Tobacco Documents Online

SEARCH FOR PAPILLOMAVIRUS DNA IN PREMALIGNANT AND MALIGNANT SQUAMOUS CELL LESIONS OF THE ORAL CAVITY AND UPPER AND LOWER RESPIRATORY TRACTS Submitted~ to: Council for Tobacco Research-USA By: A. Bennett Jenson, MD Department of Pathology Georgetown University Schools of Medicine and Dentistry Washington, DC 20007

Background Human papillomavirusPs (HPV) are associated with benign and malignant squamous lesions of the skin, oral cavity, upper and lower aerodigestive ti.act and anogenital areas. These associations are based on the fnllowing observations: i) Visualization of HPV partacles in lesions by electtron microsoc.py; iU demonstration of HPV-DNA in lesions by DNA-DNA or DNA-RNA hytaridi2ati on; iii) demonstration of PV genus-specific antigens in ]esons, and iv) experimental transEer of H P V-associated/infected squamous ce]1l explants to the congenitally immuno-de5cient mouse animal model. Because HPVs can neither be experimentally transmittei to animal mode]s nor repli'catsd' in tissve culture systems pPrmissive fior expression of biological activity, research has been limited for the most part to physi.cal and cfiemical characteiizatinn: of viribns obtained from warts. Recent advances in molecular virology, allowing cloning of HPV DNA extracted directly from~ cutaneous and mucosal ]esions, has led to the discovery of muhiple minimally related HPV types with anatomic site preference. Although most cutaneotropic HPV are associated with lesions that remain betLign~ mucostropic viruses appear to have varying degrees of malignant potentiaL Recently, H P V(par+; cularly H P V-16 and -18 ) has been associated with up: to 85 $ of advanced dysp]asas and squamous carcinomas of the anogenital tract, and at least some squamous carcinomas of the oral cavity, larynx and lung. In fact, HPV was detected in approximately 10% of carcinomas of the lung in~ one study. Detec,-~tion of HPV DNA sequences in prniiferative squamous lesions is generally accepte3 as the besE and frequently the only virologic evidence for identifying HPV as the ebologi.c agent. A]though a variety of molecular hyhddization~ techniques have been used to detect HPV DNA sequences, most require tissue which has not been d5emically fixed, i.e. fresh or fresh frozen

tissue. Recently, however, in situ tech CRIES have been devp].oped which permit detection of viral DNA sequences in formalin-fixed tissues used for rrxztine hist:cpatlxila3ical studies. Alt.tiough not yet as senstive as "wet" met,hods of molecular hybridization, these newer techniques are technically sLmpler and f.astt enough~ to a11bw large numbers of tissue specimens to be processed and examined efficiently. There are several other distinct advantages to the use of in situ hybridization techniques: i) these techniques can be used with formalin-fixed tissue ma)ang pasmb1E retsospet.~tive analyss of archived czaIlections of formalinr5xed tissue for which there are no fresh or frozen aounterparts, and ii) HPV-transformed cancer cP11s can be identified by the in~ situ hyhridiaation reaction product that is visible by routine light microsc.~opy. Preliminary Data During the last 4 years my ]aboratr,xy has been investigating the role of HPV in the etiology of cutaneous and muccsal proliferative squamous IesAns. Usung research funded piimarily from a Speoial Project for the Council for Tobacco Research, we have focused on the immunology of infection with HPV. These studies were directed at umng polyclonal antibodi,es (prepared agaan..~tt detergent-disrupted H P V-1 and BPV-1 structural antigens) using im m unofluorescence QP ) on acetone-fixe3 tissue and im m unogeroxidase QP ) on formalin-fixe3 tissue. Using these techniques, PV genus-specific antigens were identified in 50r-70 $ of plantar and com mon warts, 40-fi0 $ of verrucae, multiple and sngie pa 'pillnmas, and condylomata of the cral cavity, 50 % of ]aryngeal papillamas, 50 $ of anogenital condylDmata, 45-60 $ of male urethral pa pil7omas/bondylomata, 29-45 $ and 208 of mild and moderate cervical dysp]asias, respectively, and 5$ of conjunctival papillomas. HPV antigens were found in 5$ of squamous carcinomas of the larynx. However, HPV antigens were not found in 50 squamous carcinomas of the lungs. It is currently unknown why viral antigens are i

