Detection of Human Papillomavirus on Papanicolaou-Stained Cervical Smears Using Indirect In Situ Polymerase Chain Reaction Hybridization.
Polymerase chain reaction is an extremely sensitive method[5,6] and can detect a single copy of a specific DNA sequence in cells or clinical samples. Polymerase chain reaction technology has been applied in diagnosing a wide range of clinical conditions, infectious diseases in particular.[2,8,9] Several authors have now directly applied PCR technology to induct cells after fixation on glass slides. The amplicon is then revealed by ISH with a specifically labeled probe allowing direct or indirect detection.[10-14] Known as IS-PCR, this procedure combines the potentials of both the PCR and the ISH methods, allowing the detection of the specific DNA target sequences and the identification of positive cells in a mixed cell population.
Nuovo et al described the direct IS-PCR method, in which labeled nucleotides are directly incorporated into the PCR product and enzyme-immunohistochemical detection. One disadvantage of this method is the high rate of false-positive results due to DNA repair artifacts found in paraffin-embedded material. Specimens exposed to dry heat are therefore inappropriate for direct IS-PCR.
In the present study, we wanted to determine whether indirect IS-PCR could reveal the presence of HPV DNA in PAP-stained smears by using solution-phase PCR (SP-PCR) as a reference method. Therefore, the description of the methods and their sensitivity levels will be limited to their application in detecting HPV DNA sequences in PAP-stained cells. The advantages and disadvantages of indirect IS-PCR will be discussed. The squamous intraepithelial lesion (SIL) grades with viral cytologic features were confirmed in a histologic report of punch biopsies or conization and were histologically classified according to Ferenczy and Winkler. The cytologic diagnosis was based on the Bethesda System. In addition, the sensitivity of indirect IS-PCR was tested in cervical carcinoma cell lines containing HPV DNA with a defined copy number per genome.
MATERIALS AND METHODS
Papanicolaou-Stained Cells, Cervical Smears, Cell Lines, and Controls for PCR
Two PAP-stained cervical smears taken by Cytobrush (Medscan AB, Malmo, Sweden) under colposcopy and 1 specimen of saline cervicovaginal lavage were obtained simultaneously from 162 patients at the Department of Gynecology, University Hospital, Vienna, Austria. Specimens were stored at room temperature and examined for HPV DNA. These PAP-stained smears were classified as low-grade SILs (LSILs; n = 81; mean age, 27 years) and high-grade SILs (HSILs; n = 81; mean age, 34 years), according to the Bethesda System. Cases only including cytologic SILs confirmed by punch biopsy or conization were also evaluated in this study. Additionally, cervical cancer cell line SiHa (HTB35, American Type Culture Collection [ATCC], Rockville, Md) containing 1 to 10 copies of HPV-16 DNA per cell, CaSki (CRL 1550, ATCC) containing more than 100 copies of HPV-16 DNA per cell, and HeLa cell line (ATCC) containing 10 to 50 copies of HPV-18 DNA per cell were PAP-stained and used as positive controls. Negative controls included (a) using PAP-stained breast carcinoma cell lines (MCF-7, HTB22, ATCC), (b) eliminating AmpliTaq DNA polymerase (Perkin Elmer Cetus, Norwalk, Conn) in the reaction mixture, and (c) omitting the primers, the most frequently used control for excluding false-positive results. As an additional control of specificity we checked the amplifiability of indirect IS-PCR using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers (GAPDH as ubiquitous gene). Sequences of oligonucleotide primers were as follows: GAPDH 1 (20-mer, upstream primer, 5'-ATC ACC ATC TTC CAG GAG CG-3') and GAPDH 2 (20-mer, downstream primer, 5'-CCT GCT TCA CCA CCT TCT TG-3'), 573 base pairs (bp).
