Journal of Medical Virology 86:1134–1144 (2014)
High-Risk Human Papillomavirus Infection in Different Histological Subtypes of Renal Cell Carcinoma Ali Farhadi,1 Abbas Behzad-Behbahani,2* Bita Geramizadeh,3 Zamberi Sekawi,1 Marjan Rahsaz,3 and Sedigheh Sharifzadeh2 1
Department of Medical Microbiology and Parasitology, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, Serdang, Malaysia 2 Diagnostic Laboratory Sciences and Technology Research Center, School of Paramedical Sciences, Shiraz University of Medical Sciences, Shiraz, Iran 3 Department of Pathology, Shiraz Transplant Research Center, Nemazee Hospital, Shiraz University of Medical Sciences, Shiraz, Iran
Limited data exist regarding whether a highrisk human papillomavirus (HR-HPV) infection increases the risk of developing renal cell carcinoma. The aim of this study was to investigate whether HPV infection has a role in the pathogenesis or development of a certain histological subtype of renal cell carcinoma. Formalin-fixed paraffin-embedded (FFPE) specimens of 122 patients with histopathologically proven renal cell carcinoma and their respective peritumoral tissues were examined. The presence of HPV-DNA was determined by a combination of MY/GPþ consensus primers and HPV-16/18 type specific nested PCRs followed by direct sequencing. Catalyzed signalamplified colorimetric in situ hybridization (CSAC-ISH) technique was applied to determine the physical status of viral genome. The expression of p16INK4a and HPV L1 capsid proteins was evaluated using immunohistochemistry. HPV genome was detected in 37 (30.3%) tumor specimens and their four (4.1%) corresponding peritumoral tissues. HPV-18 was the most common viral type identified followed by HPV-16 and 58. Immunoexpression of p16INK4a was detected in 24 (20.3%) cases. Data analysis showed a significant correlation between p16INK4a expression and the presence of HR-HPV DNA (P < 0.001). CSAC-ISH analysis confirmed HR-HPV infection in 45% of tumors, which were previously tested positive for HPV-DNA. Diffuse signal pattern was identified in 15 (83.3%) samples whereas a mixed pattern of diffuse and punctate signals was only detectable in three cases. The results indicate an association of HR-HPV types with renal cell carcinoma. It is proposed that HPV infection in high-grade tumors might precede C 2014 WILEY PERIODICALS, INC.
disease progression in a number of tumors, particularly of the papillary subtype. J. Med. Virol. 86:1134–1144, 2014. # 2014 Wiley Periodicals, Inc.
KEY WORDS:
human papillomavirus; renal cell carcinoma; in situ hybridization; p16INK4a immunohistochemistry
INTRODUCTION In recent decades, the occurrence of renal cell carcinoma has increased, accounting for about 3.8% of adult malignancies and approximately 90% of the entire renal neoplasms [Jemal et al., 2010]. Generally, it is believed that various subtypes of renal cell carcinoma arise from different specialized epithelial cells along the nephron [Shanks, 1999]. Renal cell carcinoma is regarded as a highly aggressive tumor. Metastasis can be found in a third of patients during diagnosis, which leads to over 40% mortality. However, for localized renal cell carcinoma, surgery is the treatment of choice [Janzen et al., 2003; Lam et al., 2005]. Recent advances in renal cell molecular biology Grant sponsor: Shiraz University of Medical Sciences; Grant number: 87-105-2009. Conflicts of interest: none. Correspondence to: Abbas Behzad-Behbahani, PhD, Diagnostic Laboratory Sciences and Technology Research Center, School of Paramedical Sciences, Shiraz University of Medical Sciences, Shiraz, Iran. E-mail: [email protected] Accepted 30 October 2013 DOI 10.1002/jmv.23945 Published online 2 April 2014 in Wiley Online Library (wileyonlinelibrary.com).
