JOURNAL OF PATHOLOGY, VOL. 178: 249-254 (1996) EVIDENCE FOR DISSOCIATION OF HISTONE mRNA EXPRESSION FROM CELLULAR PROLIFERATION IN CUTANEOUS HUMAN PAPILLOMAVIRUS INFECTION STEPHEN A. HINCHLIFFE", MARTIN D. SMITH?,MATIIILDE E. JONATHAN BOON$,c. VIVYAN HOWARD§, DICK L. R E E S ~ VAN V E L Z E N ~AND *Department of Pathology, University of Newcastle upon Tyne, U.K.; ?Department of Dermatology, University of Newcastle upon Tyne, U.K.; fLeiden Cytology and Pathology Laboratories, Leiden, The Netherlands; $Department of Fetal and Infant Pathology, University of Liverpool, U.K. SUMMARY Replication of human papillomavirus (HPV) is thought to require the expression of components of the host cell DNA replication apparatus. The expression of the mRNA encoding the DNA replication-dependent histones H2B, H3, and H4 was investigated in a series of verrucae using an FITC-labelled oligonucleotide cocktail for in situ hybridization, to explore the possibility of HPV inducing the transcription of host cell histones. In contrast to normal epidermis, in which the majority of histone mRNA labelling was confined to basal cells, a bimodal distribution of histone positivity was observed in all verrucae. In addition to basal hyperproliferation-associated labelling, numerous positive suprabasal cells were found within the stratum spinosum. By contrast, in another hyperproliferative lesion, psoriasis, the labelling pattern was similar to that found in the normal controls, with characteristic basal hyperproliferation but no evidence of suprabasal labelling. The presence of HPV replication in every verruca and its absence from all controls of normal and psoriatic skin were demonstrated immnnohistochemically, using a polyclonal antibody to the viral capsid proteins. It is proposed that the presence of histone mRNAs in differentiated suprabasal keratinocytes of verrucae is due to the virus turning on parts of the cellular apparatus in order for it to replicate in the absence of cellular proliferation. The results also emphasize that when assessing proliferation with in situ hybridization for histone mRNA, the possibility of HPV causing the labelling of non-proliferating cells should be considered. KEY WORDS-DNA replication; histone mRNA; in situ hybridization; human papillomavirus; skin INTRODUCTION Human papillomaviruses (HPVs) are epitheliotropic, double-stranded DNA viruses associated with a variety of proliferative squamous lesions of the skin and mucosal surfaces lining the anogenital tract, oral cavity, and larynx. The oncogenic potential of certain HPV types is reflected in their association with squamous cell carcinoma of the skin' and uterine cervix.2 In contrast to other DNA viruses, the process of replication of HPV is poorly understood, largely as a consequence of the inability, until very recently, to propagate the virus in cell culture and reproduce the infectious ~ y c l e . ~ There , ~ is evidence, however, that vegetative viral DNA replication occurs only in squamous epithelial cells undergoing terminal different i a t i ~ n . In ~ . basal ~ cell nuclei, the viral genome is stably maintained as low copy number, monomeric, extrachromosomal plasmids. As the infected cell undergoes terminal differentiation, the viral DNA undergoes vegetative replication, producing thousands of copies per cell. Late genes encoding the viral capsid proteins are subsequently expressed and virions are a ~ s e m b l e d . ~ , ~ It has become apparent that of the many proteins necessary for replication of viral DNA, HPV encodes only the replication origin binding proteins El and E2.x,9Therefore it has been postulated that HPV must Addressee for correspondence: S. A. Hinchliffe, MB, PhD, Department of Pathology, University of Newcastle upon Tyne, Royal Victoria Infirmary, Newcastle upon Tyne NEI 4LP, U.K. CCC 0022-34 17/96/030249-06 0 1996 by John Wiley & Sons, Ltd. have a mechanism for reactivation of the host cell DNA replication apparatus, to