) 3 (5%) BK Ia 3 3 (1%) 3 (1%) BK Ia 4 1 (1%) 3 (10%) BK Ia 5* 0 (0%) 3 (20%) BK Ib 6 2 (5%) 3 (25%) BK ATN 7 2 (1%) 3 (10%) BK IIa 8 3 (5%) 3 (10%) BK BC 9 3 (5%) 3 (20%) BK ATN 10 3 (1%) 3 (30%) BK Ia & AMR 11 3 (5%) 3 (20%) BK Rec IgA 12 1 (1%) 3 (1%) IIb & BK Ia 13 # focal 1 (1%) 0 (0%) suspicious BC 14 3 (40%) 3 (70%) BK Ib 15 2 (1%) 3 (40%) BK Ib Ia = T-cell mediated rejection Ia; Ib = T-cell mediated rejection 1b; IIa = T cell mediated rejection IIa; BC, borderline changes; Rec IgA = recurrent IgA nephropathy; ATN = acute tubular necrosis; AMA = antibody mediated rejection. * and # : See the results section for information about these two cases. The percentage of nuclei with positive staining is given in parentheses.

3 326 Annals of Clinical & Laboratory Science, vol. 40, no. 4, 2010 Fig. 1. SV40 and p53 stains in control and BK groups. (A) Reactive enlarged epithelial cells in the control group stained negative for SV40 (B) but weakly positive for p53. (C) Enlarged cells with histologic features of BK virus infection in the BK group show focal positive staining for SV40 (C) but strong positivity for p53 in a higher percentage of cells. Arrows indicate positive staining. Magnification for A-D is 400x. slides were treated with a 20 min heat-induced antigen retrieval protocol (Target Retrieval Solution, DakoCytomation). Each primary antibody was applied for one hr. The secondary antibody was applied for 30 min (peroxidase labeled goat anti-mouse, Env+ kit, DakoCytomation). DAB solution (3,3 -diaminobenzidine, Env+ kit, DakoCytomation) was applied for ten min to produce a brown color reaction for either BK or p53 staining. To demonstrate nuclear co-staining for both p53 and SV40, SV40-stained sections (nuclear staining, 4 cases) were co-stained with p53 primary antibody (diluted 1:100, Biocare Medical, Concord, CA) for 30 min. The slides were rinsed with TBST, and MACH-2 goat anti-mouse alkaline phosphatase secondary antibody (Biocare Medical) was applied for 30 min. Slides were rinsed with TBST, and Vulcan Fast Red Chromagen Substrate (Biocare Medical) was applied for 10 min to produce positive p53 staining in pink color. The slides were counterstained with Gill-II hematoxylin and coverslips were applied. Nuclear SV40 staining was brown whereas nuclear staining of p53 was pink; thus, co-stained nuclei for both SV40 and p53 showed purple nuclear staining. Analysis and statistics. SV40 and p53 staining intensity and percentages of positive staining in the renal parenchyma were evaluated manually. The staining intensity scores of targeted and control epithelial cells were graded from 0 to 3+ (0, no staining, or focal weak fine granular staining; 1+, weak fine granular staining; 2+, moderate granular staining; and 3+, strong granular staining). Results were expressed as mean ± SE. Data for SV40 staining and p53 staining were compared using unpaired t test; a p value <0.05 was considered statistically significant. Results During acute tubular injury in the control group, reactive epithelial cells were enlarged without intranuclear inclusions and might stain weakly positive for p53 (Fig. 1A), as shown by us previously [6]. These cases stained negatively for SV40, however (Fig. 1B). BK virus infected cells were

p53 immunostaining in detecting BK infection in renal allograft biopsies 327 often enlarged and could be confirmed by positive SV40 staining (Fig. 1C).

