(CANCER RESEARCH 53.481 MS 16. October 15. 199.1]

HPV-16, Tobacco-specific 7V-Nitrosamine, and Af-Methyl-7V'-nitro-jV-nitrosoguanidine

in Oral Carcinogenesis1

Myong Soo Kim, Ki-Hyuk Shin, Jeong-Hwa Hack, Henry M. Cherrick, and No-Hee Park2

School of Dentistry and Jonsson Comprehensive Cancer Center, University of California, Los Angeles, California 90024

ABSTRACT

We previously immortalized human oral keratinocytes by transfectionwith recombinant human papillomavirus type 16 (HPV-16) DNA and

established two cell lines. These transfected cells were morphologicallydifferent from the normal counterpart, contained intact HPV-16 DNA in

an integrated form, and expressed numerous viral genes. These cells contained lower levels of wild-type p53 protein and higher levels of c-myc

mRNAs compared to normal cells. However, they proliferated only inkeratinocyte growth medium containing a low level of calcium and werenot tumorigenic in nude mice.

A HPV-16-immortalized cell line was exposed to either 4-(methylnitro-samino)-l-(3-pyridyl)-l-butanone or JV-methyI-/V"-nitro-JV-nitrosoguani-

dine. Four chemically transformed cell colonies were isolated. These cellsproliferated well in Dulbecco's minimum essential medium containing a

physiological level of calcium. They contained, similar to the immortalizedcounterpart, integrated HPV-16 sequences and lower levels of both wild-

type p53 protein and IXC messages compared to normal cells. Among thechemically transformed cells, two colonies obtained from 4-(methyl-nitro-samino)-l-(3-pyridyl)-l-butanone exposure demonstrated an enhanced

proliferation capacity in nude mice and transcribed a substantially higheramount of HPV-16 E6/E7, epidermal growth factor receptors, and c-myc

genes compared with the immortalized counterpart. These experimentsindicate that malignant transformation of oral keratinocytes can becaused by a sequential combined effect of "high risk" HPV and tobacco-

related carcinogens.

INTRODUCTION

HPV3 infection is closely associated with the development of fe

male genital epithelial cancers. Over 90% of cervical cancer biopsiescontain "high risk" HPV DNA such as types 16 (HPV-16) and 18

(HPV-18) (1, 2). Similarly, HPV infection is also closely linked to

benign and malignant oral lesions (3). Recent studies show that up to30-40% of oral cancer biopsies contain the viral DNA (4). Inasmuch

as the oral mucosal epithelium resembles the female genital tract andis continuously challenged by innumerable environmental influences,close association between HPV infection and oral malignancies is notsurprising.

Further evidence of the role of HPV in carcinogenesis derives fromits transforming capacity. Transformation of normal human foreskinand exocervical, cervical, and oral epithelial cells, the primary in vivotarget cells for HPV infection, were established with cloned HPV-16and -18 DNA (5-8). These transformed cells are immortal and harbor

integrated viral DNA expressing various HPV messages includingE6/E7 mRNAs. Although chronic propagation of HPV-immortalized

human skin keratinocytes can lead to a malignant phenotype (9),

Received 4/26/93; accepted 8/11/93.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1Supported in part by USPHS Research Grant ROI DE10049 from the National

Institute of Dental Research. NIH, and Grant 0231 from the Smokeless Tobacco ResearchCouncil, Inc.

2 To whom requests for reprints should be addressed, at Section of Oral Biology, UCLASchool of Dentistry, 43-033 CHS, 1Ü833Le Conte Avenue. Los Angeles, CA 90024.

3 The abbreviations used are: HPV, human papillomavirus; NNK, 4-(methylnitrosa-mino)-l-(3-pyridyl)-l-butanone; NHOK, normal human oral keratinocytes; HOK-16B,HPV-16-immortalized oral keratinocyte cell line; MNNG, A^-methyl-W-nitro-W-nitroso-guanidine; TGF-n, transforming growth factor a; EGFR, epidermal growth factor receptor; DMEM, Dulbecco's minimum essential medium; kbp. kilobase pair; nt, nucleotide;

ATCC. American Type Culture Collection; cDNA, complementary DNA.

HPV-immortalized epithelial cells are, for the most part, nontumori-genic in nude mice. Furthermore, the nontumorigenicity of HPV-

immortalized human oral keratinocytes is not altered by propagationof the cell cultures in more than 80 passages (10).

The ubiquity of HPV infections, the regression of most HPV-in-

duced dysplasias, and the long incubation period between initial infection and development of cancer indicate that HPV infection byitself may not be sufficient for neoplastic conversion of normal oralkeratinocytes (11). Various other environmental agents, includingchemical carcinogens, are likely to participate in the malignant progression of HPV-infected cells in the oral cavity (10). HPV-immortalized human oral keratinocytes harboring "high risk" HPV DNA

may therefore be a suitable model system for investigating the sequential combined tumorigenicity of "high risk" HPV and other en

vironmental factors.Many epidemiological studies indicate that smokeless tobacco is

linked to an increased incidence of oral and pharyngeal cancer inhumans (12-15). This linkage is further supported by the presence oftobacco-specific Af-nitrosamines in smokeless tobacco (16, 17).Among tobacco-specific N-nitrosamines, smokeless tobacco containsa high level of A'-nitrosonornicotine and NNK, which are strong

carcinogens in animals (18). However, although chronic exposure tochemical carcinogens in tobacco users is still the major known causeof oral cancer, only a minority of tobacco users develop oral cancer.Thus, an involvement of other contributing factors, e.g., ethanol andviruses, have been proposed to explain the development of smokelesstobacco-related oral malignancies (19, 20).

