Mutation Research 762 (2014) 17–23

Contents lists available at ScienceDirect

Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis journal homepage: www.elsevier.com/locate/molmut Community address: www.elsevier.com/locate/mutres

A novel mutation in the thyroglobulin gene that causes goiter and dwarfism in Wistar Hannover GALAS rats Akira Sato a , Kuniya Abe b , Misako Yuzuriha b , Sakiko Fujii c , Naofumi Takahashi a , Hitoshi Hojo a , Shoji Teramoto a , Hiroaki Aoyama a,∗ a

Toxicology Division, Institute of Environmental Toxicology, 4321 Uchimoriya-machi, Joso, Ibaraki 303-0043, Japan Technology and Development Team for Mammalian Genome Dynamism, RIKEN BioResource Center, 3-1-1 Koyadai, Tsukuba, Ibaraki 305-0074, Japan c Safety Research Division, Safety Research Institute for Chemical Compounds Co., Ltd., 363-24 Shin-ei, Kiyota-ku, Sapporo, Hokkaido 004-0839, Japan b

a r t i c l e

i n f o

Article history: Received 25 November 2013 Received in revised form 23 January 2014 Accepted 18 February 2014 Available online 28 February 2014 Keywords: Outbred Thyroid Intron Acceptor site Splicing abnormality Reproduction study

a b s t r a c t Outbred stocks of rats have been used extensively in biomedical, pharmaceutical and/or toxicological studies as a model of genetically heterogeneous human populations. One of such stocks is the Wistar Hannover GALAS rat. However, the colony of Wistar Hannover GALAS rat has been suspected of keeping a problematic mutation that manifests two distinct spontaneous abnormalities, goiter and dwarfism, which often confuses study results. We have successfully identified the responsible mutation, a guanine to thymine transversion at the acceptor site (3 end) of intron 6 in the thyroglobulin (Tg) gene (Tgc.749−1G>T ), that induces a complete missing of exon 7 from the whole Tg transcript by mating experiments and subsequent molecular analyses. The following observations confirmed that Tgc.749−1G>T /Tgc.749−1G>T homozygotes manifested both dwarfism and goiter, while Tgc.749−1G>T /+ heterozygotes had only a goiter with normal appearance, suggesting that the mutant phenotypes inherit as an autosomal semi-dominant trait. The mutant phenotypes, goiter and dwarfism, mimicked those caused by typical endocrine disrupters attacking the thyroid. Hence a simple and reliable diagnostic methodology has been developed for genomic DNA-based genotyping of animals. The diagnostic methodology reported here would allow users of Wistar Hannover GALAS rats to evaluate their study results precisely by carefully interpreting the data obtained from Tgc.749−1G>T /+ heterozygotes having externally undetectable thyroidal lesions. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Outbred stocks of experimental animals such as rats, mice and rabbits have been used extensively in the field of biomedical, pharmaceutical and/or toxicological researches. Our laboratory has also been using such outbred experimental animals for reproductive and developmental toxicity studies [1–9] mainly because their genetic background is believed to match the highly polymorphic human population. Another major reason for the use of outbred animals may be their excellent reproductive abilities: researchers often use outbred stocks of rats and/or mice for maintaining the characteristics of their interests with reduced fecundity caused by a knock out or a spontaneous mutation of a certain gene. Chia et al. estimated that approximately 85% of all rat papers and 33% of all

∗ Corresponding author. Tel.: +81 297 27 4532; fax: +81 297 27 4519. E-mail address: [email protected] (H. Aoyama). http://dx.doi.org/10.1016/j.mrfmmm.2014.02.003 0027-5107/© 2014 Elsevier B.V. All rights reserved.

mouse papers involved outbred stocks from January 2002 to July 2005 [10]. One of such stocks is the Wistar Hannover GALAS rat that has been released as a global standard outbred rodent model for the use in pharmacological and/or toxicological studies. This stock has several advantages over other outbred stocks according to the breeder’s publicity (http://www.galas.org/animalmodels.htm); i.e., each GALAS member breeder produces the animals with minimal genetic drift by means of a periodical revitalization of the stock with cryopreserved embryos and a rotational breeding system so that researchers can use these animals with constant quality in many countries in the world. The system is also expected to keep the historical control data of this stock constant over years in user laboratories and indeed, the number of toxicological studies using Wistar Hannover GALAS rats seems to be rapidly increasing. We have confirmed several beneficial characteristics of this stock of rats such as a good reproductive performance and an adequate litter size [11] and have begun to use the rats in reproductive and developmental toxicity studies [1].

