A new missense mutation in the paired domain of the mouse Pax3 gene.

Mice with dominant white spotting occurred spontaneously in the C3.NSY-(D11Mit74-D11Mit229) strain. Linkage analysis indicated that the locus for white spotting was located in the vicinity of the Pax3 gene on chromosome 1. Crosses of white-spotted mice showed that homozygosity for the mutation caused tail and limb abnormalities and embryonic lethality as a result of exencephaly; these phenotypes were analogous to those found in other Pax3 mutants. Sequence analysis identified a missense point mutation (c.101G>A) in exon 2 of Pax3 that resulted in a methionine to isoleucine conversion at amino acid 62 of the PAX3 protein. This mutation site was located in the N-terminal HTH (helix-turn-helix) motif of the paired domain of Pax3, which is necessary for binding to DNA and is highly conserved in vertebrate species. Alteration of DNA binding affinity was responsible for embryonic lethality in homozygotes and white spotting in heterozygotes. We named the mutant allele as Pax3Sp-Nag. The C3H/HeN-Pax3Sp-Nag strain may be useful for analyzing the function of Pax3 as a new model of the human disease, Waardenburg Syndrome.

Introduction

A number of gene mutations are associated with white spotting in mice, and mutations of orthologous genes have been shown to be responsible for human piebaldism and Waardenburg Syndrome (WS), which causes pigmentation defects and deafness [3, 12]. Many alleles of mouse genes have been identified, and functional analyses of them have provided considerable insights into the regulatory networks of melanocyte development [3, 12]. The present report describes a spontaneous mutation that was first identified in a female mouse that showed small white spotting on its belly; the mouse belonged to the C3.NSY-(D11Mit74-D11Mit229) congenic strain. This new white spotting phenotype showed autosomal dominant inheritance, and we provisionally named the mutation Dws(dominant white spotting). We backcrossed mutant mice to C3H/HeN, the recipient strain of the C3.NSY-(D11Mit74-D11Mit229) congenic strain, and generated the C3H/HeN-Dws strain (Fig. 1). These mice were used to identify and characterize the Dws mutation.

Fig. 1.

Coat color of white-spotted (left: Dws/+) and wild-type (right: +/+) mice. (A) Ventral view, (B) dorsal view.

Materials and Methods

Mice

C3.NSY-(D11Mit74-D11Mit229) mice were obtained from the Experimental Animal Division, Riken BioResource Center (Tsukuba, Japan). C3H/HeN (C3H) mice were purchased from Charles River Laboratories Japan (Yokohama, Japan), and C57BL/6J (B6J) mice were purchased from Japan SLC (Hamamatsu, Japan). All mice were fed a commercial CE-2 diet (CLEA Japan, Tokyo) and had ad libitum access to water. The mice were bred in a pathogen-free facility at the Institute for Laboratory Animal Research, Graduate School of Medicine, Nagoya University, and maintained under a controlled temperature of 23 ± 1°C, humidity of 55 ± 10%, and a light cycle of 12 h light (from 09:00 to 21:00)/12 h dark (from 21:00 to 09:00). Animal care and all experimental procedures were approved by the Animal Experiment Committee, Graduate School of Medicine, Nagoya University, and were conducted according to the Regulations on Animal Experiments of Nagoya University.

Genetic mapping

Approximately half of the (B6J × C3H/HeN-Dws) F1 male and female mice showed white belly spotting, confirming that this character is controlled by autosomal dominant inheritance. White-spotted (B6J × C3H/HeN-Dws) F1 mice were backcrossed to B6J mice. Sixty-four white-spotted and 89 normal (no white spotting) mice were obtained from a total of 153 progeny in the backcross generation. Genomic DNA was prepared from tail tissue by salt/ethanol precipitation. PCR genotyping of simple sequence length polymorphism (SSLP) markers was performed according to standard methods. PCR products were separated by electrophoresis on a 4% NuSieve agarose gel (FMC, Rockland, ME, USA) and visualized by UV light after ethidium bromide staining. Linkage of markers to white spotting was evaluated by χ2-tests. Genotype distribution was compared with the theoretical expectation based on Mendelian segregation.