4 p53 immunostaining in detecting BK infection in renal allograft biopsies 327 often enlarged and could be confirmed by positive SV40 staining (Fig. 1C). At the same location from the same case (BK case #10 in Table 1), more BK infected nuclei showed strongly positive staining for p53, often with peripheral nuclear staining and central nuclear halos (Fig. 1D). Table 1 lists all of the control and BK cases. Control cases all showed negative SV40 staining with various intensity of p53 staining in the tubular epithelium (from 0 to 2+ staining). There were various diagnoses in the control group, ranging from acute cellular rejection to acute tubular necrosis. All cases in the BK virus group showed 3+ nuclear p53 staining with the previously mentioned unique staining pattern. Case #5 in the BK group was a tiny biopsy that showed focal and strongly positive p53 staining, but the atypical nuclei were lost on the SV40 stained section. However, one week later, a urine cytology specimen confirmed BK virus infection. Case #13 in the BK group had positive SV40 staining in two normal sized tubular epithelial cells, but failed to show strongly positive p53 staining on a deep level Fig. 2. Top panel: Dual stains for SV40 and p53 in renal biopsy with BK virus section and thus failed to support a infection in renal tubules and a vessel. Some nuclei stained for only SV40, definitive diagnosis of BK infection. giving a brown color (a, initial early phase). Some nuclei showed positive dualthis case was called suspicious for BK staining for SV40 and p53, giving a purple color (b, early phase). Some enlarged virus infection. In the remaining cases nuclei stained only for p53, giving a pink color (c, later phase). Lower panel. in the BK group, the percentage of Together with the results from [6] and [8], this diagram illustrates the life cycle of a BK virus in relation to the double-staining patterns of SV40 and p53 seen p53 stained nuclei (with histologic in our study. The percentages of time at the initial early phase (a), the remaining features of BK virus infection), early phase (b), and the late phase (c) are indicated in the diagram. appeared greater than the percentage of SV40 stained nuclei (Table 1). In the BK group, positively for both SV40 and p53 (purple color), the mean percentage of p53 stained nuclei (16.6 ± whereas other nuclei with histologic features of BK 4.8 %) was higher than the mean percentage of virus infection stained only for p53 (Fig. 2). SV40 stained nuclei (5.4 ± 2.7%) (p <0.05). To confirm the presence of concurrent staining Discussion of SV40 and p53, we performed a dual staining procedure for p53 and SV40. Some nuclei stained This study was performed to characterize the only for SV40 (brown color) (Fig. 2). Some nuclei immunohistochemical staining pattern of p53 in with histologic features of BK infection stained BK virus infected cells and to illustrate how p53