HPV Infection in Renal Cell Carcinoma
and genetics have associated the disease with various hereditary and non-hereditary risk factors; however, they cannot explain all renal cell carcinoma cases. Based on the current understanding, it has been estimated that some 15% of the global cancer burden can be linked to oncogenic tumor viruses [Zur Hausen, 2001]. While epitheliotropic high-risk human papillomavirus (HPV) is the established aetiological agent of cervical cancer, several studies have suggested the involvement of HPV in other human malignancies such as breast, respiratory tract, bladder, colorectal, and oesophageal cancer [Walboomers et al., 1999; Syrja¨nen, 2002; Damin et al., 2004; Barghi et al., 2005; Giuliani et al., 2008; Klein et al., 2009]. Although the role of HPV in development of renal cell carcinoma has been investigated less rigorously, research has indicated conflicting results. The possible involvement of HPV in renal cell carcinoma tumors was first hypothesized by Rotola et al. [1992]. Since then, a few reports with conflicting results have been published on this subject. On the whole, a comprehensive investigation of the relationship between HPV and different histologic subtypes of renal cell carcinoma is still sporadic. The related literature on malignant lesions of kidney is lacking, inconsistent, and only limited to certain histologic subtypes of renal cell carcinoma cases. Thus, this study was designed to include a large number of renal cell carcinoma tumors with different histological grades and subtypes to examine the presence of HRHPV infection and its association with clinicopathological features of renal cell carcinoma. MATERIALS AND METHODS Sample Selection and Tumor Classification Surgical report profiles of patients who underwent surgery in Namazi hospital, Shiraz University of Medical Sciences, Iran from January 2004 to December 2009 were reviewed and histologically confirmed renal cell carcinoma cases were selected. There were no restrictions on age, gender, ethnicity, or cancer stage at recruitment. Hematoxylin and eosin (H&E) slides were retrieved and classification of different histologic subtypes of renal cell carcinoma tumors was reconfirmed by pathologist according to the Heidelberg classification system [Kovacs et al., 1997]. Histopathologic grading of the nuclei of the tumor cells was assessed using the four-group Fuhrman nuclear classification system [Fuhrman et al., 1982]. The primary tumor staging was performed according to the recent version of the 2010 TNM staging of the American Joint Committee on Cancer (AJCC) and the Union for International Cancer Control (UICC) classification system [Guinan et al., 1997; Edge and Compton, 2010]. Formalinfixed paraffin-embedded (FFPE) tissue blocks of related slides were collected and examined to have sufficient tissue to perform all experiments; disqualified cases from each group were excluded from the study. Of 138 renal cell carcinoma tumors in this
1135
period, 127 samples were selected as the best representative cases. In addition, 104 specimens of their corresponding peritumoral kidney tissue and 23 samples of normal kidney tissue from patients with blunt renal injuries were available for this study as well. Ethics Approval The study was approved by the Medical Ethics Committee of the Shiraz University of Medical Sciences, Iran (nr 1388/321), and by the Medical Research Ethics Committee, the Faculty of Medicine and Health Sciences, Universiti Putra Malaysia (UPM/FPSK/ PADS/T7-MJKEtikaPer/F01 (JMPP_Aug (08) 01). Sample Preparation and DNA Extraction For each case, four 5 mm thick slices were cut and collected carefully in autoclaved Eppendorf microcentrifuge tubes (1.5 ml). Precautions were taken to avoid contamination among specimens including cutting one case at a time, changing the microtome blade before cutting each new case and cleaning the microtome and work area thoroughly with ethanol between each case. Additionally, paraffin blocks without tissue were sectioned after cutting each real specimen and included in every step of the experiment as negative controls of DNA extraction procedures. Genomic DNA was extracted from the FFPE tissue sections using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. Extracted DNA was eluted with 50 ml ATE buffer and stored at 70˚C until analyzed. The total amount of DNA was quantified using a NanoDrop (ND-1000) spectrophotometer (peQLab Biotechnologie, Erlangen, Germany). The quality of the extracted DNA for each sample was evaluated by PCR amplification using b-globin gene specific primers, PC03 (50 -ACACAACTGTGTTCACTAGC-30 ) and PC04 (50 -CAACTTCATCCACGTTCACC-30 ) generating a 110 bp PCR product as described by Saiki et al. [1988]. Only b-globin positive samples were included in further analyses. Broad Spectrum PCR Assays and Sequencing To enhance the sensitivity of viral detection, a nested general-primer-mediated PCR assay using MY09/MY11 degenerated primer set [Manos et al., 1989] followed by GP5þ/GP6þ consensus primers [de Roda Husman et al., 1995] was carried out for detection of 23 mucosotropic HPV genotypes, as shown in Table I. All HPV positive tissue specimens were amplified and analyzed in duplicate to ensure consistency and reliability of the results. To determine the HPV genotypes, the positive PCR products were submitted for automated DNA sequencing (MilleGen, Labe`ge, France) in the presence of GP5þ and GP6þ as both directions sequencing primers. The obtained sequences were aligned with documented HPV sequences that were available in GeneBank J. Med. Virol. DOI 10.1002/jmv
1136
Farhadi et al. TABLE I. Primer Names and Sequences Used for Detection of HPV-DNA in this Study
Primer name
Sequence (50 –30 )a
Nucleotide position
Targetb
Amplicon size (bpc)
Reference
MY 11 MY09 GP5þ GP6þ 16E6F1 16E6R1 16E6F2 16E6R1
GCMCAGGGWCATAAYAAYTGG CGTCCMARRGGAWACTGATC TTTGTTACTGTGGTAGATACTAC GAAAAATAAACTGTAAATCATATTC CTGCGACGTGAGGTATATGACTTT TTGTCTCTGGTTGCAAATCTAACA GGTCGGTGGACCGGTCGATG TTGTCTCTGGTTGCAAATCTAACA
6583–6602 7015–7034 6624–6649 6719–6746 215–238 596–619 491–510 596–619
L1
452
[Manos et al., 1989]
L1
140–150
HPV-16 E6 gene
405
[de Roda Husman et al., 1995] [Tornesello et al., 2004]
HPV-16 E6 gene
129
18E6F1 18E6R1 18E6F2 18E6R2
CACTTCACTGCAAGACATAGA GTTGTGAAATCGTCGTTTTTCA ATGCTGCATGCCATAAATGT CACCGCAGGCACCTTATTA
HPV-18 E6 gene
322
[Biedermann et al., 2004; Sotlar et al., 2004] [Sotlar et al., 2004]
HPV-18 E6 gene
139
Designed for this study
170–190 470–491 289–308 409–427
Degenerate base code: H ¼ A, T, or C; M ¼ A or C; R ¼ A or G; V ¼ G, A or C; W ¼ A or T; Y ¼ T or C. L, late region of HPV genome; E: early region of HPV genome. bp: base pair.
a b c
databases using the NCBI BLAST software program (http://www. blast.ncbi.nlm.nih.gov/blast/html).
extracted DNA from infected human cells (HSV-1, CMV, and EBV) mixed together and tested using HPV PCR assays.