facilitate its own replication in cells undergoing terminal differentiation which do not replicate their chromosomal DNA and thus presumably no longer transcribe those genes essential for DNA synthesis.l o In this regard, several authors have demonstrated the induction of proliferating cell nuclear antigen (PCNA), an accessory protein of host DNA polymerase delta and an essential part of the cellular replication machinery, in differentiated suprabasal cells of HPV-infected skin (verruca vulgaris)' and cervix (condyloma acuminatum and low-grade cervical intraepithelial neoplasia),1° which was independent of the oncogenic risk potential of the infecting HPV genotype. These data support the hypothesis of an interaction between HPV and host cell replication processes and provides impetus for a study of other cellular genes associated with DNA synthesis. Recently, a technique has been described for the detection of the messenger RNA (mRNA) transcripts of several cellular histones by in situ hybridization (ISH). l37l4 Expression of the highly conserved H2A, H2B, H3, and H4 histone genes,I5 encoding small basic proteins involved in the folding of eukaryotic DNA, is tightly coupled to the replicative stage of the cell ~ y c l e . ~The ~.'~ number of mRNA transcripts encoding these replication-dependent proteins increases 30- to 100-fold on entering S-phase, due to the combination of an approximately two- to three-fold increase in the rate of histone gene transcription, together with an increase in the mRNA half-life from 8-12 to 45-60 min.1x,19 As Received I December 1994 Accepted 7 July 1995 250 S.A. HINCHLIFFE E T A L the cell enters G2, and throughout M, G1, and GO, transcripts are rapidly destabilized owing to the presence of a unique stem-loop sequence in the 3' non-coding region of the mRNA2" and the absence of a 3' polyadenylate tail.'* Non-S-phase cells thus contain a small but undetectable pool of transcripts. Detection of these histone transcripts by ISH may therefore be considered a specific marker for cell cycle-related DNA replication.*' This technique has been proposed as an additional assessment of cellular proliferation in routinely processed tissue section^.'^.'^.^^ Results suggest that histone mRNA ISH provides an estimation of the proportion of cells in S-phase which is of comparable sensitivity to that determined by flow cytometry,13 [3H]thymidine labelling,'4,23or BrdU i n c ~ r p o r a t i o n . ~ ~ Following alkaline disruption of HPV particles purified from human plantar warts, four low-molecularweight polypeptides closely associated with the viral capsids were resolved and noted to be similar to the cellular histones H2A, H2B, H3, and H4.25 As we have previously demonstrated the excellent cellular resolution obtained using non-isotopic ISH (NISH) for the detection of mRNA species in skin,26 we considered it of interest to investigate the expression of DNA replication-dependent histone mRNAs in a series of verrucae to explore the possibility of HPV inducing the transcription of host cell histones. MATERIALS AND METHODS Specimen selection The study group comprised 12 consecutive excision specimens of verruca vulgaris, which had been referred for routine histopathological examination at either the LCPL Laboratories, Leiden ( n = 2) or the Department of Fetal and Infant Pathology, Liverpool (n= 10). All lesions were derived from skin of the hand or foot [five males, seven females; mean age 15.4 (4-42) years]. In addition, control biopsies were taken at the Department of Fetal and Infant Pathology, Liverpool and at the Department of Dermatology, Newcastle of normal skin [n= 16; ten males, six females; mean age 16.4 ( 8 4 9 ) years] and of psoriatic plaques, as an example of a non-HPV-associated hyperproliferative disorder [n= 5; three males, two females; mean age 48.8 (28-66) years]. None of these individuals had a history of cutaneous HPV infection. Informed consent was obtained in all