- March 6, 2012 – Online Registration Opens
- March 12, 2012 – Abstract submission Closes (all abstracts due at this time)
- March 12, 2012 - New Investigator Award Applications Due
- April 16, 2012 - Accepted abstracts for Poster Session, New Investigators Announced
- May 4, 2012 - Hotel Reservations Close
- May 21, 2012 - Online Registration Closes
Examinations on gene expression in tumor and normal tissue on Chernobyl thyroid cancer
*Michael Abend, Bundeswehr Institute of Radiobiology
Keywords: PTC, gene expression, Chernobyl, ionizing radiation
The study of radiation carcinogenesis in humans can take advantage of instances where relationships between ionizing radiation exposure and a health outcome are strong and consistent, as in the situation where the thyroid gland is irradiated at a young age. Despite differences in study populations, designs, and dosimetric approaches, reported estimates of relative risk (RR) per unit of absorbed radiation dose in gray (Gy) for those exposed during childhood are remarkably similar among most studies (RRs of 3-6 per Gy) (1). The opportunity for molecular research of radiation-related thyroid cancer is facilitated by the resources of the Chernobyl Tissue Bank (CTB), which systematically collects biological samples from patients with Chernobyl-related thyroid pathology. Recent studies using the CTB materials and high throughput technologies have identified new ‘radiation-specific’ genes. However, findings at the gene level are generally inconsistent across studies potentially due to small sample sizes, use of controls from different populations, lack of methodological validation in independent samples, and different analytic approaches (2). More importantly, none of the previous studies had individual radiation doses; the studies assumed that all exposed cases had essentially equivalent doses. Depending on the true dose-gene expression relationship, this assumption could cause false positive or false negative associations. To improve understanding of the molecular consequences of I-131 exposure, we evaluated differential gene expression in thyroid tissue, defined as a difference in gene expression levels between tumor and corresponding normal thyroid tissue, in relation to individual I-131 thyroid dose estimates. We hypothesized that if dose-related gene expression patterns in tumor tissue truly reflect an important event in radiation carcinogenesis, they should differ from patterns observed in normal tissue; hence we chose this novel approach to analyze differential dose-expression relationships in tumor relative to the corresponding paired normal sample (T/N sample pairs). Our study used RNA specimens from the CTB of patients who underwent thyroid surgery for thyroid cancer in the Ukrainian-American cohort study composed of approximately 13,000 Ukrainian residents < 18 years at the time of the accident with individual radioactivity measurements taken within two months after the accident (3). In the current study, we first conducted an initial screen in half (32 T/N sample pairs) of our 63 cases to identify promising gene candidates that were differentially expressed in T/N sample pairs in relation to I-131 dose based on whole genome RNA microarrays (phase I). We then validated the top candidates in T/N sample pairs from the second half (31 T/N sample pairs) of the cases using qRT-PCR (phase II). For the validated candidates we additionally characterized the relationship of gene expression separately in tumor and normal tissue. For analysis of whole genome microarray gene expression data we used the non-parametric Kruskall-Wallis test (P kruskall) to compare differential gene expression across three dose categories (= 0.30, 0.31-1.0, > 1.0 Gy) with cut-off points approximately corresponding to the tertiles of dose distribution among cases, and linear regression models with trend test (P linear) for continuous dose. Only those gene transcripts that had a call “present” in at least 50% of RNA specimens from tumor and normal tissue were included in the analysis of differential gene expression (~15,000 genes). As none of the P values remained significant after FDR correction, we adopted the following criteria to select promising candidates for validation and quantification by qRT-PCR: (P kruskall = 0.001) or (0.001< P kruskall = 0.01 and P linear = 0.01) or (0.01< P kruskall < 0.05 and P linear = 0.001). Overall, 106 candidates out of about 2,500 differentially expressed gene transcripts in relation to dose satisfied the relaxed selection criteria. We further narrowed this list to 41 genes that exhibited the lowest P values (P kruskall < 0.0005 or P linear < 0.0005) and had at least a two-fold difference (increase/decrease) in differential gene expression between highest relative to lowest dose category. An additional 34 genes with P kruskall < 0.005 were included in validation, because there was evidence in the literature for gene amplification (4). Thus, 75 genes were selected for validation by means of qRT-PCR. To validate our microarray findings, we evaluated gene expression by qRT-PCR (TaqMan primer probe assays) on 62 paired RNA specimens from the remaining 31 individuals. Regarding the qRT-PCR data, we first computed residuals from standard linear models fitted to gene expression values y adjusting for age at thyroid surgery (3 categories), sex, and oblast or state of residence (Chernigov, Zhytomyr, Kiev), log y = µ + age + sex + oblast + e, (1) where µ is the overall mean expression level. We then assessed the relationship between I-131 dose and gene expression in linear mixed models with individual specimens as the outcome variable. These models account