18

A. Sato et al. / Mutation Research 762 (2014) 17–23

In contrast to several beneficial characteristics, the colony of Wistar Hannover GALAS rats is suspected of accumulating problematic spontaneous mutations due to a rotational breeding system as the nature of outbred, which may occasionally lead scientists to misunderstanding their study results. The most serious concern is spontaneous goiter and dwarfism [12–14] because their aberrant thyroidal phenotypes including tumor formations in aged individuals are quite similar to those caused by goitrogens [15–18]. According to Shimoi et al., 8.5% (10/118) of Wistar Hannover GALAS rats obtained from CLEA Japan Inc. (BrlHan:WIST@Jcl[GALAS] rats) exhibited an enlargement of the thyroid gland [13]. Doi et al. further reported that the cross of externally normal BrlHan:WIST@Jcl[GALAS] rats yielded dwarf offspring with low serum T3 and T4 levels accompanied by a high serum TSH level [12]. An increase in size of the thyroid was also noted in both dwarves and their externally normal littermates in their study, suggesting that either the same mutation or two distinct mutations was involved in the etiology of goiter and dwarfism in this stock of rats. However, the responsible gene(s) has not yet been disclosed, although an autosomal recessive mode of inheritance is suggested for the dwarfism in the original stock at RCC breeding colony [19]. The present study aimed at elucidating the underlying genetic mechanisms of goiter and dwarfism in the Wistar Hannover GALAS rat. We first confirmed the mode of inheritance of dwarfism by mating experiments with a mutant line derived from a pair of externally normal rats purchased from the breeder, which was followed by investigations of a possible mutation in thyroglobulin (Tg) gene, a strong candidate gene for the dwarfism. Then, we developed a genomic DNA-based simple and reliable diagnostic methodology for precisely identifying the mutants to examine a genetic association between the polymorphism in Tg locus and two distinct phenotypes such as dwarfism and goiter development in the colony of Wistar Hannover GALAS rats. 2. Materials and methods

Fig. 1. Retarded growth of dwarf offspring. Pups derived from the cross between heterozygotes were identified within a litter by tattooing on postnatal day 4. They were weighed individually on postnatal days 4, 7, 14 and 21. Data were analyzed on a per litter basis (N = 6–7) using Student’s t-tests. Bars indicate standard deviations. The differences in body weights between the dwarves and their normal littermates were statistically significant on postnatal days 14 and 21 for both males (**P < 0.01) and females (†† P < 0.01). , wild-type male; ♦, dwarf male; 䊉, wild-type female; and , dwarf female.

No dwarf female was used because they were not expected to be fertile. The dwarf male was postulated as a mutant homozygote, while SD rats as wild type homozygote. Externally normal littermates were presumed as heterozygotes when they produced dwarf offspring before or during the mating experiments. Dwarfism of pups was judged on lactation day 14 and thereafter when their growth retardation became evident according to the result of preliminary examination (Fig. 1). The acceptability of observed data to the expected segregation ratio was analyzed by chi-square test.

2.1. Background of the dwarfism 2.4. RT-PCR Two dwarf pups, a male and a female, arose from a cross of externally normal Wistar Hannover GALAS (BrlHan:WIST@Jcl[GALAS]) rats during the course of conducting a reproduction toxicity study with a certain test compound. They were born with normal appearances but exhibited retarded growth after two weeks of age during the lactation. The affected male and his externally normal female littermates were excluded from the study as founders for brother–sister matings to examine the heritability of this phenotype and to establish a new mutant strain. All animals used in the present study were derived from these mating pairs.

Rats were kept and sacrificed in accordance with the Guidelines for Animal Experimentation issued by the Japanese Association for Laboratory Animal Science [20]. They were housed in a barriersustained animal room with controlled temperature (22 ± 2 ◦ C), relative humidity (55 ± 15%), ventilation (at least 10 times/hr, allfresh-air system) and 12-hr light/dark cycles, and were supplied with commercial diets (MF pellet, Oriental Yeast Co., Ltd., Tokyo) and local tap water ad libitum.

The thyroids were immediately frozen in liquid nitrogen after removal from two dwarfs, two heterozygotes and two SD rats. The tissues were pulverized and briefly sonicated in 1 mL ISOGEN (Nippon gene, Tokyo). Then, 200 ␮l of chloroform was added to the suspension, mixed and centrifuged at 14,000 rpm for 15 min. The aqueous layer was carefully transferred to a fresh micro-tube to which an equal volume of isopropanol was added. The precipitates of centrifugation at 14,000 rpm for 10 min were washed with 70% ethanol and the total RNA pellets obtained were dissolved in TE buffer. Reverse transcription and subsequent PCR analysis of the first strand cDNAs were conducted with TaKaRa RNA PCRTM Kit(AMV)Ver.3.0 (Takara Bio Inc., Seta) according to the manufacturer’s direction. PCR amplification was conducted with the 12 primer pairs (Supplementary Table 1) designed to cover the entire Tg mRNA sequence (NCBI I.D.: NM 030988) under the following condition: 94 ◦ C for 2 min for pre heating, followed by denaturation at 94 ◦ C for 30 s, annealing at 55 ◦ C for 30 s, and extension at 72 ◦ C for 1 min for 25 cycles. Each amplicon was separated by 1.5% agarose gel electrophoresis and visualized by ethidium bromide staining.

2.3. Mating tests

2.5. Direct sequence

A dwarf male, his externally normal littermates and descendants (5 males and 13 females), and Jcl:SD rats (2 males and 5 females) purchased from CLEA Japan Inc. (Tokyo) were used in the mating tests for examining the mode of inheritance for the dwarfism.