Slide and Sample Preparation (Decolorization)
We removed the stain to distinguish the dark-blue stained, HPV DNA--positive nuclei from the hematoxylin-stained nuclei. The slides were placed in N-butylacetate (Merck, Darmstadt, Germany) until the coverslips and permanent medium were detached. For decolorization, slides were washed with 96% alcohol for 5 minutes, with 3% hydrochloric ethanol for 8 minutes, and with 3% sodium carbonate for 2 minutes. The alcohol-fixed PAP smears were fixed in 7.5% neutral-buffered formalin for 12 hours to attach the nucleic acids. This postfixation procedure prevented loss of the amplification product. The slides were then put in a humidified box and incubated with 2 mg/mL pepsin for 10 minutes at 37 [degrees] C. To stop digestion, the slides were washed in a solution of 0.1 mol/L sodium chloride and 0.1 mol/L Tris/HCl (pH 7.5) and dehydrated through a graded ethanol series.
PCR and Primers
Consensus primers MY09 (20-mer, 5'-CGT CCM ARR GGA WAC TGA TC-3'; degenerate code M = A + C, R = A + G, W = A + T, Y = C + T) and MY11 (20-mer, 5'-GCM CAG GGW CAT AAY AAT GG-3') of the L1 region (GeneAmplimer HPV, Perkin Elmer Cetus) were used. They detected a 410-bp product of the following HPV types: low-risk types 6 and 11, and high-risk types 16 and 18.
The amplification was performed using an Omnigene Thermal Cycler (Hybaid, Teddington, Middlesex, United Kingdom) with a solid heating block. The amplification solution consisted of the following: 1 [micro]mol/L of each consensus primer; 200 [micro]mol/L each of deoxyadenosine triphosphate, deoxycytidine triphosphate, and deoxyguanosine triphosphate; and 140 [micro]mol/L of deoxythymidine triphosphate. On each slide, we deposited 5 ng of probe in 20 [micro]L of hybridization buffer (2x SSC, 50% deionized formamide, 10% dextran sulfate, and 250 [micro]g/mL salmon sperm DNA) and added a coverslip. After denaturing target and probe DNA simultaneously in a water bath at 96 [degrees] C for 5 minutes, amplified products were detected with biotinylated oligonucleotide probes specific for HPV low-risk types 6 and 11 or high-risk types 16 and 18.
Oligonucleotide probes for HPV-L1 (6/11; 16/18) amplification products were as follows: MY125 (5'-ACA ATG AAT CCY TCT GTT TTG G-3'; degenerate code Y = C + T), MY95 (5'-GAT ATG GCA GCA CAT AAT GAC-3'), and MY130 (5'-GGG CAA TAT GAT GCT ACC AAT-3') (all supplied by Vienna Biocenter, IMP, Vienna, Austria).
In Situ Hybridization
The ISH was performed overnight in a humidified atmosphere at 42 [degrees] C. We removed the coverslips by dipping slides in 4 x SSC at room temperature for 1 to 2 minutes. After washing at high stringency in 2x SSC at room temperature, then in 2x SSC, 0.5x SSC, and 0.1x SSC bath at 42 [degrees] C, we added 100 [micro]L streptavidin alkaline phosphatase (Boehringer Mannheim, Mannheim, Germany) with buffer in a 1:200 dilution, then incubated the slides for 30 minutes at 37 [degrees] C. After rinsing twice in 1x phosphate-buffered saline for 5 minutes, we prepared a substrate buffer by diluting nitroblue tetrazolium (0.33 mg/mL) and 5-bromo-4-chloro-3-indolyl phosphate (0.16 mg/mL) and added 0.24 mg of Levamisole per milliliter of substrate buffer (Sigma, St Louis, Mo). After mixing, 100 [micro]L of the solution was placed on the slides, followed by incubation at room temperature in the dark for 1 to 3 hours (depending on the color development). The reaction was stopped by transferring the slides twice to a staining dish containing deionized water for 5 minutes. Before adding 1 drop of a permanent histologic mounting medium (Pertex, Medite, Burgdorf, Germany), the slides were counterstained with nuclear fast red, soaked in N-butylacetate, covered with a coverslip, and examined under a light microscope. A blue signal was observed in positive cells.