cases prior to biopsy. Tissue processing The specimens received at the LCPL Laboratories, Leiden were fixed by immediate immersion in Kryofix (Merck, Darmstadt, Germany), a non-crosslinking agent,z7followed by processing to paraffin in a conventional 650 W microwave oven.28The specimens received at the Department of Fetal and Infant Pathology, Liverpool and at the Department of Dermatology, Newcastle were fixed by immersion in 0.1 M phosphate buffered, pH 7.4 per cent formaldehyde for a minimum of 6 h and a maximum of 12 h, and subsequently routinely processed to paraffin. Non-isotopic in situ hybridization (NISH) for histone mRNA Each of the 12 verruca vulgaris specimens, 16 normal skin biopsies, and five biopsies from areas of psoriatic plaque formation were analysed for the presence of histone mRNA by NISH using a newly available commercial, fluoroscein-labelled oligonucleotide cocktail complementary to unique sequences on the mRN A transcripts encoding the replication-dependent human histones H2B, H3, and H4. NISH was performed as previously described,26 with the following modifications: (i) Pretreatment-Following hydration, 5 pm sections mounted on 3-aminopropyltriethoxysilane-coated slides were incubated with proteinase K [5pglml in 100 mM Tris-HCI, pH 8.0, 50 mM ethylenediamine tetraacetic acid (EDTA); 15 min at 37"C] followed by immersion in 0.2 per cent glycine (in PBS, 5 rnin), and two changes of PBS (5 min each). Slides were then acetylated in 0.25 per cent acetic anhydride in 0.1 M triethanolamine (pH 8.0) for 10 min followed by two further rinses in PBS (5 min each). (ii) Hybridization-Sections were allowed to air-dry prior to the application of 2Opl of FITC-labelled histone oligonucleotide probe cocktail (Novacastra Laboratories Ltd., Newcastle upon Tyne, U.K.). The probe was supplied in hybridization buffer at a working concentration, to which freshly denatured salmon sperm DNA was added to a final concentration of lOOpg/ml. Sections were hybridized for 16 h at 37°C in a chamber humidified with 3 0 0 m ~NaCl, 30 mM sodium citrate (2 x SSC)/30 per cent formamide. (iii) Post-hybridization wushes and signal detection-Following hybridization, coverslips were removed and slides washed in two changes of 2 x SSC/30 per cent formamide at 37°C (15 min each). Following rinses in 2 x SSC (2 x 5 min) and PBS (5 min), sections were incubated with 10 per cent normal rabbit serum (NRS) in PBS for 20 rnin at room temperature. NRS was then replaced with alkaline phosphatase-conjugated rabbitanti-FITC Fab' fragments (Dako, Denmark) diluted 1 : l O O in 1 per cent NRS. Following incubation for 60 min, slides were washed in two changes of PBS (5 min each) and one change of substrate buffer (100 m M NaCl, 50mM MgCl,, p H 9.5; 5 rnin). FITC-labelled hybrids were visualized by incubation in a chromogen solution (substrate buffer containing 0.34 mg/ml nitrobluetetrazolium and 0.17 mg/ml5-bromo-4-chloro-3-indolylphosphate). ( i v ) Controls-Positive control was human tonsil which was known to exhibit characteristic strong signal positivity. Negative controls for hybridization specificity consisted of adjacent sections which were 'sham' 25 1 NISH HISTONE mRNA AND HPV hybridized with a buffer containing no probe, but otherwise treated identically. Detection of HP V capsid proteins by immunohistochemistry Each of the 12 verruca vulgaris specimens, 16 normal skin biopsies, and five biopsies from areas of psoriatic plaque formation were analysed for the presence of HPV replication using a rabbit anti-bovine polyclonal antibody (DAKO B0580, dilution 1:lOOO) which reacts specifically against papillomavirus genus (common) capsid antigens, regardless of the host specie^."^.