Extensive epidemiologic and laboratory investigations have established

Transcription

1 2199 Quantitation of Human Papillomavirus 16 E6 and E7 DNA and RNA in Residual Material from ThinPrep Papanicolaou Tests Using Real-Time Polymerase Chain Reaction Analysis Feng Wang-Johanning, M.D. 1 Danielle W. Lu, M.D. 2 Yueying Wang, M.D. 1 Martin R. Johnson, Ph.D. 3 Gary L. Johanning, Ph.D. 4 1 Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama. 2 Division of Anatomic Pathology, Lauren V. Ackerman Laboratory of Surgical Pathology, Washington University Medical Center, St. Louis, Missouri. 3 Department of Pharmacology and Toxicology, University of Alabama at Birmingham, Birmingham, Alabama. 4 Department of Nutrition Sciences, University of Alabama at Birmingham, Birmingham, Alabama. Presented in part as an abstract at the annual meeting of the American Association for Cancer Research, New Orleans, Louisiana, March 24 28, Supported by grant R21 CA81507 from the National Cancer Institute. Address for reprints: Feng Wang-Johanning, M.D., Department of Medicine, Division of Hematology/ Oncology, 542 Wallace Tumor Institute, The University of Alabama at Birmingham, Birmingham, AL ; Fax: (205) ; fengi@uab.edu Received September 6, 2001; revision received November 8, 2001; accepted November 21, BACKGROUND. The detection of specific human papillomavirus 16 (HPV-16) E6 and E7 oncogene transcripts may be a sensitive indicator of the direct involvement of viral oncogenes in the development of cervical neoplasia and carcinoma. The goal of this study was to determine the potential clinical uses of real-time polymerase chain reaction (PCR) and reverse transcription-pcr (RT-PCR) methods for evaluating HPV-16 E6 and E7 oncogene expression. METHODS. ThinPrep cervical samples were tested for expression of oncogenes of HPV-16 by real-time PCR or RT-PCR analysis and were compared with detection of expression by conventional PCR and RT-PCR analysis. Both sets of results were correlated with the cytologic diagnosis of the cervical samples. RESULTS. The presence of HPV-16 E6 and E7 DNA and RNA was observed only in HPV-16 positive cervical carcinoma cell lines but not in HPV-18 positive or HPV negative cell lines. The percentage positive for HPV-16 E6 or E7 DNA in a series of ThinPrep cervical cytologic samples (n 348 samples) was 0% for negative samples (n 45 samples), 9.7% for atypical squamous cells of undetermined significance (ASCUS; n 144 samples), 16.9% for low-grade squamous intraepithelial lesion (LSIL; n 118 samples), and 51.2% for high-grade intraepithelial lesion (HSIL; n 41 samples). The copy numbers per nanogram for both DNA and RNA E6 and E7 were increased significantly as severity of the lesions progressed from ASCUS to HSIL, and RNA copy numbers were a more sensitive indicator of HPV-16 E6 and E7 expression than DNA copy numbers. The increase in copy numbers took place in a stepwise fashion from ASCUS, to LSIL, to HSIL. CONCLUSIONS. The detection of HPV-16 E6 and E7 expression by real-time RT-PCR or PCR analysis in ThinPrep cervical cytologic specimens may serve as a quick, reliable, and sensitive tool to identify a subset of patients who express HPV-16 oncoproteins. Cancer 2002;94: American Cancer Society. DOI /cncr KEYWORDS: human papillomavirus 16, E6 and E7 oncoproteins, real-time polymerase chain reaction analysis, ThinPrep. Extensive epidemiologic and laboratory investigations have established a strong association between infection with certain human papillomaviruses (HPVs) and cervical and anogenital carcinomas. 1 Nearly 80 different HPVs have been characterized, and about onethird specifically target genital epithelial cells. 2 Approximately 30 distinct HPV types preferentially infect the anogenital mucosa, resulting in lesions that range from condyloma to cervical intraepithelial neoplasia (CIN)/squamous intraepithelial lesion (SIL) and invasive 2002 American Cancer Society

2 2200 CANCER April 15, 2002 / Volume 94 / Number 8 carcinoma. 3 These HPV types in the cervix currently are grouped as HPV-16, other oncogenic types, condyloma-associated types, and low-risk cervical types (HPV-53, HPV-61, and HPV-71). Currently, HPV infections are monitored primarily by HPV DNA detection using polymerase chain reaction (PCR) analysis with consensus specific and type specific primers 4 6 and by the Hybrid Capture System DNA detection test. 7 Several studies have evaluated the clinical sensitivity and specificity of the Hybrid Capture assay against PCR and histology Recently, the analytic sensitivity of the second-generation Hybrid Capture assay was increased to 1000 HPV DNA copies by the reformulation of hybridization reagents and by the addition of new probes for four high-risk or intermediate risk HPV types. 14 However, this assay has some deficiencies as a predictor of cervical carcinoma risk, because it can only distinguish between high-risk types (including HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, HPV-58, HPV-59, and HPV-68) and low-risk types (including HPV-6, HPV-11, HPV-42, HPV-43, and HPV- 44) of HPV infection, and it cannot provide details about specific HPV types. Information relative to the predictive value of different high-risk types of HPV may enable clinicians to decide how aggressively to treat an individual patient and to avoid overtreatment. Amplification-based methods, mainly PCR, are currently the most sensitive methods for detection of HPV DNA or RNA. However, because of frequent contamination problems and consequent false-positive results, PCR results are not always reliable. Therefore, a new urgency has been placed on the need for establishing accurate methods to diagnose HPV infection. HPVs, and particularly type 16, are associated with invasive cervical carcinoma, and persistent high-risk HPV gene expression is considered a marker for progression from low-grade SIL (LSIL), to high-grade SIL (HSIL), to invasive squamous cell carcinoma. Expression of the HPV E6 and E7 oncoprotein open reading frames of the viral genome is considered critical in carcinogenesis. Transformation studies demonstrated that the oncogenic potential of high-risk HPV infections was due to the viral-transforming genes E6 and E7. 15 The E6 and E7 oncoproteins are expressed consistently in high-risk infection, 16 which suggests that they contribute to the initiation and/or progression of tumors. 