Phenotype of homozygous (Dws/Dws) embryos

Crosses between heterozygous (Dws/+) mice failed to produce any overt homozygous offspring in a preliminary study. To confirm the embryonic lethality of the homozygous mutant genotype, we examined 18 pregnant white-spotted (B6J × C3H/HeN-Dws) F1 females that had been crossed with white-spotted (B6J × C3H/HeN-Dws) F1 males. The pregnant females were euthanized by isoflurane inhalation on days 14.5 to 17.5 of pregnancy, and embryos were rapidly extracted for analysis. The genotypes of the embryos were determined as described above.

Sequence analysis

Our mapping analysis identified Pax3 as a candidate for the mutation. To confirm this, we analyzed the gene structure in the mutant by amplifying the coding region (9 exons) and splice junctions using the primer pairs listed in Table S1. Primer sequences were designed using genome assembly data (GRCm38.p4) as the reference sequence. PCR amplification was performed using a KOD Plus (TOYOBO, Osaka, Japan), and amplification products were purified using a QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany). PCR products were sequenced using the dideoxy chain-termination method with a BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and then analyzed on an ABI 3500 (Applied Biosystems) automated DNA sequencer.

Results

SSLP markers that showed polymorphisms between B6J and C3H were used to scan eight candidate genes (Edn3, Ednrb, Kit, Kitl, Mitf, Pax3, Sox10, and Snai2) for linkage with the white spotting phenotype [3]. The segregation data for the 64 white-spotted mice in the backcross generation are shown for each candidate gene in Table 1. All of the white-spotted mice were heterozygous (B6J/C3H) at the D1Mit215 locus, which is close to Pax3. The other seven markers showed no indication of linkage to white spotting. Our results indicated that the Dws locus was likely located in the vicinity of Pax3. Five SSLP markers on chromosome 1 were used to confirm the location of the mutation and map it more precisely (Table 2). D1Mit215, which was the closest of these markers to Pax3, showed the largest deviation from the expected ratio. These results suggested that Dws was likely a mutation of the Pax3 gene. The presence of 11 heterozygous (B6J/C3H) mice at the D1Mit215 locus with a normal (no white spotting) phenotype in the backcross generation indicated the incomplete gene penetrance of Dws. The penetrance appeared to be about 85.3% (64/64+11; Table 2).

Table 1.

Segregation ratios of neighboring markers at 8 candidate genes in the 64 white-spotted backcross progeny

Marker	Chr.	Position (Mb)	Candidate gene (Position Mb)	Ratio (homo: hetero)	χ2-value
D1Mit215	1	78.21	Pax3 (78.10–78.20)	0 : 64	64.00
D2Mit229	2	168.78	Edn3 (174.76–174.78)	32 : 32	0.00
D5Mit355	5	71.9	Kit (75.57–75.67)	26 : 38	2.25
D6Mit149	6	106.01	Mitf (97.81–98.02)	29 : 35	0.56
D10Mit96	10	99.56	Kitl (100.02–100.10)	34 : 30	0.25
D14Mit92	14	91.11	Ednrb (103.81–103.84)	29 : 35	0.56
D15Mit63	15	65.25	Sox10 (79.15–79.16)	35 : 29	0.56
D16Mit131	16	7.32	Snai2 (14.71–14.71)	28 : 36	1.00

Table 2.