HPV DNA: Solution-Phase PCR
Saline Cervicovaginal Lavage.--Specimens were collected as saline lavage from the same 162 patients (typically containing 1-10 [micro]g DNA). To avoid cross-contamination, it was important to minimize handling of the samples at the clinic at which they were collected. The specimens were collected in phosphate-buffered saline.
We used digestion buffer (50 mmol/L Tris-HCl [pH 8.5], 1 mmol/L EDTA) and added proteinase K (20 mg/mL) and Tween 20 nonionic detergent in sterile distilled water.
We prepared 2x digestion solution containing 400 [micro]g/mL of proteinase K and 2% Tween 20 in digestion buffer. Fifty micro-liters of the sample were added to a microcentrifuge tube containing 50 [micro]L of 2x digestion solution, incubated at 55 [degrees] C for 1 hour, and then spun briefly in a microcentrifuge at 12 000g. Proteinase K was inactivated by incubating the tubes at 95 [degrees] C for 10 minutes. The samples were stored at -20 [degrees] C.
Amplification of HPV-L1 Fragment (410 bp).--Each PCR reaction contained Taq PCR buffer (10 mmol/L Tris-HCl, 50 mmol/L KCl, 4 mmol/L Mg[Cl.sub.2]); 200 mmol/L each of deoxynucleoside triphosphate; 2.5 units of Taq polymerase; 0.5 mmol/L each of HPV primer, MY09, and MY11; and an aliquot of mineral oil in a 0.5-mL microcentrifuge tube. The final reaction volumes varied up to 10% without affecting amplification efficiency. Finally, we added the sample DNA and centrifuged the samples in a clean PCR microcentrifuge for 10 seconds to force the aqueous liquid under the oil. The thermal cycler program was as follows: 35 cycles at 95 [degrees] C for 1 minute of denaturation, 55 [degrees] C for 1 minute of annealing, and 72 [degrees] C for 1 minute of extension. The extension cycle was at 72 [degrees] C for 5 minutes, and samples were then soaked at 15 [degrees] C.
Polyacrylamide Gel Electrophoresis (PAGE) Analysis of PCR Products.--We prepared a 1-mm-thick, 7% bis-acrylamide (PAGE) gel in TBE buffer (50 mmol/L Tris-HCl, 67 mmol/L boric acid, 1 mmol/L EDTA) with a 14-well comb and combined 5 [micro]L of the PCR product with 5 [micro]L of a 2x gel-loading dye (0.5 mg/ mL bromophenol blue, 0.5 mg/mL xylene cyanol, 50 mg/mL Ficoll 400, 0.1 mol/L EDTA [pH 8.0], and 0.1% of sodium dodecyl sulfate), filled the sample wells with this solution, and included 1 lane with a molecular weight marker (250-500 ng) on every gel. The gels ran until the bromophenol blue dye passed out from the bottom of the gel. We then stained the gel in a dilute ethidium bromide solution (10 [micro]g/mL sterile distilled water) for 2 minutes. Photographs were made using an orange filter and a UV transilluminator.
The histologic results were obtained from 162 women with cervical intraepithelial neoplasia (CIN) grades I through III (CIN I-III). In 2 cases, the histology (CIN I) did not confirm the cytologic finding (HSIL). Compared to the histologic results, the diagnostic efficiency of the cytologic diagnosis was 98.8%. The 8 cases of mixed HPV (6/11 and 16/18) infection found by indirect IS-PCR were HSILs in the cytologic results and CIN III in the histologic results.
The nuclei of the SiHa cells (limit of sensitivity, positive control, 1-10 copies of HPV-16 DNA per cell) showed blue staining, which was also used as the criterion for HPV DNA positivity. In both groups (ie, LSIL and HSIL), HPV DNA was found to be positive (Figure 1), showing blue staining. This finding was evident in koilocytes (Figure 2), binucleated cells, and dysplastic cells, as well as in normal squamous epithelium (Figure 3). The primer independent pathway of DNA amplification was excluded by performing the indirect IS-PCR method without primers on SiHa cells. The nuclei of this cell line assumed a reddish counterstain either without primers or without AmpliTaq. The nuclei of the HTB22 cells were also stained reddish, which indicated a negative reaction.