Determination of entire protein coding sequences of Tg mRNA in each genotype was done by using the first strand cDNA prepared in the RT-PCR analysis (one sample/genotype). The RT-PCR was conducted by the same way as described above with 20 primer

2.2. Animal husbandry

A. Sato et al. / Mutation Research 762 (2014) 17–23

19

Fig. 2. Histology of the thyroid gland. (A) Hematoxylin and eosin-stained section of the thyroid from a 19-week-old normal Wistar Hannover GALAS rat. (B and C) Those from a dwarf and an externally normal littermate, respectively (15–16-week-old). Note poor colloid formation, decreased follicular sizes and/or huge vacuoles in mutant rats.

pairs (Supplementary Table 2). Each amplicon was purified with MonoFas® DNA Kit I (GL Sciences Inc., Tokyo) and sequenced directly in both directions with BigDye® Terminator v3.1 Cycle Sequencing Kit (Life Technologies Japan, Tokyo) by Applied Biosystems 3100 Genetic Analyzer (Life Technologies Japan) according to the manufacturer’s instruction. For investigating the predicted Tg mutation at the boundary region between intron 6 and exon 7, genomic DNA was prepared from the pinna of two dwarfs, two heterozygotes and two normal SD rats by standard phenol–chlorophorm extraction. Target fragments including the intron 6/exon 7 border were amplified with the following primer pair: forward, 5 caggctgcttttgtagattt 3 ; reverse, 5 cggaatctgccagagatgac 3 . After purification with MonoFas® DNA Kit I (GL Sciences Inc.), each PCR product was directly sequenced in both directions by Applied Biosystems 3730xl DNA Analyzer (Life Technologies Japan) according to the manufacturer’s direction.

2.6. DNA-based rapid diagnosing methodology for the mutant Tg allele Small pieces of the pinna (2 mm) were lysed with 200 ␮g/ml proteinase K (Sigma–Aldrich Japan, Tokyo) in 200 ␮l digestion buffer (10 mM Tris–HCl pH8.0, 25 mM EDTA, 100 mM NaCl, 0.5% SDS) at 55 ◦ C for more than two hours. After brief centrifugation, the supernatant was diluted 40-fold with sterile water and the resultant was used as a sample for PCR. A total of 15.0 ␮l reaction solution composed of 1.5 ␮l lysate, 1× Ampdirect® Plus (Shimazu Biotech, Kyoto), 0.375U Nova TaqTM Hot Start DNA Polymerase (EMD Biosciences, Inc., San Diego, CA) and 0.5 ␮M primer pairs. Forward primers were designed for separately amplifying genomic DNA fragment containing mutant- or wild-type sequence: they intentionally contained one mismatch with the target sequence at the −1 position from 3 end, while they had two mismatches with opposite sequence at its 3 ends for amplifying only the target sequence. The forward primer for amplifying only the wild type sequence was 5 ccctgaaatgctatctacatttgtttcg 3 and that for the mutant 5 ccctgaaatgctatctacatttgtttct 3 (intentional mismatches are underlined). The same reverse primer, 5 cggaatctgccagagatgac 3 , was used for both reactions. The PCR condition was 94 ◦ C for 10 min for pre heating (hot start), followed by denaturation at 94 ◦ C for

30 s, annealing at 55 ◦ C for 30 s, and extension at 72 ◦ C for 30 s for 30 cycles. Genotypes of rats were determined by agarose gel electrophoresis of each amplicon: rats were judged as heterozygotes when both amplicons were detected. 2.7. Pathological examination of the thyroid for analyzing correlation between goiter development and Tg genotype The thyroid glands of 15-week-old externally normal Wistar Hannover GALAS rats (13 males and 12 females) were observed grossly at necropsy, carefully removed, weighed and fixed in neutral buffered 10% formalin. These rats were derived from crosses between presumed heterozygotes, so that they should have either heterozygous or +/+ genotype. The thyroids were routinely dehydrated, paraffin embedded, sectioned, stained with hematoxylin and eosin (H/E), and examined histopathologically under a light microscope. The SNP at the acceptor site of the 6th intron was determined by the result of triplicated allele specific PCR described above. 3. Results 3.1. Histological observations of the thyroid Preliminary histopathological examinations of the thyroids revealed poor colloid formation and decreased follicular sizes in dwarf rats (Fig. 2B). Follicular epithelia had huge vacuoles and their nuclei shifted to the luminal side. Thyroidal follicular epithelia of externally normal littermates also exhibited huge vacuoles and nuclear luminal shifts, although the sizes of some follicles were comparable to those of normal rats (Fig. 2A and C). These observations suggest that the dwarfism accompanied by goiter in our experiment was caused by the same mutation(s) as those found previously in the Wistar Hannover GALAS rat colonies [12–14,19]. 3.2. Mating experiments demonstrating an autosomal recessive inheritance of the dwarfism Results of mating experiments are summarized in Table 1. The cross between the dwarf male and presumably heterozygous

20

A. Sato et al. / Mutation Research 762 (2014) 17–23

Table 1 Results of mating experiments to examine the mode of inheritance for dwarfism. Mating pair (presumed genotype)

No. of litters examined

Female

Male

Externally normal (Tgc.749−1G>T /+) Externally normal (Tgc.749−1G>T /+) Externally normal (Tgc.749−1G>T /+) Externally normal (+/+)

Dwarf (Tgc. 749−1G>T /Tgc.749−1G>T ) Externally normal (Tgc. 749−1G>T /+) Externally normal (+/+) Externally normal (Tgc.749−1G>T /+)

8 12 4 5

No. of progenies observed (expected) Normal

Dwarf

2 value

27 (23.5) 78 (77.25) 54 (54) 53 (53)

20 (23.5) 25 (25.75) 0 (0) 0 (0)

0.39 0.01

A dwarf male and both sexes of rats presumably carrying the mutant gene at a heterozygous condition were derived from the Wistar Hannover GALAS stock (BrlHan:WIST@Jcl[GALAS] rats), while males and females with wild genotype (+/+) were derived from the SD stock (Jcl:SD rats).