5 328 Annals of Clinical & Laboratory Science, vol. 40, no. 4, 2010 staining can improve the detection rate of BK virus infection in renal allograft biopsies. At our institutions, BK virus infection is identified in approximately 2% of renal allograft biopsies. Using paraffin-embedded renal allograft biopsies, we found a unique staining pattern of p53 in BK virus infected cells (strong granular nuclear staining with a central halo). For the first time, dual staining of SV40 and p53 in BK virus infected cells was illustrated using the immunohistochemical method in paraffin-embedded tissues. This confirmed previous studies that demonstrated co-expression of SV40 and p53 in BK virus infected cells using an immunofluorescent method on frozen sections of tissue [6]. Our study highlighted three staining patterns in BK virus infected cells that were stained for both SV40 and p53 (Fig. 2, lower panel). This observation can be explained by the findings of Low et al [8] and Seemayer et al [6], who estimated the length of the BK virus life cycle in human kidney epithelial cells. There is a 24 to 36 hr interval between virus reception in the cell and the onset of viral transcriptional activity [8]. This period is followed by 3 hr of T-antigen expression at the initial early phase (SV40 positive only), 7 hr of early phase expression when T-antigen and p53 are both expressed (SV40 and p53 co-staining), followed by 8 hr of late phase when VP1 and p53 are co-expressed (p53 staining only), before virusdependent cell lysis [6]. It was estimated that the BK virus life cycle ranges from 42 to 54 hr [6]. The life cycle of BK virus infection with corresponding dual-staining patterns seen in our study is depicted in Fig. 2. Using an immunofluorescent method, Seemayer et al [6] observed co-expression of VP1 and p53 at the late phase, but we were unable to test for co-expression of VP1 and p53 in BK virus infected cases because an antibody for VP1 antigen is not commercially available. In the BK group, we found that the percentage of p53 positive cells was much higher than the percentage of SV40 positive cells in two thirds of the cases (10/15). As the BK virus life cycle evolves, proteins (T-antigen and VP1 protein) of BK virus may co-exist in the infected renal tubular cells. SV40 staining is a standard assay for BK virus infection in most pathology laboratories. SV40 staining alone, however, can miss BK virus infected cells at late phases of infection, when VP1 protein dominates. Thus, p53 staining, used together with SV40 staining, can highlight more BK virus infected cells in renal biopsies. BKVN in renal transplants can be histologically classified into three patterns: BKVN A through BKVN C [2,3,9]. These patterns are based on viral cytopathic changes for interstitial fibrosis, tubular atrophy, and inflammation. BKVN A shows minimal to mild viral cytopathic changes, and insignificant inflammatory infiltrates, tubular atrophy, and fibrosis. BKVN B shows mild to severe viral cytopathic changes, significant inflammatory infiltrates (moderate to severe), but only mild tubular atrophy and fibrosis. BKVN C shows moderate to severe tubular atrophy and interstitial fibrosis, and variable viral cytopathic changes and inflammatory infiltrates [2,3,9]. In our study, in 10/15 cases in the BK group in which a high percentage of p53 positivity was seen, p53 staining helped to identify the viral load in BKVN B. In 3 of 15 BK cases (#2, #3, and #12), p53 helped to confirm the presence of BK virus infection when SV40 staining was focal (stage A of BKVN). In two cases of the BK group, p53 staining was not helpful in establishing the presence or absence of BK virus infection. Identification of BKVN in renal allograft biopsies indicates that interstitial nephritis is secondary to BK virus infection, even when some tubulitis is present. In the absence of BK virus infection, the interstitial nephritis is usually due to acute cellular rejection. Our study indicates that p53 was very sensitive (92%) for identifying BK virus infected cells. The specificity of using p53 to detect BK virus infected cells in was 86%, taking into account the staining intensity and unique staining pattern (central halo). Although the gold standard for diagnosis of BKVN in renal allografts is immunohistochemical SV40 staining, there are pitfalls to this method. BK virus infection can be missed if the renal biopsy is too small or if the biopsy does not include the renal medulla, since the BK virus is found in medullary collecting ducts as an ascending infecting pattern more often than in cortical tubules [10]. Based on our study, p53 staining can be very useful in screening for minimum infection by BK virus, particularly when SV40 staining is not immediately available.

