differencesinsmad4expressioninhumanpapillomavirus …and 1.5 units of taq polymerase. dna was...

Differences in Smad4 Expression in Human PapillomavirusType16^Positive and Human PapillomavirusType16^NegativeHead and Neck Squamous Cell CarcinomaAdriana Ba¤ ez,1,2 Alan Cantor,6 Sharon Fonseca,2 Maria Marcos-Martinez,3 Linda A. Mathews,5

Carlos A.Muro-Cacho,4 andTeresitaMun‹ oz-Antonia4,5

Abstract TheSMADs are agroupof interrelatedproteins thatmediate transforminggrowth factorh (TGF-h)signaling. UponTGF-h binding theTGF-h type I receptor phosphorylates Smad2 and Smad3,which then complex with Smad4 and translocate to the nucleus, with subsequent activation oftarget genes. Disruption of TGF-h signaling is thought to contribute to the development of headand neck squamous cell carcinomas (HNSCC). Alterations in the function of the DPC4/Smad4tumor suppressor gene have been found to inactivateTGF-h signaling in several tumor types. Forexample, DPC4/Smad4 is lost or mutated in colorectal, pancreatic, and esophageal cancers. Inaddition, DPC4/Smad4 transcriptional activity and TGF-h ability to inhibit DNA synthesis isblockedby the E7 protein of the humanpapillomavirus type16 (HPV16) in cervical carcinoma celllines. HPV16 infection is a risk factor for the development of a subset of HNSCC.This study wasundertaken to investigate a potential correlation between expressionof components of theTGF-hsignaling pathway and HPV16 status in HNSCC tumors.We examined the expression of TGF-hsignalingproteins Smad2, Smad2-P, andSmad4by immunohistochemistry in 27HPV16-negativeand 16 HPV16-positive HNSCCs.We compared the expression patterns and assessed their rela-tionship to HPV16 status. No significant differences were detected between HPV16-positive andHPV16-negative tumors in the expression of Smad2 and Smad2-P. Smad4 expression, however,was decreased in 56% of the HPV16-positive tumors and in 39% of HPV16-negative tumors.Thisdifference was statistically significant (P = 0.01) suggesting that loss of Smad4 expression maybe involved in HPV16-induced carcinogenesis of HNSCC.

Transforming growth factor h (TGF-h) belongs to a multi-gene family with multiple biological activities (1). TGF-hproteins are potent endogenous modulators of cellulargrowth, differentiation, migration, extracellular interactions,immune system function, and apoptosis. These activities aremediated by a ligand-induced cell surface complex formed byTGF-h receptors type I (ThR-I) and type II (ThR-II; ref. 2).TGF-h signals are transduced by the TGF-h receptors andtheir substrates, the Smad proteins, which maintain the cellin a nonproliferative state in G1 (3, 4). Specific TGF-hreceptor kinases phosphorylate Smad2 and Smad3 uponligand binding. These Smads then associate with Smad4 and

the resulting complex translocates to the nucleus where itactivates transcription of target genes.

The TGF-h receptor complex, including the downstreamsignaling molecules, has been implicated in the pathogenesisof many human cancers including head and neck squamouscell carcinoma (HNSCC; refs. 5–12) and have been suggestedto act as a tumor suppressor pathway (13). Alterations ingenes involved in the TGF-h growth inhibitory pathway wouldallow tumor cells to escape TGF-h-mediated growth regula-tion. Previous studies have shown a direct relationshipbetween reduced levels of ThR-II, resistance to the growthinhibitory control of TGFh-1, and tumorigenicity, thusdefining a role for ThR-II as a tumor suppressor gene. Theexact mechanisms for down-regulation of ThR-II expressionare not fully understood; however, ThR-II is frequentlymutated in a wide variety of tumor types including HNSCC(5, 7). Mutations in ThR-I are less frequent (11, 12).Consistent with a tumor suppressor role for the TGF-hsignaling pathway, alterations have been identified in othercomponents of the pathway (14–24). For example, mutationsin the DPC4/Smad4 (14–19) and MADR2/Smad2 (21, 22)genes have been reported in pancreatic, cervical, esophageal,and colon cancer.

In addition to genetic alterations, other mechanisms such asinfection with human papillomavirus (HPV) have beensuggested to inactivate the TGF-h signaling pathway in certain

www.aacrjournals.org Clin Cancer Res 2005;11(9) May1, 20053191

Authors’ Affiliations: Departments of 1Otolaryngology-Head and Neck Surgery,2Pharmacology and 3Pathology, University of Puerto Rico School of Medicine,SanJuan, Puerto Rico and Departments of 4Interdisciplinary Oncology, 5MolecularOncology Program, and 6Biostatistics and Informatics Core, H Lee Moffitt CancerCenter Research Institute,Tampa, FloridaReceived 7/1/04; revised 11/23/04; accepted 12/13/04.Grant support: Department of Otolaryngology and NIH grant P20 CA91402.The costs of publication of this article were defrayed in part by the payment ofpage charges.This article must therefore be hereby marked advertisement in accor-dance with 18 U.S.C. Section 1734 solely to indicate this fact.Requests for reprints: Adriana Ba¤ ez, Department of Otolaryngology, School ofMedicine, University of Puerto Rico, P.O. Box 365067, San Juan, PR 00936-5067. Phone: 787-758-2525; Fax: 1-787-759-6722; E-mail: [email protected].