females produced a total of 20 dwarves and 27 externally normal pups, which agreed with the expected ratio of 1:1 (␹2 = 0.39, P > 0.05). The cross between presumable heterozygotes also yielded 78 externally normal and 25 dwarf offspring, of which segregation ratio was closely fitted in with the expected ratio of 3:1 (␹2 = 0.01, P > 0.05). No dwarf pup was found among a total of 107 offspring in the reciprocal outcrosses between presumably heterozygous rats and normal SD rats with wild genotype. These results suggest that the dwarfism is inherited as an autosomal recessive trait as speculated previously [19]. 3.3. A point mutation in the thyroglobulin gene (Tgc.749−1G>T ) as a cause of dwarfism Our hypothesis was that a certain mutant allele of the Tg gene, which encodes a precursor peptide of thyroid hormone, might be a cause of the dwarfism because mutant alleles of this gene such as rdw in rats and cog in mice had been reported to induce dwarfism with or without goiter in a homozygous condition [21–24]. We first examined for alterations in the amplicon length of Tg mRNA using total RNA samples extracted from the thyroids of dwarves, heterozygotes (determined by mating tests) and SD rats with normal thyroid. The analysis revealed an amplicon length polymorphism among the samples when RT-PCR was conducted with a primer set for the 5 ends of 6th exon and the middle part of 9th exon, respectively, suggesting a partial deletion of Tg mRNA between the 6th and 9th exons in dwarf and heterozygous rats (Fig. 3). The size of deleted sequence seemed to be approximately 150 base pairs (bp), which was almost equivalent to the whole length of 7th exon (144 bp). There was no difference in the size of amplicons covering the remaining region of Tg mRNA (8.5 kb) among samples. Subsequent bidirectional sequencing analyses of the whole coding sequence of Tg mRNA (cDNA) clearly demonstrated the complete missing of exon 7 in both dwarf and heterozygous rats (Accession Nos. AB682739 and AB682738, Supplementary Fig. 1), suggesting that the mutation induces complete deletion of 48 amino acids without changing the following reading frame. The analyses also disclosed that there was no difference in cDNA sequences between a dwarf rat derived from the Wistar Hannover GALAS colony and a SD rat

Fig. 3. Partial deletion of Tg mRNA between the 6th and 9th exons in dwarf and heterozygous rats. Lanes 1 and 3, heterozygotes; lanes 2 and 4, homozygotes (dwarves); lanes 5 and 6, wild-type homozygotes (SD rats); lane 7, negative control; M, molecular weight marker. White arrow head shows RT-PCR products (approximately 480 bp) amplified with the primer pair no. 3 described in Supplementary Table 1, while the solid indicates truncated products, of which size is approximately 150 bp shorter than expected.

except for the deletion, while 5 SNPs including both synonymous and non-synonymous ones were found between a heterozygote carrying the deletion and the SD rat. These findings suggest that all the SNPs were derived from the wild-type allele of heterozygote. We then sequenced the genomic DNA at the boundary region of the 6th intron and the 7th exon to confirm the hypothesis that either a genomic deletion including the whole 7th exon or other small mutation such as SNP that could cause the skipping of this exon would be the primary cause of aberrant Tg transcript. As shown in Fig. 4, the analyses revealed a guanine (G) to thymine (T) transversion at the 3 end of intron 6 in dwarves and their littermates with a goiter (a wild type intron with AG at the 3 end was replaced by a mutant allele with AT). We propose the term Tgc.749−1G>T as a symbol of this mutant gene.

3.4. Rapid diagnosing method for predicting Tgc.749−1G>T /+ genotype Forward primers designed for separately amplifying a part of genomic DNA containing Tgc.749−1G>T or wild-type sequence intentionally contained one mismatch with the target sequence at the −1 position from 3 end, while they had two mismatches with opposite sequence at its 3 end. These primers should enable us to amplify only each target sequence when they were used for PCR with a mutual reverse primer. The allele specific PCR did demonstrate that both of the primer pairs yielded the same size of amplicons when genomic DNA of Tgc.749−1G>T /+ heterozygotes was amplified, while they amplified only each target sequence when samples from Tgc.749−1G>T /Tgc.749−1G>T or +/+ homozygotes were applied (Fig. 5).

Fig. 4. Point mutation at c.749-1 of thyroglobulin gene. Sequencing analysis of the flanking region between intron 6 and exon 7 in Tg gene revealed a G > T transversion at the 3 end of intron 6 (arrowhead) in both the dwarf and heterozygous rats. Nucleotide positions are provided at the bottom.

A. Sato et al. / Mutation Research 762 (2014) 17–23

Fig. 5. Rapid diagnosis of Tgc.749−1G>T carriers. Allele specific PCR was performed on genomic DNA from Tgc.749−1G>T /+ (lane 1), Tgc.749−1G>T /Tgc.749−1G>T (dwarf; lane 2) and +/+ (SD; lane 3) rats by using diagnostic primer pairs for the mutant or wild-type allele. Both primer pairs yielded single products with expected size of 170 bp. Lane 4, negative control; M, molecular weight marker.