6 p53 immunostaining in detecting BK infection in renal allograft biopsies 329 Previous studies showed that BKVN affects up to 10% of renal transplant biopsies [2,11]. Since there is no specific antiviral therapy, the goal in BKVN is early diagnosis and reduction of immunosuppression to eliminate the virus without increasing the risk of rejection. Various strategies of immunosuppression reduction with or without investigational agents such as leflunomide or cidofovir have been reported [12]. Discontinuation of calcineurin inhibitors (CNI) with maintenance of sirolimus and low-dose prednisone was associated with a 72% reduction in risk of graft loss while avoiding acute rejection [13]. Leflunomide therapy with mycophenolic acid discontinuation and CNI reduction in 26 patients has been reported with some success [13]. In another study, neither iv immunoglobulin nor cidofovir was independently associated with graft survival in addition to immunosuppression reduction [14]. Despite various methods to minimize immunosuppression, renal allograft outcome is still sub-optimal. In summary, our study indicates that there is a unique p53 immunostaining pattern (strong nuclear staining with central halo) in BK virus infected cells. SV40 staining remains as a gold standard to confirm the BK virus infection, since the antibody against SV40 is a more specific than the antibody against p53. However, staining for p53 along with SV40 increases the chance of detecting BK virus in the renal allograft biopsies (Fig. 2). Thus, p53 staining can play a sensitive and specific role as an adjuvant to SV40 staining for the pathologic diagnosis of BK viral infection in renal allograft biopsies. References 1. Bonvoisin C, Weekers L, Xhingnesse P, Grosch S, Milicevic M, Krzesinski J-M. Polyomavirus in renal transplantation: a hot problem. Transplantation 2008; 85: S42-S Hirsch HH, Brennab DC, Drachenberg CB, Ginevri F, Gordon J, Limaye AP, Mihatsch MJ, Nickeleit V, Ramos E, Randhawa P, Shapiro R, Steiger J, Suthanthiran M, Trofe J. Polyomavirus associated nephropathy in renal transplantation: interdisciplinary analyses and recommendations. Transplantation 2005:79: Drachenberg CB, Hirsch HH, Romos E, Papadimitriou JC. Polyomavirus disease in renal transplantation: review of pathological findings and diagnostic methods. Hum Pathol 2005;26: Shivakumar CV, Das GC. Interaction of human polyomavirus BK with the tumor-suppressor protein p53. Oncogene 1996:13: Harris KF, Christensen JB, Imperiable MJ. BK virus large T antigen: interactions with the retinoblastoma family of tumor suppressor proteins and effects on cellular growth control. J Virol 1996;70: Seemayer CA, Seemayer NH, Durmuller U, Gudat F, Schaub S, Hirsch HH, Mihatsch MJ. BK virus large T and VP-1 expression in infected human renal allografts. Nephol Dial Transplant 2008;23: McLaren BK, Zhang PL, Herrera GA. P53 protein is a reliable marker in identification of renal tubular injury. Appl Immunohistochem Mol Morphol 2004;12: Low J, Humes HD, Szczypka M, Imperiale M. BKV and SV40 infection of human kidney tubular epithelial cells in vitro. Virology 2004;323: Drachenberg CB, Papadimitriou JC, Hirsch HH, Wali R, Crowder C, Nogueira J, Cangro CB, Mendley S, Mian A, Ramos E. Histological patterns of polyomavirus nephropathy: correlation with graft outcome and viral load. Am J Tranplant 2004;4: Meehan SM, Kraus MD, Kadambi PV, Chang A. Nephron segment localization of polyoma virus large T antigen in renal allografts. Human Pathol 2006;37: Gardner SD, Mackenzie EF, Smith C. Prospective study of the human polyomavirus BK and JC and cytomegalovirus in renal transplant recipients. J Clin Pathol 1984; 37: Weiss AS, Gralla J, Chan L, Klem P, Wiseman AC. Aggressive immunosuppression minimization reduces graft loss following diagnosis of BK virus-associated nephropathy: a comparison of two reduction strategies. Clin J Am Soc Nephrol 2008;3: Josephson MA, Gillen D, Javaid B, Kadambi P, Meehan S, Foster P, Harland R, Thistlethwaite RJ, Garfinkel M, Atwood W, Jordan J, Sadhu M, Millis MJ, Williams J. Treatment of renal allograft polyoma BK virus infection with leflunomide. Transplantation 2006;81: Wadei HM, Rule AD, Lewin M, Mahale AS, Khamash HA, Schwab TR, Gloor JM, Textor SC, Fidler ME, Lager DJ, Larson TS, Stegall MD, Cosio FG, Griffin MD. Kidney transplant function and histological clearance of virus following diagnosis of polyomavirusassociated nephropathy (PVAN). Am J Transplant 2006; 6:

Clinical Grand Rounds. May 2016

Clinical Grand Rounds. May 2016 Clinical Grand Rounds May 2016 The Case 51M African American w/ HTN, ESRD (on HD for 13 years) who underwent a DDRTx in June 2015. Complications: BK viremia with AKI in Jan 2016---MMF held and tacro dose