17 The viral oncoprotein E6 initiates degradation of the cellular tumor suppressor protein p53, 18 whereas oncoprotein E7 leads to the inactivation of another cellular tumor suppressor protein, the retinoblastoma gene product prb. 19 It is known that these synergistic effects are important steps in carcinogenesis, leading to a loss of cell cycle control Using PCR analysis, Cornelissen et al. 23 showed that the same complement of E6-E7 spliced mrnas is found in HPV-16 transformed fibroblasts, premalignant lesions, and carcinomas. Because it also has been observed that the degree of cellular proliferation is correlated with expression of the E6 and E7 proteins, 24 it would be of interest to determine whether quantitative differences in E6-E7 spliced mrnas show an association with disease progression. It has been suggested that the detection of specific HPV oncogene transcripts may be a sensitive indicator of the direct involvement of viral oncogenes in carcinogenesis. 25,26 In the current study, we developed a quantitative reverse transcriptase-pcr (RT- PCR) assay that facilitates quantitation of HPV-16 E6 and E7 oncogene transcripts in ThinPrep cervical cytologic samples with preinvasive cervical lesions. The presence of HPV-16 E6 and/or E7 DNA was correlated highly with the grade of cervical SILs and also was predictive of morphologic SIL subtypes. MATERIALS AND METHODS Cell Lines and Cultures The CaSki (squamous cell carcinoma), SiHa (squamous cell carcinoma), C33A (squamous cell carcinoma), and HeLa (adenocarcinoma) cervical carcinoma cell lines were obtained from the American Type Culture Collection (Rockville, MD). Cells were cultured under standard conditions in Dulbecco modified Eagle medium (for the C33A, HeLa, and SiHa cell lines) and in RPMI 1640 medium (for the CaSki cell line) supplemented with 10% fetal bovine serum. All cell lines were maintained at 37 C in a humidified 5% CO 2 atmosphere. ThinPrep Material Preparation The ThinPrep Papanicolaou (Pap) test is a liquidbased preparation for use in routine screening for the presence of cervical carcinoma and its precursor lesions. Cervical cytologic material was obtained with collection devices and rinsed into a vial containing preservative solution (PreservCyt; Cytyc Corporation, Boxborough, MA). Abnormal ThinPrep Pap test samples from December 2000 to May 2001 were collected from the Cytopathology Laboratory at the Barnes-Jewish Hospital of Washington University Medical Center. The samples were collected from patients ranging in age from 17 years to 70 years who came for routine Pap smear tests. The Pap smears were reviewed and signed by a cytopathologist in the Division of Anatomic Pathology and Immunology at Washington University Medical Center. There was 60% correlation between the cytologic (Pap test) and histologic (biopsy) diagnoses in these samples. A total of 348 samples

3 Real-Time PCR Quantitation of HPV-16 E6 and E7/Wang-Johanning et al FIGURE 1. Human papilloma virus 16 genomic DNA and RNA target sequences for reverse transcriptase-polymerase chain reaction (RT-PCR) amplification, real-time PCR amplification, or real-time RT-PCR amplification (nt: nucleotide; URR: upstream regulatory region). with atypical squamous cells of undetermined significance (ASCUS; n 144 samples), LSIL (n 118 samples), HSIL (n 41 samples), and within normal limits (negative control; n 45 samples) were tested. All specimens were tested within 3 4 months of collection, and all samples were stored at room temperature. Residual liquid material on each normal control and abnormal ThinPrep cervical cytologic sample was used and divided into two parts: One part was used for DNA isolation, and the other part was used for RNA isolation. Extraction of DNA and RNA Residual ThinPrep cervical cytologic samples were centrifuged at 1800 rpm for 5 minutes, and the concentrated cell pellet was used for DNA and RNA extraction and purification. Total RNA from cell lines and ThinPrep specimens was extracted using STAT-60 reagent (Tel-Test Inc., Friendswood, TX) and was treated to eliminate DNA contamination, as described previously. 27 The RNA and DNA were purified using Qiagen RNeasy and DNeasy Tissue kits (Qiagen Inc., Valencia, CA) following the manufacturer s directions. TABLE 1 Sequence of Polymerase Chain Reaction Primers and Probes for Human Papillomavirus 16 E6, E7, S9, and -Actin a A. Taqman Probe and Primers specific to HPV-16 E6 and E7 open reading frames: HPV-16 E6 (nt ) Forward primer: CTGCAATGTTTCAGGACCCA Reverse primer: TCATGTATAGTTGTTTGCAGCTCTGT Probe: FAM-AGGAGCGACCCGGAAAGTTACCACAGTT-BHQ HPV-16 E7 (nt ) Forward primer: AAGTGTGACTCTACGCTTCGGTT Reverse primer: GCCCATTAACAGGTCTTCCAAA Probe: FAM-TGCGTACAAAGCACACACGTAGACATTCGTA-BHQ S9 (nt ) Forward primer: ATCCGCCAGCGCCATA Reverse primer: TCAATGTGCTTCTGGGAATCC Probe: FAM-AGCAGGTGGTGAACATCCCGTCCTT-TAMRA -actin (nt ) Forward primer: TCACCCACACTGTGCCCATCTACGA Reverse primer: CAGCGGAACCGCTCATTGCCAATGG Probe: FAM-ATGCCCTCCCCCATGCCATC-TAMRA B. Conventional RT-PCR or PCR primers specific to HPV-16 E6 and E7 open reading frames: E6 forward primer (nt ) CTCTGAATTCGCCACCATGCACCAAAAGAGAACTGCA (EcoRI site underlined) E6 reverse primer (nt ) CCCTCGAGGTATCTCCATGCATGATTACA (XhoI site underlined) E7 forward primer (nt ) CTCTGAATTCGCCACCATGCATGGAGATACACCTACA (EcoRI site underlined) E7 reverse primer (nt ) CCCTCGAGGATCAGCCATGGTACATTATGG (XhoI site underlined) HPV-16: human papillomavirus 16; nt: nucleotide; RT-PCR: reverse transcriptase-polymerase chain reaction. a All primer sequences are written from 5 to 3. Oligounucleotide Primer and Probe Design The HPV-16 E6 (nucleotides [nt] ), HPV-16 E7 (nt ; GenBank accession no. NC ) (Fig. 1, Table 1A), -actin, and S9 primers and probes (Table 1A) were designed using the Primer Express software package (Perkin-Elmer, Foster City, CA) and were synthesized by Biosearch Technologies, Inc. (Novato, CA). Homo sapiens ribosomal protein S9 (GenBank accession no. BC BC000802) and Homo sapiens -actin (GenBank accession no. BC BC004251; Perkin-Elmer) were used as endogenous controls for quantitation of RNA and DNA, respectively. These were chosen as controls, because we confirmed with dot blot analysis that there was no difference in the level of their expression in normal tissues and tumor tissues from the cervix (data not presented). The HPV-16 probes for E6 and E7 do not cross-react with other HPV types.