3.5. Correlation between the Tgc.749−1G>T allele and goiter development We further investigated the correlation between the Tg genotype and goiter development using both sexes of 15-week-old externally normal Wistar Hannover GALAS rats. They were genotyped by the allele specific PCR described above after the necropsy. The examinations revealed that the heaviest 8 thyroids in males as well as the heaviest 7 in females were derived from those with Tgc.749−1G>T /+ genotype and exhibited vacuolar change in their follicular epithelium as found in the preliminary observations (Fig. 2). The average thyroid weights of Tgc.749−1G>T /+ males and females were approximately as twice as heavier than those of +/+ individuals (Fig. 6). These results demonstrated that animals having Tgc.749−1G>T /+ genotype always developed goiter under our experimental condition. 4. Discussion Thyroglobulin is initially synthesized by ribosomes on the rough-surfaced endoplasmic reticulum (rER) of thyroidal follicular epitheliums, folded within rER by catalytic effects of molecular chaperones, and exported to Golgi apparatus. The process is often disrupted by missense or nonsense mutations occurred in the protein coding regions of thyroglobulin (Tg) gene to cause goiter and/or

Fig. 6. Correlation between the SNP genotype and thyroid status. Thyroid weights of +/+ and Tgc.749−1G>T /+ rats were statistically analyzed by Student’s or Aspin-Welch’s t-test following F-test (***, significantly different at P < 0.001). Mean thyroidal weights and standard deviations of +/+ males and females were 17.1 ± 1.3 mg (N = 5) and 15.8 ± 2.9 mg (N = 5), respectively, while those of Tgc.749−1G>T /+ were 35.6 ± 4.4 mg (N = 8) and 27.8 ± 2.4 mg (N = 7) for males and females, respectively. ♦, +/+ males; , Tgc.749−1G>T /+ males; ◦, +/+ females; 䊉, Tgc.749−1G>T /+ females.

21

dwarfism in humans [25,26] and other species of animals including mice, rats, goats and cattle [21–24,27–29]. A missplicing of Tg pre-mRNA due to a transversion of nucleotide at the acceptor site of intron 3 is also known to induce a severe congenital goiter and hypothyroidism in humans [30], although this type of mutation is less common than those occurred in the coding regions. These mutations give rise to a distention of rER in thyroidal follicular epitheliums, a typical ultra-structural abnormality of the thyroid called as endoplasmic reticulum storage disease (ERSD), in both humans and rodents [31–33]. Misfolding of thyroglobulin due to amino acid changes is suggested to prevent the protein export from rER to Golgi by a certain quality control system to induce the accumulation of thyroglobulin within rER, which in turn causes dilation of rER accompanied by dislocations of nuclei toward the luminal side in some serious cases [34,35]. Since pathologic features of the thyroid in affected Wistar Hannover GALAS rats [12,13,19] were the same as those of Tg mutants reported previously, these rats were postulated to carry a mutation somewhere within the Tg gene. The present analyses of Tg transcripts obtained from the thyroid gland of affected individuals successfully demonstrated that dwarves and their externally normal littermates expressed aberrant Tg mRNA with complete deletion of exon 7 in both homozygous and heterozygous conditions. Determinations of genomic sequences further revealed a G to T transversion at the 3 end of intron 6. The 3 end of common introns is known to have an almost invariant AG sequence and a point mutation occurred within this site often causes missplicing of pre-mRNA, which in turn manifests a variety of abnormal phenotypes [36,37]. These results suggest that the missplicing of Tg mRNA in our mutant line was caused by this point mutation. The relationship between the transversion and the thyroidal morphology was also confirmed in the present study that rats with normal thyroid always had AG/AG (+/+) sequence at the 3 end of intron 6, whereas all rats with AT/AG (Tgc.749−1G>T /+) genotype developed the goiter. These observations, together with the results of mating experiments, lead to the conclusion that this mutant allele causes dwarfism with goiter and goiter alone in homozygous and heterozygous conditions, respectively: in other words, the mutant phenotypes inherit as an autosomal semi-dominant trait. The mutant strain named DWH (Dwarfism derived from Wistar Hannover GALAS rats) has been established from a pair of the founder dwarf male and his female littermate as a unique animal model of congenital goiter and dwarfism due to missplicing of Tg gene. It is now maintained by full-sib matings of Tgc.749−1G>T /+ heterozygotes diagnosed by the allele-specific PCR and has reached the 25th generation at this writing. We further confirmed the involvement of Tgc.749−1G>T mutation in the etiology of goiter in the colony of commercially available Wistar Hannover GALAS rats when a total of 156 BrlHan:WIST@Jcl[GALAS] rats were purchased from the same breeder. The examinations demonstrated that 7 individuals judged as Tgc.749−1G>T /+ heterozygotes exhibited goiter, while the remaining rats with +/+ genotype had apparently normal thyroid (unpublished observation at the Institute of Environmental Toxicology). These observations suggest that the Tgc.749−1G>T mutation was kept in the breeder’s colony. Nuisance mutations kept in commercially available outbred colonies of rats and mice often confuse the results of any kinds of experiments with such animals. Examples include spontaneous mutations in mdr1a gene in CF-1 and CD-1 mice [38,39] and ahr gene in Wistar Hannover rats [40], which may seriously modulate the sensitivity of mutant individuals to xenobiotics. We also have experienced several genetic malformation syndromes that mimic the effects of teratogens during the course of reproductive and developmental toxicity studies in common outbred rats purchased from commercial breeders [41–43]. The present mutation