[Figures 1-3 ILLUSTRATION OMITTED]
The Table shows that up to 60.5% (98/162) of PAP-stained cell smears were positive for HPV DNA by indirect IS-PCR, including 8 cases (8.2%) of mixed HPV infection. At least 8 cases were found to be negative by indirect IS-PCR, but were positive according to the native SP-PCR method. The Table also shows the diagnostic accuracy of HPV DNA by the native SP-PCR method as compared to indirect IS-PCR. Regardless of the HPV genotypes, the positive reference method corresponded to a positive HPV DNA diagnosis by indirect IS-PCR in 91.8% (90/98) of the cases. Irrespective of the HPV typing and sampling method, the reference method confirmed the HPV DNA-positive results of indirect IS-PCR in 94.8% (182/192) of the cases.
Comparison of the Solution-Phase Polymerase Chain Reaction (PCR) (Reference) Method Versus Indirect In Situ PCR of Papanicolaou Smears for Detection of Human Papillomavirus(*)
Solution-Phase PCR (Reference) LSIL HSIL Positive Negative Positive Negative Indirect in situ PCR 48 2 42 6 Positive 8 104 0 150 Negative Diagnostic accuracy Sensitivity, % 85.7 100 Specificity, % 98.1 96.1 Positive predictive 96.1 87.5 value, % Negative predictive 92.9 100 value, % Efficiency, % 93.8 96.9
(*) LSIL indicates low-grade squamous intraepithelial lesion; HSIL, high-grade squamous intraepithelial lesion.
Solution-phase PCR is a sensitive method for detecting single copy genes, but cells and tissue need to be disrupted in order to isolate DNA. The extraction procedure is incompatible with the preservation of cell morphology, and the samples consist of a cell suspension or histologic tissue specimens. By comparison, ISH allows a direct association between the molecular result and its morphologic characteristics, although the sensitivity of ISH is limited.
Indirect IS-PCR allows an association to be drawn between HPV DNA-positive cells and their cytologic characteristics, such as koilocytes, dyskeratocytes, and irregular nuclei. This technique is almost as sensitive as SPPCR. This sensitivity has been demonstrated by several authors, including Zehbe et al, who used IS-PCR to detect SiHa cells. The use of IS-PCR on PAP-stained smears has been reported in the literature,[3,4,19,20] but cells with a low HPV DNA copy number were not detected. The 2 advantages of the indirect IS-PCR are excellent preservation of cytologic features and minimization of cell loss.
We found that sufficient digestion of cells by proteinase K was crucial for the successful application of indirect ISPCR. If the cell layers are too thick, HPV DNA-positive cells can only be identified in cell monolayers, where the reaction mixture has diffused into the cell nucleus. Papanicolaou staining and destaining do not interfere with indirect IS-PCR for HPV DNA detection. This was demonstrated in our positive controls under the same conditions. We found that indirect IS-PCR facilitated the diagnosis of PAP smears, particularly if cellular changes associated with HPV were evident.
In Situ Amplification
For IS-PCR performed directly on glass slides, the cellular material is overlaid with PCR mixture, and steps are taken to prevent evaporation of the PCR reaction mixture. To avoid such evaporation, the edges of the coverslips were sealed with rubber cement (Fixogum, Marabuwerke GmbH & Co, Tamm, Germany) or the coverslips were completely covered with mineral oil. Evaporation could alter the concentration of the reagents and affect the efficiency of the procedure.
Detection of the Intracellular PCR Product
By using probes, the indirect methods recognize the amplified sequence and provide maximum specificity in the detection of the intracellular PCR product. At the theoretical level, probes targeting regions between the primers are ideal in achieving the highest specificity. Oligonucleotide probes and PCR-generated probes meet this standard. However, full-length probes or genomic probes are often used for the sake of convenience or because of the increased number of reporter molecules that produce a higher signal intensity.
Diffusion artifacts represent a significant problem for indirect IS-PCR performed on cells in suspension. Polymerase chain reaction products and DNA leak out of template-positive cells and serve as templates for extracellular amplification, which is probably far more efficient than intracellular amplification.