In the present study, we propose a hypothesis that oral cancer isinduced by sequential exposure of normal oral keratinocytes to "highrisk" HPV and tobacco-related carcinogen(s). To test the hypothesis,

NHOK and an HPV-16-immortalized oral keratinocyte cell line,HOK-16B (8), were exposed to either NNK or MNNG. Four chemically transformed cell colonies were cloned from the HOK-16B cells

exposed to chemical carcinogens. The in vitro proliferation characteristics, in vivo tumorigenicity, and the expression of HPV-16 E6/E7,TGF-a, EGFR, c-myc, p53, and the gene deleted in colon cancer

(DCC) were studied from the chemically transformed cells. SinceNNK mutates p53 and c-Ki-ras2 genes resulting in cell transformation(21-23), mutations of these genes were also investigated from the

chemically transformed cells. Such transformants proliferate well in aculture medium containing a physiological level of calcium (1.5 min).At least two transformants demonstrated enhanced growth capacity innude mice, whereas primary human oral keratinocytes exposed tochemical carcinogens failed to show evidence of transformation. Cellswith enhanced in vivo growth capacity contained a significantly higheramount of HPV-16 E6/E7, EGFR, and c-myc messages compared withthe immortalized counterpart. No mutations in p53 and Ki-ras genes

were found from the chemically transformed cells. These data indicatethat HPV-16 and tobacco-related carcinogens sequentially transform

oral keratinocytes into a malignant phenotype.

MATERIALS AND METHODS

Culture of Primary NHOK and HOK-16B Cells. Excised retromolar

tissue from the oral cavity of a healthy male volunteer was washed in calciumand magnesium-free Hanks' balanced salt solution (GIBCO/BRL, Grand

Island, NY). Monolayer cultures of primary NHOK from the tissue were

4811

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

HPV AND TOBACCO IN ORAL CANCER

prepared as described elsewhere (8). An HPV-16-immortalized oral keratino-cyte line, HOK-16B, was cultured in keratinocyte growth medium (Clonetics,

San Diego, CA) supplemented with pituitary extract as described previously(8).

Exposure of NHOK and HOK-16B Cells to NNK or MNNG. The primary NHOK and HOK-16B cells were plated at 2 x IO5 cells/60-mm Petri

dish. When the cultures reached 70% confluency, the cultures were exposed toNNK (5.0-20.0 /j.g/ml; Chemsyn Science Laboratories, Lenexa, KS) for 4

weeks or MNNG (1.0 ng/ml;Sigma Chemical Co., St. Louis, MO) for 3 h. Thecultures were then maintained in a selection medium DMEM, containing 1.5mM of calcium and supplemented with 10% fetal bovine serum to collectchemically transformed cell colonies. Because NHOK and HOK-16B cellscannot proliferate in DMEM, NHOK and HOK-16B cells were cultured in

DMEM after exposure to chemical carcinogens to select chemically transformed cells. Four colonies were isolated from the HOK-16B cells exposed to

chemical carcinogens (three from NNK exposure and one from MNNG treatment) and named 16NNK-1, 16NNK-2, 16NNK-3, and 16MNNG-1.

Determination of Cell Proliferation Rate in DMEM. Confluent cellmonolayers in 100-mm Petri dishes were trypsinized and counted. The cells

were suspended in DMEM supplemented with 10% fetal bovine serum and 2X 10s cells were plated onto one 60-mm Petri dish. The number of cells wascounted after 2, 4, and 6 days of incubation at 37°C.There were three cultures

in each group at each time.Determination of in Vivo Tumorigenicity of Cells. The in vivo tumori-

genicity of cells was determined in nude mice. Monolayer cultures weretrypsinized and resuspended in culture medium and 0.1 ml of phosphate-buffered saline containing 1 x IO7 cells were injected s.c. into athymic nude

mice (nu/nu; UCLA Nude Mice Facility, Los Angeles, CA) at 24 hours afterX-irradiation (300 rads). All mice received injections in the right flank and

were monitored twice weekly for the appearance of tumors over a period of 4months. Tumors that developed were excised for histological examination.

Northern Analysis. To determine the transcription of HPV-16 E6/E7,EGFR, TGF-a, c-myc, p53, DCC and ß-actingenes, cytoplasmic polyadenyl-

ated RNA was extracted from cells using standard procedures. Probes used foranalysis included: 1.2-kbp fragment (nucleotides 24-880, 3357-3820) representing the major early HPV-16 message including E6/E7 genes, human EGFRcDNA (ATCC, Rockville, MD), human TGF-a cDNA (ATCC), v-myc provimi

DNA (ATCC), p53 cDNA (from Dr. E. Harlow, MGH Cancer Center,Charlestown, MA), DCC cDNA (from Dr. B. Vogelstein, Johns Hopkins University, Baltimore, MD), and the human ß-actingene (from Dr. L. Kedes,Stanford University, Palo Alto, CA). All were labeled with [32P]dCTP