^^ Details of the immunohistochemical technique are reported elsewhere.' Estimation of HPV capsid protein and histone mRNA labelling indices Histological sections orthogonal to the basement membrane/epidermal surface were selected. The characteristic architectural stratification of the epidermis allowed the identification of individual keratinocyte layers parallel to the basement membrane. In order to incorporate in each case at least part of the stratum granulosum, ten such layers, starting at the basal layer, were separately assessed for the presence of HPV capsid proteins and histone mRNA. The labelling indices of a particular cell layer were estimated by counting the number of positive cells within a systematic series of randomly located samples of ten cells in that layer. The total number of samples of ten cells to be assessed in any particular layer in a given specimen was determined using the continuous running mean method. In all layers of every verruca, fewer than ten samples of ten cells were found to be sufficient to reach a mean within 5 per cent of the final mean. Therefore for the estimation of labelling indices in this group, ten randomly located samples of ten cells were assessed for the number of cells positive for HPV capsid proteins and histone mRNA signal (a total of 100 nuclei per cell layer). In the analysis of histone mRNA positivity in normal and psoriatic biopsies, a stable mean within 5 per cent of the final mean required the counting of a maximum of 30 samples of ten cells per layer. Therefore estimation of the labelling index in these specimens was based on the analysis of 30 such samples, a total of 300 nuclei per cell layer. No staining for HPV capsid proteins was seen in any of the normal or psoriatic biopsies. Labelling index was expressed as the number of positive cells/total number of epidermal cells assessed x 100 per cent. Reproducibility analysis Intra- and inter-observer reproducibility were studied by repeated, blind assessment of the labelling index by two different observers. Statistical analysis Comparison of labelling indices was performed using paired t-tests. RESULTS Histone NISH gave a high cellular resolution with little non-specific staining; all 'no probe' controls were clear of signal background. There was no appreciable difference in the staining quality or the labelling index between the two groups of verrucae processed in the Liverpool and Leiden laboratories. Intra- and interobserver reproducibility of the labelling index estimation were greater than 93 per cent for every sample. Photomicrographs of representative areas of HPV-infected, psoriatic, and normal skin are shown in Fig. 1. The results of histone NISH labelling index estimation are shown in Fig. 2. A similar pattern of positivity was observed in normal and psoriatic skin, with labelling confined to the basal and immediately suprabasal cells. The mean labelling index was significantly greater in the psoriatic skin (mean layers 1-3,3.10 per cent versus 1.04 per cent WO.01). In marked contrast, the verrucae displayed a bimodal pattern of histone positivity: the mean labelling index decreased from 15.1 per cent (layer 1) to 0 per cent (layer 3) (P<O.OOl), increased to 30.5 per cent (layer 5 ) (P<O.OOl), then decreased to 5.9 per cent (layer 6) (P<O.OOl). Histone positivity in the lowermost layers of the verrucae (mean layers 1-3, 7-30 per cent) greatly exceeded that of the normal skin (mean layers 1-3, 1.04 per cent) (P<O.OOl). Immunohistochemistry using the antibody to HPV capsid proteins produced staining of suprabasal cell nuclei in all verrucae, as reported in detail elsewhere." HPV capsid proteins were first detected in epidermal layer 3; the labelling index increased in the more superficial layers, with positivity of the vast majority of nuclei in layers 7-10. The relationship between the labelling indices for HPV capsid proteins and histone mRNA is shown in Fig. 3. No staining for HPV capsid proteins was observed in any of the normal or psoriatic control biopsies. DISCUSSION This study confirms the applicability of non-isotopic in situ hybridization for the detection of histone mRNA transcripts in archival formalin-fixed, paraffinembedded tissue, as reported by other^.'