4 2202 CANCER April 15, 2002 / Volume 94 / Number 8 In Vitro Transcription and Purification of HPV-16 E6 and E7 crna For E7 crna synthesis, the p734 plasmid described previously, 28 corresponding to HPV-16 E7 (312 base pairs [bp]; nt ) was used as a template for in vitro transcription. For E6 crna synthesis, p186c6, an RT-PCR product of the E6 gene (492 bp; nt ) that was derived from a patient with cervical carcinoma, was amplified using primers specific to HPV-16 E6 open reading frames (see Table 1B). The PCR product was cloned into a pcdna3 vector (Invitrogen, Carlsbad, CA), and the sequence was confirmed by sequence analysis (99% homology to HPV-16 E6; Gen- Bank accession no. NC ). The p734 and p186c6 plasmids were linearized with XhoI, and the linearized plasmids were used as templates for in vitro transcription using a MAXIscript in vitro transcription kit (Ambion, Austin, TX) following the manufacturer s instructions. The resulting crna was incubated for 20 minutes with 4 units DNase and purified on a 5% acrylamide/8 M urea gel. Gel purification was necessary to remove incomplete transcripts and unincorporated dntps. The eluted HPV E6 or E7 crna was precipitated and quantitated by absorbance at 260 nm. Because the exact size of the crna was known, the exact copy number of E6 or E7 crna could be calculated according to the following formula: copy number mass ( g) /molecular weight Nuc length, where molecular weight for RNA, Nuc 1 for single-strand crna and 2 for double-strand crna, and length the length of the target in base pairs. Quantitation of HPV-16 E6 and E7 mrna using Real-Time PCR One-step RT-PCR was performed using an ABI PRISM 7700 sequence detector as follows. Homo sapiens ribosomal protein S9 (see Table 1A) (nt ; Gen- Bank accession no. XM ) was used as an endogenous control. The amplification reactions were performed in a 25- L final volume containing 1 TaqMan buffer (Perkin-Elmer) plus dntps (0.3 mm each), units of AmpliTaq Gold, RNase inhibitor (5 units), 2% glycerol, and units of MuLV reverse transcriptase. All RT-PCR reactions were performed in optical reaction tubes (Applied Biosystems, Foster City, CA) designed for the ABI PRISM 7700 sequencedetector system. Reverse transcription and thermal cycling conditions were 30 minutes at 48 C followed by 10 minutes at 95 C and 40 cycles of 15 seconds at 95 C and 1 minute at 60 C. The optimized concentrations of E6 and E7 forward and reverse primers and probe (Table 1A) were 100 nm, 100 nm, and 150 nm, respectively. PCR premixes containing all reagents except for total RNA were used as a no-template control. Ribosomal protein S9 with optimized concentrations of primers and probe (forward, reverse, and probe concentrations were 200 nm, 200 nm, and 100 nm, respectively) was amplified to generate a standard curve with known amounts of CaSki cell RNA (10.0 ng, 5.0 ng, 1.0 ng, and 0.5 ng total RNA). The unknown samples were amplified in different sets of reactions using the E6 or E7 primers and probe (Table 1A). Linear extrapolation of the cycle threshold (CT) values was carried out using the equation of the line obtained from the E6 standard curve or the E7 standard curve. These values were then divided by the relative amounts of S9, which were quantitated by linear extrapolation from the CT values of the unknown samples using the equation of the line obtained from the S9 standard curve. Known amounts of E6 or E7 crna molecules ( copies, copies, copies, and copies) were used to generate an absolute standard curve. The copy number of E6 and E7 mrna was then determined by linear extrapolation of the CT values using the equation of the line obtained from the absolute E6 or E7 standard curve. These values were then divided by the relative amounts of S9. SiHa or CaSki cell RNA (HPV-16 positive) was used as a positive control, and an equal amount of C33A or HeLa cell RNA (HPV-16 negative) was used as a negative control. Quantitation of HPV-16 E6 and E7 DNA using Real-Time PCR For real-time DNA quantitation, -actin primer and probe were used as internal controls. The amplification reactions were performed in a 25- L final volume containing 1 Universal PCR Master Mix (Applied Biosystems) and the optimized concentration of E6 or E7 primers and probe (100 nm forward primer, 100 nm reverse primer, and 150 nm probe). All PCR assays were performed in optical reaction tubes (Applied Biosystems) designed for the ABI PRISM 7700 sequencedetector system. Thermal cycling conditions were 1 minute at 94 C followed by 40 cycles of 30 seconds at 94 C, 30 seconds at 55 C, and 30 seconds at 72 C. The optimized concentrations of -actin forward and reverse primers and probe were 300 nm, 300 nm, and 200 nm, respectively. Using these concentrations, -actin was amplified to generate a standard curve using known amounts of genomic DNA from a healthy human female (200.0 ng, 20.0 ng, 2.0 ng, and 0.2 ng; Promega, Madison, WI). The unknown samples were amplified in separate reactions using the E6 or E7 primers and probe (Table 1A). Linear extrapolation of