Type-Specific PCR Assays for HPV-16/18 To decrease the chance of missing an infection caused by the two most common oncogenic HPV types 16/18, a type-specific nested PCR system was designed for detecting each type. The primer sets used for detection of HPV-16 and 18 were 16E6F1/16E6R1 and 18E6F1/18E6R1 as outer primers for the primary round of nested PCR followed by 16E6F2/16E6R1 and 18E6F2/18E6R2 primers as inner primers for the secondary PCR respectively, as described in Table I [Biedermann et al., 2004; Sotlar et al., 2004; Tornesello et al., 2004]. When the nested PCR was carried out, the product of the negative control in the first step of the reaction was used as template in the second round of PCR amplification as the negative control to detect any possible contamination. Plasmids and Cell Line All primers used for broad-spectrum and typespecific nested PCR assays were tested for their sensitivity using plasmids containing HPV-16 and HPV-18 DNA cloned in pBluescript (Manassas, VA) and pBR322 (Manassas, VA) respectively, which were purchased from American Type Culture Collection (ATCC). Tenfold dilutions of HPV DNA-containing plasmids were made in a background of 300 ng human cellular DNA extracted from the K562 cells which is a human erythroleukaemia line cell line obtained from the National Cell Bank of Iran (Pasteur Institute, Tehran, Iran). The dilution series (seven dilution steps) started with 30 pg (corresponding to 2.56 106 for HPV-16 and 2.28 106 for HPV18 target copies) and ended with 30 ag of HPV-DNA (corresponding to approximately two viral target copies). The specificity of all assays was determined using 0.05 pg of each plasmids containing viral DNA (HHV-6, BKV, JCV, HHV-8) and 300 ng of the J. Med. Virol. DOI 10.1002/jmv
Immunostaining for p16INK4a and L1 Capsid Protein Paraffin-embedded tissue samples were sectioned at 5 mm thickness and mounted onto pre-coated slides with 300 ml of poly-L-lysine solution (0.1% w/v, Sigma–Aldrich; St. Louis, MO). Slides were baked for 20 min at 60˚C to improve tissue adhesion. After deparaffinization, endogenous peroxidase activity quenching was performed by incubating sections in 3% H2O2 solution in 1 PBS at room temperature for 20 min. Antigen retrieval for p16INK4a and L1 capsid protein was performed by boiling tissue sections in Tris-EDTA buffer (10 mM Tris–HCL, 1 mM EDTA solution, 0.05% Tween 20, pH 9.0) and sodium citrate buffer (10 mM sodium citrate, 0.05% Tween 20, pH 6.0) in a water bath at 95–99˚C for 40 min, respectively. After washing with PBS, the samples were incubated with monoclonal mouse antihuman p16INK4a antibody, Clone JC8 (Santa Cruz Biotechnology Inc., Santa Cruz, CA) and monoclonal mouse anti-HPV antibodies Clone K1H8, DAKO (Glostrup, Denmark) which detects various HPV genotypes including 6, 11, 16, 18, 31, 33, 42, 51, 52, 56, and 58 in dilution 1:50 with DAKO antibody Diluents for 60 min at room temperature in a humidified chamber. The sections without primary antibody served as the negative control with each run. After buffer washes, sections were covered with EnVision þ Dual Link System-HRP solution (Dako) for 30 min at room temperature. The reaction was visualized using 3,30 -diaminobenzidine (DAB) staining. Sections from FFPE HeLa cell block and a breast cancer case previously known as negative for p16INK4a marker were used as positive and negative controls for p16INK4a immunostaining, respectively. Furthermore, for HPV major capsid protein staining, sections
HPV Infection in Renal Cell Carcinoma
from a previously confirmed cervix condyloma acuminatum case and a normal human kidney tissue served as positive and negative controls, respectively. Evaluation of Immunohistochemical Staining Immunostaining with p16INK4a and anti-HPV antibodies was examined in a double-blind protocol and scored under 400 magnification using standard light microscopy. For immunostaining of p16INK4a, the percentage of tumor cells that were reactive with p16INK4a antibody was evaluated in a semi-quantitative scoring system adopted from previous studies [Klaes et al., 2001; Kalof et al., 2005]. Briefly, the percentage of positive cells (negative: 1% of cells positive; sporadic: 5% of cells positive; focal: 5–25% of cells positive; and diffuse: 25% of cells positive) was assessed. The cytoplasmic and nuclear signals were scored independently. Strong nuclear as well as cytoplasmic staining was interpreted as a positive reaction. Samples showing a diffuse distribution were defined to have