^,'^,^^,^^ In normal skin, labelling for the replication-dependent histones was restricted to occasional basal and immediately suprabasal cells, in agreement with our previous The verrucae showed a bimodal labelling pattern (Fig. 2). In keeping with the epidermal hyperproliferation characteristic of this lesion, significantly more basal cells contained histone mRNAs than in normal, age-matched skin. The second peak of histone labelling in the differentiated suprabasal keratinocytes (layers 4 6 ) of the stratum spinosum was a striking finding and not observed in the psoriatic specimens. In the latter lesion, epidermal hyperproliferation, a feature of both psoriasis and verrucae, was present but restricted, as reported by others, to the lowermost cell layers.31 Similarly, on histological examination of the verrucae, the very 252 S . A. HINCHLIFFE ET A L Fig. 1-Photomicrographs of non-isotopic in situ hybridization for the mRNA transcripts encoding cellular histones H2B, H3, and H4 in HPV-infected (A), psoriatic (B), and normal (C) skin. Note the bimodal pattern of the signal in HPV-infected skin (A), with positive cells present in the stratum spinosum (layers 4 and 5) in addition to the most basal layers. The signal in sporiatic (B) and normal (C) skin is confined to the most basal layers ++- ~ W verruca vulgaris - histone A- psoriarir HPV ,*-k--A / cn = -c 40 al n a i 20 0 1 2 3 4 5 6 7 8 B 10 Epidermal Cell Layer -I L 0 / 4 I I II 1 2 3 4 5 6 7 8 L 9 10 Epidermal Cell Layer Fig. 2-The relationship between histone mRNA labelling index and epidermal cell layer in HPV-infected, psoriatic, and normal skin. Data are expressed as mean ( f SEM) Fig. 3-The relationship between histone mRNA labelling index and HPV capsid protein labelling index in vet-rucae. HPV data are published elsewhere" occasional mitotic figure was noted in the stratum spinosum, but the vast majority of mitotic activity resided within basal cells. Suprabasal histone positivity in verrucae thus appears to be associated with the presence of HPV and is not merely a reflection of epidermal hyperproliferation. This conclusion is supported by Penneys et al., who used seborrhoeic keratoses as a control for non-HPV-related epidermal hyperproliferation in their study of PCNA expression in cutaneous HPV infection. I 2 Although they detected PCNA in cells of the stratum spinosum in verrucae, staining in the seborrhoeic keratoses was confined to the basal layers. In this study, it was not possible to determine which specific HPV genotypes were present 253 NISH HISTONE mRNA AND HPV in each verruca. However, we consider that the association between suprabasal histone positivity and HPV is unlikely to be type-specific, since others have demonstrated different viral types, notably 1, 2, and 4, in individual lesions.32 We propose that the presence of histone mRNA transcripts in differentiated suprabasal keratinocytes of verrucae is associated with HPV replication as distinct from cellular proliferation, although the possibility of the latter cannot be entirely excluded. HPV must have a mechanism to reactivate this and other host cell replication genes, to facilitate the amplification and subsequent packaging of its own DNA in cells that have exited from the cell cycle and are undergoing terminal differentiation. Examination of Fig. 3 shows, using data from our previous study,' that in verrucae the appearance of large amounts of HPV capsid antigens detectable by immunohistochemistry is spatially tightly associated with the suprabasal peak of replication-dependent histone mRNAs. Vegetative viral replication would thus appear to be restricted to a specific subpopulation of keratinocytes at a particular stage of differentiation. Subsequently the more superficial keratinocytes, whilst continuing to contain large amounts of virus, no longer transcribe detectable levels of histone mRNA and have presumably ceased to replicate viral