- Email: [email protected]
Interferon, alpha 17 (IFNA17) Ile184Arg polymorphism and cervical cancer risk
Cancer Letters 189 (2003) 183–188 www.elsevier.com/locate/canlet
Interferon, alpha 17 (IFNA17) Ile184Arg polymorphism and cervical cancer risk Jae Weon Kim a,b,*, Ju Won Roh c, Noh Hyun Park a,b, Yong Sang Song a,b, Soon Beom Kang a,b, Hyo Pyo Lee a,b a
Department of Obstetrics and Gynaecology, College of Medicine, Seoul National University, 28 Yungun-Dong, Chongno-Ku, Seoul, 110-744, South Korea b Cancer Research Institute, College of Medicine, Seoul National University, 28 Yungun-Dong, Chongno-Ku, Seoul, 110-744, South Korea c Center for Uterine Cancer, National Cancer Center, Goyang, Gyeonggi, 411-764, South Korea Received 14 May 2002; received in revised form 26 August 2002; accepted 27 August 2002
Abstract The aim of this study was to estimate whether the Ile184Arg polymorphism of the Interferon, alpha 17 (IFNA17) gene might be correlated with the risk and prognosis of cervical squamous cell carcinoma (SCCA). Genomic DNA from 173 patients with invasive cervical SCCA and from 198 non-cancer controls (NCC) was examined for the Ile184Arg polymorphism with a polymerase chain reaction-restriction fragment length polymorphism-based assay. All P values are from two-sided tests. The distribution of IFNA17 Ile184Arg genotype among SCCA cases (Ile/Ile, 21%; Arg/Ile, 57%; and Arg/Arg, 22%) was different significantly from that among NCC (Ile/Ile, 32%; Arg/Ile, 56%; and Arg/Arg, 12%) (P ¼ 0:0345). A significant increase in 184Ile homozygote frequency was found in the SCCA patients compared with the NCC group (odds ratio 2.96; 95% confidence interval 1.53–5.73; P ¼ 0:001). However, differences in clinicopathologic prognostic factors and in the 5 year survival rate were not significant among the subjects of each genotype group (P . 0:05). The results of this study suggest that the 184Ile homozygote of the IFNA17 gene may represent an important susceptibility biomarker for cervical cancer risk. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Interferon, alpha 17; Cervical cancer; Polymorphism
1. Introduction Carcinoma of the cervix is the third most common cancer among women worldwide and in Korea [1,2]. In cervical cancer, human papillomavirus (HPV) has been demonstrated to be the central etiologic agent of carcinogenesis across epidemiologic and virologic studies worldwide [3]. In a recent international study, * Corresponding author. Tel.: 182-2-760-3511; fax: 182-2-7623599. E-mail address: [email protected] (J.W. Kim).
HPV DNA was found in nearly 100% of cervical tumour specimens examined, irrespective of the nationality of the patients from whom the samples were obtained [4]. However, because HPV infections are so widespread in the general population and because HPV-immortalized cell lines are generally not tumorigenic, other factors in addition to HPV infection, such as viral load, viral persistence, lifestyle and host factors, also play an important additional role in cervical cancer development [5–8]. In particular, several lines of evidence have suggested that the host’s immune responses are important in controlling both
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(02)00548-7
184
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
In this article, we report results from a hospitalbased case-control study examining the association of IFNA17 Ile184Arg polymorphisms with the risk and prognosis of cervical cancer. 2. Materials and methods 2.1. Patients and specimens
Fig. 1. Analysis of IFNA17 polymorphism by PCR-RFLP. A 432 bp PCR fragment amplified and digested with SspI. A 123 bp ladder was used as a size marker (SM). 184Arg homozygotes without an SspI site gave no further fragment (lane 3 and 6). 184Arg homozygotes with an SspI site showed one band with the size of 380 bp (lane 4 and 5). The smaller band (52 bp) migrated out of the gel. Ile/ Arg heterozygotes showed two bands (lane 1, 2 and 7).
HPV infections and HPV-associated cervical lesions [9–11]. A decrease in the host’s immune response is prerequisite for a prolonged HPV infection and for the progression of HPV-associated cervical lesions [12]. Interferon (IFN) is major contributor to the first line of these antiviral defences and it exerts many other important effects on cells in addition to inhibiting virus replication. IFNs belong to the network of cytokines that become actively engaged in host defence during virus infection [13]. Interferon, alpha (IFNA) constitutes a complex gene family with 14 genes clustered on the short arm of chromosome 9. At the IFNA17 locus it has been shown that the 184Ile and 184Arg variants are allelic since they segregated in families and showed a close agreement with the Hardy–Weinberg equilibrium [14–16]. Individuals homozygous for the 184Ile allele contain the SspI site which is lost with 184Arg polymorphism [16,17]. The IFNA proteins appear to be functionally equivalent but may differ in specific activity [18]. Whether the polymorphic IFNA proteins differ in their specific activity has not yet been determined definitely. However, among nasopharyngeal cancer patients, there was a significant increase in the frequency of the 184Ile allele relative to the 184Arg allele [15]. Our hypothesis is that some individuals with a specific IFNA17 genotype might suffer a disadvantage in fighting against HPV infection.
