Introduction

Cataracts are one of the most common eye defects and causes of lens opacity (Liu et al. 2006), which may lead to blindness (Kang et al. 2008). According to the latest assessment, cataracts are responsible for 51 % of blindness worldwide, representing about 20 million people (Quinlan 2015). Causes of cataracts include congenital defects, senility, metabolic disorders and exposure to a variety of physical and chemical agents (Song et al. 1997). Mutations affecting the lens in mice can be identified easily by visual inspection, and a remarkable number of mutant lines have been characterized (Graw 2009). The establishment of an animal model of cataracts is an effective method to elucidate human cataractogenesis (Graw 2004). In particular, mouse models for cataracts are useful for isolating cataract genes and analyzing the mechanism of cataract development (Kondo et al. 2014a), and many types of inherited cataracts exist in mice and have been evaluated developmentally, histologically and genetically (Kohale et al. 2004; Okamura et al. 2003). Cataract lenses display various morphologic features, including the posterior dislocation of the nucleus (Teramoto et al. 2000), swollen lens fibers (Kondo et al. 2011), vacuolation of the epithelium (Maeda et al. 2001) and vacuolation of the lens cortex (Song et al. 1997).

In the present study, we histologically and genetically investigated the characteristics of CF#1/lr mice, a novel mouse cataract model that originated from the CF1 outbred strain.

Materials and methods

Mouse husbandry

This study utilized CF#1/lr mice, which are a new cataract strain derived from the CF#1 strain of mice. In the Central Research Division (Takeda Chemical Industries, Osaka, Japan), mice with cataracts were identified in a colony of CF#1 mice (outbred colony). The cataract mice were isolated from the colony, named CF#1/lr (lens rupture) and bred by inbreeding (more than 100 generations). These animals were obtained in 2003 and are currently bred inhouse at Osaka Prefecture University. Normal BALB/c mice were purchased from CLEA Japan (Tokyo, Japan) and were used for the mating experiment, genomewide screening and histological studies. MSM/Ms mice were kindly donated by Dr. Moriwaki (National Institute of Genetics, Mishima, Japan), and ddY mice were purchased from Japan SLC, Inc. (Hamamatsu, Japan). MSM/Ms, ddY and CF1/b cac mice, which have been previously described (Kondo et al. 2014b), were used as controls in direct sequencing experiments. All mice were maintained under controlled conditions of room temperature (24 ± 1 °C), humidity (55 ± 5 %) and lighting (lights on 0800–2000). The mice received a commercial diet (CE2; CLEA Japan) and water ad libitum. The present study was performed in accordance with the Guidelines for Animal Experimentation of Osaka Prefecture University, Japan.

Observations and histological studies

The mice were observed with the naked eye under non-anesthetic conditions more than once a week after their eyes opened at approximately 2 weeks old. For the histological studies, 2-, 3-, 4-, 5-, 10- and 14-week-old CF#1/lr mice and 14-week-old BALB/c mice were used. Under isoflurane anesthesia, the mice were infused with heparin–saline followed by 10 % neutral-buffered formalin. The eyes were then removed and immersed in the same fixative for 2 days. The eyes were dehydrated by a graded series of ethanol, soaked in butyl alcohol and embedded in paraffin (Tissue Prep; Fisher Scientific, NJ, USA). Sections (4 μm thick) were cut in a plane perpendicular to the anteroposterior axis of the eye and stained with hematoxylin and eosin.

Mating experiment

To elucidate the mode of inheritance, cataract mice were mated with wild-type BALB/c mice in order to obtain heterozygous mutants. Heterozygous mutants were then mated in order to obtain the segregation ratio of affected and unaffected mice. The incidence of lens opacity was determined with the naked eye by visual examination more than once a week from 2 weeks until 5 months of age.

Linkage analysis

Genomic DNA was extracted from the tail of affected backcrossed progeny [CF#1/lr mouse × (CF#1/lr mouse × BALB/c mouse)]. For linkage analysis of up to 2–4 per chromosome, polymorphic microsatellite markers (Sigma Aldrich Japan, Tokyo, Japan) were selected. These markers had typed polymorphisms between CF#1/lr and BALB/c mice. For the primary check, 21–61 DNA samples from affected mice were used. For the secondary check, 5 microsatellite markers on chromosome 3 (Chr 3) and 250 DNA samples from affected mice were used (Table 1). PCR was performed with a thermocycler (GeneAtlas; Astec, Fukuoka, Japan), and products were analyzed by electrophoresis on a 1.5 % agarose gel. The significance of linkage was evaluated by a Chi-square test of independence (degree of freedom = 1) of frequencies of hetero- and homozygote in the affected backcross mice.