overexpression of p16INK4a immunohistochemically. Focal pattern was considered as moderate expression and sporadic staining was regarded as low expression, as previously classified by Fregonesi et al. [2003]. The immunocytochemical staining of HPV L1 capsid protein was considered positive when the neoplastic cells showed specific immunoreactivity for major capsid protein in the nucleus and occasionally in the cytoplasm, as well. Catalyzed Signal-Amplified Colorimetric In Situ Hybridization (CSAC-ISH) In situ hybridization was performed on 6 mm thick FFPE tissue sections of renal cell carcinoma cases using GenPointTM HPV biotinylated DNA probe (Dako Cytomation) and the GenPointTM Catalyzed Signal Amplification kit (Dako Cytomation) for detection of HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 68 according to the manufacturer’s manual. HeLa cells, which contain up to 50 copies of HPV-18 genome integrated at five sites of human chromosomes, were grown to confluence and suspended cells were aliquoted and pelleted [Meissner, 1999]. Formalin fixed, paraffin embedded cell blocks were prepared according to Li et al. [2007] protocol and served as positive control with each experiment. In addition, 50 ng of Hind III digested biotin-labeled Lambda DNA (Fermentas, Lithuania) was used as negative control instead of HPV biotinylated DNA probe in order to determine the specificity of the reaction. Nuclear staining corresponding to the areas of hybridization were viewed and scored as positive or negative under 400 magnifications using standard light microscopy. The brown DAB staining in the nuclei was interpreted as positive for presence of HR-HPV DNA. Furthermore, positive nuclear staining was further characterized as punctate (indicative of HPV genome integration), diffuse (indicative of episomal HPV genome) or mixed as well.
1137
Statistical Analysis Statistical analysis was performed using the SPSS statistical package software (Version 18.0; SPSS, Chicago, IL). The associations among the presence of HR-HPV as detected by PCR, the overexpression of p16INK4a and clinicopathologic characteristics were analyzed using Chi-square test or two-sided Fisher’s exact test, where appropriate. To assess the significance of the difference between detection of viral DNA and p16INK4a protein in tumor tissue specimens when compared to peritumoral corresponding tissues, Wilcoxon signed-rank test or McNemar test were used. In addition, Mann–Whitney analytical test was applied to evaluate the significance difference between detection of viral DNA in renal cell carcinoma tissue specimens and normal kidney tissue blocks. Differences were considered to be statistically significant when P-values were less than 0.05. RESULTS Evaluation of Sensitivity Assays The primary PCR assay with MY09/MY11 primers detected HPV-16 and 18 at 2.56 103 and 2.28 102 viral DNA copies, respectively. The secondary PCR assay with GP5þ and GP6þ primers using the amplicons from the first round PCR of each dilution showed detection of 256 copies for HPV-16 and 22 copies for HPV-18 viral DNA targets. The two assays did not show equal sensitivity between HPV-16 and 18. The MY/GPþ system was about 10 times more sensitive for detecting HPV-18 viral DNA copies compared to HPV-16. The primary PCR assays detected HPV-16 and 18 at 2.56 104 and 2.28 103 viral DNA copies, respectively. The secondary PCR assay using the amplicons from the first round PCR of each dilution showed detection of approximately two copies for both HPV16 and HPV-18 viral DNA targets. The sensitivity assay results demonstrated that although at the first round of assay, detection of HPV-18 DNA was about 10 times more sensitive compared to detection of HPV-16, in the secondary PCR step, the two assays showed equal sensitivity between detection of HPV16 and 18. The amplification of the mixed viral DNAs gave negative results, confirming the specificity of each assay. Study Set A total of 122 tissue blocks of renal cell carcinoma patients with the mean age of 54 years (range: 3–81) and 96 specimens of their related surrounding normal kidney tissue blocks in addition to 19 tissue blocks from patients with renal trauma including 12 men and 7 women with mean age of 39 were available for further analysis. DNA fragments of bglobin gene from five cases of the renal cell carcinoma tissue blocks, eight cases of their related peritumoral tissue blocks and four specimens from renal trauma J. Med. Virol. DOI 10.1002/jmv
1138
Farhadi et al.