IFNA17 polymorphism was examined in 176 women with squamous cell carcinoma (SCCA) of the uterine cervix and 201 non-cancer controls (NCCs). The patients with cervical cancer were treated at the Department of Obstetrics and Gynaecology, Seoul National University Hospital during the period from March 1999 to April 2001 and were consecutively recruited in our study. None of the individuals in the control group exhibited any evidence of cervical lesions when tested by Papanicolaou cervical smears. The age distribution was uniformly matched in both groups, and all women were born in Korea. Peripheral blood samples were taken with informed consent. Research protocol used was approved by the institutional review board of Seoul National University Hospital. Clinicopathologic parameters (Clinical stage, invasion depth, lymph node involvement, parametrial involvement, tumour size, and 5 year survival rate) were obtained by retrospective review of medical records. Survival was defined as the time between primary treatment and patients’ death due to the tumour. 2.2. Determination of IFNA17 Ile184Arg genotype Genomic DNA was extracted from ethylenediaminetetra-acetic acid (EDTA) preserved peripheral venous blood using the standard proteinase-K digestion and phenol/chloroform method. IFNA17 genotypes were determined with a polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP)-based assay. The primers, based on the published sequence of human complementary DNA of the interferon, alpha 17 gene (GenBank accession no. NM_021268), were as follows: F ¼ 5 0 -CAATCAGGATCATTGCCATG-3 0 , R ¼ 5 0 GCTTTGGACTTCCCCAGG-3 0 [17,19]. The DNA was amplified in a Perkin-Elmer GeneAmp System 9600 (The Perkin-Elmer Corp., Norwalk City, CT)
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
185
Table 1 Genotype and allele frequencies of IFNA17 Ile184Arg in cervical cancer patients and non-cancer controls
Observed Patients with cervical cancer Control subjects Expected proportions Patients with cervical cancer Control subjects OR c 95% confidence intervals P value a b c
No.
Arg/Arg
Arg/Ile
Ile/Ile
184Ile allele
173 198
36 (0.21) b 63 (0.32)
98 (0.57) 112 (0.56)
39 (0.22) 23 (0.12)
0.509 0.399
42 71 1.0 (Reference)
86 95 1.53 0.93, 2.50 0.089
45 32 2.96 1.53, 5.73 0.001
HW a
0.44 0.19
x 2 analysis used for Hardy–Weinberg equilibrium. The relative frequency of each genotype is indicated in parentheses. Odds ratio relative to Arg/Arg genotype.
according to the manufacturer’s protocol. These reactions were carried out in a 20 ml mixture containing 10 mM Tris–HCl (pH 9.0), 40 mM KCl, 1.5 mM MgCl2, all four deoxynucleoside triphosphates (each at 0.2 mM), and 1 U of Taq polymerase (Bioneer Corp., Korea). The reaction mixtures were heated to 94 8C for 5 min followed by 30 cycles of 94 8C for 1 min, 59 8C for 1 min, and 72 8C for 1 min. Finally, the reactions were extended 10 min at 72 8C. The PCR products (432 base pairs [bp]) were digested with SspI (New England Biolabs Inc., Beverly, MA) and separated by 2% agarose gel electrophoresis. When a SspI restriction site was present, the fragment of 432 bp was digested into two lengths of 52 and 380 bp. Individuals with Ile/Ile had the 52 and 380 bp fragment, heterozygous individuals had three fragments; and individuals with Arg/Arg retained the larger parent bands (432 bp) (Fig. 1). The IFNA17 genotypes were identified for all control subjects and case patients except three from each group, allowing the analysis of data from 173 case patients and 198 control subjects. To ensure assay quality control 30 randomly selected samples were reassayed to evaluate the consistency of results from the first and second assays. The DNAs from patients whose IFNA17 sequence had been determined by direct sequencing were used as positive controls. 2.3. Statistical analysis Chi-square statistics were used to evaluate the
distribution of IFNA17 allele types and genotypes among the cancer patients and control subjects. Odds ratios (ORs) and 95% confidence intervals (CIs), obtained from unconditional logistic regression, were used to measure the strength of the association between IFNA17 polymorphism and cervical cancer risk. The patients were censored at the time of last follow-up. Survival analysis was done by the Kaplan–Meier method with a log-rank test. All P values are from two-sided tests. Statistical calculations were performed using the SPSS software package (version 9.0 for Windows 98; SPSS Inc, Chicago, IL).
3. Results The genotype distribution results are presented in Table 1. Expected binomial Hardy–Weinberg proportions are described and in two groups the distribution of alleles fit the Hardy–Weinberg equilibrium. The proportions of individuals homozygous for the 184Arg allele, heterozygous, and homozygous for the 184Ile allele were 32%, 56%, and 12% among the control group, and 21%, 57%, and 22% in the women with cervical cancer (x2 ¼ 10:79, P ¼ 0:0345, d:f: ¼ 2), respectively. One clear difference was observed: cancer patients were more likely than the control subjects to carry the 184Ile allele (0.509 vs. 0.399, P ¼ 0:0041). Furthermore, compared with women with the Arg/Arg genotype,
186
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
women who had either the Arg/Ile or Ile/Ile genotype had an elevated risk of cervical cancer (OR 1.78; 95% CI 1.11–2.85; P ¼ 0:017) and this risk was elevated to an even greater extent among women who were homozygous for the 184Ile allele (OR 2.96; 95% CI 1.53–5.73, P ¼ 0:001). However, differences in clinicopathologic factors (data not shown) and the 5 year survival rate were not significant among the patients of each genotype group (P . 0:05).