F2005 American Association for Cancer Research.

Human Cancer Biology

Research. on April 3, 2020. © 2005 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

tumor types (25–27). Transfection with HPV16 DNA immor-talizes human keratinocytes in vitro, suggesting a role for HPVin carcinogenesis (28). This HPV-induced carcinogenesis ismediated by the E6 and E7 viral oncoproteins, which exertoverlapping effects in cell cycle control (29, 30). Both proteinsare consistently expressed in cervical tumors and in a subset ofHNSCCs (29–32). E7 forms complexes with members of theretinoblastoma family, pRb, p107, and p130 and enhancesdegradation of pRb and release of E2F, which activates cellcycle progression (33, 34). The oncogenic potential of E6 islinked to its antiapoptotic effect through its binding to p53and to activation of host-cell telomerases (35–37). Binding ofE6 to p53 mediates its degradation by proteasomes, affectingG1-S checkpoint control (36) and leading to chromosomalinstability (37). Yet, malignant transformation by E6/E7 seemsto require further alterations of cell cycle regulatory genes,including components of the TGF-h signaling pathway(38, 39). In early stages of cervical cancer, malignant prog-ression in HPV-infected cells has been linked to the loss ofsensitivity to TGF-h (38–45). This resistance has been linkedto alterations in the expression of cell cycle regulatory genes(38). Whereas TGF-h1 has been shown to down-regulatetranscription of E6 and E7 genes in nontumorigenic HPV-infected genital epithelial cell lines, the HPV16-infected cells,SiHa and CasKi, show resistance to TGF-h regulation(41, 46). TGF-h resistance in HPV16-transformed humankeratinocytes has been associated with loss of expression ofThR-I (42). In addition, four different mutations in ThR-Iand a polymorphism that may predispose to cervical cancerhave been detected in HPV16-infected cervical cells (43).More recently, it has been shown that E7 blocks the Smadtranscriptional activity by binding to the complex formedby Smad2, Smad3 and Smad4 and preventing themfrom binding with DNA (25). These results suggest aninteraction between HPV16 E6 and E7 oncoproteins andTGF-h signaling components in the tumorigenicity of certaintumor types.

Tobacco and alcohol uses are the major risk factors forHNSCC (47); however, recent epidemiologic and molecularstudies provide evidence that HPV may be associated with asubset of HNSCC (30–32). Infection with high risk HPV16DNA immortalizes oral keratinocytes, and the E6 and E7oncoproteins are mutagenic in human oral keratinocytes,suggesting a role for HPV in oral carcinogenesis (48).Furthermore, HPV infection has been associated with benign,

premalignant, and malignant lesions in the head and neckregion with HPV16 being the most common subtype (31).Recent reports using PCR-based assays have estimated afrequency of HPV16 in HNSCC of 20% to 30% (32). However,using similar PCR-based assays we found in a series of 118Puerto Rican HNSCC cases that 44% were infected withHPV16 (49).

To examine whether aberrant expression of TGF-h signalingcomponents could be contributing to oncogenesis in HPV-positive HNSCC, we examined by immunohistochemistry thedifferences in patterns of expression of several key molecules ofthe TGF-h pathway in HPV16-negative and HPV16-positiveHNSCC. Our results show that differences in expression ofSmad4 between HPV16-positive and HPV 16-negative HNSCCare statistically significant, suggesting that in this subgroupof HNSCC, Smad4 may play a role in the development ofHNSCC.

Materials andMethods

Patient and tissue specimens. Tissue samples were obtained from 43patients with HNSCC who underwent curative surgery. Patients wereaccrued through the Otolaryngology-Head and Neck Surgery CancerClinics at the University of Puerto Rico School of Medicine. Beforeenrollment in the study, a signed consent form was obtained from allpatients to allow the collection of tissues and demographic and clinicaldata. The study was approved by the Institutional Review Boards of theUniversity of Puerto Rico Medical Sciences Campus and the MoffittCancer Center and Research Institute. The data recorded for each patientincluded age, sex, site, stage, and pathologic grade.

DNA isolation. High molecular weight DNA from the tumor wasisolated using the DNA Isolation kit for cells and tissues (Roche AppliedSciences, Indianapolis, IN) according to the manufacturer’s instruc-tions. DNA concentration was measured using a CytoFluor Series 4000(Applied Biosystems, Foster City, CA).