patients were not amplifiable and were excluded from the study. The clinocopathological features of the patients are shown in Table II. PCR Amplification and Typing A nested PCR using the MY09/MY11 and GP5þ/ GP6þ primer set (MY/GPþ) by a touchdown approach was carried out to support the detection of low-copy number HPV DNA in a background of human DNA sequences. HPV DNA was amplified in 13 (10.7%) of 122 renal cell carcinoma cases. However, it was not detected in any subject of two other study groups. Sequence analysis of the PCR amplicons generated by the nested PCR with GP5þ/GP6þ primers allowed the identification of HPV-16 in five (4.1%) cases, HPV-18 in five (4.1%) cases, and HPV58 in three (2.4%) of renal cell carcinoma cases. When the same samples were analyzed by nested TABLE II. Summary of Clinicopathologic Characteristics of Renal Cell Carcinoma Patients Characteristic Total Gender Male Female Age 54
High-Risk Human Papillomavirus Infection in Different Histological Subtypes of Renal Cell Carcinoma Ali Farhadi,1 Abbas Behzad-Behbahani,2* Bita Geramizadeh,3 Zamberi Sekawi,1 Marjan Rahsaz,3 and Sedigheh Sharifzadeh2 1
Department of Medical Microbiology and Parasitology, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, Serdang, Malaysia 2 Diagnostic Laboratory Sciences and Technology Research Center, School of Paramedical Sciences, Shiraz University of Medical Sciences, Shiraz, Iran 3 Department of Pathology, Shiraz Transplant Research Center, Nemazee Hospital, Shiraz University of Medical Sciences, Shiraz, Iran
Limited data exist regarding whether a highrisk human papillomavirus (HR-HPV) infection increases the risk of developing renal cell carcinoma. The aim of this study was to investigate whether HPV infection has a role in the pathogenesis or development of a certain histological subtype of renal cell carcinoma. Formalin-fixed paraffin-embedded (FFPE) specimens of 122 patients with histopathologically proven renal cell carcinoma and their respective peritumoral tissues were examined. The presence of HPV-DNA was determined by a combination of MY/GPþ consensus primers and HPV-16/18 type specific nested PCRs followed by direct sequencing. Catalyzed signalamplified colorimetric in situ hybridization (CSAC-ISH) technique was applied to determine the physical status of viral genome. The expression of p16INK4a and HPV L1 capsid proteins was evaluated using immunohistochemistry. HPV genome was detected in 37 (30.3%) tumor specimens and their four (4.1%) corresponding peritumoral tissues. HPV-18 was the most common viral type identified followed by HPV-16 and 58. Immunoexpression of p16INK4a was detected in 24 (20.3%) cases. Data analysis showed a significant correlation between p16INK4a expression and the presence of HR-HPV DNA (P < 0.001). CSAC-ISH analysis confirmed HR-HPV infection in 45% of tumors, which were previously tested positive for HPV-DNA. Diffuse signal pattern was identified in 15 (83.3%) samples whereas a mixed pattern of diffuse and punctate signals was only detectable in three cases. The results indicate an association of HR-HPV types with renal cell carcinoma. It is proposed that HPV infection in high-grade tumors might precede C 2014 WILEY PERIODICALS, INC.
disease progression in a number of tumors, particularly of the papillary subtype. J. Med. Virol. 86:1134–1144, 2014. # 2014 Wiley Periodicals, Inc.
KEY WORDS:
human papillomavirus; renal cell carcinoma; in situ hybridization; p16INK4a immunohistochemistry
INTRODUCTION In recent decades, the occurrence of renal cell carcinoma has increased, accounting for about 3.8% of adult malignancies and approximately 90% of the entire renal neoplasms [Jemal et al., 2010]. Generally, it is believed that various subtypes of renal cell carcinoma arise from different specialized epithelial cells along the nephron [Shanks, 1999]. Renal cell carcinoma is regarded as a highly aggressive tumor. Metastasis can be found in a third of patients during diagnosis, which leads to over 40% mortality. However, for localized renal cell carcinoma, surgery is the treatment of choice [Janzen et al., 2003; Lam et al., 2005]. Recent advances in renal cell molecular biology Grant sponsor: Shiraz University of Medical Sciences; Grant number: 87-105-2009. Conflicts of interest: none. Correspondence to: Abbas Behzad-Behbahani, PhD, Diagnostic Laboratory Sciences and Technology Research Center, School of Paramedical Sciences, Shiraz University of Medical Sciences, Shiraz, Iran. E-mail: [email protected] Accepted 30 October 2013 DOI 10.1002/jmv.23945 Published online 2 April 2014 in Wiley Online Library (wileyonlinelibrary.com).