4. Discussion Despite substantial advances in the understanding of the molecular events involved in the development of many epithelial cancers, knowledge on the molecular pathogenesis of cervical cancer is far from complete. Although various oncogenic HPV types have been found in almost all of the cervical carcinomas, only a small fraction of women harbouring oncogenic HPV in their lower genital tract will actually develop cervical cancer [7]. This fact means that the majority of women infected with this common virus are capable of controlling the virus and hence never develop cancer. Although HPV infection is the major known risk factor for the development of cervical cancer, it alone is not sufficient [5]. Additional determinants of these events that have been raised are exogenous and lifestyle factors (such as parity, oral contraceptive use, and cigarette smoking), viral factors (such as viral load, viral persistence, and viral variants), and host factors, particularly the host immune response to HPV infection [5]. It is well known that cancer has a complex etiology, and that genetic susceptibility factors, environmental risk factors, as well as the interplay of these two sets of causative agents, all work to play a defined role in its development [20]. If we apply this generally accepted theory to the case of cervical cancer, the environmental risk factors might be HPV infection, while host immune response to HPV infection might come under the heading of genetic susceptibility factors. Although the contribution of a single gene may account for only a portion of the susceptibility to the disorder, determining the genetic causes of cancers nevertheless has immense public health importance, ranging from prevention to earlier detection and treatment of disease [20].
In this study, we genotyped 173 Korean cervical cancer patients and 198 ethnically identical controls and found that the IFNA17 184Ile allele was associated with an increased risk of cervical cancer. To our knowledge this is the first report of such a finding and suggests that IFNA17 polymorphism may be an important susceptibility biomarker for the risk of cervical cancer. In the association between the Ile/ Ile genotype and cervical cancer, the causative mechanism is difficult to assess. Association between genetic markers and cancer may in principle depend on two different mechanisms. The first is the direct functional involvement of the polymorphic gene product [15]. Previous studies have shown that genetic factors substantially influence production of cytokines and that the innate anti-inflammatory cytokine profile may contribute to the disease process [21– 23]. Polymorphisms in the regions of cytokine genes that affect transcription may account for some of the inter-individual variation in cytokine production [24]. Danis and his colleagues have reported the results on the relationship between IL-1 production and an 86 bp variable repeat polymorphism in intron two of the IL1Ra gene. A less common allele of this polymorphism (allele two) was associated both with increased production of IL-1Ra protein and reduced production of IL-1 alpha protein by monocytes. The IFNA proteins appear to be functionally equivalent but may differ in specific activity [18]. Whether the polymorphism of the IFNA gene results in the production of proteins differing in their specific activity or amount has not yet been determined. However, our findings of an association between cervical cancer and the Ile/Ile allele suggest that this polymorphism may be functionally important. Based on these findings we now suggest that an innate antiviral IFNA profile may contribute to the continuation of HPV-infection and even the development of cervical cancer. While the direct functional involvement of IFNA protein variants cannot be excluded, such an association could also be due to linkage disequilibrium with an adjacent gene influencing cell-cycle control such as p16, which is located close to the IFN gene cluster [25,26]. It has been shown that genetic alteration of the IFN gene cluster and p16 deletion sometimes concurred. The present study is one of the hospital-based retrospective case-control studies. A major concern of such
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
studies is proper case and control selection. Ideally the selection of cases in a case-control study is based on all individuals within a population who develop the disease, or on a truly representative sample of them [27]. However, identifying either a random sample from the general population or alternatively the source population for the case subjects presenting at a particular hospital may be difficult. Selection of a representative control group is another of the difficult tasks in many case-control studies because factors related to inclusion, such as social class, education and lifestyle, may also be related to the exposure variables under consideration [27]. For example, in the case of cervical cancer, if controls that had undergone regular cervical cytologic screening were more likely to participate in a study, subsequent comparisons of cases and controls would be biased. This is because the majority of squamous cell carcinoma of the uterine cervix progress through a series of well-defined preinvasive lesions and hence, the disease can be easily prevented by regular cervical cytologic screening during this usually lengthy process [28]. Therefore, in casecontrol studies addressing susceptibility to cervical cancer, it is necessary that all other information including cervical cytologic screening history should be analyzed together with genetic polymorphism data in order to produce more precise and bias-free results. Unfortunately in this study the data regarding cervical cytologic screening history was unavailable. All of the subjects in this study belong to a single ethnic group; Korean. Therefore, the potentially confounding effect of ethnicity was not a major concern in this study. The frequency of control subjects homozygous for the 184Ile allele was lower than expected under the Hardy–Weinberg equilibrium, but not to a statistically significant degree (P ¼ 0:19). The prevalence of the IFNA17 Ile/Ile genotypes in our control group was 12%, substantially lower than observed (46–80%) in a previous study of Swedes, Finns, Indians, and African blacks but in close agreement with that found in Chinese population [16]. These data are discrepant with the higher incidence of cervical cancer among Korean women than Swedish and Finnish women. But given the shortage of systematically organized cervical cytologic screening programs in Korea, the potential role of IFNA17 polymorphism in cervical cancer cannot be excluded.