only detected in approximately 50 % of benign lesions and only up to 5 % of malignant lesons known to be a_sgociated with HPV although all leswns oontain actively replicating, viral DNA. The only way to prove an association of HPV with ]esions that are not expressing viral antigens is by the appropriate molecular D N A hylmdization studaes on fresh or forma ]urfixe3 tissue. O bjectives We will examine a variety of squamous and' adenocarcinomas of the lung to determine if HPV DNA sequences are present by molecular hybridization studies. Our initial~ studies will be directed towards farma]nn-fixed, exophytic, endobronchial squamous carcinomas, snce this is the lung lesian that appears identical to HPV-associated velsucous carcinomas from other stes. We have access to cancPSous lung tissue from one of the laargest pathology departments in the United States, Baptist Memorial Hospital, Memphis, TenneSSee (see attached letter from Dr. Pn J. Dean), and from the Armed Forces b'tute of Pathology (AFIP), who is sending a mic:rohioingist from the navy to my ]aboratory to work on this project if I obtain funding from C T R. Expedmental Desgn A sampling of formalin-fixed paraffin-embedded specimens of lung tissue from patients with exaphytic, endobronchial squamous carcinomas of the L,ung will be selected from the archival fil~s of Baptist Memorial Hcspital and AFIP. The specimens will be examined by in stu hybaidizatwn under stringent aonditi,ons u,sing, a probe prepared from HPV types 6, 11 and 16. These particu]ar HPV types have been chosen because of their known: association with lung carcinomas, other probes wiltl be prepare3 from HPV types 18, 31 and 35. Lncreasi.rx3 the stringency of the hyta.dization oonditions minimizes cross hyhridizatiQn~ between partially related viruses, thus a positive signal with one of these HPV probes would confirm the presence of DNA from~ that particular HPV in the tissue specimen.

M ethodnlogy Preparation of focmalin-fxed ti.ssue for in-stu hybrtidization Sections 3 um thick wi71 be cut from formalin~fixed, paraffin-embedded lumg tissues and floated on the surface of a water bath containing warm, protein free tap water. eCnn will be removed from the water bath on pnly-D lysine coated glass slides, air-dried vertically at room temperature, and baked at 80°C for 10 minutes. Poly-lysine coated slides will be prepared by immeTsing clean glass mic,xoscope slides in 0.1$ paly-Dlysine ( M W=50,000; Sigma), and air drying at room temperature. Sections will then be dewaxed by a 10 minute im~merson in xylene, followed by two 10 minute immersons in absnlute alcohol, then air dried. Freshly-prepared pronase snlution (CalBiochem; 82,000 PUK/gm), 0.3 mg/ml in 50 m M Tn5-C1 pH 7.4 containing 5 m M EDTA, will be spotted onto the dry dewaxed tissues and pesmi tted to digest at room: temperature for 15 minutes. Exces.s pronase salution wi11 be removed and the tissues washed twice for 5 minutes in Tri.s-Sa]dne (0.1 K Tris-C1 pH 7.5, 0.1 M NsCl) containing 2 mg/ml glycine. Follawing the second wash the slides will be dehydrated through a graded alcohol series (30 %, 60 %, 95 %, absoLute) and air dried. Frepaiatian of hybridiaaticn probes Probes for H P V-6 and other H P V types have been molecularly cloned and are available to us already inserted into a bactp..rial plasmid. Passenger DNA will be recovered by isopycni.c centrifugation using oesium triflt,nroacetate (CsTFAtM , Pharmacia) and recovered by ethanol ~ecipitati,on. The probes will be bintin-]abeled by nick translatinn with the thymidine analog Bio ]1-dUTP (Bethesda Research Iabora rIF] ~ , Gaithersburg,, M D). The hybridizatian cocktail will oonsdst of 45 $ deionized formamide 5XSSC (0.75 M NaCl, 0.75 sodium c3trats), 25 m M sodium phosphate pH 6.5, 1x Denhardt's solution (0.02 $ each polyvinyl pyrr(:didone-40, Ficoll 400 and BSA fraction V), 250 ug/ml sheared denatured herriM spesm DNA, 10 $(wtJvnU dextran suffa te, and 0.5 ug/ml