(ICN Radiochemicals, Irvine, CA) by multiprime labeling (Amersham Corp.,Arlington Heights, IL). Specific radioactivities of labeled probes were alwayshigher than 5 X IO8 cpm/ftg of DNA. Northern blot hybridization was carried

out under stringent conditions as described previously (10). After hybridization, the filter was autoradiographed on SB-5 X-ray film (Eastman Kodak,Rochester, NY) for 12 h at -70°C. After exposure, the probe was stripped off

the filter for rehybridization to the next radiolabeled probe.Southern Analysis. High molecular weight cellular DNA was extracted

from cells by phenol:chloroform:isoamyl alcohol (25:24:1) and ethanol precipitation. To determine the presence and physical state, if any, of HPV-16

DNA, 10 /j.g of DNA were digested with Bam\\\ or EcoRV restriction enzymes. BamHl enzyme separates vectors from HPV-16 sequences, whereas£coRVdoes not digest pMHPV-16d plasmid that was originally transfected toprimary NHOK to establish the HOK-16B cell line (8). The fragmented DNAs

were run in a 0.8% agarose gel, transferred to a nitrocellulose filter, andhybridized to 12P-labeled 7.9-kbp HPV-16 DNA under stringent conditions.

After a washing, the filter was exposed to SB-5 X-ray film.

Western Analysis. Cells grown in Petri dishes were lysed in lysis buffer[10 mM Na2HPO4 (pH 7.2), 9 mg/ml NaCl, 1% Triton X-100, 5 mg/ml de-

oxycholate, l mg/ml SDS, 2 mg/ml sodium azide, and 40 fig/ml sodiumfluoride] on ice for 30 min. The cell lysate was centrifuged at 14,000 rpm for20 min and the supernatant containing 1.0 mg/ml of protein was denatured byboiling for 2 min in a sample buffer [62.5 mMTris-HCl (pH 6.8), 10% glycerol,1% sodium dodecyl sulfate, 1% ß-mercaptoethanol, and 0.001% bromophenol

blue]. An aliquot of the denatured supernatant containing 100 fig of proteinwas electrophoresed in a 10% sodium dodecyl sulfate-polyacrylamide gel andtransferred onto an immobilon-P membrane (Millipore Corp., Bedford, MA).After incubation in blocking buffer (0.2% I-block, phosphate buffered saline,

and 0.05% Tween 20) for 3 h at room temperature, the membrane was exposed

to mouse anti-human monoclonal antibody for p53 (AB-2; Oncogene Sciences,

Manhasset, NY) at room temperature for 1 h. The membrane was then treatedwith anti-mouse IgG-alkaline phosphatase conjugate (Tropix, Inc., Bedford,

MA), washed with blocking buffer and assay buffer (0.1 Mdiethanolamine and1.0 mM MgCli), incubated in Nitroblock Reagent, washed again with assaybuffer, and incubated in chemiluminescent substrate solution using the Western-Light kit (Tropix, Inc.). The membrane was autoradiographed on SB-5X-ray film for 4 min at room temperature.

Reverse Transcription-PCR Analysis, Cloning, and DNA Sequencing.The synthesis and amplification of p53, c-Ki-ra.s2, DCC, and cellular ß-actincDNAs were carried out using a RNA PCR kit (Perkin-Elmer Cetus, Irvine,

CA). In a total volume of 20 jil, 100 ng of polyadenylated RNA were dissolvedin a solution containing 10 HIMTris-HCl (pH 8.3), 50 mM KC1, 5 mM MgCl2,

0.01% gelatin, 1 mM concentrations each of the four deoxyribonucleotidetriphosphates, 2.5 JAMoligodeoxythymidylate primer, 1 unit of RNase inhibitor,and 2.5 units of reverse transcriptase. Reverse transcription mixture was incubated at 42°Cfor 15 min, at 99°Cfor 5 min, and at 5°Cfor 5 min.

For p53 and c-Ki-ras2, the resulting cDNA product was amplified by theaddition of 80 ¡úof PCR mixture containing 10 mM Tris-HCl (pH 8.3), 50 mM

KC1, 5 mM MgCl2, 2.5 units of recombinant Taq DNA polymerase, and 0.5 /J.Mp53 or c-Ki-7YJs2 primers. The p53 sense primer extends from nt 296 to nt 315

and the antisense primer extends from nt 1015 to nt 996 of the cDNA (senseprimer: 5'-CCCAGAAAACCTACCAGGGC-3'; antisense primer: 5'-CGA-AGCGCTCACGCCCACGG-3'). The c-Ki-ras2 sense primer extends from nt3 to nt 22 (5'-GACTGAATATAAACTTGTGG-3') and the antisense primerextends from nt 351 to nt 332 (5:-CACAAAGAAAGCCCTCCCA-3') of the

cDNA. The amplified c-Ki-ras2 cDNA includes codons which are most fre

quently mutated in cancer cells, i.e., codons 11, 12, 13, 59, and 61. Eachamplification cycle consisted of 1 min of denaturation at 94°C,followed by 2min of annealing at 60°Cand 3 min of extension at 72°C.A total of 35 cycleswere run with a final extension step at 72°Cfor 7 min. Amplified p53 cDNA

was ligated to pCRII vector using the TA Cloning Kit (Invitrogen, San Diego,CA) under the conditions recommended by the manufacturer. The nucleotidesequence of the amplified cDNA was determined by the primer extensionmethod as described previously (10). Five p53 and four c-Ki-ras2 clones were

sequenced for each cell line.The DCC cDNA was amplified using sense primer (5'-AACAGAG-