187
In summary, this study found that women homozygous for the IFNA17 184Ile allele had a statistically significantly increased risk for squamous cell carcinoma of the cervix. Acknowledgements This work was supported by grants from the Postdoctoral Fellowship Program of Korea Science & Engineering Foundation to J.W.K. and Seoul National University Hospital (04-2000-022), Seoul, Korea. References [1] D.M. Parkin, P. Pisani, J. Ferlay, Estimates of the worldwide incidence of 25 major cancers in 1990, Int. J. Cancer 80 (1999) 827–841. [2] Central cancer registry center in Korea, Annual report of the central cancer registry in Korea (1998.1 , 1998.12), Ministry of Health and Welfare, Seoul, (2000). [3] IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Human Papillomavirus, WHO, Lyon, Vol. 64 (1995). [4] J.M. Walboomers, M.V. Jacobs, M.M. Manos, F.X. Bosch, J.A. Kummer, K.V. Shah, P.J. Snijders, J. Peto, C.J. Meijer, N. Munoz, Human papillomavirus is a necessary cause of invasive cervical cancer worldwide, J. Pathol. 189 (1999) 12–19. [5] A. Hildesheim, Human papillomavirus variants: implications for natural history studies and vaccine development efforts, J. Natl. Cancer Inst. 89 (1997) 752–753. [6] F.X. Bosch, N. Munoz, S. de Sanjose, I. Izarzugaza, M. Gili, P. Viladiu, M.J. Tormo, P. Moreo, N. Ascunce, L.C. Gonzalez, Risk factors for cervical cancer in Colombia and Spain, Int. J. Cancer 52 (1992) 750–758. [7] H. Haverkos, M. Rohrer, W. Pickworth, The cause of invasive cervical cancer could be multifactorial, Biomed. Pharmacother. 54 (2000) 54–59. [8] N. Denko, C. Schindler, A. Koong, K. Laderoute, C. Green, A. Giaccia, Epigenetic regulation of gene expression in cervical cancer cells by the tumour microenvironment, Clin. Cancer Res. 6 (2000) 480–487. [9] T.C. Wu, Immunology of the human papillomavirus in relation to cancer, Curr. Opin. Immunol. 6 (1994) 746–754. [10] I.H. Frazer, Immunology of papillomavirus infection, Curr. Opin. Immunol. 8 (1996) 484–491. [11] M. Murakami, K.J. Gurski, M.A. Steller, Human papillomavirus vaccines for cervical cancer, J. Immunol. 22 (1999) 212– 218. [12] D. Labeit, S. Labeit, M. Berger, H. Gallati, R. Rosenberg, K. Friese, Interferon-alpha controls HPV infection in cervix epithelium, Zentralbl. Gynakol. 117 (1995) 566–577.
188
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
[13] E.D. Maeyer, D.M. Maeyer-Guignard, Interferons, in: A.W. Thomson (Ed.), The Cytokine Handbook, Academic Press, Cambridge, 1994, pp. 265–288. [14] I. Golovleva, M. Kandefer-Szerszen, L. Beckman, E. Lundgren, Polymorphism in the interferon-alpha gene family, Am. J. Hum. Genet. 59 (1996) 570–578. [15] I. Golovleva, R. Birgander, A. Sjalander, E. Lundgren, L. Beckman, Interferon-alpha and p53 alleles involved in nasopharyngeal carcinoma, Carcinogenesis 18 (1997) 645–647. [16] I. Golovleva, N. Saha, L. Beckman, Ethnic differences in interferon-alpha allele frequencies, Hum. Hered. 47 (1997) 185–188. [17] J. Mizoguchi, P.M. Pitha, N.B. Raj, Efficient expression in Escherichia coli of two species of human interferon-alpha and their hybrid molecules, DNA 4 (1985) 221–232. [18] K. Henco, J. Brosius, A. Fujisawa, J.I. Fujisawa, J.R. Haynes, J. Hochstadt, T. Kovacic, M. Pasek, A. Schambock, J. Schmid, Structural relationship of human interferon alpha genes and pseudogenes, J. Mol. Biol. 185 (1985) 227–260. [19] O.I. Olopade, S.K. Bohlander, H. Pomykala, E. Maltepe, E. Van Melle, M.M. Le Beau, M.O. Diaz, Mapping of the shortest region of overlap of deletions of the short arm of chromosome 9 associated with human neoplasia, Genomics 14 (1992) 437–443. [20] D.J. Schaid, K. Buetow, D.E. Weeks, E. Wijsman, S.W. Guo, J. Ott, Discovery of cancer susceptibility genes: study designs, analytic approaches, and trends in technology, in: Innovative study designs and analytic approaches to the genetic epidemiology of cancer, J. Natl. Cancer Inst. 26 (1999) 1–16. [21] R.G. Westendorp, J.A. Langermans, T.W. Huizinga, A.H. Elouali, C.L. Verweij, D.I. Boomsma, J.P. Vandenbroucke,
[22]
[23]
[24]
[25]
[26]
[27]
[28]
J.P. Vandenbrouke, Genetic influence on cytokine production and fatal meningococcal disease, Lancet 349 (1997) 170–173. J.C. Knight, D. Kwiatkowski, Inherited variability of tumour necrosis factor production and susceptibility to infectious disease, Proc. Assoc. Am. Physicians 111 (1999) 290–298. D. Heresbach, M. Alizadeh, A. Dabadie, N. Le Berre, J.F. Colombel, J. Yaouanq, J.F. Bretagne, G. Semana, Significance of interleukin-1beta and interleukin-1 receptor antagonist genetic polymorphism in inflammatory bowel diseases, Am. J. Gastroenterol. 92 (1997) 1164–1169. V.A. Danis, M. Millington, V.J. Hyland, D. Grennan, Cytokine production by normal human monocytes: inter-subject variation and relationship to an IL-1 receptor antagonist (IL1Ra) gene polymorphism, Clin. Exp. Immunol. 99 (1995) 303–310. G. Sternik, M.G. Pittis, M. Gutierrez, R.A. Diez, L. Sen, In acute lymphoblastic leukemia deletion of the tumour suppressor gene P16 is associated with abnormal interferon genes, Medicina (B Aires) 58 (1998) 463–468. R.C. Aguiar, H. Sill, J.M. Goldman, N.C. Cross, The commonly deleted region at 9p21-22 in lymphoblastic leukemias spans at least 400 kb and includes p16 but not p15 or the IFN gene cluster, Leukemia 11 (1997) 233–238. S. Wacholder, D.T. Silverman, J.K. McLaughlin, J.S. Mandel, Selection of controls in case-control studies III; Design options, Am. J. Epidemiol. 135 (1992) 1042–1050. J.W. Kim, S.B. Kang, D.K. Seo, S.H. Choi, C.M. Lee, Y.B. Kim, N.H. Park, Y.S. Song, S.B. Kang, H.P. Lee, Cytologic screening history of 249 patients with cervical cancer, J. Korean Cancer Assoc. 30 (1998) 1198–1206.