22

A. Sato et al. / Mutation Research 762 (2014) 17–23

experiment with genetically uniform inbred animals. We hope the genomic DNA-based simple and reliable diagnostic methodology described here would be helpful for the breeders and users of Wistar Hannover GALAS rats to distinguish precisely Tgc.749−1G>T /+ heterozygotes with normal external appearance from +/+ individuals. Conflict of interest statement The authors declare that there are no conflicts of interest. Acknowledgements The authors express their thanks to Mses. Mayumi Mitsutani and Michiko Harigae at the Institute of Environmental Toxicology for their helpful assistance in maintaining the DWH strain. They also deeply appreciate Dr. Katsushi Suzuki, professor emeritus of, and Dr. Hiroetsu Suzuki, professor of Graduate school of Veterinary Medicine and Life Science, Nippon Veterinary and Life Science University, for their assistances in reconfirming the genotyping results. This study was supported in part by the Ministry of the Environment, Japan. Fig. 7. Apparent effects of Kelthane on the weights of maternal thyroid in a reproduction toxicity study. (A) Original data (mean ± standard deviations). Note the apparent bimodal dose–response curve and large standard deviations in the 10, 30 and 100 ␮g/kg/day groups, suggesting the contribution of Tgc.749−1G>T /+ individuals to the increases of group means. Asterisks (* and **) on each column represent statistically significant differences from the control (P < 0.05 and P < 0.01, respectively). (B) Re-evaluated data in which heterozygotes are excluded. A total of 6 females (2, 3 and 1 in the 10, 30 and 100 ␮g/kg/day groups, respectively) were judged to be Tgc.749−1G>T /+ heterozygotes based on the results of DNA sequencing and/or histopathologic observations of the thyroid. Note the monotonic dose-response curve and minimized standard deviations in the low-dose groups, suggesting that Kelthane did not increase thyroidal weights of females at a dose level of 300 ␮g/kg/day or less.

is one of such cases, by which we nearly misunderstood the result of reproduction toxicity study in Wistar Hannover GALAS rats with kelthane, a suspected endocrine disrupter attacking the thyroid. As shown in Fig. 7A (unpublished observation at the Safety Research Institute for Chemical Compounds Co., Ltd.), there was a bimodal increase in thyroidal weights of kelthane-treated females as if kelthane caused a typical low-dose effect of endocrine disrupters [44–46]. However, identification of Tgc.749−1G>T mutation allowed us to re-evaluate these data since variations in thyroidal weights were relatively large in the low-dose groups and histopathologic features of the extremely heavy thyroids observed in that study were the same as those of Tgc.749−1G>T /+ heterozygotes. Subsequent sequencing analyses of genomic DNA retrieved from the formalin-fixed liver revealed that at least some of goitrous rats carried Tgc.749−1G>T mutation, although the remaining rats could not be genotyped precisely. These observations, together with the result of statistical re-evaluation that mean thyroidal weights in the low-dose groups were comparable to that in the control group when Tgc.749−1G>T /+ females exhibiting mutant phenotypes were excluded (Fig. 7B), led to the conclusion that the test compound did not cause any low-dose effects. Our experience should be a warning that an apparent nonmonotonic or bimodal dose-response is not always the indication of endocrine-disrupters’ low-dose effects but can be attributable to a spontaneous mutation of certain genes and/or genetic polymorphisms kept in the colony of outbred experimental animals as suggested previously [39,40]. The scientific countermeasures should include identifications of such mutant genes to provide reliable methodology for precisely genotyping the suspected animals. Otherwise, the result should be reconfirmed by the following

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mrfmmm. 2014.02.003. References [1] H. Aoyama, H. Hojo, K.L. Takahashi, N. Shimizu, M. Araki, M. Harigae, N. Tanaka, N. Shirasaka, M. Kuwahara, N. Nakashima, E. Yamamoto, M. Saka, S. Teramoto, A two-generation reproductive toxicity study of 2,4-dichlorophenol in rats, J. Toxicol. Sci. 30 (Spec No.) (2005) 59–78. [2] H. Aoyama, H. Hojo, K.L. Takahashi, N. Shimizu-Endo, M. Araki, Y. TakeuchiKashimoto, M. Saka, S. Teramoto, Two-generation reproduction toxicity study in rats with methoxychlor, Congenit. Anom. (Kyoto) 52 (2012) 28–41. [3] H. Hojo, H. Aoyama, K.L. Takahashi, N. Shimizu, M. Araki, Y. Takizawa, K. Sakasai, M. Kuwahara, M. Saka, S. Teramoto, Two-generation reproduction toxicity study in rats with 1,1,1-trichloro-2,2-bis(4-chlorophenyl)ethane (p,p -DDT), Congenit. Anom. (Kyoto) 46 (2006) 105–114. [4] N. Shimizu, H. Aoyama, N. Hatakenaka, M. Kaneda, S. Teramoto, An in vitro screening system for characterizing the cleft palate-inducing potential of chemicals and underlying mechanisms, Reprod. Toxicol. 15 (2001) 665–672. [5] N. Shimizu, H. Aoyama, H. Hojo, K.L. Takahashi, M. Kikuta, H. Kushida, H. Kawakatsu, S. Teramoto, Prenatal developmental toxicity studies of 1,1,1trichloro-2,2-bis(4-methoxyphenyl)ethane (methoxychlor) in rats and rabbits, Congenit. Anom. (Kyoto) 41 (2001) 329–337. [6] K.L. Takahashi, H. Aoyama, H. Hojo, N. Shimizu, M. Kikuta, H. Kushida, H. Kawakatsu, K. Ebino, S. Teramoto, Prenatal developmental toxicity studies of 1,1,1-trichloro-2,2-bis(4-chlorophenyl)ethane (p, p‘-DDT) in rats and rabbits, Congenit. Anom. (Kyoto) 41 (2001) 321–328. [7] K.L. Takahashi, H. Aoyama, K. Kawashima, S. Teramoto, Effects of dinoseb, 4,6dinitro-o-cresol, and 2,4-dinitrophenol on rat Sertoli-germ cell co-cultures, Reprod. Toxicol. 17 (2003) 247–252. [8] K.L. Takahashi, H. Hojo, H. Aoyama, S. Teramoto, Comparative studies on the spermatotoxic effects of dinoseb and its structurally related chemicals, Reprod. Toxicol. 18 (2004) 581–588. [9] K.L. Takahashi, N. Takahashi, H. Hojo, M. Kuwahara, H. Aoyama, S. Teramoto, Pathogenetic transition in the morphology of abnormal sperm in the testes and the caput, corpus, and cauda epididymides of male rats after treatment with 4,6-dinitro-o-cresol, Reprod. Toxicol. 22 (2006) 501–507. [10] R. Chia, F. Achilli, M.F. Festing, E.M. Fisher, The origins and uses of mouse outbred stocks, Nat. Genet. 37 (2005) 1181–1186. [11] H. Aoyama, M. Kikuta, N. Shirasaka, H. Hojo, K.L. Takahashi, N. Shimizu, M. Harigae, F. Taguchi, S. Teramoto, Historical control data on reproductive abilities and incidences of spontaneous fetal malformations in Wistar Hannover GALAS rats, Congenit. Anom. (Kyoto) 42 (2002) 194–201. [12] T. Doi, M. Namiki, M. Ashina, N. Toyota, H. Kokoshima, T. Kanno, Y. Wako, M. Tayama, Y. Nakashima, M. Nasu, M. Tsuchitani, Morphological and endocrinological characteristics of the endocrine systems in Wistar Hannover GALAS rats showing spontaneous dwarfs, J. Toxicol. Pathol. 17 (2004) 197–203. [13] A. Shimoi, C. Kuwayama, i.M. Miyauch, C. Kakinuma, M. Kamiya, T. Harada, T. Ogihara, M. Kurokawa, K. Mizuguchi, Vacuolar change in the thyroid follicular cells in BrlHan: WIST@Jcl (GALAS) rats, J. Toxicol. Pathol. 14 (2001) 253–257.