Detection of HPV16. DNA from tumors was tested with aglyceraldehyde-3-phosphate dehydrogenase primer set to confirm theintegrity of the genomic DNA. Samples were screened for the presenceof HPV16 DNA by amplification with HPV16 E7-type specific primersets 5V-GATGAAATAGATGGTCCAGC-3V and 5V-GTACCTACGTGT-GTGCTTTGT-3V, which generate a 120-bp amplicon and E6-typespecific primer sets 5V-CGAAACCGGTTAGTATAA-3V and 5V-GTATT-CTCCATGCATGATT-3V, which generate a 523-bp amplicon. Eachreaction was carried out in a 25-AL reaction volume with finalconcentration of 10 mmol/L Tris-HCl (pH 9), 1.5 mmol/L MgCl2,50 mmol/L KCl, 0.4 Amol/L of each primer, 200 Amol/L of eachdeoxynucleotide triphosphate, 0.1 to 0.2 Ag of genomic DNA template,

www.aacrjournals.orgClin Cancer Res 2005;11(9) May1, 2005 3192

Fig. 1. PCR detection of HPV16 E6 and E7 in HNSCC tissues. Representative results of PCR reactions using HPV16 E6- and E7-specific primers. DNA samples wereelectrophoresed on 2% agarose gel and viewed by ethidium bromide staining. Size of PCR products was measured relative to molecular weight marker (M). Positive control(SiHa DNA) andnegative control (K567 DNA) gave the expected results. Lanes 2-18, results of E6 primers (523 bp) from16 tumors with positive results (lanes 3, 6, 8,12,15,16).The121-bp amplification products with E7 primers of the same tumor samples in the same order. Samples 3, 6, 8,12,15, and16 are HPV16 E7 positive.




and 1.5 units of Taq polymerase. DNA was amplified in a GeneAmp2700 System (Applied Biosystems). Conditions for the E6 primer setconsisted of an initial denaturation step of 3 minutes at 94jC and 35cycles of 1 minute at 94jC, 1 minute at 55jC, 1.5 minutes at 72jC, andone final extension step of 7 minutes at 72jC for the 120-bp amplicon.For the E7 primer set, after the initial denaturation step, conditions

consisted of 35 cycles of 30 seconds at 94jC, 30 seconds at 55jC,1minute at 72jC, and a final extension step of 7 minutes at 72jC. PCRproducts were separated by gel electrophoresis in a 2% ethidium bromideagarose gel in Tris/borate buffer [89 mmol/L Tris-borate, 89 mmol/Lboric acid (pH 7.8), and 0.8 mmol/L EDTA]. The amplified productswere detected using the Gel Doc EQ System with Molecular Analysis


Table1. Expression of smads in HPV16+ and HPV16�HNSCC

HPV Site Gender Age Differentiation Stage N-Smad4 T-Smad4 N-Smad2-P T-Smad2-P

1 1 Larynx M 38 MD I 0.1 0 0.8 0.12 1 Oropharynx F 56 MD II 0.8 0.2 2 13 1 Oral Cavity M 68 MD II 0.6 0.4 1.8 2.54 1 Oral Cavity M 56 WD II 0.4 0.2 1.8 2.45 1 Oral Cavity M 60 PD II 1.2 1.2 NA NA6 1 Oral Cavity M 79 MD II 0.4 0.1 1.8 2.17 1 Oral Cavity F 71 WD III 0.1 0 0.4 1.68 1 Oral Cavity M 67 WD III 0 0 0.8 1.29 1 Hypopharynx M 82 MD IV 0.3 0.1 1.2 2.810 1 Larynx M 67 MD IV 0.1 0.3 0.8 1.411 1 Larynx M 57 MD IV 0 0 2.4 2.412 1 Oropharynx M 53 MD IV 0 0 0 0.113 1 Larynx M 70 WD IV 0.1 0 2.4 2.714 1 Oral Cavity M 55 MD IV 0.3 0.6 0 015 1 Larynx M 56 WD IV 2.4 1.8 NA NA16 1 Hypopharynx M 50 MD IV 2.4 0.8 1.2 0.5