HPV Infection in Renal Cell Carcinoma
and genetics have associated the disease with various hereditary and non-hereditary risk factors; however, they cannot explain all renal cell carcinoma cases. Based on the current understanding, it has been estimated that some 15% of the global cancer burden can be linked to oncogenic tumor viruses [Zur Hausen, 2001]. While epitheliotropic high-risk human papillomavirus (HPV) is the established aetiological agent of cervical cancer, several studies have suggested the involvement of HPV in other human malignancies such as breast, respiratory tract, bladder, colorectal, and oesophageal cancer [Walboomers et al., 1999; Syrja¨nen, 2002; Damin et al., 2004; Barghi et al., 2005; Giuliani et al., 2008; Klein et al., 2009]. Although the role of HPV in development of renal cell carcinoma has been investigated less rigorously, research has indicated conflicting results. The possible involvement of HPV in renal cell carcinoma tumors was first hypothesized by Rotola et al. [1992]. Since then, a few reports with conflicting results have been published on this subject. On the whole, a comprehensive investigation of the relationship between HPV and different histologic subtypes of renal cell carcinoma is still sporadic. The related literature on malignant lesions of kidney is lacking, inconsistent, and only limited to certain histologic subtypes of renal cell carcinoma cases. Thus, this study was designed to include a large number of renal cell carcinoma tumors with different histological grades and subtypes to examine the presence of HRHPV infection and its association with clinicopathological features of renal cell carcinoma. MATERIALS AND METHODS Sample Selection and Tumor Classification Surgical report profiles of patients who underwent surgery in Namazi hospital, Shiraz University of Medical Sciences, Iran from January 2004 to December 2009 were reviewed and histologically confirmed renal cell carcinoma cases were selected. There were no restrictions on age, gender, ethnicity, or cancer stage at recruitment. Hematoxylin and eosin (H&E) slides were retrieved and classification of different histologic subtypes of renal cell carcinoma tumors was reconfirmed by pathologist according to the Heidelberg classification system [Kovacs et al., 1997]. Histopathologic grading of the nuclei of the tumor cells was assessed using the four-group Fuhrman nuclear classification system [Fuhrman et al., 1982]. The primary tumor staging was performed according to the recent version of the 2010 TNM staging of the American Joint Committee on Cancer (AJCC) and the Union for International Cancer Control (UICC) classification system [Guinan et al., 1997; Edge and Compton, 2010]. Formalinfixed paraffin-embedded (FFPE) tissue blocks of related slides were collected and examined to have sufficient tissue to perform all experiments; disqualified cases from each group were excluded from the study. Of 138 renal cell carcinoma tumors in this
1135
period, 127 samples were selected as the best representative cases. In addition, 104 specimens of their corresponding peritumoral kidney tissue and 23 samples of normal kidney tissue from patients with blunt renal injuries were available for this study as well. Ethics Approval The study was approved by the Medical Ethics Committee of the Shiraz University of Medical Sciences, Iran (nr 1388/321), and by the Medical Research Ethics Committee, the Faculty of Medicine and Health Sciences, Universiti Putra Malaysia (UPM/FPSK/ PADS/T7-MJKEtikaPer/F01 (JMPP_Aug (08) 01). Sample Preparation and DNA Extraction For each case, four 5 mm thick slices were cut and collected carefully in autoclaved Eppendorf microcentrifuge tubes (1.5 ml). Precautions were taken to avoid contamination among specimens including cutting one case at a time, changing the microtome blade before cutting each new case and cleaning the microtome and work area thoroughly with ethanol between each case. Additionally, paraffin blocks without tissue were sectioned after cutting each real specimen and included in every step of the experiment as negative controls of DNA extraction procedures. Genomic DNA was extracted from the FFPE tissue sections using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. Extracted DNA was eluted with 50 ml ATE buffer and stored at 70˚C until analyzed. The total amount of DNA was quantified using a NanoDrop (ND-1000) spectrophotometer (peQLab Biotechnologie, Erlangen, Germany). The quality of the extracted DNA for each sample was evaluated by PCR amplification using b-globin gene specific primers, PC03 (50 -ACACAACTGTGTTCACTAGC-30 ) and PC04 (50 -CAACTTCATCCACGTTCACC-30 ) generating a 110 bp PCR product as described by Saiki et al. [1988]. Only b-globin positive samples were included in further analyses. Broad Spectrum PCR Assays and Sequencing To enhance the sensitivity of viral detection, a nested general-primer-mediated PCR assay using MY09/MY11 degenerated primer set [Manos et al., 1989] followed by GP5þ/GP6þ consensus primers [de Roda Husman et al., 1995] was carried out for detection of 23 mucosotropic HPV genotypes, as shown in Table I. All HPV positive tissue specimens were amplified and analyzed in duplicate to ensure consistency and reliability of the results. To determine the HPV genotypes, the positive PCR products were submitted for automated DNA sequencing (MilleGen, Labe`ge, France) in the presence of GP5þ and GP6þ as both directions sequencing primers. The obtained sequences were aligned with documented HPV sequences that were available in GeneBank J. Med. Virol. DOI 10.1002/jmv
1136
Farhadi et al. TABLE I. Primer Names and Sequences Used for Detection of HPV-DNA in this Study
Primer name
Sequence (50 –30 )a
Nucleotide position
Targetb
Amplicon size (bpc)
Reference
MY 11 MY09 GP5þ GP6þ 16E6F1 16E6R1 16E6F2 16E6R1
GCMCAGGGWCATAAYAAYTGG CGTCCMARRGGAWACTGATC TTTGTTACTGTGGTAGATACTAC GAAAAATAAACTGTAAATCATATTC CTGCGACGTGAGGTATATGACTTT TTGTCTCTGGTTGCAAATCTAACA GGTCGGTGGACCGGTCGATG TTGTCTCTGGTTGCAAATCTAACA
6583–6602 7015–7034 6624–6649 6719–6746 215–238 596–619 491–510 596–619
L1
452
[Manos et al., 1989]
L1
140–150
HPV-16 E6 gene
405
[de Roda Husman et al., 1995] [Tornesello et al., 2004]
HPV-16 E6 gene
129
18E6F1 18E6R1 18E6F2 18E6R2
CACTTCACTGCAAGACATAGA GTTGTGAAATCGTCGTTTTTCA ATGCTGCATGCCATAAATGT CACCGCAGGCACCTTATTA
HPV-18 E6 gene
322
[Biedermann et al., 2004; Sotlar et al., 2004] [Sotlar et al., 2004]
HPV-18 E6 gene
139
Designed for this study
170–190 470–491 289–308 409–427
Degenerate base code: H ¼ A, T, or C; M ¼ A or C; R ¼ A or G; V ¼ G, A or C; W ¼ A or T; Y ¼ T or C. L, late region of HPV genome; E: early region of HPV genome. bp: base pair.
a b c
databases using the NCBI BLAST software program (http://www. blast.ncbi.nlm.nih.gov/blast/html).
extracted DNA from infected human cells (HSV-1, CMV, and EBV) mixed together and tested using HPV PCR assays.