A. Sato et al. / Mutation Research 762 (2014) 17–23 [14] H. Kokoshima, T. Doi, N. Yamada, M. Tsuchitani, Proliferative lesions in thyroid follicular cells of dwarfs derived from Wistar Hannover GALAS rats, Toxicol. Pathol. (2013). [15] C.C. Capen, S.L. Martin, The effects of xenobiotics on the structure and function of thyroid follicular and C-cells, Toxicol. Pathol. 17 (1989) 266–293. [16] A. Hood, J. Liu, C.D. Klaassen, Effects of phenobarbital, pregnenolone-16alphacarbonitrile, and propylthiouracil on thyroid follicular cell proliferation, Toxicol. Sci. 50 (1999) 45–53. [17] A. Hood, Y.P. Liu, V.H. Gattone, C.D. 2nd, Klaassen, Sensitivity of thyroid gland growth to thyroid stimulating hormone (TSH) in rats treated with antithyroid drugs, Toxicol. Sci. 49 (1999) 263–271. [18] T. Yamada, T. Kunimatsu, K. Miyata, S. Yabushita, T. Sukata, S. Kawamura, T. Seki, Y. Okuno, N. Mikami, Enhanced rat Hershberger assay appears reliable for detection of not only (anti-)androgenic chemicals but also thyroid hormone modulators, Toxicol. Sci. 79 (2004) 64–74. [19] K. Weber, R. Ernst, H. Fankhauser, J.F. Hardisty, W. Heider, K. Stevens, Thyroid dysplasia in Wistar Hannover GALAS Rats, J. Toxicol. Pathol. 22 (2009) 247–254. [20] Japanese Association for Laboratory Animal Science (JALAS), Guideline for animal experimentation, Exp. Anim. 36 (1987) 285–288 (in Japanese). [21] W.G. Beamer, L.J. Maltais, M.H. DeBaets, E.M. Eicher, Inherited congenital goiter in mice, Endocrinology 120 (1987) 838–840. [22] P.S. Kim, M. Ding, S. Menon, C.G. Jung, J.M. Cheng, T. Miyamoto, B. Li, S. Furudate, T. Agui, A missense mutation G2320R in the thyroglobulin gene causes non-goitrous congenital primary hypothyroidism in the WIC-rdw rat, Mol. Endocrinol. 14 (2000) 1944–1953. [23] P.S. Kim, S.A. Hossain, Y.N. Park, I. Lee, S.E. Yoo, P. Arvan, A single amino acid change in the acetylcholinesterase-like domain of thyroglobulin causes congenital goiter with hypothyroidism in the cog/cog mouse: a model of human endoplasmic reticulum storage diseases, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 9909–9913. [24] M. Umezu, S. Kagabu, J. Jiang, E. Sato, Evaluation and characterization of congenital hypothyroidism in rdw dwarf rats, Lab. Anim. Sci. 48 (1998) 496–501. [25] H. Grasberger, S. Refetoff, Genetic causes of congenital hypothyroidism due to dyshormonogenesis, Curr. Opin. Pediatr. 23 (2011) 421–428. [26] S.M. Park, V.K. Chatterjee, Genetics of congenital hypothyroidism, J. Med. Genet. 42 (2005) 379–389. [27] M.H. Ricketts, K. Schulz, A. van Zyl, A.J. Bester, C.D. Boyd, H. Meinhold, P.P. van Jaarsveld, Autosomal recessive inheritance of congenital goiter in Afrikander cattle, J. Hered. 76 (1985) 12–16. [28] M.H. Ricketts, M.J. Simons, J. Parma, L. Mercken, Q. Dong, G. Vassart, A nonsense mutation causes hereditary goitre in the Afrikander cattle and unmasks alternative splicing of thyroglobulin transcripts, Proc. Natl. Acad. Sci. U. S. A. 84 (1987) 3181–3184. [29] G.J. Veenboer, J.J. de Vijlder, Molecular basis of the thyroglobulin synthesis defect in Dutch goats, Endocrinology 132 (1993) 377–381. [30] T. Ieiri, P. Cochaux, H.M. Targovnik, M. Suzuki, S. Shimoda, J. Perret, G. Vassart, A 3’ splice site mutation in the thyroglobulin gene responsible for congenital goiter with hypothyroidism, J. Clin. Invest. 88 (1991) 1901–1905. [31] A. Mayerhofer, A.G. Amador, W.G. Beamer, A. Bartke, Ultrastructural aspects of the goiter in cog/cog mice, J. Hered. 79 (1988) 200–203.