Average 0.58 0.36 1.1 1.317 2 Oral Cavity F 56 MD I 0.4 0.1 1.8 2.418 2 Oral Cavity M 49 MD II 0.8 1.8 1 119 2 Oropharynx M 62 MD II 0.2 0.4 NA NA20 2 Larynx M 55 MD II 1.2 1.4 0.6 221 2 Larynx M 44 MD II 0.5 0.4 0.8 2.222 2 Larynx M 56 WD II 1.2 0.5 1.8 1.623 2 Oral Cavity M 47 MD II 1.8 1.8 0.8 2.424 2 Larynx M 64 MD III 0.1 0 0.8 025 2 Oral Cavity M 84 MD III 0 0.4 1.2 0.726 2 Larynx M 46 WD III 0.1 0.4 1.8 2.427 2 Oropharynx M 43 MD III 1.8 1.2 NA NA28 2 Larynx M 75 MD III 0.1 0.1 NA NA29 2 Larynx M 73 MD IV 0 0 1.8 1.830 2 Hypopharynx M 48 MD IV 0 0 1.2 031 2 Oral Cavity M 55 MD IV 0.6 0.6 1.6 2.732 2 Oral Cavity M 58 WD IV 0.4 0.2 1.8 2.733 2 Oropharynx M 51 WD IV 0.8 1.6 1.2 2.434 2 Hypopharynx M 74 MD IV 1.2 1.8 NA 2.735 2 Hypopharynx F 74 PD IV 0.2 0.2 1.8 0.236 2 Larynx M 60 WD IV 0.2 0.4 0 NA37 2 Larynx M 65 WD IV 0.2 0.8 NA NA38 2 Oral Cavity M 51 WD IV 1.2 1.2 1.2 2.439 2 Oral Cavity M 49 WD IV 0.8 0.6 0.6 0.640 2 Oral Cavity F 70 MD IV 0.8 1.8 0.4 0.641 2 Oral Cavity M 72 WD IV 0.8 1.6 0.4 0.842 2 Larynx M 58 MD III 0.4 0.2 NA NA43 2 Nasopharynx F 81 MD IV 0.1 0.8 0.8 0.1

Average 0.59 0.75 0.87 1.17

NOTE: Forty-three carcinomas with different degrees of differentiation were evaluated by immunohistochemistry for the expression of Smad2-P and Smad4. Cases aretabulated according to HPV status (1 = positive, 2 = negative), site, gender, and differentiation.Abbreviations:WD, well differentiated; MD, moderately differentiated; PD, poorly differentiated.

Loss of Smad4 Signaling in HPV16-Positive HNSCC



Software (Bio-Rad Laboratories, Hercules, CA). Reaction mixtures withSiHa DNA and K562 DNA were used as positive control and negativecontrols, respectively.

Immunohistochemistry. To minimize heterogeneity and tissuevariability across the different tissue sections, consecutive 3-Amsections were prepared without discarding intervening tissue. Thefirst section was stained with H&E and the rest of the sections wereused for immunohistochemistry. Formalin-fixed, paraffin-embeddedtissue sections were dried at 37jC overnight. Sections were deparaffi-nized by an initial warming to 60jC, followed by two xylene changes10 minutes each, two series of 30 dips in absolute alcohol, 30 dipsin 95% alcohol, and 20 dips in deionized water. Antigen retrieval orenzyme digestion procedures were done as previously described(8, 9). Slides were placed for 5 minutes in TBS/Tween and processedon a Dako Autostainer using the Dako LSAB+ Peroxidase detectionkit (Dako, Carpinteria, CA). Endogenous peroxidase was blockedwith 3% aqueous hydrogen peroxide followed by two 20-dip washesin deionized water. Anti-Smad2P and anti-Smad4 primary antibodies(both from Upstate Biotechnology, Lake Placid, NY) were applied atroom temperature at a 1:100 dilution. On two other sections, theprimary antibody was replaced with an isotype-match antibody ofunrelated specificity and the primary antibody was omitted. In bothsituations, signal was not observed. After 1 hour, slides were rinsedwith PBS. Detection was made using the Vectastain Elite avidin-biotincomplex kit series (Vector Laboratories, Burlingame, CA) usingdiaminobenzidine as chromogen. Counterstain was done withmodified Mayer’s hematoxylin. Slides were dehydrated throughgraded alcohol, cleared with xylene and mounted with resinous

mounting medium. In each case, quantification of protein expressionwas done visually by an experience immunopathologist. Signalintensity was classified as 0 (no intensity), 1+ (weak intensity), 2+(moderate), and 3+ (strong). The percentages of positive cells in eachof the four intensity groups were estimated and multiplied by theircorresponding intensity scores. The four partial values obtained ineach section were added and expressed as a final score. The averageof three of these estimations is provided.

Statistical analysis. The Wilcoxon rank sum (Mann-Whitney U)test was used to compare protein expression between the differentgroups.

Results and Discussion

Recent reports using PCR-based assays have estimated anHPV16 frequency in HNSCC of 20% to 30% (32). Morerecently, we reported a significantly higher HPV16 frequencyof 44% in a series of 118 HNSCC Puerto Rican patients (49).In the present study, we investigated the presence of HPV16in 43 HNSCC specimens selected from the Otolaryngology-Head and Neck Surgery Cancer Clinics at the University ofPuerto Rico School of Medicine. High molecular weight DNAfrom the tumors was isolated and screened for the presenceof HPV16 DNA by PCR amplification with HPV16 E6 andE7-type specific primer sets. The results of the PCRamplification can be seen in Fig. 1. Of the 43 samples


Fig. 2. Smad4 expression in HPV16-positive (A-D)and HPV16-negative (E-H) squamous cell carcinoma.A and C, nonneoplastic squamous epithelium.Strong expression in basal, parabasal, and lowintermediate layers. B andD, marked reduction inexpression in carcinoma. E andG, nonneoplasticsquamous. Prominent expression in epithelium inmostof the cells. F andH, marked expression in carcinoma.




tested by PCR, 27 samples (63%) were HPV16 negative and16 samples (37%) were HPV16 positive. Eleven of thesesamples were positive for E6 and E7. In five of these samples,the DNA available was only sufficient to test for E6. Thisfrequency of 37% positivity, although higher than those ofother series, is lower than the 44% previously reported by usin the Puerto Rican population.