Type-Specific PCR Assays for HPV-16/18 To decrease the chance of missing an infection caused by the two most common oncogenic HPV types 16/18, a type-specific nested PCR system was designed for detecting each type. The primer sets used for detection of HPV-16 and 18 were 16E6F1/16E6R1 and 18E6F1/18E6R1 as outer primers for the primary round of nested PCR followed by 16E6F2/16E6R1 and 18E6F2/18E6R2 primers as inner primers for the secondary PCR respectively, as described in Table I [Biedermann et al., 2004; Sotlar et al., 2004; Tornesello et al., 2004]. When the nested PCR was carried out, the product of the negative control in the first step of the reaction was used as template in the second round of PCR amplification as the negative control to detect any possible contamination. Plasmids and Cell Line All primers used for broad-spectrum and typespecific nested PCR assays were tested for their sensitivity using plasmids containing HPV-16 and HPV-18 DNA cloned in pBluescript (Manassas, VA) and pBR322 (Manassas, VA) respectively, which were purchased from American Type Culture Collection (ATCC). Tenfold dilutions of HPV DNA-containing plasmids were made in a background of 300 ng human cellular DNA extracted from the K562 cells which is a human erythroleukaemia line cell line obtained from the National Cell Bank of Iran (Pasteur Institute, Tehran, Iran). The dilution series (seven dilution steps) started with 30 pg (corresponding to 2.56 106 for HPV-16 and 2.28 106 for HPV18 target copies) and ended with 30 ag of HPV-DNA (corresponding to approximately two viral target copies). The specificity of all assays was determined using 0.05 pg of each plasmids containing viral DNA (HHV-6, BKV, JCV, HHV-8) and 300 ng of the J. Med. Virol. DOI 10.1002/jmv
Immunostaining for p16INK4a and L1 Capsid Protein Paraffin-embedded tissue samples were sectioned at 5 mm thickness and mounted onto pre-coated slides with 300 ml of poly-L-lysine solution (0.1% w/v, Sigma–Aldrich; St. Louis, MO). Slides were baked for 20 min at 60˚C to improve tissue adhesion. After deparaffinization, endogenous peroxidase activity quenching was performed by incubating sections in 3% H2O2 solution in 1 PBS at room temperature for 20 min. Antigen retrieval for p16INK4a and L1 capsid protein was performed by boiling tissue sections in Tris-EDTA buffer (10 mM Tris–HCL, 1 mM EDTA solution, 0.05% Tween 20, pH 9.0) and sodium citrate buffer (10 mM sodium citrate, 0.05% Tween 20, pH 6.0) in a water bath at 95–99˚C for 40 min, respectively. After washing with PBS, the samples were incubated with monoclonal mouse antihuman p16INK4a antibody, Clone JC8 (Santa Cruz Biotechnology Inc., Santa Cruz, CA) and monoclonal mouse anti-HPV antibodies Clone K1H8, DAKO (Glostrup, Denmark) which detects various HPV genotypes including 6, 11, 16, 18, 31, 33, 42, 51, 52, 56, and 58 in dilution 1:50 with DAKO antibody Diluents for 60 min at room temperature in a humidified chamber. The sections without primary antibody served as the negative control with each run. After buffer washes, sections were covered with EnVision þ Dual Link System-HRP solution (Dako) for 30 min at room temperature. The reaction was visualized using 3,30 -diaminobenzidine (DAB) staining. Sections from FFPE HeLa cell block and a breast cancer case previously known as negative for p16INK4a marker were used as positive and negative controls for p16INK4a immunostaining, respectively. Furthermore, for HPV major capsid protein staining, sections
HPV Infection in Renal Cell Carcinoma
from a previously confirmed cervix condyloma acuminatum case and a normal human kidney tissue served as positive and negative controls, respectively. Evaluation of Immunohistochemical Staining Immunostaining with p16INK4a and anti-HPV antibodies was examined in a double-blind protocol and scored under 400 magnification using standard light microscopy. For immunostaining of p16INK4a, the percentage of tumor cells that were reactive with p16INK4a antibody was evaluated in a semi-quantitative scoring system adopted from previous studies [Klaes et al., 2001; Kalof et al., 2005]. Briefly, the percentage of positive cells (negative: 1% of cells positive; sporadic: 5% of cells positive; focal: 5–25% of cells positive; and diffuse: 25% of cells positive) was assessed. The cytoplasmic and nuclear signals were scored independently. Strong nuclear as well as cytoplasmic staining was interpreted as a positive reaction. Samples showing a diffuse distribution were defined to have overexpression of p16INK4a immunohistochemically. Focal pattern was considered as moderate expression and sporadic staining was regarded as low expression, as previously classified by Fregonesi et al. [2003]. The immunocytochemical staining of HPV L1 capsid protein was considered positive when the neoplastic cells showed specific immunoreactivity for major capsid protein in the nucleus and occasionally in the cytoplasm, as well. Catalyzed Signal-Amplified