GATTCAGCCAAT-3') from exon 22 and antisense primer (5'-AGCAGC-TAAACTTTGACATT-3') from exon 23 (24). The cellular ß-actincDNA wasamplified using sense primer (5'-ATCATGTTTGAGACCTTCAA-3') and antisense primer (5'-CATCTCTTGCTCGAAGTCCA-3') (25). Each amplification cycle consisted of 1 min of denaturation at 94°C,followed by 2 min ofannealing at 56°Cand 3 min of extension at 72°C.A total of 35 cycles wererun with a final extension step at 72°Cfor 7 min. The amplified cDNA was

electrophoresed in 1.5% agarose gel, transferred to a nylon filter, and hybridized to internal oligonulceotide probes (DCC cDNA, 5'-GGATGATCTGAGT-GAACA-3'; and ß-actincDNA, 5'-GACCTGGCTGGCCCGGACCTGACT-GACTAC-3') after 3'-end labeling with ¡7-32P]ATPusing T4 polynucleotide

kinase (GIBCO/BRL). The hybridization was carried out as described in"Southern Analysis." The oligonucleotide primers and probes obtained from

Bio-Synthesis, Inc. (Dentón, TX) were purified and desalted before use.

RESULTS

Morphology and Proliferation Rate. The chemically transformedcells were morphologically similar to their immortalized counterpartand proliferated well in DMEM supplemented with 10% fetal bovineserum. Population doubling times of the 16NNK-1, 16NNK-2,16NNK-3, and 16MNNG-1 cells in DMEM were approximately 32,38, 48, and 42 h, respectively, but the HOK-16B line did not proliferate in this medium (Fig. 1). The HOK-16B cells eventually differ

entiated and detached from the culture dishes. Primary NHOK wereable to be subcultured for up to the 4th passage in keratinocyte growthmedium but the cells began to differentiate and completely detachedfrom the culture plates by passage 5. Further, NHOK could not betransformed by NNK or MNNG exposure. This experiment was repeated five times and the same results were consistently obtained.

In Vivo Tumorigenicity. No tumors developed in mice receivingNHOK, HOK-16B, 16NNK-1, or 16MNNG-1 lines. However, 60 and

4812

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

HPV AND TOBACCO IN ORAL CANCER

2000000-'S

1600000-5I„

1200000-Ì"5

800000 -O—

HOK-16B•

—¿� 16NNK-1D—16NNK-2•—

16NNK-3ft— 16MNNG-1

400000 -

2 4

Period of Culture (Days)

Fig. 1. Chemically transformed HOK-16B cells have obtained the ability to grow inDMEM supplemented with 10% fetal bovine serum.

80% of animals receiving the 16NNK-2 and 16NNK-3 cell lines

developed tumors, respectively. Tumors began to appear at 3 weeksafter injection and reached their peak at 0.5 cm in diameter by 7weeks. The tumors then regressed, leaving only necrotic tumor tissueby 12 weeks. Microscopically, the tumors exhibited a differentiatedsquamous cell carcinoma histology.

Viral DNA and RNA. Southern analysis after BamHl digestionshowed that the immortalized and chemically transformed cells contained 7.9-kbp HPV-16 DNA genome, indicating the presence ofintact HPV-16 DNA in these cells. In addition to the 7.9-kbp HPVDNA genome, all of the cells contained rearranged HPV-16 DNA

sequences; the hybridization of cellular DNA digested with BamHlshowed HPV-16-specific bands larger than 7.9 kbp (Fig. 2A). Afterdigestion with EcoRV, an enzyme that does not cut the piuHPV-lodplasmid (8), the Southern analysis showed HPV-16 specific bands

larger than 30 kbp, suggesting that HPV DNA exists as an integrated

form, not as an episomal form, in both the parental (HOK-16B) and all

chemically transformed oral keratinocytes (Fig. 2B).Northern analysis using the probe containing HPV-16 E6/E7 gene

showed that HOK-16B cells expressed the 1.6-kilobase HPV-16

E6/E7 mRNAs. Chemically transformed cells also expressed numerous mRNAs including the predominant 1.6-kilobase E6/E7 messages.The densitometric intensity of the 1.6-kilobase E6/E7 bands of thetumorigenic 16NNK-2 and 16NNK-3 cell lines was notably greaterthan that of the parental cell line but when compared to the HOK-16B

cell line, the intensity of these bands of the nontumorigenic16MNNG-1 and 16NNK-1 cells was similar for 16MNNG-1 butsignificantly lower for 16NNK-1 (Fig. 2C; Table 1).

Analysis of p53 and c-K.i-ra.s2. The levels of 2.7-kilobase p53

mRNA transcript in the immortalized and chemically transformedcells were similar to each other (Fig. 3C; Table 1). No mutations werefound in the region of p53 and c-Ki-ras2 cDNA amplified from

NHOK or any of the established cell lines. Western blot analysisshowed that p53 protein levels in the immortalized and chemicallytransformed cells were similar but notably lower than that fromNHOK (Fig. 4).