Table 1 Linkage analysis of cataract gene on chromosome 3 of affected backcross progenies
Full size table

Direct sequencing

Total RNA was extracted from the lens of CF#1/lr, MSM/Ms, ddY and CF1/b cac mice using TRIzol reagent (Life Technologies, NY, USA) according to the manufacturer’s protocol. Total RNA was resuspended in diethyl-pyrocarbonate-treated water. Single-strand cDNA was synthesized using a PrimeScript 1st strand cDNA Synthesis Kit (Takara Bio, Otsu, Japan). Primers were designed from known sequences (http://www.ncbi.nlm.nih.gov/) to amply contain the coding regions of the mouse Bcar3 (Table 2). PCR was performed using PrimeSTAR GXL DNA Polymerase (Takara Bio). PCR products were purified using a NucleoSpin Gel and PCR Clean-up (Macherey–Nagel; Duren, Germany) and prepared for DNA sequencing on an ABI 3130 Genetic Analyzer (Applied Biosystems, CA, USA) using a BigDye Terminator v 3.1 Cycle Sequencing Kit (Applied Biosystems). To confirm mutations in the genomic DNA, primers were designed to span exon7 of the Bcar3 gene and internal primers were used for sequencing (Table 2).

Table 2 Sequencing primers of BCAR3
Full size table

Results

Macroscopic observation

CF#1/lr mice are shown in Fig. 1. Characteristics of this cataract mouse were opacity of the lens (Fig. 1a) and white fragments in the posterior region of the eye (Fig. 1b). Until 3 weeks of age, CF#1/lr mice did not show opacity of the lens. Opacity appeared in mice at 4 weeks of age, and the rate of opacity was 22.2 %; this gradually increased to more than 50 % at 6 weeks of age (50.9 %). All mice were completely affected by 14 weeks of age. There were no significant differences in incidence and the age of onset between male and female mice.

Fig. 1
figure 1

Macroscopic features of CF#1/lr mice. a Extensive opacity is observed in the eye. b White fragments (arrow) are observed in the posterior region of the eye

Full size image

Histological findings

The lens of 2-week-old CF#1/lr mice showed no obvious pathologic changes (Fig. 2a, b). In the equator region of the lens of 3-week-old mice, lens fibers were vacuolated (Fig. 2c, d). At 4 weeks of age, vacuole denaturation increased in size and some of the lens fibers were swollen (Fig. 2e, f). At 5 weeks of age, the lens ruptured at the posterior pole and the nucleus was dislocated (Fig. 2g, h). In 14-week-old CF#1/lr mice, the lens nucleus was completely extruded (Fig. 2i), while in BALB/c mice, the lens was normal (Fig. 2j). Morphologic abnormalities were not observed in areas besides the lens, such as the retina or the cornea (Fig. 3).

Fig. 2
figure 2

Lens of CF#1/lr mice (a through i) and BALB/c mice (h). a, b Lens of a 2-week-old CF#1/lr mouse is normal in appearance. c, d Equator region of the lens of a 3-week-old CF#1/lr mouse demonstrates vacuoles (asterisks) in the lens cortex. e, f A 4-week-old CF#1/lr mouse exhibits bigger vacuole denaturation (asterisks) in the lens cortex. In addition, swelling (arrows) and abnormal arrangement of the lens fibers are present. g, h Posterior region of the lens of a 5-week-old CF#1/lr mouse. The posterior pole is ruptured and the lens nucleus is dislocated. i A 14-week-old CF#1/lr mouse shows complete extrusion of the lens nucleus. j The eye of a 14-week-old BALB/c mouse is normal. C cornea, LC lens cortex, N lens nucleus, R retina. Scale bars 500 μm (a, c, e, g, i, j) or 100 μm (b, d, f, h)

Full size image
Fig. 3
figure 3

Eyes of 10-week-old CF#1/lr mice. a Whole eye, b cornea, c iris, d ciliary body, e retina. Morphologic abnormalities were not observed in areas besides the lens. Scale bars, 500 μm (a) or 100 μm (b–e)

Full size image

Linkage analysis

Heterozygous progeny from CF#1/lr and wild-type BALB/c mice was phenotypically normal. The ratio of affected to unaffected mice in the offspring of heterozygous mutants was approximately 1:3 [36 (female, 16; male, 20):107 (female, 46; male, 61)]. This result indicates that the mode of inheritance was autosomal recessive and the causative mutation was single. As such, affected backcrossed progeny [CF#1/lr × (CF#1/lr × BALB/c)] was used for linkage analysis. As the result of a genome-wide linkage analysis, the mutation was mapped to chromosome 3, close to the marker D3Mit348. This indicates that there is a linkage between the mutant gene and Chr 3. Therefore, further linkage analyses were carried out with 5 microsatellite markers (D3Mit158, D3Mit79, D3Mit216, D3Mit254 and D3Mit348). The results of further mapping are given in Table 1. The χ2 values for these five polymorphic microsatellite loci ranged from 200.70 to 246.01. All 24 offspring arose as a result of a single crossover event (Fig. 4a). Analysis of the haplotype distribution pattern allowed for the mutated gene to be localized at the 2.4 cM region between D3Mit79 and D3Mit216. The physical distance between D3Mit79 and D3Mit216 was 4.8 Mb (Fig. 4b).

