[CANCER RESEARCH 63, 4777– 4780, August 15, 2003] Modulation of Medullary Thyroid Carcinoma Penetrance Suggests the Presence of Modifier Genes in a RET Transgenic Mouse Model 1 Aaron N. Cranston 2 and Bruce A. J. Ponder University of Cambridge and Cancer Research UK Department of Oncology, CIMR, Hills Road, Cambridge, CB2 2XY, United Kingdom ABSTRACT We described previously a thyroid phenotype for transgenic mice ex- pressing an activated Ret oncogene driven from a human calcitonin promoter. Medullary thyroid carcinomas (MTC), a tumor of the thyroid parafollicular C cells, occur in this transgenic line with a pathology analogous to that seen in patients with multiple endocrine neoplasia type 2 (MEN2). When the transgene was introgressed onto four different genetic backgrounds, between 0 and 98% of transgenic progeny developed thyroid tumors by 10 months of age, indicating that tumor penetrance could be modulated by genetic background. Furthermore, tumors on the CBA/ca and C57BL/6J backgrounds were significantly larger than those arising in BALB/c transgenic mice. These results are relevant to human MEN2 disease, because this model system may be used to study genes modifying thyroid tumor penetrance in the dominantly inherited human cancer syndrome. INTRODUCTION The RET proto-oncogene encodes a receptor tyrosine kinase and, in conjunction with the GFR  coreceptors, 3 binds the GDNF family of neurotrophic factors (reviewed in Refs. 1–3). Ligand-induced dimer- ization results in autophosphorylation of RET and the initiation of downstream signaling cascades (4). Gain-of-function mutations in RET have been identified as the causative factor in the inherited cancer syndrome, MEN2, and familial medullary thyroid carcinoma (5– 8). Previously, we created a number of transgenic lines of mice expressing the 3' long isoform of hRET carrying an activating muta- tion at codon 634 in which arginine was substituted for cysteine, analogous to the most common human MEN2A mutation, and driven from a human calcitonin promoter (9). One of these lines, CT-2A-3, expressed a single copy of the transgene from a single insertion site. A proportion of transgenic mice from this line developed MTC. Characterization of the pathological lesions by immunohistochemistry confirmed similarities between this transgenic mouse model and the MEN2A human cancer syndrome. We observed that the number of mice developing MTC increased over time and that the proportion of mice with bilateral MTC was different depending on the strain back- ground (9), raising the possibility of modifier effects attributable to genetic background. However, a thorough analysis of strain back- ground effects was not conducted. The existence of modifying genes in the human population may be inferred from genetic studies that show the MEN2 cancer syndrome to be characterized by variable phenotypic expression both within and between families (5). The transgenic lines presented here may therefore provide a suitable model system in which to identify and elucidate the effects of modifier genes in the etiology of this dominantly inherited cancer syndrome. MATERIALS AND METHODS Mice. To address the question of potential modifier genes, we introgressed our transgene onto four different genetic backgrounds by backcrossing our mixed background (C57BL/6J; FVB/N;CBA/ca) CT-2A-3 line (see Ref. 9) to CBA/ca, C57BL/6J (B6) and FVB/N and also by outcrossing it to BALB/c. Each line was maintained with the transgene in the heterozygous state and backcrossed to each of the four inbred strains for four further sequential generations (N). The N5 progeny were then aged for 43 weeks (10 months). According to the equation, 1 - (1/2) N-1 , where N is the Nth generation (10), each of these lines is predicted to be 94% homozygous for the inbred strain alleles. We called these lines, CT-2A-3 CBA(N5), CT-2A-3 B6(N5), CT-2A-3 FVB(N5), and CT-2A-3 BALB(N5). Animals were handled according to the appropriate United Kingdom Home Office License, UKCCCR guidelines (11), veterinary best practice, and local university regulations. Histology and Immunohistochemistry. At 43 weeks of age, mice were necropsied, and the thyroids were removed en bloc from age-matched trans- genic and nontransgenic littermate mice. The esophagus and trachea were trimmed close to the thyroid glands. The thyroid glands were measured in three dimensions using Vernier scale calipers, and the entire thyroid block weighed to quantitate thyroid volume and mass, respectively. To demonstrate that increases in thyroid volume were indeed attributable to the development of MTC, immunohistochemistry using a rabbit polyclonal antibody to calcitonin (DAKO; 1:3000) was performed on transverse histological sections as de- scribed previously (9). RESULTS AND DISCUSSION Of the original CT-2A line, 65% of the transgenics developed MTC, indicating that the tumor phenotype was not fully penetrant (9). These mice ranged in age from 6 months to 2 years and segregated alleles from three genetic backgrounds (CBA/B6/FVB). We expected that alleles from different genetic backgrounds would differ in the effects on tumor penetrance. We chose an experimental end point of 43 weeks (10 months) to assess tumor development across the various strain backgrounds. When the transgene was introgressed to N5 on four different genetic backgrounds, the tumor phenotype, assessed macroscopically at 43 weeks of age (Fig. 1), varied in penetrance from 0% in the FVB/N background (0 of 16) to almost complete penetrance (47 of 48; 98%) in the CBA/ca background (Fig. 2A). We confirmed that these tumors were in fact medullary thyroid carcinomas by immunohisto- chemical staining for calcitonin (Fig. 3). By 43 weeks, the majority of transgenic mice on a C57BL/6J background (9 of 14; 64%) but only 1 in 7 (4 of 29; 14%) of the BALB/c transgenics had developed MTC. These differences are highly significant (P  0.0001). Gener- ally, tumors were bilateral, although only one lobe had detectable tumor in 8 and 3% of CBA/ca and BALB/c transgenic mice, respec- tively. As expected, no tumors were observed at the same age in any of the nontransgenic mice on any of the strain backgrounds (CBA/ca NTg:19; C57BL/6J NTg: 12; BALB/c NTg: 26; FVB/N NTg: 11). To quantify the phenotypic differences, we measured thyroid weights and volumes (Fig. 2B). To see if there was a correlation, one was plotted against the other (Fig. 2C). The correlation coefficient (r) was calculated to be 0.63, whereas the SD of residuals from the line was 17.98 (n = 35), indicating that there was not a significant departure from linearity (P = 0.27), and demonstrating, as expected, a strong correlation between thyroid weight and volume. One caveat Received 6/3/03; revised 6/3/03; accepted 7/2/03. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by Cancer Research UK. 2 To whom requests for reprints should be addressed, at University of Cambridge and Cancer Research UK Department of Oncology, CIMR, Hills Road, Cambridge, CB2 2XY, United Kingdom. Phone: 44 (0) 1223-336900; Fax: 44 (0) 1223-336902; E-mail: arron.cranston@ntlworld.com. 3 The abbreviations used are: MEN2, multiple endocrine neoplasia type 2; MTC, medullary thyroid carcinoma; GFR, GDNF family receptors; GDNF, glial cell line- derived neurotrophic factor. 4777 Research. on November 29, 2021. © 2003 American Association for Cancer cancerres.aacrjournals.org Downloaded from