To test whether aberrant expression of TGF-h signalingcomponents could be contributing to the tumorigenicmechanism of HPV16-positive HNSCC, we used immuno-histochemistry to examine the expression of TGF-h signalingmolecules. Both HPV16-positive and HPV16-negative HNSCCcases were examined for the ThR-II expression and TGF-hintracellular signal transducers (Smads). The HNSCC caseswere evaluated using different antibodies on consecutivesections of the same tissue block as described in MATERIALSAND METHODS. The decrease in ThR-II expression observedwas comparable to results obtained previously (8). Well tomoderately differentiated carcinomas showed heterogeneousexpression of variable intensity, and poorly differentiatedcarcinomas were completely devoid of ThR-II (data notshown), suggesting that aberrant ThR-II expression is acontributing factor to the pathogenesis of HNSCC in ourseries. This is in agreement with what has been reported byus and others, suggesting that reduced expression of TGF-hreceptors is an unfavorable prognostic factor in humanHNSCC (8). However, we found no apparent relationshipbetween loss of ThR-II and HPV16 positivity (data notshown).

To further examine if alterations in the TGF-h signalingpathway are involved in the tumorigenicity of HPV16-positiveHNSCC, we examined by immunohistochemistry the expres-sion of Smads, the TGF-h intracellular signal transducers.As stated before, 43 squamous cell carcinomas were includedin this study (see Table 1). The age of the patients ranged from38 to 84 years with an average of 62 for the HPV16-positivecases and of 60 for the HPV16-negative cases. Thirty-seven

patients (86%) were male and six (14%) were female.The anatomic location of the tumors included larynx,hypopharynx, oropharynx, nasopharynx, and oral cavity.We found no statistically significant difference in expressionbetween normal tissue and the corresponding carcinoma forSmad2, Smad2-P, or Smad3 (data not shown). This contrastswith our previous report of 13 HNSCC from non-Hispanicwhites, where we found, in f70% of the tumors, a loss inthe expression of the activated and phosphorylated formof Smad2 (9).

Interestingly, when the patterns of expression wereanalyzed in the context of HPV16 status, the averageexpression of Smad4 in the normal tissue of both HPV16-positive and HPV16-negative cases was similar (0.58 versus0.59). However, the average expression of Smad4 in HPV16-positive carcinomas was reduced (0.36) compared with thatof HPV16-negative carcinomas (0.75; Fig. 2; Table 1). Thedifference between normal and tumor was statisticallysignificant only in HPV16-positive cells (P = 0.01). Thesedifferences were independent of age, sex, differentiation, ortumor stage. In contrast, there was no statistically significantdifference in the average expression of other Smads. Forexample, as can be seen in Fig. 3 and Table 1, there is nostatistically significant difference in the expression of Smad2-P in HPV16-positive cases between normal tissue (1.1) andtumor (1.3).

Loss of Smad4 has been reported in several human cancertypes, including esophageal squamous cell carcinoma (50),where expression of Smad4 seems to be associated with depthof invasion. The mechanism for this alteration in theexpression of Smad4 in HNSCC is not fully understood. Forexample, even in those instances where the protein levels aredecreased, the DPC4/Smad4 mRNA levels remain relativelysteady, suggesting that the defect is not at the level oftranscription (50). The DPC4/Smad4 and MADR2/Smad2 geneshave been mapped to 18q.21, and genetic alterations in DPC4/Smad4 and MADR2/Smad2 genes have been reported in human


Fig. 3. Expression of Smad2-P detected byimmunohistochemistry in squamous cell carcinoma.Comparison between HPV-positive (A and C) andHPV-negative (B andD) tissue samples.A, HPV-positive, nonneoplastic squamous mucosashowing koilocytotic changes and strong nuclearexpression of Smad2-P. B, HPV-negative,nonneoplastic squamous mucosa with strong nuclearexpression of Smad2-P. C, corresponding squamouscell carcinomawith marked decrease in Smad2-Pexpression.D, corresponding squamous cellcarcinomawith no decrease in Smad2-P expression.




pancreatic and colon cancers, in which 18q.21 is frequentlylost (14, 16, 17, 21). In head and neck cancers, however,genetic analysis revealed minimal loss of this chromosomalregion (51). Mutations in DPC4/Smad4 , Smad 3 , and MADR2/Smad2 genes have been reported in pancreatic (14–16), colon(17, 18, 21), lung (52), and cervical cancers (22). However, instudies of head and neck, hepatocellular, breast, and prostatecancers mutations in the ThR-II, MADR2/Smad2 and DPC4/Smad4 genes were found to be rare (9, 52–57). This suggeststhat genetic instability in the Smad genes does not account forthe widespread resistance to TGF-h in HNSCC, and that othermechanism(s) may be involved.