Interferon, alpha 17 (IFNA17) Ile184Arg polymorphism and cervical cancer risk Jae Weon Kim a,b,*, Ju Won Roh c, Noh Hyun Park a,b, Yong Sang Song a,b, Soon Beom Kang a,b, Hyo Pyo Lee a,b a
Department of Obstetrics and Gynaecology, College of Medicine, Seoul National University, 28 Yungun-Dong, Chongno-Ku, Seoul, 110-744, South Korea b Cancer Research Institute, College of Medicine, Seoul National University, 28 Yungun-Dong, Chongno-Ku, Seoul, 110-744, South Korea c Center for Uterine Cancer, National Cancer Center, Goyang, Gyeonggi, 411-764, South Korea Received 14 May 2002; received in revised form 26 August 2002; accepted 27 August 2002
Abstract The aim of this study was to estimate whether the Ile184Arg polymorphism of the Interferon, alpha 17 (IFNA17) gene might be correlated with the risk and prognosis of cervical squamous cell carcinoma (SCCA). Genomic DNA from 173 patients with invasive cervical SCCA and from 198 non-cancer controls (NCC) was examined for the Ile184Arg polymorphism with a polymerase chain reaction-restriction fragment length polymorphism-based assay. All P values are from two-sided tests. The distribution of IFNA17 Ile184Arg genotype among SCCA cases (Ile/Ile, 21%; Arg/Ile, 57%; and Arg/Arg, 22%) was different significantly from that among NCC (Ile/Ile, 32%; Arg/Ile, 56%; and Arg/Arg, 12%) (P ¼ 0:0345). A significant increase in 184Ile homozygote frequency was found in the SCCA patients compared with the NCC group (odds ratio 2.96; 95% confidence interval 1.53–5.73; P ¼ 0:001). However, differences in clinicopathologic prognostic factors and in the 5 year survival rate were not significant among the subjects of each genotype group (P . 0:05). The results of this study suggest that the 184Ile homozygote of the IFNA17 gene may represent an important susceptibility biomarker for cervical cancer risk. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Interferon, alpha 17; Cervical cancer; Polymorphism
1. Introduction Carcinoma of the cervix is the third most common cancer among women worldwide and in Korea [1,2]. In cervical cancer, human papillomavirus (HPV) has been demonstrated to be the central etiologic agent of carcinogenesis across epidemiologic and virologic studies worldwide [3]. In a recent international study, * Corresponding author. Tel.: 182-2-760-3511; fax: 182-2-7623599. E-mail address: [email protected] (J.W. Kim).
HPV DNA was found in nearly 100% of cervical tumour specimens examined, irrespective of the nationality of the patients from whom the samples were obtained [4]. However, because HPV infections are so widespread in the general population and because HPV-immortalized cell lines are generally not tumorigenic, other factors in addition to HPV infection, such as viral load, viral persistence, lifestyle and host factors, also play an important additional role in cervical cancer development [5–8]. In particular, several lines of evidence have suggested that the host’s immune responses are important in controlling both
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(02)00548-7
184
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
In this article, we report results from a hospitalbased case-control study examining the association of IFNA17 Ile184Arg polymorphisms with the risk and prognosis of cervical cancer. 2. Materials and methods 2.1. Patients and specimens
Fig. 1. Analysis of IFNA17 polymorphism by PCR-RFLP. A 432 bp PCR fragment amplified and digested with SspI. A 123 bp ladder was used as a size marker (SM). 184Arg homozygotes without an SspI site gave no further fragment (lane 3 and 6). 184Arg homozygotes with an SspI site showed one band with the size of 380 bp (lane 4 and 5). The smaller band (52 bp) migrated out of the gel. Ile/ Arg heterozygotes showed two bands (lane 1, 2 and 7).
HPV infections and HPV-associated cervical lesions [9–11]. A decrease in the host’s immune response is prerequisite for a prolonged HPV infection and for the progression of HPV-associated cervical lesions [12]. Interferon (IFN) is major contributor to the first line of these antiviral defences and it exerts many other important effects on cells in addition to inhibiting virus replication. IFNs belong to the network of cytokines that become actively engaged in host defence during virus infection [13]. Interferon, alpha (IFNA) constitutes a complex gene family with 14 genes clustered on the short arm of chromosome 9. At the IFNA17 locus it has been shown that the 184Ile and 184Arg variants are allelic since they segregated in families and showed a close agreement with the Hardy–Weinberg equilibrium [14–16]. Individuals homozygous for the 184Ile allele contain the SspI site which is lost with 184Arg polymorphism [16,17]. The IFNA proteins appear to be functionally equivalent but may differ in specific activity [18]. Whether the polymorphic IFNA proteins differ in their specific activity has not yet been determined definitely. However, among nasopharyngeal cancer patients, there was a significant increase in the frequency of the 184Ile allele relative to the 184Arg allele [15]. Our hypothesis is that some individuals with a specific IFNA17 genotype might suffer a disadvantage in fighting against HPV infection.