Expression of TGF-a, c-myc, EGFR, and the Gene DCC. The4.5-kilobase mRNA is the common transcript of TGF-a gene in nor

mal, immortalized, and chemically transformed oral keratinocytes.The 1.5-kilobase TGF-a mRNAs were also seen in the immortalized

and chemically transformed cells. Among these transcripts, theamount of 1.5-kilobase mRNA was notably enhanced in 16NNK-2,16NNK-3, and 16MNNG-1 cells (Fig. 3A; Table 1). Two c-myc tran

scripts with sizes of 2.4 and 1.1 kilobases were expressed from boththe HOK-16B and the chemically transformed cells, but the amount of

this gene transcription was substantially higher in the chemicallytransformed cells than in the HOK-16B line (Fig. 35; Table 1). Theexpression of both 10.0- and 5.0-kilobase EGFR messages from the

chemical transformants was also significantly increased compared tothe parental counterpart (Fig. 3D and Table 1). The DCC mRNAswere not determined from any type of cells when analyzed by Northern analysis (data not shown), but they were detected by reversetranscriptase-PCR analysis from all cell types. The densitometric in-

Fig. 2. Physical state and expression of HPV-16DNA in HOK-16B and chemically transformedcell lines 16NNK-1. lnNNK-2. 16NNK-3, and16MNNG-1. A. Southern blot hybridization of highmolecular weight cellular DNA digested withBamH\ restriction enzyme to 32P-HPV-16 DNA. B.

Southern blot hybridization of high molecularweight cellular DNA digested with EcoRV restriction enzyme to ['2P|HPV-16 DNA. C, Northern

blot hybridization of cellular polyadenylated RNAsto 1.2-kbp "P-HPV-16 DNA fragment representingthe major early HPV-16 messagesincluding E6 andE7. Kb, kilobase.

Ato •¿� T—

Kbp23.1-9.4-6.6-

B.

Kbp23.1-

9.4-

m T-<o i

CM mi i

iO _C. <em

«o

Z Z Z ZIO ^ ^ «È •¿�*~ CD CD CO CO

I*:O

,- CMn *i i i o

* x * zz z z zz z z 5CD CD CD CD

Kb

ÃŽ&4.4-

2.4-

1.4- -1.6 (HPV-16 E6/E7)

m T- CM CO f'n

CO I I I ^

I Z Z Z Zn z z z 2ÌJÌCO <D «O «O -2.0 </3-actin>

4813

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

HPV AND TOBACCO IN ORAL CANCER

Table 1 Relative degree of transcription ofHPV-16 E6/E7, EGFR, TGF-a, c-myc, p53, and DCC genes from the HPV-immortalizedand chemically transformed oral keratinocytesAmounts of transcription of HPV-16 E6/E7, EGFR, TGF-a, c-myc,andp53 mRNAs from the chemically transformed cells (16NNK-1, 16NNK-2, 16NNK-3,and 16MNNG-1)were

quantitated relative to that of HOK-16B after normalization of them to the levels of human ß-actinmRNAs. The relative amounts of the DCC transcription from different cell lineswere determined from the reverse transcription-pCR experiment relative to that from NHOK after normalization of ß-actingene expression.

Relative expression ofgenesHPV

E6/E7Cell

lines(1.6)"NHOK

ND*HOK-16B1.016NNK-10316NNK-21.716NNK-31.516MNNG-1

1.0"Numbers in parentheses, size of the messagesinhND, notdone.A

m ,- <N « 1 r>"••IllsD.^

y V V^¿

z z z z£z z z2y

CO CD COCDKb

Kb9.5-7.5-

9.5-•Til

::2-4-1.4-EGFR(10.0)ND1.02.72.41.82.2kilobases.m

T-CMCD11^â„¢

\^^^*"O

CDCD-

-((5.0)1.04.11.71.41.6?s*

ZiiCO

CDPC.TGF-a

c-myc p53DCC(4.5)

(1.5) (2.4) (1.1)(2.7)ND

ND ND1.01.01.0 1.0 1.0 1.00.60.70.7 2.0 2.7 1.00.61.21.9 2.0 3.5 1.20.40.91.1 2.7 4.7 1.20.71.02.1 2.0 3.2 1.20.8m

,_ CM mi<D1 1 1O^

v" \S VZv>1 jC Jb. JLA*i

à i i iI2z S $ 9^Kd97-68-

43-

-p53

0.24- 0.24-

-2.0</3-actln)

c. 2 T 7 n

Kb4.4-

2.4-

1.4-

IXO

z z zz z zCO CO CO

oz

co

D S T? ?o7 * * * z¿z z z zO z z z s2 «O CD CD CD

Kb

4.4-

Fig. 3. Transcription of TGF-a, c-myc,p53, and EGFR from HOK-16B and chemicallytransformed cell lines 16NNK-1, 16NNK-2, 16NNK-3,and 16MNNG-1.A, Northern blothybridization of cellular polyadenylated RNAs to 32P-TGF-a cDNA. B, Northern blothybridization of cellular polyadenylated RNAs to *2P-v-mycproviral DNA. C, Northernblot hybridization of cellular polyadenylated RNAs to 32P-p53cDNA. D, Northern blothybridization of cellular polyadenylated RNAs to "P-EGFR cDNA. Kb, kilobase.

tensities of amplified DCC cDNA bands of the immortalized andchemically transformed cells were similar but notably lower than thatof NHOK, indicating that the immortalized and the chemically transformed cells contained fewer DCC messages than the normal counterpart (Fig.5; Table 1).

DISCUSSION

The hypothesis that oral cancer is induced by multiple factors,including HPV infection and chemical carcinogens, is supported bythe following observations, (a) Certain types of HPV are consistentlyassociated with squamous cell carcinoma of the human oral cavity (3,

Fig. 4. Western blot analysis of p53 expression from HOK-16B and chemically transformed cell lines 16NNK-1, 16NNK-2, 16NNK-3,and 16MNNG-1.Kd, molecular weightin thousands.