Colorimetric In Situ Hybridization (CSAC-ISH) In situ hybridization was performed on 6 mm thick FFPE tissue sections of renal cell carcinoma cases using GenPointTM HPV biotinylated DNA probe (Dako Cytomation) and the GenPointTM Catalyzed Signal Amplification kit (Dako Cytomation) for detection of HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 68 according to the manufacturer’s manual. HeLa cells, which contain up to 50 copies of HPV-18 genome integrated at five sites of human chromosomes, were grown to confluence and suspended cells were aliquoted and pelleted [Meissner, 1999]. Formalin fixed, paraffin embedded cell blocks were prepared according to Li et al. [2007] protocol and served as positive control with each experiment. In addition, 50 ng of Hind III digested biotin-labeled Lambda DNA (Fermentas, Lithuania) was used as negative control instead of HPV biotinylated DNA probe in order to determine the specificity of the reaction. Nuclear staining corresponding to the areas of hybridization were viewed and scored as positive or negative under 400 magnifications using standard light microscopy. The brown DAB staining in the nuclei was interpreted as positive for presence of HR-HPV DNA. Furthermore, positive nuclear staining was further characterized as punctate (indicative of HPV genome integration), diffuse (indicative of episomal HPV genome) or mixed as well.
1137
Statistical Analysis Statistical analysis was performed using the SPSS statistical package software (Version 18.0; SPSS, Chicago, IL). The associations among the presence of HR-HPV as detected by PCR, the overexpression of p16INK4a and clinicopathologic characteristics were analyzed using Chi-square test or two-sided Fisher’s exact test, where appropriate. To assess the significance of the difference between detection of viral DNA and p16INK4a protein in tumor tissue specimens when compared to peritumoral corresponding tissues, Wilcoxon signed-rank test or McNemar test were used. In addition, Mann–Whitney analytical test was applied to evaluate the significance difference between detection of viral DNA in renal cell carcinoma tissue specimens and normal kidney tissue blocks. Differences were considered to be statistically significant when P-values were less than 0.05. RESULTS Evaluation of Sensitivity Assays The primary PCR assay with MY09/MY11 primers detected HPV-16 and 18 at 2.56 103 and 2.28 102 viral DNA copies, respectively. The secondary PCR assay with GP5þ and GP6þ primers using the amplicons from the first round PCR of each dilution showed detection of 256 copies for HPV-16 and 22 copies for HPV-18 viral DNA targets. The two assays did not show equal sensitivity between HPV-16 and 18. The MY/GPþ system was about 10 times more sensitive for detecting HPV-18 viral DNA copies compared to HPV-16. The primary PCR assays detected HPV-16 and 18 at 2.56 104 and 2.28 103 viral DNA copies, respectively. The secondary PCR assay using the amplicons from the first round PCR of each dilution showed detection of approximately two copies for both HPV16 and HPV-18 viral DNA targets. The sensitivity assay results demonstrated that although at the first round of assay, detection of HPV-18 DNA was about 10 times more sensitive compared to detection of HPV-16, in the secondary PCR step, the two assays showed equal sensitivity between detection of HPV16 and 18. The amplification of the mixed viral DNAs gave negative results, confirming the specificity of each assay. Study Set A total of 122 tissue blocks of renal cell carcinoma patients with the mean age of 54 years (range: 3–81) and 96 specimens of their related surrounding normal kidney tissue blocks in addition to 19 tissue blocks from patients with renal trauma including 12 men and 7 women with mean age of 39 were available for further analysis. DNA fragments of bglobin gene from five cases of the renal cell carcinoma tissue blocks, eight cases of their related peritumoral tissue blocks and four specimens from renal trauma J. Med. Virol. DOI 10.1002/jmv
1138
Farhadi et al.
patients were not amplifiable and were excluded from the study. The clinocopathological features of the patients are shown in Table II. PCR Amplification and Typing A nested PCR using the MY09/MY11 and GP5þ/ GP6þ primer set (MY/GPþ) by a touchdown approach was carried out to support the detection of low-copy number HPV DNA in a background of human DNA sequences. HPV DNA was amplified in 13 (10.7%) of 122 renal cell carcinoma cases. However, it was not detected in any subject of two other study groups. Sequence analysis of the PCR amplicons generated by the nested PCR with GP5þ/GP6þ primers allowed the identification of HPV-16 in five (4.1%) cases, HPV-18 in five (4.1%) cases, and HPV58 in three (2.4%) of renal cell carcinoma cases. When the same samples were analyzed by nested TABLE II. Summary of Clinicopathologic Characteristics of Renal Cell Carcinoma Patients Characteristic Total Gender Male Female Age 54