IFNA17 polymorphism was examined in 176 women with squamous cell carcinoma (SCCA) of the uterine cervix and 201 non-cancer controls (NCCs). The patients with cervical cancer were treated at the Department of Obstetrics and Gynaecology, Seoul National University Hospital during the period from March 1999 to April 2001 and were consecutively recruited in our study. None of the individuals in the control group exhibited any evidence of cervical lesions when tested by Papanicolaou cervical smears. The age distribution was uniformly matched in both groups, and all women were born in Korea. Peripheral blood samples were taken with informed consent. Research protocol used was approved by the institutional review board of Seoul National University Hospital. Clinicopathologic parameters (Clinical stage, invasion depth, lymph node involvement, parametrial involvement, tumour size, and 5 year survival rate) were obtained by retrospective review of medical records. Survival was defined as the time between primary treatment and patients’ death due to the tumour. 2.2. Determination of IFNA17 Ile184Arg genotype Genomic DNA was extracted from ethylenediaminetetra-acetic acid (EDTA) preserved peripheral venous blood using the standard proteinase-K digestion and phenol/chloroform method. IFNA17 genotypes were determined with a polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP)-based assay. The primers, based on the published sequence of human complementary DNA of the interferon, alpha 17 gene (GenBank accession no. NM_021268), were as follows: F ¼ 5 0 -CAATCAGGATCATTGCCATG-3 0 , R ¼ 5 0 GCTTTGGACTTCCCCAGG-3 0 [17,19]. The DNA was amplified in a Perkin-Elmer GeneAmp System 9600 (The Perkin-Elmer Corp., Norwalk City, CT)
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
185
Table 1 Genotype and allele frequencies of IFNA17 Ile184Arg in cervical cancer patients and non-cancer controls
Observed Patients with cervical cancer Control subjects Expected proportions Patients with cervical cancer Control subjects OR c 95% confidence intervals P value a b c
No.
Arg/Arg
Arg/Ile
Ile/Ile
184Ile allele
173 198
36 (0.21) b 63 (0.32)
98 (0.57) 112 (0.56)
39 (0.22) 23 (0.12)
0.509 0.399
42 71 1.0 (Reference)
86 95 1.53 0.93, 2.50 0.089
45 32 2.96 1.53, 5.73 0.001
HW a
0.44 0.19
x 2 analysis used for Hardy–Weinberg equilibrium. The relative frequency of each genotype is indicated in parentheses. Odds ratio relative to Arg/Arg genotype.
according to the manufacturer’s protocol. These reactions were carried out in a 20 ml mixture containing 10 mM Tris–HCl (pH 9.0), 40 mM KCl, 1.5 mM MgCl2, all four deoxynucleoside triphosphates (each at 0.2 mM), and 1 U of Taq polymerase (Bioneer Corp., Korea). The reaction mixtures were heated to 94 8C for 5 min followed by 30 cycles of 94 8C for 1 min, 59 8C for 1 min, and 72 8C for 1 min. Finally, the reactions were extended 10 min at 72 8C. The PCR products (432 base pairs [bp]) were digested with SspI (New England Biolabs Inc., Beverly, MA) and separated by 2% agarose gel electrophoresis. When a SspI restriction site was present, the fragment of 432 bp was digested into two lengths of 52 and 380 bp. Individuals with Ile/Ile had the 52 and 380 bp fragment, heterozygous individuals had three fragments; and individuals with Arg/Arg retained the larger parent bands (432 bp) (Fig. 1). The IFNA17 genotypes were identified for all control subjects and case patients except three from each group, allowing the analysis of data from 173 case patients and 198 control subjects. To ensure assay quality control 30 randomly selected samples were reassayed to evaluate the consistency of results from the first and second assays. The DNAs from patients whose IFNA17 sequence had been determined by direct sequencing were used as positive controls. 2.3. Statistical analysis Chi-square statistics were used to evaluate the
distribution of IFNA17 allele types and genotypes among the cancer patients and control subjects. Odds ratios (ORs) and 95% confidence intervals (CIs), obtained from unconditional logistic regression, were used to measure the strength of the association between IFNA17 polymorphism and cervical cancer risk. The patients were censored at the time of last follow-up. Survival analysis was done by the Kaplan–Meier method with a log-rank test. All P values are from two-sided tests. Statistical calculations were performed using the SPSS software package (version 9.0 for Windows 98; SPSS Inc, Chicago, IL).
3. Results The genotype distribution results are presented in Table 1. Expected binomial Hardy–Weinberg proportions are described and in two groups the distribution of alleles fit the Hardy–Weinberg equilibrium. The proportions of individuals homozygous for the 184Arg allele, heterozygous, and homozygous for the 184Ile allele were 32%, 56%, and 12% among the control group, and 21%, 57%, and 22% in the women with cervical cancer (x2 ¼ 10:79, P ¼ 0:0345, d:f: ¼ 2), respectively. One clear difference was observed: cancer patients were more likely than the control subjects to carry the 184Ile allele (0.509 vs. 0.399, P ¼ 0:0041). Furthermore, compared with women with the Arg/Arg genotype,
186
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
women who had either the Arg/Ile or Ile/Ile genotype had an elevated risk of cervical cancer (OR 1.78; 95% CI 1.11–2.85; P ¼ 0:017) and this risk was elevated to an even greater extent among women who were homozygous for the 184Ile allele (OR 2.96; 95% CI 1.53–5.73, P ¼ 0:001). However, differences in clinicopathologic factors (data not shown) and the 5 year survival rate were not significant among the patients of each genotype group (P . 0:05).
4. Discussion Despite substantial advances in the understanding of the molecular events involved in the development of many epithelial cancers, knowledge on the molecular pathogenesis of cervical cancer is far from complete. Although various oncogenic HPV types have been found in almost all of the cervical carcinomas, only a small fraction of women harbouring oncogenic HPV in their lower genital tract will actually develop cervical cancer [7]. This fact means that the majority of women infected with this common virus are capable of controlling the virus and hence never develop cancer. Although HPV infection is the major known risk factor for the development of cervical cancer, it alone is not sufficient [5]. Additional determinants of these events that have been raised are exogenous and lifestyle factors (such as parity, oral contraceptive use, and cigarette smoking), viral factors (such as viral load, viral persistence, and viral variants), and host factors, particularly the host immune response to HPV infection [5]. It is well known that cancer has a complex etiology, and that genetic susceptibility factors, environmental risk factors, as well as the interplay of these two sets of causative agents, all work to play a defined role in its development [20]. If we apply this generally accepted theory to the case of cervical cancer, the environmental risk factors might be HPV infection, while host immune response to HPV infection might come under the heading of genetic susceptibility factors. Although the contribution of a single gene may account for only a portion of the susceptibility to the disorder, determining the genetic causes of cancers nevertheless has immense public health importance, ranging from prevention to earlier detection and treatment of disease [20].