26). Although some of these viruses can occasionally be detected inhistologically normal tissues, only a small fraction of HPV-infected

lesions progress to cancer (27, 28). (b) Even though chronic exposureto chemical carcinogens in tobacco is still the major known cause oforal cancer, only a minority of tobacco users develop oral cancer.These observations suggest that these risk factors are involved in thedevelopment of oral cancer and/or the progression of this disease.Thus we hypothesize that oral cancer is developed by sequentialexposure of normal oral epithelial cells to "high risk" HPV and

tobacco-related chemical carcinogens.To test this hypothesis, HPV-16-immortalized oral keratinocytes (8)

were exposed to NNK, one of the most potent alkylating carcinogensfound in smokeless tobacco, or MNNG, an alkylating agent that doesnot require metabolic activation (29). Inasmuch as (a) both NNK andMNNG are genotoxic agents and can alkylate DNA, often resulting inG to A transitions in rodents, and (b) NNK can pyridyloxobutylateDNA and can induce p53 and c-Ki-ras2 mutations in laboratory animals (21-23), exposure of the immortalized oral keratinocytes to these

carcinogens could generate more transformed cells.Our results show that sequential in vitro exposure of normal human

oral keratinocytes to HPV-16 and NNK can generate cells with limited

tumorigenicity. Although the tumors had spontaneously regressed,these cells were different from their immortal counterparts. The precise reason of tumor regression is not clear but may be due to thecontinued expression of wild-type c-Ki-ras2 in the chemically treated

cells. However, the chemically transformed cells proliferated fasterand were calcium resistant, whereas the immortalized cells terminallydifferentiated and died in culture medium containing a physiologicallevel of calcium. These results demonstrate that the immortalized cellsare further transformed by exposure to chemical carcinogens. Themolecular mechanisms of cell transformation by these chemical car-

4814

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

HPV AND TOBACCO IN ORAL CANCER

CDCO

7

T- CVJ COI I I

IG

co co co coBp872-»

603-»

310-271-

Although the reasons for enhanced HPV-16 E6/E7 messages in the

tumorigenic cells remain speculative, mutation of the E2 gene mayplay an important role in the overexpression of viral E6/E7 (37).Mutations of the £2gene at its DNA-binding domain induce the loss

of the represser activity of E2 that may result in an overexpression ofE6/E7 messages. Recent studies have also implicated mutations of theviral El gene as having a possible role in carcinogenesis progression(28). Since HPV-immortalized and chemically transformed cells wereoriginally transfected with piuHPV-lód, a plasmid containing linearHPV-16 DNA with an interruption of the El gene, it is unlikely that

these cells express intact El messages.Detection of lower amounts of DCC messages in both the HPV-

immortalized and the chemically transformed cells is also interesting.Since DCC mRNA is frequently absent or reduced in most colorectalcarcinoma cell lines (38), low expression of DCC gene in these cellsmay be closely linked to different phenotypic characteristics of thesecell lines from NHOK.

REFERENCES

317-(y#-actin)

Fig. 5. Southern blot hybridization analysis of amplified DCC cDNA fragment derivedfrom NHOK, HOK-16B, and chemically transformed cell lines 16NNK-1, 16NNK-2.16NNK-3, and 16MNNG-1. The cDNAs were constructed from mRNAs using reversetranscription and the DCC fragment was amplified with polymerase chain reaction. Theamplified cDNA fragments were then electrophoresed, transferred to a nylon filter, andhybridized to [32P]DCC oligonucleotide. Bp, base pair.

cinogens remain unknown but a higher level of HPV-16 E6/E7,TGF-a, c-myc, and EGFR transcripts may be associated with thephenotypic changes of 16NNK-2 and 16NNK-3 cells. Although the16NNK-1 and 16MNNG-1 lines proliferate well in DMEM, these

cells did not show tumorigenicity in nude mice. Inasmuch as variouscell growth factors promote cell proliferation through binding toEGFR (30), the high level of EGFR may be related to the proliferationcapacity of these cell lines in DMEM. The reason for the nontumori-genicity of the 16NNK-1 and 16MNNG-1 lines remains unknown but

we speculate that the nontumorigenicity of these lines may be linkedto lower expression of HPV-16 E6/E7 messages compared to thetumorigenic 16NNK-2 and -3 lines.

Inasmuch as the amount of p53, DCC, and TGF-a transcripts from

the chemical transformants is similar to that from the parental counterpart, it seems that these genes do not play a role in the conversionof the immortalized cells to chemical transformants. Furthermore,since (a) all chemically transformed cells contained a higher level ofEGFR and c-myc transcripts compared to the HOK-16B line and (b)

only two of four chemical transformants showed in vivo tumorigenicity, an enhanced level of EGFR and c-myc transcripts is not likely to

be solely responsible for the tumorigenicity of the two chemicaltransformants. However, since a higher level of HPV-16 E6/E7 tran

scripts is observed only in the tumorigenic lines, an enhanced level ofthese viral messages in combination with increased EGFR and c-myc

transcripts may be responsible for the chemical transformation of thecells. The association of enhanced viral E6/E7 transcripts with thetumorigenic transformants is supported by numerous observations.HPV-16 E6 and E7 gene products (a) are oncogenic substances (31-33), (6) are associated with the immortalization of human keratino-

cytes (31), (c) cooperate with other cytoplasmic oncoproteins such asras to fully transform primary rat cells, (d) induce DNA synthesis ingrowth arrested cells, (e) transform established rodent cells, (/) modulate transcription from certain promoters, and (g) combine with andinactivate the cellular tumor suppressor p53 and Rb gene product(34-36).