for correlations of the tumor and normal tissue measurements taken on the same individual and also accommodate duplicate measurements on the same person for the same tissue type. The models were rij = µr + tumor + dosetumor + dosenormal + eij , (2) where rij denotes the residual from model (1) for subjects i (i=1, 2,…,31) on the jth sample (j=1, 2 for tumor and normal tissue), and eij is normally distributed error term that also incorporates correlations from repeated measurements on the same sample. The dose effect in tumor samples is quantified by dosetumor, and the dose effect in normal tissue is given by dosenormal. To assess differences in dose effect by tissue type, we tested the null hypothesis H0: dosetumor = dosenormal. Again, I-131 dose was used in two ways: in 3 independent dose groups, which leads to a 2 degree of freedom Wald test P value for H0, and using the 3 ordered dose groups, corresponding to a 1 degree of freedom test for H0. Parameters for the mixed models were estimated using the restricted maximum likelihood method incorporated in PROC MIXED, SAS 9.1.3. Of 75 genes assayed, 26 genes could be not examined, such as lack of amplification or other technical issues. For 11 of the remaining 49 genes, the dose-related expression in tumor tissue was statistically significantly different from the dose-related expression in normal tissue when dose was used as a categorical or ordinal variable in the linear mixed models. Associations significant for both categorical and ordinal dose were found for the following 6 genes: ACVR2A, CA12, CDK12, FAM38A, LMO3, MTA1. Three genes (SLC19A1, SLC43A3, ZNF493) had statistically significant differences in differential gene expression for categorical dose (2 degree of freedom tests), but not for ordinal dose, suggesting non-linear dose-expression relationships; and two genes (AJAP1, GALNT7) had a statistically significant difference in dose-expression relationship for ordinal dose, but not for categorical dose. For SLC43A3 the associations between gene expression and dose were restricted to tumor tissue (2 degree of freedom test) and for FAM38A (2 degree of freedom and 1 degree of freedom tests), SLC43A3 (2 degree of freedom test), LMO3 (1 degree of freedom test), MTA1 (1 degree of freedom test) to normal tissue). For CA12, GALNT7, LMO3, and SLC43A3 there is good separation in expression between tumor and normal tissue at each dose level as well as a suggestion of opposing trends with dose. However, in most instances, significant differences in dose-response by tissue type appeared to be due to the heterogeneity of dose-response patterns in the absence of a monotone dose-response in either tissue type. The 11 genes were all located on different chromosomes and belong to different biological processes/pathways, including cell adhesion (AJAP1, FAM38A), energy metabolism (CA12), transcription or DNA methylation (LMO3, ZNF493, MTA1, SLC19A1), and growth/differentiation (CDK12, ACVR2A). All these biological processes/pathways have been implicated in cellular responses to ionizing radiation or were previously reported to be uniquely deregulated in post-Chernobyl thyroid cancers (CA12, GALNT7, LMO3, and SLC43A3) (4, 5). Dose-dependency for differential gene expression detected years after exposure likely represents a late and/or long lasting effect of radiation. Mechanisms by which radiation-induced changes are sustained over time could be related to gene amplification and/or epigenetic modifications. We plan to further pursue the significance of gene amplification and epigenetic modifications of these genes by performing detailed examinations on corresponding DNA regions, post-transcriptional regulation (microRNA) and at the protein level using immuno-histochemistry. Our work has implications for being able to discriminate radiation-related from sporadic thyroid cancers using gene expression or DNA measurements. As we report one of the first efforts in this area using individual dose estimates, it remains that the methodology must be confirmed in sporadic vs. radiation exposed thyroid cancers.
Acknowledgement: This work has been done in cooperation and would have been not possible without the important work performed in joint collaboration by Alina Brenner, Ruth Pfeiffer, Christian Ruf, Maureen Hatch, Tetiana Bogdanova and Mykola Tronko to name but a few.
1. A.V. Brenner, M.D. Tronko, M. Hatch, T.I. Bogdanova, V.A. Oliynik, J.H. Lubin, L.B. Zablotska, V.P. Tereschenko, R.J. McConnell, E.Ron, I-131 dose response for incident thyroid cancers in Ukraine related to the Chornobyl accident. Environ Health Persp. 119, 933-9 (2011).
2. C. Maenhaut, V. Detours, G. Dom, D. Handkiewicz-Junak, M. Oczko-Wojciechowska, B. Jarzab, Gene expression profiles for radiation-induced thyroid cancer. Clin Oncol (R Coll Radiol). 23, 282-8 (2011).
3. V.A. Stezhko, E.E. Buglova, L.I. Danilova, V.M. Drozd, N.A. Krysenko, N.R. Lesnikova, V.F. Minenko, V.A. Ostapenko, S.V. Petrenko, L.B. Zablotska, A cohort study of thyroid cancer and other thyroid diseases after the Chornobyl accident: objectives, design and methods. Radiat Res. 161, 481-92 (2004).
4. L. Stein, J. Rothschild, J. Luce, J.K. Cowell, G. Thomas, T.I. Bogdanova, M.D. Tronko, L. Hawthorn, Copy number and gene expression alterations in radiation-induced papillary thyroid carcinoma from chernobyl pediatric patients. Thyroid. 20, 475-87 (2010).
5. L.B. Zablotska, E. Ron, A.V. Rozhko, M. Hatch, O.N. Polyanskaya, A.V. Brenner, J. Lubin, G. N. Romanov, R.J. McConnell et al., Thyroid cancer risk in Belarus among children and adolescents exposed to radioiodine after the Chornobyl accident. Brit J Cancer. 104, 181-187 (2011).