Decreased expression of Smad4 could further contribute toTGF-h signaling abnormalities at a post-receptor level intumors with abnormal receptor expression and/or be respon-sible for defective signaling in tumors with normal receptorexpression. Decrease in Smad4 expression is associated witha defective TGF-h response (14–20). Loss of sensitivity to the

antiproliferative effects of TGF-h in pancreatic cells has beenassociated with loss of Smad4 expression and subsequent lossof p21 (58). We are currently examining the status of p21 inthese tumors.

It could be speculated that a step in the HNSCC tumor-igenic mechanism is the acquisition of resistance to TGF-hwhich, in turn, makes cells more susceptible to infection byHPV16 and more tumorigenic. Alternatively, it can bespeculated that HPV16 infection is the first event in HNSCC,and that in analogy to what has been reported in cervicalcancers (which also exhibit significantly low Smad4 expres-sion), binding of E7 (or E6) inhibits Smad biological activity(25). Inhibition of Smad biological activity could be amechanism of resistance to the antiproliferative effects ofTGF-h. Experiments are in progress in our laboratory toinvestigate whether TGF-h resistance makes cells more suscep-tible to HPV16 infection, or whether HPV16 infection makescells resistant to TGF-h.



References

1. Massague J. TGF-h signal transduction. Annu RevBiochem1998;67:753^91.

2. Massague J, Attisano L,Wrana JL. TheTGF-h familyand its composite receptors. Trends Cell Biol 1994;4:172^8.

3. Alexandrow MG, Moses HL. Transforming growthfactor h and cell cycle regulation. Cancer Res 1995;55:1452^7.

4. Attisano L,Wrana JL. Smads as transcriptional co-modulators. Curr Opin Cell Biol 2000;12:235^43.

5. Garrigue-Antar L, Mun‹ oz-Antonia T, Antonia SJ,Gesmonde J, Vellucci VF. Missense mutations of thetransforming growth factor h type II receptor in hu-man head and neck squamous carcinoma cells. Can-cer Res 1995;55:3982^7.

6. Eisma RJ, Spiro JD, von Biberstein SE, Lindquist R,Kreutzer DL. Decreased expression of transforminggrowth factor h receptors on head and neck squa-mous cell carcinoma tumor cells. Am J Surg 1996;172:641^5.

7.Mun‹ oz-AntoniaT, Li XC, Reiss M, Antonia SJ. A mu-tation in theTGF-h receptor gene promoter associatedwith loss of gene expression. Cancer Res 1996;56:4831^5.

8. Muro-Cacho CA, Anderson M, Cordero J,Mun‹ oz-Antonia T. Expression of transforminggrowth factor h type II receptors in head andneck squamous cell carcinoma. Clin Cancer Res1999;5:1243^8.

9. Muro-Cacho C, Rosario-Ortiz K, Livingston S,Mun‹ oz-Antonia T. Defective transforming growthfactor h signaling pathway in head and neck squa-mous cell carcinoma is evidenced by the lack of ex-pression of activated Smad2. Clin Cancer Res 2001;7:1618^26.

10. LogulloAF, Nonogaki S,Miguel RE, et al. Transform-ing growth factor h1 (TGFh1) expression in head andneck squamous cell carcinoma patients as related toprognosis. JOral Pathol Med 2003;32:139^45.

11. ChenT, YanW,Wells RG, et al. Novel inactivatingmutations of transforming growth factor-h type I re-ceptor gene in head-and-neck cancer metastases. IntJCancer 2001;93:653^61.

12. KnoblochTJ, Lynch MA, Song H, et al. Analysis ofTGF-h type I receptor for mutations and polymor-phisms in head and neck cancers. Mutat Res 2001;479:131^9.

13. Kim SJ, ImYH, Markowitz SD, BangYJ. Molecularmechanisms of inactivation ofTGF-h receptors duringcarcinogenesis. Cytokine Growth Factor Rev 2000;11:159^68.

14. Hahn SA, Schutte M, HoqueAT, et al. DPC4, a can-

didate tumor suppressor gene at human chromosome18q21.1. Science1996;271:350^3.

15. SutoT, SugaiT, HabanoW, et al. Allelotype analysisof the PTEN, Smad4 and DCC genes in biliary tractcancer. Anticancer Res 2002;22:1529^36.

16. Iacobuzio-Donahue CA, Song J, Parmiagiani G,Yeo CJ, Hruban RH, Kern SE. Missense mutations ofMADH4: characterization of the mutational hot spotand functional consequences in human tumors. ClinCancer Res 2004;10:1597^604.

17. Takagi Y, Kohmura H, Futamara M, et al. Somaticalterations of the DPC4 gene inhuman colorectal can-cers in vivo. Gastroenterology1996;111:369^72.