5 Real-Time PCR Quantitation of HPV-16 E6 and E7/Wang-Johanning et al CT values was obtained using the equation of the line obtained from the HPV-16 E6 or E7 standard curve. These values were then divided by the relative amounts (ng) of -actin. Known amounts of E6 or E7 PCR products obtained from CaSki cell DNA ( copies, copies, copies, and copies) were used to generate an absolute standard curve. The copy number of E6 or E7 DNA was determined by linear extrapolation of the CT values using the equation of the line obtained from the absolute E6 or E7 standard curve. These values were then divided by the relative amounts of -actin. SiHa or CaSki cell DNA (HPV-16 positive) was used as a positive control, and an equal amount of C33A or HeLa cell DNA (HPV-16 negative) was used as a negative control. Conventional RT-PCR and PCR Amplification RNA was reverse transcribed to synthesize DNA using Ready-To-Go first-strand cdna synthesis beads (Pharmacia), as described previously. 27 The sequences of the primers used for E6 and E7 amplification are shown in Table 1B and Figure 1. The reverse transcribed samples were amplified in a volume of 50 L containing 3.3 L ofthefirst-strand cdna synthesis mixture (corresponding to 1 g of input RNA), 5 L of 10 PCR buffer (Qiagen), 0.5 L of Ampli-Taq DNA polymerase (2.5 U; Qiagen), and various sense and antisense oligonucleotide primer pairs at 50 pmol each. Each sample was analyzed in parallel with human -actin primers. To assure that observations were not due to DNA contamination, all RNA samples were treated with DNase before cdna synthesis. In addition, 1 g of RNA from each sample without reverse transcription was PCR amplified to control for genomic DNA contamination. PCR reactions initially were denatured at 94 C for 4 minutes followed by 30 cycles of denaturation at 94 C for 1 minute, annealing at 55 C for 1 minute, and extension at 72 C for 1 minute. Amplified products were analyzed on a 2% agarose gel. SiHa or CaSki cell RNA (HPV-16 positive) was used as a positive control, and an equal amount of C33A or HeLa cell RNA (HPV-16 negative) was used as a negative control. For DNA PCR, approximately 200 ng of DNA isolated from ThinPrep cervical cytologic samples or cervical carcinoma cells were used to directly amplify the HPV-16 E6 or E7 DNA. The PCR reactions were carried out in a manner similar to that used for RT-PCR, except that samples were not treated with DNase or reverse transcribed. The amplified products were analyzed on a 2% agarose gels, as described previously. FIGURE 2. Amplification products of human papilloma virus 16 E6, E7, or -actin using the cell lines CaSki (lane 1), SiHa (lane 2), C33A (lane 3), and HeLa (lane 4). The lane designated 100 bp is a 100-base-pair DNA ladder (Life Technologies). (A) Polymerase chain reaction (PCR) amplification. (B) Reverse transcriptase-pcr amplification. RESULTS Detection of HPV-16 E6 and E7 DNA and RNA in Cervical Cell Lines HPV-16 E6 and E7 DNA and RNA were detected in CaSki and SiHa cells (HPV-16 positive cervical cell lines) but not in HeLa or C33A cells (HPV-16 negative cervical cell lines) (Fig. 2A,B, respectively) by conventional PCR using E6 specific and E7 specific primers (see Fig. 1, Table 1B). The results show that HPV-16 E6 and E7 are expressed in cervical carcinoma cells that are HPV-16 positive. Figure 2 shows that the HPV-16 positive cell lines CaSki and SiHa expressed HPV-16 E6 mrna (492-bp band), whereas the HPV-18 positive HeLa cells and the HPV negative C33A cells did not express E6. Alternative splicing of E6 can result in up to three RT-PCR products (E6, E6f*1, and E6*2; see Fig. 1), and we observed three bands after amplifying the RNA of CaSki and SiHa cells (Fig. 2B). E7 also was amplified by RT-PCR in the HPV-16 positive cell lines (Fig. 2; 312-bp band). These results suggest that RT- PCR is an accurate indicator of E6 and E7 oncoprotein status in cervical carcinoma cells. However, we could not quantitate the relative differences between expression of the two oncogenes in these two HPV-16 positive cell lines, which was possible using real-time PCR. HPV-16 E6 or E7 DNA was detected by real-time PCR only in the HPV-16 positive cervical carcinoma cell lines (CaSki and SiHa) but not in the HPV-18 positive cell line (HeLa) or in the HPV negative cell line