In this study, we genotyped 173 Korean cervical cancer patients and 198 ethnically identical controls and found that the IFNA17 184Ile allele was associated with an increased risk of cervical cancer. To our knowledge this is the first report of such a finding and suggests that IFNA17 polymorphism may be an important susceptibility biomarker for the risk of cervical cancer. In the association between the Ile/ Ile genotype and cervical cancer, the causative mechanism is difficult to assess. Association between genetic markers and cancer may in principle depend on two different mechanisms. The first is the direct functional involvement of the polymorphic gene product [15]. Previous studies have shown that genetic factors substantially influence production of cytokines and that the innate anti-inflammatory cytokine profile may contribute to the disease process [21– 23]. Polymorphisms in the regions of cytokine genes that affect transcription may account for some of the inter-individual variation in cytokine production [24]. Danis and his colleagues have reported the results on the relationship between IL-1 production and an 86 bp variable repeat polymorphism in intron two of the IL1Ra gene. A less common allele of this polymorphism (allele two) was associated both with increased production of IL-1Ra protein and reduced production of IL-1 alpha protein by monocytes. The IFNA proteins appear to be functionally equivalent but may differ in specific activity [18]. Whether the polymorphism of the IFNA gene results in the production of proteins differing in their specific activity or amount has not yet been determined. However, our findings of an association between cervical cancer and the Ile/Ile allele suggest that this polymorphism may be functionally important. Based on these findings we now suggest that an innate antiviral IFNA profile may contribute to the continuation of HPV-infection and even the development of cervical cancer. While the direct functional involvement of IFNA protein variants cannot be excluded, such an association could also be due to linkage disequilibrium with an adjacent gene influencing cell-cycle control such as p16, which is located close to the IFN gene cluster [25,26]. It has been shown that genetic alteration of the IFN gene cluster and p16 deletion sometimes concurred. The present study is one of the hospital-based retrospective case-control studies. A major concern of such
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
studies is proper case and control selection. Ideally the selection of cases in a case-control study is based on all individuals within a population who develop the disease, or on a truly representative sample of them [27]. However, identifying either a random sample from the general population or alternatively the source population for the case subjects presenting at a particular hospital may be difficult. Selection of a representative control group is another of the difficult tasks in many case-control studies because factors related to inclusion, such as social class, education and lifestyle, may also be related to the exposure variables under consideration [27]. For example, in the case of cervical cancer, if controls that had undergone regular cervical cytologic screening were more likely to participate in a study, subsequent comparisons of cases and controls would be biased. This is because the majority of squamous cell carcinoma of the uterine cervix progress through a series of well-defined preinvasive lesions and hence, the disease can be easily prevented by regular cervical cytologic screening during this usually lengthy process [28]. Therefore, in casecontrol studies addressing susceptibility to cervical cancer, it is necessary that all other information including cervical cytologic screening history should be analyzed together with genetic polymorphism data in order to produce more precise and bias-free results. Unfortunately in this study the data regarding cervical cytologic screening history was unavailable. All of the subjects in this study belong to a single ethnic group; Korean. Therefore, the potentially confounding effect of ethnicity was not a major concern in this study. The frequency of control subjects homozygous for the 184Ile allele was lower than expected under the Hardy–Weinberg equilibrium, but not to a statistically significant degree (P ¼ 0:19). The prevalence of the IFNA17 Ile/Ile genotypes in our control group was 12%, substantially lower than observed (46–80%) in a previous study of Swedes, Finns, Indians, and African blacks but in close agreement with that found in Chinese population [16]. These data are discrepant with the higher incidence of cervical cancer among Korean women than Swedish and Finnish women. But given the shortage of systematically organized cervical cytologic screening programs in Korea, the potential role of IFNA17 polymorphism in cervical cancer cannot be excluded.
187
In summary, this study found that women homozygous for the IFNA17 184Ile allele had a statistically significantly increased risk for squamous cell carcinoma of the cervix. Acknowledgements This work was supported by grants from the Postdoctoral Fellowship Program of Korea Science & Engineering Foundation to J.W.K. and Seoul National University Hospital (04-2000-022), Seoul, Korea. References [1] D.M. Parkin, P. Pisani, J. Ferlay, Estimates of the worldwide incidence of 25 major cancers in 1990, Int. J. Cancer 80 (1999) 827–841. [2] Central cancer registry center in Korea, Annual report of the central cancer registry in Korea (1998.1 , 1998.12), Ministry of Health and Welfare, Seoul, (2000). [3] IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Human Papillomavirus, WHO, Lyon, Vol. 64 (1995). [4] J.M. Walboomers, M.V. Jacobs, M.M. Manos, F.X. Bosch, J.A. Kummer, K.V. Shah, P.J. Snijders, J. Peto, C.J. Meijer, N. Munoz, Human papillomavirus is a necessary cause of invasive cervical cancer worldwide, J. Pathol. 189 (1999) 12–19. [5] A. Hildesheim, Human papillomavirus variants: implications for natural history studies and vaccine development efforts, J. Natl. Cancer Inst. 89 (1997) 752–753. [6] F.X. Bosch, N. Munoz, S. de Sanjose, I. Izarzugaza, M. Gili, P. Viladiu, M.J. Tormo, P. Moreo, N. Ascunce, L.C. Gonzalez, Risk factors for cervical cancer in Colombia and Spain, Int. J. Cancer 52 (1992) 750–758. [7] H. Haverkos, M. Rohrer, W. Pickworth, The cause of invasive cervical cancer could be multifactorial, Biomed. Pharmacother. 54 (2000) 54–59. [8] N. Denko, C. Schindler, A. Koong, K. Laderoute, C. Green, A. Giaccia, Epigenetic regulation of gene expression in cervical cancer cells by the tumour microenvironment, Clin. Cancer Res. 6 (2000) 480–487. [9] T.C. Wu, Immunology of the human papillomavirus in relation to cancer, Curr. Opin. Immunol. 6 (1994) 746–754. [10] I.H. Frazer, Immunology of papillomavirus infection, Curr. Opin. Immunol. 8 (1996) 484–491. [11] M. Murakami, K.J. Gurski, M.A. Steller, Human papillomavirus vaccines for cervical cancer, J. Immunol. 22 (1999) 212– 218. [12] D. Labeit, S. Labeit, M. Berger, H. Gallati, R. Rosenberg, K. Friese, Interferon-alpha controls HPV infection in cervix epithelium, Zentralbl. Gynakol. 117 (1995) 566–577.