18. FukushimaT, Mashiko M,Takita K, et al. Mutationalanalysis of TGF-h type II receptor, Smad2, Smad3,Smad4, Smad6 and Smad7 genes in colorectal can-cer. JExp Clin Cancer Res 2003;22:315^20.

19. Blaker H, von HerbayA, Penzel R, Gross S, Otto HF.Genetics of adenocarcinomas of the small intestine:frequent deletions at chromosome18q and mutationsof the SMAD4 gene. Oncogene 2002;21: 158^64.

20.Natsugoe S, Xiangming C, Matsumoto M, et al.Smad4 and transforming growth factor h 1 ex-pression in patients with squamous cell carcinomaof the esophagus. Clin Cancer Res 2002;8:1838^42.

21. Eppert K, Scherer SW, Ozcelik H, et al. MADR2maps to 18q21and encodes aTGFh regulated MAD-related protein that is functionally mutated in colorec-tal carcinoma. Cell 1996;86:543^52.

22. Maliekal TT, Antony ML, Nair A, Paulmurugan R,Karunagaran D. Loss of expression, and mutations ofSmad2 and Smad4 in human cervical cancer. Onco-gene 2003;22:4889^97.

23. BoulayJL, Mild G, LowyA, et al. SMAD7 is a prog-nostic marker in patients with colorectal cancer. Int JCancer 2003;104:446^9.

24. Xu G, Chakraborty C, Lala PK. Expression ofTGF-hsignaling genes in the normal, premalignant, and ma-lignant human trophoblast: loss of smad3 in chorio-carcinomacells.BiochemBiophysResCommun2001;287:47^55.

25. Lee DK, Kim BC, Kim IY, Cho EA, Satterwhite DJ,Kim SJ. The human papillomavirus E7 oncoproteininhibits transforming growth factor-h signaling byblocking binding of the Smad complex to its target se-quence. JBiol Chem 2002;277:38557^64.

26. Lee DK, Kim BC, Brady JN, Jeang KT, Kim SJ.Human T-cell lymphotropic virus type 1 tax inhibitstransforming growth factor-h signaling by blockingthe association of Smad proteins with Smad-bindingelement. J Biol Chem 2002;277:33766^75.

27. Saha D, Datta PK, Beauchamp RD. Oncogenicras represses transforming growth factor-h/Smadsignaling by degrading tumor suppressor Smad4.J Biol Chem 2001;276:29531^7.

28. Pirisi L, Creek KE, Doniger J, DiPaolo JA. Con-tinuous cell lines with altered growth and differenti-ation properties originate after transfection of humankeratinocytes with human papillomavirus type 16DNA. Carcinogenesis 1988;9:1573^9.

29. zur Hausen H. Papillomavirus infections: a majorcause of human cancers. Biochim Biophys Acta1996;1288:F55^78.

30. BrandsmaJL, AbramsonAL. Associationofpapillo-maviruses with cancers of the head and neck. ArchOtolaryngol Head Neck Surg1989;115:621^62.

31.McKaigRG, Baric RS,OlshanAF.Humanpapilloma-virus and head and neck cancer: epidemiology andmolecular biology. Head Neck1998;20:250^65.

32. Gillison ML, KochWM, Capone RB, et al. Evidencefor causal association between human papillomavirusand a subset of head and neck cancers. JNatl CancerInst 2000;92:709^20.

33.DysonN, Howley PM,Mu« nger K, HarlowE.Thehu-man papillomavirus 16 E7 protein is able to bind theretinoblastoma gene product. Science (Washington,DC) 1989;243:934^7.

34. Boyer SN,Wazer DE, BandV. E7 protein of humanpapillomavirus-16 induces degradation of retinoblas-toma protein through the ubiquitin-proteasome path-way. Cancer Res1996;56:4620^4.

35.Werness BC, Levine AJ, Howley PM. Associationof human papilloma virus type 16 and 18 E6 proteinswith p53. Science (Washington, DC) 1990;248:76^9.

36. Scheffner M, Huibregtse JM,Vierstra RD, HowleyPM.The HPV-16 E6 and E6-AP complex functions asa ubiquitin-protein ligase in the ubiquitination of p53.Cell1993;75:495^505.

37. Filatov L, GolubovskayaV, HurtJC, Byrd LL, PhillipsJM, KaufmannWK. Chromosomal instability is corre-lated with telomere erosion and inactivation of G2

checkpoint function in human fibroblasts expressinghuman papillomavirus type 16 E6 oncoprotein. Onco-gene1998;16:1825^38.

38. HasskarlJ, Butz K,Whitaker N, Ullmann A, Durst M,Hoppe-Seyler F. Differential cell cycle response ofnontumorigenic and tumorigenic human papillo-mavirus-positive keratinocytes towards transforminggrowth factor-h1. JMol Med 2000;78:94^101.

39. zur Hausen H. Papillomaviruses causing cancer:evasion from host-cell control in early events in carci-nogenesis. JNatl Cancer Inst 2000;92:690^8.