10.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

4815

Meanwell, C. A. The epidemiology of human papillomavirus infection in relation tocervical cancer. Cancer Surv., 7: 481^97, 1988.Schwarz, E., Freese, U. K., Gissmann, L., Mayer, W., Roggenbuck, B., Stremlau, A.,and zur Hausen, H. Structure and transcription of human papillomavirus sequences incervical carcinoma cells. Nature (Lond.), 314: 111-114, 1985.

Kellokoski, J. K., Syrjänen, S. M., Chang, F., Yliskoski, M., and Syrjänen, K. J.Southern blot hybridization and PCR in detection of oral human papillomavirus(HPV) infections in women with genital infections. J. Oral Pathol. Med., 21: 459-464, 1992.Syrjänen,S. Viral infections in oral mucosa. Scand. J. Dent. Res., 700: 17-31, 1992.

Woodworth, C. D., Doniger, J., and DiPaolo, J. A. Immortalization of human foreskinkeratinocytes by various human papillomavirus DNAs corresponds to their association with cervical carcinoma. J. Virol., 63: 159-164, 1989.

Pecoraro, G., Morgan, D., and Defendi, V. Differential effects of human papillomavirus types 6, 16, and 18 DNAs on immortalization and transformation of humancervical epithelial cells. Proc. Nati. Acad. Sci. USA, 86: 563-567, 1989.

Woodworth, C. D., Bowden, P. E., Doniger, J., Pirisi, L., Barnes, W., Lancaster, W. D.,and DiPaolo, J. A. Characterization of normal human exocervical epithelial cellsimmortalized in vitro by papillomavirus types 16 and 18 DNA. Cancer Res., 48:4620-4628, 1988.Park, N-H., Min, B-M., Li, S-L., Huang, M. Z., Cherrick, H. M., and Doniger, J.

Immortalization of normal human oral keratinocytes with type 16 human papillomavirus. Carcinogenesis (Lond.), 12: 1627-1631, 1991.Hurlin, P. J., Kaur, P., Smith, P. P., Perez-Reyes, N., Blanton, R. A., and McDougall,J. K. Progression of human papillomavirus type 18-immortalized human keratinocytes to a malignant phenotype. Proc. Nati. Acad. Sci. USA. 88: 570-574, 1991.Li, S-L., Kim, M. S., Cherrick, H. M., Doniger, J., and Park, N-H. Sequentialcombined tumorigenic effect of HPV-16 and chemical carcinogens. Carcinogenesis(Lond.), 13: 1981-1987, 1992.

zur Hausen, H. Intracellular surveillance of persisting viral infections. Lancet, 2:489-491, 1986.

Hoffmann, D., Djordjevic, M. V, and Brunnemann, K. D. New brands of oral snuff.Food Chem. Toxicol., 29: 65-68, 1991.

Stich, H. F., Parida, B. B., and Brunnemann, K. D. Localized formation of micronuclei in the oral mucosa and tobacco-specific nitrosamines in the saliva of "reverse"

smokers, Khaini-tobacco chewers and gudakhu users. Int. J. Cancer, 50: 172-176,

1992.Preston-Martin, S. Evaluation of the evidence that tobacco-specific nitrosamines(TSNA) cause cancer in humans. Crii. Rev. Toxicol., 21: 295-298, 1991.Winn, D. M., Blot, W. J., Shy, C. M., Pickle, L. W., Toledo, A., and Fraumeni, J. F.,Jr. Snuff dipping and oral cancer among women in the Southern United States. N.Engl. J. Med., 304: 745-749, 1981.Hecht, S. S., Chen, C.-H. B., Hirota, N., Ornaf, R. M., Tso, T. C., and Hoffmann, D.Tobacco-specific nitrosamines: formation from nicotine in vitro and during tobaccocuring and carcinogenicity in strain-A mice. J. Nati. Cancer Inst., 60: 819-824, 1978.

Hoffmann, D., Brunnemann, K. D.. Adams, J. D., and Hecht, S. S. Formalion andanalysis of N-nitrosamines in tobacco products and their endogenous formation inconsumers. In: I. K. O'Neill, R. C. von Borsteel, C. T. Miller, J. Long, and H. Bartsch

(eds.), /V-Nitroso Compounds: Occurrence, Biological Effects and Relevance to Human Cancer, pp. 743-762. New York: Oxford University Press, 1984.

Hoffmann, D., Brunnemann, K. D., Adams, J. D., Rivenson, A., and Hecht, S. S.Ai-Nitrosamines in tobacco carcinogenesis. In: P. N. Magee (ed), Nitrosamines andHuman Cancer (Banbury Report No. 12), pp. 211-225. Cold Spring Harbor, NY: Cold

Spring Harbor Laboratory, 1982.Blot, W. J., McLaughlin, J. K., Winn, D. M., Austin. D. F., Greenberg, R. S.,Preston-Martin, S., Bernstein, L., Schoenberg, J. B., Stemhagen. A., and Fraumeni, J.

F., Jr. Smoking and drinking in relation to oral and pharyngeal cancer. Cancer Res.,48: 3282-3287, 1988.