6 2204 CANCER April 15, 2002 / Volume 94 / Number 8 TABLE 2 Expression of Human Papillomavirus-16 E6 and E7 DNA and RNA in Cervical Carcinoma Cell Lines Detected by Real-Time Polymerase Chain Reaction and Reverse Transcriptase-Polymerase Chain Reaction Analysis Cell line HPV type DNA copy no. a relative to HeLa Fold change RNA copy no. Fold change relative to HeLa E6 C33A Negative CaSki HeLa SiHa E7 C33A Negative CaSki HeLa SiHa HPV: human papillomavirus. a Values are mean standard deviation. FIGURE 3. Comparison of human papillomavirus 16 E6 and E7 DNA and RNA copy numbers in cervical carcinoma cell lines using real-time polymerase chain reaction analysis. (C33A) (Table 2, Fig. B). We assigned HeLa cells a value of 1, with all other copy number values expressed relative to HeLa cells. These data indicate that the relative level of HPV-16 E6 or E7 DNA was CaSki SiHa C33A HeLa cells (see Table 2). These data indicate that the relative level of HPV-16 E6 RNA expression was SiHa CaSki C33A HeLa cells, and the relative expression level of HPV-16 E7 RNA was SiHa CaSki HeLa C33A cells. The DNA copy number per ng DNA in CaSki cells was times greater (E6) and times greater (E7) than the copy number in SiHa cells (see Table 2). It is interesting to note that RNA copy numbers of E6 or E7 in both cell lines were not appreciably different from each other (see Fig. 3). The RNA copy number per ng RNA in SiHa cells was 1.5 times greater (E6) and 2.15 times greater (E7) than the copy number in CaSki cells. Detection of HPV-16 E6 and E7 DNA and RNA in ThinPrep Cervical Cytologic Samples We evaluated 144 ASCUS samples, 118 LSIL samples, 41 HSIL samples, and 45 negative control samples for HPV-16 E6 and E7 DNA levels using real-time PCR analysis with primers and probes specific for HPV-16 E6 and E7 (see Table 1). The samples were considered positive when the E6 or E7 copy number per ng DNA was equal to or greater than the number obtained from the HPV-16 positive cell line SiHa. The percentages of these ThinPrep cervical samples that were positive for HPV-16 E6 or E7 DNA according to the above criterion using real-time PCR analysis were 0% for negative samples, 9.7% for ASCUS samples, 16.9% for LSIL samples, and 51.2% for HSIL samples (see Table 3). The percentages of samples that were positive for HPV-16 E6 or E7 DNA by conventional PCR

7 Real-Time PCR Quantitation of HPV-16 E6 and E7/Wang-Johanning et al TABLE 3 Detection of Human Papillomavirus 16 E6 and E7 DNA in ThinPrep Samples by Real-Time Polymerase Chain Reaction Analysis Compared with Conventional Polymerase Chain Reaction Analysis E6 positive E7 positive E6 or E7 positive Taqman Con. PCR Taqman Con. PCR Taqman Con. PCR Diagnosis No. a No % No % No % No. % No. % No. % Normal ASCUS LSIL HSIL PCR: polymerase chain reaction; Taqman: detection of DNA expression by real-time PCR using Taqman analysis; Con. PCR: detection of DNA expression using conventional PCR; ASCUS: atypical squamous cells of undetermined significance; LSIL: low-grade squamous intraepithelial lesion; HSIL: high-grade squamous intraepithelial lesion. TABLE 4 Human Papillomavirus 16 E6 and E7 DNA Copy Number in Human Papillomavirus 16 Positive Cervical Samples by Real-Time Polymerase Chain Reaction Analysis Copy no. a > 10 5 No. of samples b Diagnosis E6 E7 E6 E7 E6 E7 E6 E7 E6 E7 E6 E7 ASCUS % LSIL % HSIL % E6: E6 DNA positive; E7: E7 DNA positive; ASCUS: atypical squamous cells of undetermined significance; %: percentage of positive samples in each copy number group; LSIL: low-grade squamous intraepithelial lesion; HSIL: high-grade squamous intraepithelial lesion. a Number of E6 or E7 DNA copies per ng DNA. Bold numbers represent the largest percentage in each copy number category. b Total number of samples positive relative to SiHa cell copy number. amplification were 0% for negative samples, 11.8% for ASCUS samples, 15.3% for LSIL samples, and 51.2% for HSIL samples. Table 4 shows the E6 and E7 DNA copy number for various grades of SIL. The percentages of samples with greater DNA copy numbers became progressively greater as the severity of lesions increased. ThinPrep cervical cytologic samples that were positive for HPV-16 DNA were evaluated for RNA expression using real-time PCR analysis. We tested for RNA expression in all samples that were HPV-16 E6 or E7 DNA positive. However, because some E6 or E7 DNA negative ThinPrep cervical samples did not have enough cells for both DNA and RNA isolation (a total of 8 ASCUS samples, 7 LSIL samples, and 1 HSIL samples were in this category), RNA expression was not evaluated in these samples. In conventional RT- PCR analysis, the expression of E7 RNA, but not E6 RNA, was detected in some LSIL samples. Expression of both E6 RNA and E7 RNA was detected in HSIL samples. Figure 4 is a sample gel illustrating E6 and E7 RNA expression in ThinPrep specimens from a normal woman and from patients with ASCUS through HSIL. This figure demonstrates that RNA prepared from ThinPrep specimens can be amplified successfully by conventional RT-PCR analysis. Table 5 shows that the cervical ThinPrep samples with an ASCUS diagnosis had from 10 2 to 10 3 E6 or E7 RNA copies per ng RNA, LSIL samples had from 10 3 to 10 4 E6 or E7 RNA copies per ng RNA, and the majority of HSIL samples had from 10 4 to 10 5 E6 or E7 RNA copies per ng RNA. These results indicate that HPV-16 E6 or E7 RNA copy number increased as the cervical SIL progressed. The data from Table 5 and Figure 4 also indicate that E6 and E7 have different patterns of mrna expression in lower grades of SIL, with E7 expression predominating over E6 prior to the development of invasive cervical carcinoma.