DNA. The concept of viral replication occurring only within a subset of keratinocytes is supported by other in vivo studies. In both skin (verruca vulgaris)" and cervix (condyloma acuminatum and low-grade cervical intraepithelial neoplasia),1° co-detection of PCNA and HPV DNA was restricted to a relatively narrow band of the epithelium; more superficial cells contained the virus, but PCNA induction could not be demonstrated. Similarly, recent in vitro experiments have demonstrated tight linkages between the expression of both low-risk HPV-11 and high-risk HPV-16 viral E6 and E7 genes, induction of PCNA, and amplification of viral DNA in epithelial raft cultures of differentiated keratin~cytes~ and in organotypic cultures of dysplastic keratinocyte~.~~ In summary, we have presented evidence confirming the applicability of in situ hybridization for replicationdependent histone transcripts to routinely fixed and embedded tissue. HPV replication in the epidermis was strongly associated with the de nova transcription of one or more of the host cellular histones H2B, H3, and H4. Viral replication would thus appear to involve not only the reactivation of host cell repIication mechanisms such as DNA polymerases and associated proteins (e.g., PCNA), but also a dependence on the host for the proteins with which, following replication, the viral chromosomes are complexed. Taken together, the results of this and other recent studies lead us to speculate that regulation of the viral life cycle is in some way intimately related to the degree of differentiation of the infected cell. Finally, when assessing cellular proliferation with histone in situ hybridization, one must be aware of the possible labelling of non-proliferating cells by HPV and perhaps other viral infections. ACKNOWLEDGEMENTS We are grateful to Dr J. Seymour for comments on a previous draft of the manuscript. MDS is supported by the Medical Research Council, U.K. REFERENCES 1. Androphy EJ. Molecular biology of human papillomavirus infection and oncogenesis. J fnvesf Dermufoi 1994; 103: 248-256. 2. Lane S, Wells M. Editorial. Human papillomaviruses, p53, and cervical neoplasia. J Przthol 1994; 172: 299-300. 3. Meyers C, Frattini MG, Hudson JB, Laimins LA. Biosynthesis of human papillomavirus from a continuous cell line upon epithelial differentiation. Science 1992; 257: 971-973. 4. Dollard SC, Wilson J, Demeter LM, e f ul. Production of human papillomavirus and modulation of the infectious program in epithelial raft cultures. Genes Dev 1992; 6 1131-1142. 5. Taichman LB, LaPorta RF. The expression of papillomaviruses in epithelial cells. In: Salzman NP, Howley PM, eds. The Papovaviridae. New York: Plenum Press, 1987; 109-136. 6. Grussendorf EI, zur Hausen H. Localization of viral DNA replication in sections of human warts by nucleic acid hybridization with complementary RNA of human papilloma virus type 1. Arch Dermutol Res 1979; 264: 55-63. 7. Orth G, Jeanteur P, Croissant 0. Evidence for and localisation of vegetative viral DNA replication by autoradiographic detection of RNA--DNA hybrids in sections of tumors induced by Shope papilloma virus. Proc Natl Acad Sci USA 1971; 6 8 1876-1880. 8. Del Vecchio AM, Romanczuk H, Howley PM, e f a/. Transient replication of human papillomaviral DNA. J Vtrol 1992; 6 6 5949-5958. 9. Chiang C-M, Ustav M, Stenlund A, e t a / . Viral El and E2 proteins support replication of homologous and heterogenous papillomaviral origins. Proc N u f l Acad Sci U S A 1992; 8 9 5799-5803. 10. Demeter LM, Stoler MH, Broker TR, Chow LT. Induction of proliferating cell nuclear antigen in differentiated keratinocytes of human papillomavirus-infected lesions. Hum Parhol 1994; 25: 343-348. 11. Boon ME, Howard CV, van Velzen D. PCNA independence of Ki-67 expression in HPV infection. Cell Biol Ini 1993; 17: 1001-1004. 12. Penneys NS, Bogaert M, Serfling U, Sisto M. PCNA expression in cutaneous keratinous neoplasms and verruca vulgaris. A m J Pufhol 1992; 141: 139-142. 