23

[32] G. Medeiros-Neto, P.S. Kim, S.E. Yoo, J. Vono, H.M. Targovnik, R. Camargo, S.A. Hossain, P. Arvan, Congenital hypothyroid goiter with deficient thyroglobulin. Identification of an endoplasmic reticulum storage disease with induction of molecular chaperones, J. Clin. Invest. 98 (1996) 2838–2844. [33] Y. Sakai, S. Yamashina, S.I. Furudate, Missing secretory granules, dilated endoplasmic reticulum, and nuclear dislocation in the thyroid gland of rdw rats with hereditary dwarfism, Anat. Rec. 259 (2000) 60–66. [34] M. Baryshev, E. Sargsyan, G. Wallin, A. Lejnieks, S. Furudate, A. Hishinuma, S. Mkrtchian, Unfolded protein response is involved in the pathology of human congenital hypothyroid goiter and rat non-goitrous congenital hypothyroidism, J. Mol. Endocrinol. 32 (2004) 903–920. [35] P.S. Kim, J. Lee, P. Jongsamak, S. Menon, B. Li, S.A. Hossain, J.H. Bae, B. Panijpan, P. Arvan, Defective protein folding and intracellular retention of thyroglobulinR19 K mutant as a cause of human congenital goiter, Mol. Endocrinol. 22 (2008) 477–484. [36] D. Baralle, M. Baralle, Splicing in action: assessing disease causing sequence changes, J. Med. Genet. 42 (2005) 737–748. [37] J.R. Ellis Jr., B. Heinrich, V.F. Mautner, L. Kluwe, Effects of splicing mutations on NF2-transcripts: transcript analysis and information theoretic predictions, Genes. Chromosomes Cancer 50 (2011) 571–584. [38] G.R. Lankas, M.E. Cartwright, D. Umbenhauer, P-glycoprotein deficiency in a subpopulation of CF-1 mice enhances avermectin-induced neurotoxicity, Toxicol. Appl. Pharmacol. 143 (1997) 357–365. [39] D.R. Umbenhauer, G.R. Lankas, T.R. Pippert, L.D. Wise, M.E. Cartwright, S.J. Hall, C.M. Beare, Identification of a P-glycoprotein-deficient subpopulation in the CF-1 mouse strain using a restriction fragment length polymorphism, Toxicol. Appl. Pharmacol. 146 (1997) 88–94. [40] A.B. Okey, M.A. Franc, I.D. Moffat, N. Tijet, P.C. Boutros, M. Korkalainen, J. Tuomisto, R. Pohjanvirta, Toxicological implications of polymorphisms in receptors for xenobiotic chemicals: the case of the aryl hydrocarbon receptor, Toxicol. Appl. Pharmacol. 207 (2005) 43–51. [41] H. Aoyama, S. Teramoto, Y. Shirasu, Heritable pulmonary lobation anomaly in the rat, Teratology 37 (1988) 159–166. [42] S. Fujii, K. Yabe, Y. Kimura, Y. Ito, M. Rokukawa, M. Furukawa, K. Ito, M. Matsuura, M. Kiguchi, Syndactyly lethal: new mutation with multiple malformations occurring in Sprague Dawley rats, Congenit. Anom. (Kyoto) 49 (2009) 262–268. [43] M. Kaneda, S. Teramoto, Y. Shirasu, Anatomical features associated with preaxial duplication (pd): a recessive mutation in the rat, Teratology 40 (1989) 77–84. [44] G.P. Daston, J.C. Cook, R.J. Kavlock, Uncertainties for endocrine disrupters: our view on progress, Toxicol. Sci. 74 (2003) 245–252. [45] R. Melnick, G. Lucier, M. Wolfe, R. Hall, G. Stancel, G. Prins, M. Gallo, K. Reuhl, S.M. Ho, T. Brown, J. Moore, J. Leakey, J. Haseman, M. Kohn, Summary of the National Toxicology Program’s report of the endocrine disruptors low-dose peer review, Environ. Health Perspect. 110 (2002) 427–431. [46] L.N. Vandenberg, T. Colborn, T.B. Hayes, J.J. Heindel, D.R. Jacobs Jr., D.H. Lee, T. Shioda, A.M. Soto, F.S. vom Saal, W.V. Welshons, R.T. Zoeller, J.P. Myers, Hormones and endocrine-disrupting chemicals: low-dose effects and nonmonotonic dose responses, Endocr. Rev. 33 (2012) 378–455.