188
J.W. Kim et al. / Cancer Letters 189 (2003) 183–188
[13] E.D. Maeyer, D.M. Maeyer-Guignard, Interferons, in: A.W. Thomson (Ed.), The Cytokine Handbook, Academic Press, Cambridge, 1994, pp. 265–288. [14] I. Golovleva, M. Kandefer-Szerszen, L. Beckman, E. Lundgren, Polymorphism in the interferon-alpha gene family, Am. J. Hum. Genet. 59 (1996) 570–578. [15] I. Golovleva, R. Birgander, A. Sjalander, E. Lundgren, L. Beckman, Interferon-alpha and p53 alleles involved in nasopharyngeal carcinoma, Carcinogenesis 18 (1997) 645–647. [16] I. Golovleva, N. Saha, L. Beckman, Ethnic differences in interferon-alpha allele frequencies, Hum. Hered. 47 (1997) 185–188. [17] J. Mizoguchi, P.M. Pitha, N.B. Raj, Efficient expression in Escherichia coli of two species of human interferon-alpha and their hybrid molecules, DNA 4 (1985) 221–232. [18] K. Henco, J. Brosius, A. Fujisawa, J.I. Fujisawa, J.R. Haynes, J. Hochstadt, T. Kovacic, M. Pasek, A. Schambock, J. Schmid, Structural relationship of human interferon alpha genes and pseudogenes, J. Mol. Biol. 185 (1985) 227–260. [19] O.I. Olopade, S.K. Bohlander, H. Pomykala, E. Maltepe, E. Van Melle, M.M. Le Beau, M.O. Diaz, Mapping of the shortest region of overlap of deletions of the short arm of chromosome 9 associated with human neoplasia, Genomics 14 (1992) 437–443. [20] D.J. Schaid, K. Buetow, D.E. Weeks, E. Wijsman, S.W. Guo, J. Ott, Discovery of cancer susceptibility genes: study designs, analytic approaches, and trends in technology, in: Innovative study designs and analytic approaches to the genetic epidemiology of cancer, J. Natl. Cancer Inst. 26 (1999) 1–16. [21] R.G. Westendorp, J.A. Langermans, T.W. Huizinga, A.H. Elouali, C.L. Verweij, D.I. Boomsma, J.P. Vandenbroucke,
[22]
[23]
[24]
[25]
[26]
[27]
[28]
J.P. Vandenbrouke, Genetic influence on cytokine production and fatal meningococcal disease, Lancet 349 (1997) 170–173. J.C. Knight, D. Kwiatkowski, Inherited variability of tumour necrosis factor production and susceptibility to infectious disease, Proc. Assoc. Am. Physicians 111 (1999) 290–298. D. Heresbach, M. Alizadeh, A. Dabadie, N. Le Berre, J.F. Colombel, J. Yaouanq, J.F. Bretagne, G. Semana, Significance of interleukin-1beta and interleukin-1 receptor antagonist genetic polymorphism in inflammatory bowel diseases, Am. J. Gastroenterol. 92 (1997) 1164–1169. V.A. Danis, M. Millington, V.J. Hyland, D. Grennan, Cytokine production by normal human monocytes: inter-subject variation and relationship to an IL-1 receptor antagonist (IL1Ra) gene polymorphism, Clin. Exp. Immunol. 99 (1995) 303–310. G. Sternik, M.G. Pittis, M. Gutierrez, R.A. Diez, L. Sen, In acute lymphoblastic leukemia deletion of the tumour suppressor gene P16 is associated with abnormal interferon genes, Medicina (B Aires) 58 (1998) 463–468. R.C. Aguiar, H. Sill, J.M. Goldman, N.C. Cross, The commonly deleted region at 9p21-22 in lymphoblastic leukemias spans at least 400 kb and includes p16 but not p15 or the IFN gene cluster, Leukemia 11 (1997) 233–238. S. Wacholder, D.T. Silverman, J.K. McLaughlin, J.S. Mandel, Selection of controls in case-control studies III; Design options, Am. J. Epidemiol. 135 (1992) 1042–1050. J.W. Kim, S.B. Kang, D.K. Seo, S.H. Choi, C.M. Lee, Y.B. Kim, N.H. Park, Y.S. Song, S.B. Kang, H.P. Lee, Cytologic screening history of 249 patients with cervical cancer, J. Korean Cancer Assoc. 30 (1998) 1198–1206.