8 2206 CANCER April 15, 2002 / Volume 94 / Number 8 FIGURE 4. Reverse transcriptase-polymerase chain reaction amplification of various cervical samples. (A) Normal cervical cells (lane 1), low-grade squamous intraepithelial lesion (LSIL; lane 2), and high-grade intraepithelial lesion (HSIL; lanes 3 and 4). (B) HSIL (lanes 1 and 8), LSIL (lanes 2 and 9), LSIL (lanes 3 and 10), atypical squamous cells of undetermined significance (ASCUS) plus lesion (lanes 4 and 11), HSIL (lanes 5 and 12), ASCUS plus lesion (lanes 6 and 13), and controls without added RNA (lanes 7 and 14). The lane designated 100 bp is a 100-base-pair DNA ladder (Life Technologies). Lanes 1 7 were amplified with the E6 primer pair, and lanes 8 14 were amplified with the E7 primer pair. Comparison of Real-Time PCR Analysis of DNA and RNA Expression in HSIL ThinPrep Samples ThinPrep cervical cytologic samples with a diagnosis of HSIL were compared for E6 and E7 DNA and RNA levels in the same patient (Table 6). Significantly, the RNA copy number, in many instances, was greater than the DNA copy number in the same patient. In some samples, even though DNA was below the limit of detection using real-time PCR analysis, the RNA copy number was detectable in the same sample. These results are in keeping with the previous findings that HSIL, which encompasses CIN II and CIN III, has a greater tendency for progression to invasive carcinoma. DISCUSSION The HPV oncoproteins E6 and E7 are expressed in SIL (also called CIN) and invasive carcinoma. 29,30 Previous reports indicated that HPV E6 and E7 are retained and expressed in cervical carcinoma tissues 31 and that they are essential for the immortalization of human primary keratinocytes 20 and the transformation of rodent cell lines. 32 In high-risk HPV types, there is conservation of sequences of the E6 and E7 genes to which p53 33 or Rb 19 bind, as well as E6 sequences that target degradation of p These studies collectively indicate that the HPV-16 E6 and E7 genes are involved in the etiology of cervical carcinoma. Recently, PCR analysis has been used to detect HPV DNA and RNA in cervical samples The content of HPV-16 mrna reportedly was significantly greater in a higher grade lesion (CIN III) or carcinoma than in a CIN I/normal diagnostic subgroup. 39 E6 and E7 transcripts of HPV were reported in virtually all HPV positive neoplasia specimens, except for the absence of E7 transcripts in some CIN specimens. 25,40 Although nested RT-PCR increased the sensitivity for detection of HPV-16 expression, 41,42 the likelihood of a false-positive result also will rise. No studies to date have provided a good methodology for quantitating E6 and E7 gene expression in cervical cells. In the study reported here, quantitation of both HPV-16 E6 and E7 DNA and RNA was investigated using real-time PCR analysis in four cervical carcinoma cell lines and in 348 residual ThinPrep cervical cytologic samples that were collected for routine Pap tests (303 samples with epithelial cell abnormalities and 45 samples that were interpreted as within normal limits). The advantages of using real-time PCR analysis in the context of evaluating E6 and E7 expression in cervical specimens is that a very small tissue sample can be used for the assay, and the oncogene expression can be quantitated. In addition, because the size of the target amplified is small, partial degradation of RNA in ThinPrep samples does not preclude its amplification by real-time RT-PCR analysis. ThinPreppreserved cells stored for 24 hours at room temperature or at 4 C had intact 28S and 18S ribosomal RNA, and both cellular and HPV messenger RNAs were amplified from preserved samples by RT-PCR. 43 In addition, we found that ThinPrep samples stored for a longer period (at least 2 months) also were capable of being amplified by RT-PCR. In our real-time RT-PCR or PCR analyses, smaller amounts of either RNA or DNA, respectively, were required to amplify E6 or E7 compared with conventional RT-PCR or PCR analyses. Because the amount of cervical cells available for analysis is sometimes limited, especially from healthy women, real-time PCR would allow for the detection of expression of greater numbers of relevant genes in cervical samples compared with other PCR methods. Another advantage of real-time PCR analysis using ThinPrep cervical cytologic samples is that samples collected in geographically remote locales, with limited access to a pathologist, can be sent at ambient temperature to a testing laboratory for analysis without special precautions to prevent nucleic acid degradation. For most degrees of SIL in this study, differences

9 Real-Time PCR Quantitation of HPV-16 E6 and E7/Wang-Johanning et al TABLE 5 Human Papillomavirus 16 E6 and E7 RNA Copy Number in Human Papillomavirus 16 Positive Cervical Samples by Real-Time Polymerase Chain Reaction Analysis Copy no. a > 10 5 No. of samples b Diagnosis E6 E7 E6 E7 E6 E7 E6 E7 E6 E7 E6 E7 ASCUS % LSIL % HSIL % E6: E6 DNA positive; E7: E7 DNA positive; ASCUS: atypical squamous cells of undetermined significance; LSIL: low-grade squamous intraepithelial lesion; HSIL: high-grade squamous intraepithelial lesion. a Number of E6 or E7 RNA copies per ng RNA. Bold numbers represent the largest percentage in each copy number category. b Total number of samples positive relative to SiHa cell copy number. TABLE 6 Comparison of Human Papillomavirus 16 E6 and E7 DNA and RNA Expression in High-Grade Intraepithelial Lesion ThinPrep Samples HPV-16 E6 HPV-16 E7 Sample no. Conventional PCR (DNA) a Real-time PCR (DNA) b Real-time RT-PCR (RNA) b Conventional PCR (DNA) a Real-time PCR (DNA) b Real-time RT-PCR (RNA) b 2 Negative p p p p N/A p N/A 17 p p p p p p p p p p p p p p p p? p Negative p p p p p p p p p p? p p p p p p p p p p p p? HPV-16: human papillomavirus 16; PCR: polymerase chain reaction; RT: reverse transcriptase; Negative: no band observed on gel; N/A: not analyzed; p?: very low expression. a Graded from weakly positive (p ) to strongly positive (p ) band intensity. b DNA or RNA copy number. in expression were observed between real-time PCR analysis and conventional PCR analysis. This may have been because some samples with low copy numbers of E6 or E7 DNA or RNA were detectable by real-time PCR analysis but not by conventional PCR amplification. Because of frequent contamination problems and consequent false-positive results using conventional PCR, there is a need to establish accurate methods to diagnose HPV infection. Comparing conventional PCR with real-time PCR and real-time RT-