Genetic mapping of the Dws locus using SSLP markers around the Pax3 gene

Marker	Position (Mb)	White spotting (homo : hetero)	Normal (homo: hetero)	χ2-value
D1Mit484	76.39	1 : 63	77 : 12	107.53
D1Mit132	77.15	0 : 64	77 : 12	111.47
Pax3	78.10–78.20
D1Mit215	78.21	0 : 64	78 : 11	114.44
D1Mit81	85.76	0 : 64	74 : 15	103.11
D1Mit49	87.2	1 : 63	74 : 15	99.17

Embryonic lethality of the homozygous genotype

We analyzed F2 embryos produced by intercrosses of white-spotted (B6J × C3H/HeN-Dws) F1 mice (Table 3). Malformed embryos, characterized by exencephaly and tail and limb abnormalities (Fig. 2), were present from E14.5 to E17.5. The phenotype of the malformed embryos was similar to those described for other Pax3 mutants, such as Pax3, Pax3, Pax3, Pax3, Pax3, and Pax3 [2, 5,6,7, 10, 14]. There was an extremely small number of homozygous (C3H/C3H) embryos, and litter size tended to be reduced, with the occasional occurrence of fetal death (data not shown) at E17.5. Genotyping analysis using D1Mit215 indicated that all the abnormal embryos were homozygous for the C3H-derived allele. The rate of malformed homozygous embryos (C3H/C3H) tended to increase with embryonic age. These facts suggested the embryonic lethality of Dws/Dws mice.

Table 3.

Incidence of malformed embryos of each genotype for D1Mit215 on different days of gestation

Fetal age	No. of embryos	Average litter size	Number of embryos(B6J/B6J:B6J/C3H:C3H/C3H)	Number of malformed embryos(B6J/B6J:B6J/C3H:C3H/C3H)	Incidence *
E14.5	33	8.3 (33/4)	9 : 18 : 6	0 : 0 : 4	67% (4/6)
E15.5	29	7.3 (29/4)	12 : 11 : 6	0 : 0 : 5	83% (5/6)
E16.5	33	8.3 (33/4)	6 : 21 : 6	0 : 0 : 5	83% (5/6)
E17.5	39	6.5 (39/6)	11 : 25 : 3	0 : 0 : 3	100% (3/3)

Fig. 2.

E15.5 embryos of the F2 generation. D1Mit215 genotypes are shown in parentheses. (A) Exencephaly, (B) tail and limb abnormalities.

Sequence of the Pax3 gene in mutants

A missense point mutation, G to A, was identified at nucleotide 101 in exon 2 of the Pax3 gene (Fig. 3). This alteration resulted in a methionine to isoleucine conversion at amino acid 62 (p.Met62Ile) in highly conserved region among vertebrate species of the PAX3 protein (Fig. 3 and Table 4 ).

Fig. 3.

(A) Missense mutation in the Pax3 gene of the Dws mutant. A c.101 G>A transition in exon 2 causes the conversion of Met to Ile at amino acid 62. (B) Schematic diagram of the mouse PAX3 protein including homeodomain and paired domain with N-terminal HTH motif. Numbers correspond to the amino acid sequence.

Table 4.

Multiple sequence alignments of PAX3 protein from various species

Discussion

Pax3 encodes a transcription factor that is important in melanocytes and influences melanocytic proliferation, resistance to apoptosis, migration, lineage specificity, and differentiation [9]. The gene possesses four structural domains: a paired domain, an octapeptide motif, a homeodomain, and a transactivation domain. The paired domain, named after the two helix-turn-helix (HTH) motif-containing sub-domains within it, binds DNA in addition to facilitating interactions with other proteins [9]. The N-terminal HTH motif (amino acid residue 54–94) is necessary and sufficient for binding to DNA, and is highly conserved among vertebrate species [9]. The mutation identified here is located within the N-terminal HTH motif in the paired domain of the Pax3 gene (Fig. 3). Several human PAX3 mutations occur as missense or frameshift mutations within the highly conserved region of exon 2, which gives rise to part of the paired domain [9, 11]. Although exactly the same mutation have not been reported, a mutation on same amino acid (p.Met62Val) within the paired domain has been reported in human PAX3. The patient with this mutation exhibited dystopia canthorum and confluent eyebrows as WS features [8]. It is likely that the amino acid change (p.Met62Ile) on the N-terminal HTH motif identified in this study would alter the DNA binding affinity of the paired domain and would underlie the loss of Pax3 function. We formally named Dws as Pax3.