In agreement with Haase et al, we observed more problems related to diffusion artifacts when short DNA sequences were amplified by IS-PCR, but others have convincingly shown that the problems persist with amplicons of up to 740 bp. It has been suggested that diffusion artifacts can be significantly reduced by optimal fixation and permeabilization, reduction of PCR cycling numbers, and generation of longer or more complex PCR products.
False-negative results on glass slides may be explained by the loss of PCR products while washing during the detection procedure or DNA polymerase cell derivative inhibitors. The poor amplification efficiency and the reproducibility of IS-PCR can also yield false-negative results. Poor quality or reduced accessibility of target sequences related to the protein DNA cross-linking properties of routinely used fixatives, such as formaldehyde, almost certainly contribute to false-negative results. These cross-links can be partially reversed by appropriate treatment of samples by proteases, hydrochloric acid, and heat.
Controls for Indirect IS-PCR Results
Additional indirect IS-PCR--specific controls (checking the amplifiability using GAPDH primers) should be performed for each experiment, including omission of DNA polymerase or use of irrelevant probes for ISH. The purpose of these controls is to detect nonspecific probes, antibody sticking, and ISH specificity.
Conclusion for Indirect IS-PCR
In view of the progress achieved so far, this new technology might well become a powerful technique for routine diagnosis in laboratories. Indirect IS-PCR is a method of intracellular PCR amplification used to detect specifically amplified DNA sequences by hybridization of labeled probes. Now we can study the course and determination of viral activation, as well as the morphologic correlates. The analysis using indirect IS-PCR will also focus on other intracellular infectious agents, with obvious diagnostic implications. The method can also be used to localize altered genes or foreign DNA in single-cell preparations and to study DNA mutations or chromosomal translocations in order to understand the effect and the latency period between DNA alterations and the morphologic characteristics of malignancy. In the future, IS-PCR will be a tool for evaluating novel gene therapies, since it will not only allow tracing of cellular differentiation of transfected cells, but also will show the incorporation of the transfected genes at the chromosomal level. In addition, this method is highly sensitive and permits the detection of single-copy species.
Regarding the disadvantages, it should be taken into consideration that thermocycling requires appropriate instrumentation. To date, the reproducibility of this method is variable, labor intensive, and usually requires ISH in addition to thermocycling. However, suitable equipment, standardized protocols, and improved reporter systems can minimize these disadvantages.
[1.] Meisels A, Morin C. Cytopathology of the Uterine Cervix. Chicago, Ill: ASCP Press; 1991:73-117.
[2.] Nuovo GJ. Applications of PCR in situ hybridization: human papillomavirus. In: Nuovo GJ, ed. PCR In Situ Hybridization: Protocols and Applications. 3rd ed. New York, NY: Lippincott-Raven Press; 1996:334-380.
[3.] Komminoth P, Long AA. In situ polymerase chain reaction: an overview of methods, applications and limitations of a new molecular technique. Virchows Arch B Cell Pathol. 1993;64:67-73.
[4.] Zehbe I, Hacker GW, Rylander E, Sallstrom J, Wilander E. Detection of single HPV copies in SiHa cells by in situ polymerase chain reaction (in situ PCR) combined with immunoperoxidase and immunogold-silver staining (IGSS) techniques. Anticancer Res. 1992;12:2165-2168.
[5.] Mullis K, Faloona F, Scharf SJ, Saiki RK, Horn GT, Erlich HA. Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harbor Symp Quant Biol. 1986;51:263-273.
[6.] Saiki RK, Gelfand DH, Stoffel S, et al. Primer directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science. 1988;239:487491.
[7.] Kawasaki ER, Saiki RK, Ehrlich HA. PCR technology. In: Ehrlich HA, ed. Principles and Applications for DNA Amplification. Stockton, NY: Stockton Press; 1989:170-242.
[8.] Mougin C, Didier JM, Bettinger D, Madoz L, Coumes-Marquet S, Lab M. In situ PCR to cells and to wax sections. In: Gosden JR, ed. PRINS and In Situ PCR Protocols. Totowa, NJ: Humana Press Inc; 1997:77-97. Methods in Molecular Biology.