10 2208 CANCER April 15, 2002 / Volume 94 / Number 8 PCR (Table 6), the data suggest that some conventional PCR signals were false-positive results. Realtime PCR, and particularly real-time RT-PCR, yielded results that tracked more accurately the progression of SIL/CIN to cervical carcinoma. Because the amounts of cells required for measurement of a target gene by real-time RT-PCR or real-time PCR are so small, the measurements also can be repeated to obtain estimates of standard deviations among samples. Two recent studies have used real-time PCR to quantitate HPV-16 and HPV-18 DNA in cervical tissues. 44,45 Those studies reported that real-time quantitation of HPV-16 and HPV-18 DNA is accurate over a large copy number range and is reproducible and type specific. Those authors also found good agreement between real-time and traditional DNA PCR analyses of HPV. However, the sample size was small in those studies, and detailed analyses of quantitation, as a function of grade of cervical lesions, was not provided. More importantly, HPV-16 RNA copy numbers were not evaluated or compared with DNA copy numbers in those studies. A significant finding in our study was that RNA copy numbers tended to be higher than DNA copy numbers in the same patient with HSIL; and, in some samples, RNA copy numbers, but not DNA copy numbers, were detectable in HSIL. This is in contrast to DNA copy numbers, which, in many samples, were not increased in cervical carcinoma. Higher RNA than DNA copy numbers also were present in ASCUS Thin- Prep specimens. We hypothesize that more cervical samples are positive for RNA than for DNA, because RNA copy number (RNA expression) may be a more sensitive indicator than DNA copy number of early changes in cervical cells from patients who are predisposed to develop higher grades of dysplasia and cervical carcinoma. This hypothesis can be tested by follow-up studies in patients with ASCUS and in patients with less severe grades of cervical dysplasia; we are in the process of following the patients in this study to determine whether this hypothesis is correct. It was reported recently that E6 and E7 act synergistically with chemical carcinogens to initiate tumor formation. 46 It was found that E6 acted weakly in the promotion stage of carcinogenesis in the formation for benign tumors but acted strongly in the progression stage, which involves the malignant conversion of benign tumors. In contrast, E7 primarily affected the promotion stage of carcinogenesis. Therefore, it appears that E7 promotes the formation of benign tumors, whereas E6 acts primarily to accelerate progression of these benign tumors to the malignant stage. That study suggested that E6 and E7 cooperate in inducing tumor formation in mice that express both oncogenes. Our results showing early expression of E7 and increased E6 expression in later stages of tumor progression support that carcinogenesis study. The sensitive real-time PCR and RT-PCR assays that we have developed can be used to assess and monitor early changes in E7 oncoprotein mrna expression, whereas the detection of E6 expression would signal later stages of disease progression in patients with SIL/CIN. The current results indicate that the application of real-time RT-PCR analysis of HPV-16 E6 and E7 oncogene expression is more sensitive and accurate than conventional PCR analysis. This likely is due to the capability of real-time RT-PCR to amplify very small amounts of nucleic acids and to quantify the results. Although the copy numbers of E6 or E7 DNA in the CaSki cell line were significantly greater compared with the SiHa cell line, the copy numbers of E6 or E7 RNA in the two cell lines were similar. This result suggests that RNA expression of both oncogenes, as measured by RNA copy number, is a better indicator of cervical carcinoma progression compared with DNA copy number. Our data show that the copy numbers of both E6 and E7 DNA and RNA increased as a function of grade of SIL, with the highest copy numbers present in the HSIL samples. Within a given sample, analysis of RNA expression is more sensitive than DNA expression. These results suggest that real-time PCR quantitation of HPV-16 E6 and E7 DNA and RNA can provide important information to incorporate in the evaluation and treatment of women with a diagnosis of SIL. It remains to be determined whether a comprehensive profile of E6 and E7 DNA and RNA levels are of value in predicting whether lower grades of SIL will progress to cervical carcinoma. REFERENCES 1. zur Hausen H. Papillomavirus infections a major cause of human cancers. Biochim Biophys Acta. 1996;1288:F Farthing A, Masterson P, Mason WP, Vousden KH. Human papillomavirus detection by hybrid capture and its possible clinical use. J Clin Pathol. 1994;47: Mitchell MF, Tortolero-Luna G, Wright T, et al. Cervical human papillomavirus infection and intraepithelial neoplasia: a review. J Natl Cancer Inst Monogr. 1996;21: Walboomers JM, Husman AM, Snijders PJ, et al. Human papillomavirus in false negative archival cervical smears: implications for screening for cervical cancer. J Clin Pathol. 1995;48: Manos M, Lee K, Greer C, et al. Looking for human papillomavirus type 16 by PCR. Lancet. 1990;335: Snijders PJ, van den Brule AJ, Schrijnemakers HF, Snow G, Meijer CJ, Walboomers JM. The use of general primers in the polymerase chain reaction permits the detection of a broad spectrum of human papillomavirus genotypes. J Gen Virol. 1990;71: Reid R, Lorincz AT. Human papillomavirus tests. Baillieres Clin Obstet Gynaecol. 1995;9:

11 Real-Time PCR Quantitation of HPV-16 E6 and E7/Wang-Johanning et al Brown DR, Bryan JT, Cramer H, Fife KH. Analysis of human papillomavirus types in exophytic condylomata acuminata by hybrid capture and Southern blot techniques. J Clin Microbiol. 1993;31: Fife KH, Katz BP, Roush J, Handy VD, Brown DR, Hansell R. Cancer-associated human papillomavirus types are selectively increased in the cervix of women in the first trimester of pregnancy. Am J Obstet Gynecol. 1996;174: Cavuslu S, Mant C, Starkey WG, et al. Analytic sensitivities of hybrid-capture, consensus and type-specific polymerase chain reactions for the detection of human papillomavirus type 16 DNA. J Med Virol. 1996;49: Cope JU, Hildesheim A, Schiffman MH, et al. Comparison of the hybrid capture tube test and PCR for detection of human papillomavirus DNA in cervical specimens. J Clin Microbiol. 1997;35: Shah KV, Solomon L, Daniel R, Cohn S, Vlahov D. Comparison of PCR and hybrid capture methods for detection of human papillomavirus in injection drug-using women at high risk of human immunodeficiency virus infection. J Clin Microbiol. 1997;35: Melbye M, Smith E, Wohlfahrt J, et al. Anal and cervical abnormality in women prediction by human papillomavirus tests. Int J Cancer. 1996;68: Poljak M, Brencic A, Seme K, Vince A, Marin IJ. Comparative evaluation of first- and second-generation digene hybrid capture assays for detection of human papillomaviruses associated with high or intermediate risk for cervical cancer. J Clin Microbiol. 1999;37: Watanabe S, Kanda T, Yoshiike K. Human papillomavirus type 16 transformation of primary human embryonic fibroblasts requires expression of open reading frames E6 and E7. J Virol. 1989;63: zur Hausen H. Human papillomaviruses in the pathogenesis of anogenital cancer. Virology. 1991;184: Jeon S, Lambert PF. Integration of human papillomavirus type 16 DNA into the human genome leads to increased stability of E6 and E7 mrnas: implications for cervical carcinogenesis. Proc Natl Acad Sci USA. 1995;92: Scheffner M, Huibregtse JM, Vierstra RD, Howley PM. The HPV-16 E6 and E6-AP complex functions as a ubiquitinprotein ligase in the ubiquitination of p53. Cell. 1993;75: Dyson N, Howley PM, Munger K, Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science. 1989;243: Hawley-Nelson P, Vousden KH, Hubbert NL, Lowy DR, Schiller JT. HPV16 E6 and E7 proteins cooperate to immortalize human foreskin keratinocytes. EMBO J. 1989;8: Munger K, Phelps WC, Bubb V, Howley PM, Schlegel R. The E6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes. J Virol. 1989;63: Vousden KH, Vojtesek B, Fisher C, Lane D. HPV-16 E7 or adenovirus E1A can overcome the growth arrest of cells immortalized with a temperature-sensitive p53. Oncogene. 1993;8: Cornelissen MT, Smits HL, Briet MA, et al. Uniformity of the splicing pattern of the E6/E7 transcripts in human papillomavirus type 16-transformed human fibroblasts, human cervical premalignant lesions and carcinomas. J Gen Virol. 1990;71: von Knebel Doeberitz M, Oltersdorf T, Schwarz E, Gissmann L. Correlation of modified human papilloma virus early gene expression with altered growth properties in C4-1 cervical carcinoma cells. Cancer Res. 1988;48: Rose BR, Thompson CH, Tattersall MH, Elliott PM, Dalrymple C, Cossart YE. Identification of E6/E7 transcription patterns in HPV 16-positive cervical cancers using the reverse transcription/polymerase chain reaction. Gynecol Oncol. 1995;56: Rose BR, Thompson CH, Jiang XM, et al. Detection of human papillomavirus type 16 E6/E7 transcripts in histologically cancer-free pelvic lymph nodes of patients with cervical carcinoma. Gynecol Oncol. 1994;52: Wang-Johanning F, Frost AR, Johanning GL, et al. Expression of human endogenous retrovirus K envelope transcripts in human breast cancer. Clin Cancer Res. 2001;7: Wang-Johanning F, Gillespie GY, Grim J, et al. Intracellular expression of a single-chain antibody directed against human papillomavirus type 16 E7 oncoprotein achieves targeted antineoplastic effects. Cancer Res. 1998;58: Pillai MR, Halabi S, McKalip A, et al. The presence of human papillomavirus-16/-18 E6, p53, and Bcl-2 protein in cervicovaginal smears from patients with invasive cervical cancer. Cancer Epidemiol Biomarkers Prev. 1996;5: Auvinen E, Kujari H, Arstila P, Hukkanen V. Expression of the L2 and E7 genes of the human papillomavirus type 16 in female genital dysplasias. Am J Pathol. 1992;141: Schwarz E, Freese UK, Gissmann L, et al. Structure and transcription of human papillomavirus sequences in cervical carcinoma cells. Nature. 1985;314: Yutsudo M, Okamoto Y, Hakura A. Functional dissociation of transforming genes of human papillomavirus type 16. Virology. 1988;166: Werness BA, Levine AJ, Howley PM. Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science. 1990;248: Crook T, Tidy JA, Vousden KH. Degradation of p53 can be targeted by HPV E6 sequences distinct from those required for p53 binding and trans-activation. Cell. 1991;67: de Villiers EM, Wagner D, Schneider A, et al. Human papillomavirus DNA in women without and with cytological abnormalities: results of a 5-year follow-up study. Gynecol Oncol. 1992;44: Schiffman MH, Bauer HM, Hoover RN, et al. Epidemiologic evidence showing that human papillomavirus infection causes most cervical intraepithelial neoplasia. J Natl Cancer Inst. 1993;85: Bernard C, Mougin C, Laurent R, Lab M. Oncogene activation: an informative marker for the human papillomavirus lesions severity. Cancer Detect Prev. 1994;18: de Roda Husman AM, Walboomers JM, Meijer CJ, et al. Analysis of cytomorphologically abnormal cervical scrapes for the presence of 27 mucosotropic human papillomavirus genotypes, using polymerase chain reaction. Int J Cancer. 1994;56: Forslund O, Lindqvist P, Haadem K, Czegledy J, Hansson BG. HPV 16 DNA and mrna in cervical brush samples quantified by PCR and microwell hybridization. J Virol Methods. 1997;69:

12 2210 CANCER April 15, 2002 / Volume 94 / Number Nakagawa S, Yoshikawa H, Yasugi T, et al. Ubiquitous presence of E6 and E7 transcripts in human papillomaviruspositive cervical carcinomas regardless of its type. J Med Virol. 2000;62: Sotlar K, Selinka HC, Menton M, Kandolf R, Bultmann B. Detection of human papillomavirus type 16 E6/E7 oncogene transcripts in dysplastic and nondysplastic cervical scrapes by nested RT-PCR. Gynecol Oncol. 1998;69: Fujii T, Tsukazaki K, Kiguchi K, Kubushiro K, Yajima M, Nozawa S. The major E6/E7 transcript of HPV-16 in exfoliated cells from cervical neoplasia patients. Gynecol Oncol. 1995;58: Dimulescu II, Unger ER, Lee DR, Reeves WC, Vernon SD. Characterization of RNA in cytologic samples preserved in a methanol-based collection solution. Mol Diagn. 1998;3: Tucker RA, Unger ER, Holloway BP, Swan DC. Real-time PCR-based fluorescent assay for quantitation of human papillomavirus types 6, 11, 16, and 18. Mol Diagn. 2001;6: Cubie HA, Seagar AL, McGoogan E, et al. Rapid real time PCR to distinguish between high risk human papillomavirus types 16 and 18. Mol Pathol. 2001;54: Song S, Liem A, Miller JA, Lambert PF. Human papillomavirus types 16 E6 and E7 contribute differently to carcinogenesis. Virology. 2000;267:141 l50.