13. Pringle JH, Baker J, Colloby PS, Lauder I . The detection of cell proliferation in normal and malignant formalin-fixed paraffin-embedded tissue sections using in situ hybridisation for histone mRNA. J Pafhol 1993; 169(Suppl): 144A. 14. Chou MY, Chang ALC, McBride J, Donoff B, Gallagher GT, Wong DTW. A rapid method to determine proliferation patterns of normal and malignant tissues by H3 mRNA in situ hybridisation. A m J Pathol 1990; 136 729-733. 15. Hanekeln T, Key1 HG, Ross R, Schmidt ER. Evolution of histone gene loci in chironomid midges. Genome 1993; 3 6 HPV852-861. 16. Schumperli D. Cell-cycle regulation of histone gene expression. Cell 1986; 4 5 471472. 17. Zahradka P, Elliot T, Hovland K, Larson DE, Saward L. Repression of histone gene transcription in quiescent 3T6 fibroblasts. Eur J Biochem 1993; 217: 683-690. 18. Marzluff W, Hanson RJ. Degradation of a nonpolyadenylated messenger: histone mRNA decay. In: Balasco J, Brawerman G, eds. Control of Messenger RNA Stability. New York: Academic Press, 1993; 267-290. 19. Stein GS, Stein JL, Lian JB, van Wijnen AJ, Wright AJ, Wright KL. Modifications in molecular mechanisms associated with control of cell cycle regulated human histone gene expression during differentiation. CeN Bwphys 1989; 1 5 201-223. 20. Birnstiel ML, Busslinger M, Strub K. Transcription termination and 3' processing: the end is in site. Ce// 1985; 41: 349-359. 21. Stein GS, Stein JL, van Wijnen AJ, Lian JB. Histone gene transcription: a model for responsiveness to an integrated series of regulatory signals mediating cell cycle control and proliferation/differentiation interrelationships. J Cell Biochem 1994; 5 4 393404. 22. Bosch FX, Udvarhelyi N, Venter E. Expression of histone H3 gene in benign, semi-malignant and malignant lesions of the head and neck: a reliable proliferation marker. Eur J Cancer 1993; 2 9 14541461. 23. Vemura RP, Aragona E, Gupta S. Analysis of hepatocellular proliferation: study of archival liver tissue is facilitated by an endogenous marker of DNA replication. Hepatology 1992; 1 6 968-973. 24. Smith MD, Healy E, Thomson V, Morley AR, Reel JL. Use of in situ detection of histone mRNA in the assessment of epidermal proliferation: comparison with the Ki67 antigen and BrdU incorporation. Br J Dermatol 1995; 132 359-366. 25. Favre M, Breitburd F, Croissant 0, Orth G. Chromatin-like structures obtained after alkaline disruption of bovine and human papillomaviruses. J Virol 1977; 21: 1205-1209. 254 S. A. HlNCHLlFFE ET A L 25. Smith MD, Rees JL. Wavelength-specific upregulation of keratin mRNA expression in response to ultraviolet radiation. J Invest Dermutol 1994; 102: 433439. 27. Boon ME, Schmidt U, Cramer-Knijnenburg GI, van Krieken JHJM. Using Kryofix as an alternative for formalin results in more optimal and standardked immunostaming of paraffin sections. Puthol Res Pruct 1992; 188 832-8 35. 28. Kok LP. Boon ME. Microwave Cookbook for Microscopists: Art and Science of Visualisation. 3rd edn. Leiden: Coulomb Press, 1992. 29. Kurman RJ. Shah KH. Lancaster WD, Jenson AB. Immunoperoxidase localization of papillomavirus antigens in cervical dysplasia and vulvar condylomas. Am J Ohstel Gy7ecol 1981; 1 4 0 931-935. 30. Ferenczy A, Braun L, Shah KV. Human papillomavirus (HPV) in condylomatous lesions o f cervix. Am J Surx Path01 1981; 5: 661-670. 31. Bata-Csorgo BZ, Hammerberg C, Voorhees JJ, Cooper KD. Flow cytometric identification of proliferative subpopulations within normal human epidermis and the localization of the primary hyperproliferative population is psoriases. J E-ur, Med 1993; 1 7 8 1271-1281. 32. Chen SL, Tsao YP, Lee JW, Sheu WC, Liu YT. Characterization and analysis o f human papillomaviruses of skin warts. Arch Derrrratol Res 1993; 285 460465. 33. Halbert CL, Demers GW, Galloway DA. The E6 and E7 genes of human papillomavirus type 6 have weak immortalizing activity in human epithelial cells. J Virol 1992; 6 6 2125-2134.