Modifier genes of Pax3 mutant alleles have been reported [1, 7], and they expand our understanding of the complex network governing melanocyte development. Large size variations in white spotting on the belly were observed among the 64 white-spotted mice of the backcross generation, but such size variation was relatively small in the C3H/HeN-Pax3 strain. This suggests that Pax3 modifier genes are present in the B6J genetic background. To screen the interaction between modifier and mutant genes, thirteen backcross mice that showed relatively large white spots were selected and genotyped using markers located in the vicinity of seven WS candidate genes (Edn3, Ednrb, Kit, Kitl, Mitf, Sox10, and Snai2). None of the markers showed a distribution disequilibrium from the expected ratio (Table S2). This suggests that the size variation in white spotting mediated by Pax3 allele was not a consequence of a direct interaction with other candidate WS genes.

During breeding of C3H/HeN- Pax3 mice, males with white spotting were crossed with wild type females. The incidence of white-spotted offspring was 41.2% (21/51), which is lower than the expected ratio of 50% of the mice with the mutation phenotype. The reduction in the incidence of white-spotted mice from the expected ratio in the C3H/HeN-Pax3 strain is in accordance with incomplete gene penetrance (85.3%). White-spotted mice with tail abnormalities, e.g., looped tails, were rarely observed (<1%) in this strain. The looped tail feature was not passed on to the next generation. Although the cause is not known, a similar phenomenon has been reported for other Pax3 mutant mice such as Pax3, Pax3, and Pax3 [6, 10, 14]. Heterozygous (Pax3/+) mice with the p.Met62Ile mutation do not appear to show deafness. Similarly, the patient with the p.Met62Val mutation of PAX3 did not show deafness [8]. The missense mutation at amino acid 62 did not directly affect hearing ability, although deafness is a frequent feature of WS patients and is observed at a rate of about 60% in case of type 1 WS, most of which are caused by PAX3 mutations [11]. The underlying genetic mutation and the phenotype of several Pax3 mutant alleles have been determined and listed in Table 5 [2, 4, 6, 7, 10, 13, 14]. Although a few Pax3 mutants, such as Pax3, Pax3, and Pax3, have a mutation in paired the domain [4, 13, 14], the Pax3 allele reported here is the first missense mutant in the N-terminal HTH motif of the paired domain of Pax3 gene in mice.

Table 5.

Pax3 mutations and mutant phenotypes

Allele	Mutation	Phenotype
		Homozygous	Heterozygous
Pax3Rwa	841 bp deletion spanning the promoter region and intron 1	Embryonic lethal	Belly white spotting
Pax3Sp-7H	Missense mutation (V38G) in exon 2 (paired domain)	Embryonic lethal	Belly white spotting
Pax3Sp-d	Missense mutation (G42R) in exon 2 (paired domain)	Perinatal lethal	Belly white spotting
Pax3Sp-Nag	Missense mutation (M62I) in exon 2 (paired domain)	Embryonic lethal	Belly white spotting
Pax3Sp-1Wli	Nonsense mutation (K107X) in exon 2 (paired domain)	Embryonic lethal	Belly white spotting
Pax3Sp	Point mutation of splice acceptor of intron 3	Embryonic lethal	Belly white spotting
Pax3Sp-2H	32 bp deletion in exon 5 (homeodomain)	Embryonic lethal	Belly white spotting
Pax3Sp-1Xzg	Missense mutation (N269D) in exon 6 (homeodomain)	Embryonic lethal	Belly white spotting

In conclusion, the C3H/HeN-Pax3 strain, which carries a new missense mutation in the paired domain of the mouse Pax3 gene, may be useful for analyzing the function of Pax3 with respect to human WS. This strain has been deposited in the Riken BioResource Center (Tsukuba, Japan) under the catalog number RBRC06568.