biotin labeled jxobe. Hybridization Prepared tissue samples will be overlaid with 50 ul of hytridi.zation °ocktail and ocversli.pped. The probe and tissue will then be denatired smu2taneously by heating the slides in an~ oven at 100 105°C for 10 minutes (nonstnrigent condytions). Slides will then be oooled to 35-37°C and the genetic ma terial allowed to hybddize at this temperature for 2 hours. Stringency will be izxreased by sequentially decreasng the temperature of heating until csoes hybridization is e3iminated. Following hybiidization, oovesslips will be removed and the slides dipped in 2XSSC to remokve the excess hy n coctail. Slides w'ill then be washed at room temperature twice in 2XSSC 0.1% SDS, 3 minutes per wash, followed by a one minute incubation in 2XSSC 0.1% SDS and 5 minute blocking incubation in 3% BSA in Txis~Saline. Slides will theni . be air-dxied. Codtyimetric Det.ection Biotinylated probes that hybridize to HPV DNA sequences in lesions will be detected by incubating washed', blocked slides for five minutss at room temperature with an avirlin-alkaline phosphatase complex. The complexes will be prepared fresh by incubating 40 ul Avidin DN (Vector Laborat°ries) with 5 ul of biotinylated modified alkaline phosphatase (synthes-ized from Boehringer-Mannheim) calf intestinal alkaline phosphatase as described by in Il ml l$ BSA in Tris-Saline-Triton (0.1 M Tris-C1 pH 7.5 0.1 M NaC1, 5 mM MgC12, 0.01$ Tziton X-100) for five minutes. Non-specifically bound complexes will be removed by three sequential 3 minute washes in Tns-Saline 9.5 (0:1 M Tris-Cl pH 9.5, 0.1 M NaCl, 50 mM MgC12. Color will be developed by incubating the slides for 2 hours at 37° in freshly prepared M cG adey reagent. The M c G adey reagent conssts of 0.2 ml of a 50 mg/ml 50$ DMF (N,N-dimethyl formamide; Sigma/distMed water) stock solution to NBT (nitro blue tetrazolium ch]bride grade IlZ, Sigma) anc] 0.1 mol of 50 mg/ml

stock salution of BCIP (5-bromo-4-dhloro-3-indoyl phosphate para-tnluiciiie salt; Amresco) in 30 ml Tris-Saline. After aolcar development, the slides will be nnsed in distillad water and cvunte.r5tained with Erlich's hematoxylin and metannl yellow, air-diied, and mounted with permount. The presesvation~ of tissue archiLecture and cellular detail in histnlogic seetivns will allow an exact ]ncalizatioa of HPV sequences in individual cells. The hybridi:zatir•wn product shou]d be confined to the nucleus of infected cells. It3 stu hylsidization of formalirr-fixed samples of carcinomas of the cervix, vulva ar,d anus known to contain spet-ific types of HPV DNA will be used as the povitive controls. Significance Ahhough epidemnn.l,ogic studies have implicated a variety of different environmental agents, the etiolagy of lung carcinomas, ,larly irrlividual cases, remains a scurce of speculation. Premalignant and malignant lesions of the lung are very similar histopathologically to those of the uterine cervix. Human papiIlbmaviru,es have been found in: up to 85 $ of squamous and adenocarcinomas of the cervix; up to 10 % of lung carcinomas may be caused by HPV. Since most lung tissue available for study is formalin-fixed ) we FropoEe to use in situ hytridizatiQn techniques on paraffin-embedded sections to determine retrospectively which lung cancers are caused by HPV. Overview I have been supported for 4 years by a special project from the CTR. As previously mentioned, all of the studies supported by CTR were immunnlogie localization of HPV stiuctural viral~ antigens in human tissues processed for hist.opathology. On November 1, 1985, I submitted a c}rant to NIF1 primarily based on data obtained from the Special Project. I now want to extend my expertise to localization of HPV DNA sequences in lung carcinomas by in s-itu hybridization of humaa ti$ue processed for hist,apathningy. Within the next several years, I would

expect tn submit a grant to NIH om in situ hytridi.zatiQn ba_Sed on data obtained from the proposed studies.