[9.] Ehrlich GD, Greenberg S, Abbott MA. Detection of human T-cell lymphoma/leukemia viruses. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR Protocols: A Guide to Methods and Applications. San Diego, Calif: Academic Press Inc; 1990:325-386.
[10.] Komminoth P, Long AA, Ray R, Wolfe HJ. In situ polymerase chain reaction detection of viral DNA, single-copy genes, and gene rearrangements in cell suspensions and cytospins. Diagn Mol Pathol. 1992;1:85-97.
[11.] Nuovo GJ, Forde A, MacConnell P, Fahrenwald R. In situ detection of PCR-amplified HIV-1 nucleic acids and tumor necrosis factor in cervical tissues. Am J Pathol. 1993;143:40-48.
[12.] Bagasra O, Pomerantz R. Human immunodeficiency virus type I provirus is demonstrated in peripheral blood monocytes in vivo: a study utilizing an in situ polymerase chain reaction. AIDS Res Hum Retroviruses. 1993;9:69-76.
[13.] Bagasra O, Seshamma T, Pomerantz R. Polymerase chain reaction in situ: intracellular amplification and detection of HIV-1 proviral DNA and other specific genes. J Immunol Methods. 1993;158:131-145.
[14.] Bagasra O, Seshamma T, Hansen J, et al. Application of in situ PCR methods in molecular biology. Cell Vision. 1994;1:324-335.
[15.] Ferenczy A, Winkler B. Cervical intraepithelial neoplasia and condyloma. In: Kurman RJ, ed. Blaustein's Pathology of the Female Genital Tract. New York, NY: Springer-Verlag; 1987:177-217.
[16.] The 1988 Bethesda System for reporting cervical/vaginal cytological diagnoses. JAMA. 1988;262:931-934.
[17.] Yee C, Krishnan-Hewlett I, Baker CC, Schlegl R, Howley PM. Presence and expression of human papillomavirus sequences in human cervical carcinoma cell lines. Am J Pathol. 1985;119:361-366.
[18.] Schwarz E, Freese UK, Gissmann L, et al. Structure and transcription of human papillomavirus sequences in cervical carcinoma cells. Nature. 1985;314: 111-114.
[19.] Choi YJ. Detection of human papillomavirus DNA on routine Papanicolaou's smears by in situ hybridization with the use of biotinylated probes. Am J Clin Pathol. 1991;95:475-480.
[20.] Liang X-M, Wieczorek RL, Koss LG. In situ hybridization with human papillomavirus using biotinylated DNA probes on archival cervical smears. J Histochem Cytochem. 1991;39:771-775.
. Haase TA, Retzel EF, Staskus KA. Amplification and detection of lentiviral DNA inside cells. Proc Natl Acad Sci U S A. 1990;87:4971-4975.
[22.] Teo IA, Shaunak S. PCR in situ: aspects which reduce amplification and generate false-positive results. Histochem J. 1995;27:660-669.
Accepted for publication October 30, 2000.
From the Department of Gynecology and Obstetrics, Division of Special Gynecology (Dr Manavi), and the Institute of Clinical Pathology, Division of Gynecopathology (Drs Bauer and Czerwenka, and Ms Pischinger), University Hospital of Vienna, Vienna, Austria.
Reprints: Mahmood Manavi, MD, Department of Gynecology and Obstetrics, Division of Special Gynecology, University Hospital, University of Vienna, Wahringer Gurtel 18-20, A-1090 Vienna, Austria.
|Printer friendly Cite/link Email Feedback|
|Author:||Manavi, Mahmood; Bauer, Margit; Pischinger, Kerstin; Czerwenka, Klaus|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Mar 1, 2001|
|Previous Article:||Novel Method for Simultaneous Analysis of p53 and K-ras Mutations and p53 Protein Expression in Single Histologic Sections.|
|Next Article:||Verumontanum Mucosal Gland Hyperplasia Is Associated With Atypical Adenomatous Hyperplasia of the Prostate.|