Steele, C., and Shillitoe, E. J. Viruses and oral cancer. Crii. Rev. Oral Biol. Med., 2:153-176, 1991.Belinsky, S. A., Devereux, T. R., White, C. M., Foley, J. F., Maronpot, R. R., and

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

HPV AND TOBACCO IN ORAL CANCER

Anderson, M. W. Role of Clara cells and type II cells in the development of pulmonary tumors in rats and mice following exposure to a tobacco-specific nitrosamine.Exp. Lung Res., 17: 263-278, 1991.

22. Orcffo, V. I., Lin, H. W., Padmanahham, R., and Witschi, H. K-ras andp53 mutationsin 4-(methylnitrosamino)-l-(3-pyridyl)-l-butanone-induced hamster lung tumors.Carcinogenesis (Lond.), 14: 451^*55, 1993

23. You, M., Candrian, U., Maronpot, R. R., Stoner, G. D., and Anderson, M. W.Activation of the Ki-ra.v protooncogene in spontaneously occurring and chemicallyinduced lung tumors of the strain A mouse. Proc Nati. Acad. Sci. USA, 86: 3070-3074, 1989.

24. Goyette, M. C, Cho, K., Fasching, C. L., Levy, D. B., Kinzler, K. W., Paraskeva, C,Vogelstein. B., and Stanhridge, E. J. Progression of colorectal cancer is associatedwith multiple tumor suppressor gene defects hut inhibition of tumorigenicity isaccomplished by correction of any single defect via chromosome transfer. Mol. Cell.Biol., 12: 1387-1395, 1992

25. Fuqua. S. A. W., Fitzgerald, S. D., and McGuire, W. L. A simple polymerase chainreaction method for detection and cloning of low-abundance transcripts. BioTech-niqucs, 9: 206-211, 1990.

26. Shah, K. V. Papillomavirus infections of the respiratory tract, the conjunctiva, and theoral cavity. In: H. Pfcister (ed.), Papillomavirus and Human Cancer, pp. 73—90.Boca

Raton, FL: CRC Press, 1990.27. Jalal. H., Sanders, C. M., Prime, S. S.. Scully. C., and Maitland, N. J. Detection of

human papiiloma virus type 1ft DNA in oral squamous from normal young adults. J.Oral Pathol. Med., 21: 465^170, 1992.

28. MacNab. J. C., Walkinshaw, S. A., Cordiner, J. W., and Clements, J. B. Humanpapillomavirus in clinically and histologically normal tissue of patients with genitalcancer. N. Engl. J. Med., 315: 1052-1058, 1986.

29. Bertram, J. S., and Heidelberger, C. Cell cycle dependency of oncogenic transforma

tion induced by /V-methyl-W-nitro-A^-nitrosoguanidine in culture. Cancer Res., 34:526-537, 1974.

30. Massague, J. Epidermal growth factor-like transforming growth factor: II. Interaction

with epidermal growth factor receptors in human placenta membranes and A431 cells.J. Biol. Chem.. 258: 13614-13620, 1983.

31. Munger. K., Phelps, W. C., Bubb, V, Howley, P. M., and Schlegel, R. The E6 and £7genes of the human papillomavirus type 16 together are necessary and sufficient fortransformation of primary human keratinocytes. J. Virol., 63: 4417-4421, 1989.

32. Romanczuk, H., Villa, L. L., Schlegel, R., and Howley, P. M. The viral transcriptionalregulatory region upstream of the £6and £7genes is a major determinant of thedifferential immortalization activities of human papillomavirus types 16 and 18.J. Virol., 65: 2739-2744, 1991.

33. Miyasaka, M., Takami, Y., Inoue, H., and Hakura, A. Rat primary embryo fibroblastcells suppress transformation by the £6and £7genes of human papillomavirus typa16 in somatic hybrid cells. J. Virol., 65: 479-482, 1991.

34. Scheffner, M., Werness, B. A., Huibregtse, J. M., Levine, A. J., and Howley, P. M. TheE6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes thedegradation of p53. Cell, 63: 1129-1136, 1990.

35. Dyson, N., Howley, P. M.. Munger. K., and Harlow, E. The human papillomavirus-16

E7 oncoprotein is able to bind the retinoblastoma gene product. Science (WashingtonDC), 243: 934-936, 1989.

36. Ruley, H. E. Transforming collaborations between ras and nuclear oncogenes. CancerCells (Cold Spring Harbor), 2: 258-268, 1990.

37. Barsoum, J., Prakash, S. S., Han, P., and Androphy, E. J. Mechanism of action of thepapillomavirus E2 repressor: repression in the absence of DNA binding. J. Virol., 66:3941-3945, 1992.

38. Cho, K. R., and Vogelstein, B. Suppressor gene alterations in the colorectal adenoma-carcinoma sequence. J. Cell. Biochem., Suppi I6G: 137-141, 1992.

4816

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

1993;53:4811-4816. Cancer Res   Myong Soo Kim, Ki-Hyuk Shin, Jeong-Hwa Baek, et al.  

-nitrosoguanidine in Oral CarcinogenesisN-nitro-′N-Methyl-N-Nitrosamine, and NHPV-16, Tobacco-specific

  Updated version

  http://cancerres.aacrjournals.org/content/53/20/4811

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/53/20/4811To request permission to re-use all or part of this article, use this link

Research. on February 15, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from