40. Creek KE, Geslani G, Batova A, Pirisi L. Progressivelossofsensitivity togrowthcontrolby retinoicacidandtransforming growth factor-h at late stages of humanpapillomavirus type16-initiated transformation of hu-mankeratinocytes.AdvExpMedBiol1995;375:117^35.

41. Kang SH,Won K, Chung HW, et al. Genetic integrityof transforming growth factor h (TGF-h) receptors incervical carcinoma cell lines: loss of growth sensitivitybut conserved transcriptional response toTGF-h. IntJCancer1998;77:620^5.

42. MiY-d, Borger D, Fernandes PR, Pirisi L, Creek KE.Loss of transforming growth factor-h (TGF-h) recep-tor type1mediatesTGF-h resistance inhumanpapillo-maviruses type-16 transformed human keratinocytesat late stages of in vitro progression.Virology 2000;270:408^16.

43. ChenT, deVries EG,HollemaH, et al. Structural alter-ations of transforming growth factor-h receptor genesin human cervical carcinoma. Int J Cancer 1999;82:43^51.

44. Nees M, Geoghegan JM, Munson P, et al. Humanpapillomavirus type 16 E6 and E7 proteins inhibitdifferentiation-dependent expression of transforminggrowth factor-h2 in cervical keratinocytes. CancerRes 2000;60:4289^98.

45. Rorke EA, Zhang D, Choo CK, Eckert RL,Jacobberger JW. TGF-h-mediated cell cycle arrest of

HPV16-immortalized human ectocervical cells corre-lates with decreased E6/E7 mRNA and increasedp53 and p21(WAF-1) expression. Exp Cell Res 2000;259:149^53.

46. Lee S, ChoYS, Shim C, et al. Aberrant expression ofSmad4 results in resistance against the growth inhibi-tory effect of transforming growth factor h in the SiHahuman cervical carcinoma cell line. Int JCancer 2001;94:500^7.

47. Spitz MR. Epidemiology and risk factors for headand neck cancer. Semin Oncol1994;21:281^8.

48. Liu X, Han S, Baluda MA, Park NH. HPV16oncogenes E6 and E7 are mutagenic in normalhuman oral keratinocytes. Oncogene 1997;4:2347^52.

49. Ba¤ ezA,Almodo¤ varJI,CantorA,etal.Highfrequencyof HPV16-associated head and neck squamous cellcarcinoma in the Puerto Rican population. Head Neck2004;26:778^84.

50. Fukuchi M, Masuda N, Miyazaki T, et al. DecreasedSmad4 expression in transforming growth factor-hsignaling pathway during progression of esophagealsquamous cell carcinoma. Cancer 2002;737^43.

51.Takebayashi S, OgawaT, Jung KY, et al. Identificationof newminimally lost regions on18q inhead and necksquamous cell carcinoma. Cancer Res 2000;60:3397^403.

52.NagatakeM,TakagiY, OsadaH, et al. Somatic in vivoalterations of the DPC4 gene at 18q21in human lungcancers. Cancer Res1996;56:2718^20.

53.Wang D, Kanuma T, Takama F, et al. Mutation analy-sis of the Smad3 gene in human ovarian cancers. IntJOncol1999;15:949^53.

54.Takakura S, OkamotoA, Saito M, et al. Allelic imbal-ance in chromosome band 18q21and SMAD4 muta-tions in ovarian cancers. Genes Chromosomes Cancer1999;24:264^71.

55. Riggins GJ, Kinzler KW,Vogelstein B,ThiagalingamS. Frequency of Smad gene mutations in human can-cers. Cancer Res1997;57:2578^80.

56. Kawate S,Takenoshita S, Ohwada S, et al. Mutationanalysis of transforming growth factor h type II recep-tor, Smad2, and Smad4 in hepatocellular carcinoma.IntJOncol1999;14:127^31.

57. Latil A, Pesche S,Valri A, Fournier G, Cussenot O,Lidereau R. Expression and mutational analysis of theMADR2/smad2 gene in human prostate cancer. Pros-tate1999;40:225^31.

58.Grau AM, Zhang L, Wang W, et al. Inductionof p21waf expression and growth inhibition bytransforming growth factor h involves the tu-mor suppressor gene DPC4 in human pancre-atic adenocarcinoma cells. Cancer Res 1997;57:3929^34.



2005;11:3191-3197. Clin Cancer Res Adriana Báez, Alan Cantor, Sharon Fonseca, et al. Negative Head and Neck Squamous Cell Carcinoma

−Positive and Human Papillomavirus Type 16−Type 16 Differences in Smad4 Expression in Human Papillomavirus

Updated version

http://clincancerres.aacrjournals.org/content/11/9/3191

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/11/9/3191.full#ref-list-1

This article cites 56 articles, 19 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/11/9/3191.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/11/9/3191To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/11/9/3191.full#ref-list-1

http://clincancerres.aacrjournals.org/content/11/9/3191.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



differencesinsmad4expressioninhumanpapillomavirus …and 1.5 units of taq polymerase. dna was...

Documents