Fig. 4
figure 4

a Distribution of haplotypes in a set of 250 affected offspring from genetic backcrossing (CF#1/lr × [CF#1/lr × BALB/c]). The typed loci are listed on the left. Columns denote specific chromosomes identified in affected backcrossed progeny. Values at the bottom of the figure are the number of progeny that inherited the indicated chromosomal haplotype from the F1 parent. Black squares represent the CF#1/BALB allele; white squares represent the CF#1/CF#1 allele. b Genetic linkage map of chromosome 3. Typed loci are listed on the left of the chromosome, and values to the right are distances between loci. The map shows the location of the cataract gene of CF#1/lr mice (arrow)

Full size image

Direct sequencing

Sequencing of the Bcar3 cDNA identified a 1-bp insertion of cytosine in exon 7 at position 1452 (Fig. 5a). In addition, further mutations, such as single nucleotide deletion, were not detected after the insertion. This insertion results in a frameshift mutation, leading to a premature stop codon and a truncated protein (Fig. 5b). This mutation was not detected in other strains (CF1/b cac, MSM/Ms and ddY). The same mutation was confirmed in the genomic DNA.

Fig. 5
figure 5

CF#1/lr mutation affects the Bcar3 gene. a Exon structure of Bcar3 cDNA is shown. Sequence analysis demonstrates a cytosine insertion at cDNA position 1452, leading to a subsequent amino acid exchange. b Amino acid exchanges from 346 to 381, and codon 382 becomes a premature stop codon

Full size image

Discussion

The present study revealed that a new cataract model, the CF#1/lr mouse, is characterized by lens rupture at the posterior pole at 4–14 weeks after birth. The mutation is inherited in an autosomal recessive fashion, and the anomaly carried a single nucleotide insertion in the Bcar3 gene, leading to a premature stop codon and a truncated protein.

The 4 major morphologies of mouse cataracts are nuclear, cortical, capsular–epithelial and lens extrusion (Smith et al. 1997). The various abnormal changes in cataract lenses include nuclear remnants in the lens fibers (Graw et al. 2002), vacuolated epithelial cells (Kondo et al. 2010; Singh et al. 1995), degeneration of cortical fibers, progressive condensation of the nucleus (Narita et al. 2002) and degeneration of the lens capsule. A lens extrusion cataract involves the thinning and rupture of the posterior lens capsule, leading to the extrusion of lens cortical material into the vitreous (Smith et al. 1997). In CF#1/lr mice, the lens fiber cells formed around the equator were vacuolated and swollen. The posterior lens capsule was ruptured and the lens nucleus was extruded. These results suggest that the abnormal lens feature of CF#1/lr mice is a lens extrusion cataract.

In the 4 major morphologies of mouse cataracts, reports of lens extrusion cataracts constitute the fewest (Smith et al. 1997). In humans, it is reported that lens rupture occurs in addition to renal failure in Alport syndrome (Agrawal et al. 2015). Moreover, external injuries have been reported as causes of lens rupture (Gray et al. 2011). Also in mice, several reports regarding lens extrusion cataracts are available. Abi2 knockout mice were found to display rupture of the posterior lens capsule at the time of birth. This anomaly occurred due to abnormal lens fiber arrangement during development and failure of lens suture formation (Grove et al. 2004). An RLC mouse was also reported to exhibit lens rupture. This mouse showed rupture of the posterior capsule around 50 days after birth and showed abnormal arrangement of the lens fibers at the newborn stage (Teramoto et al. 2000). We did not observe abnormal morphologic changes in the 2 weeks after birth, suggesting that CF#1/lr mice have normal lens fiber arrangement during development.

In this study, we found that the causative gene of cataract in CF#1/lr mice lies on chromosome 3, between D3Mit79 and D3Mit216 (Fig. 4). In this region, we picked 8 genes (Cnn3, Abcd3, Arhgap29, Abca4, Dnttip2, Bcar3, Fnbp1 l, Pde5a) associated with ocular disease (Table 3). Sequence analysis was performed with 8 genes. We discovered a 1-bp insertion of cytosine in exon 7 at position 1452 bp of Bcar3. This mutation was not detected in other strains, and the other 7 genes showed no mutations in their protein coding regions; thus, it is assumed that Bcar3 is the causative gene of CF#1/lr mice. This insertion results in a frameshift mutation and produces a premature stop codon and a truncated protein.

Table 3 List of picked 8 genes located in refined cataract locus between 118.6 and 123.4 Mb on chromosome 3
Full size table

BCAR3 is a 95-kDa protein with an amino-terminal SH2 domain and carboxyl-terminal domain homologous to Ras family guanine nucleotide exchange factors (GEF-like domain) (Gotoh et al. 2000; Vanden Borre et al. 2011). BCAR3 binds to p130Cas (Bcar1), a focal adhesion adapter protein. p130Cas and BCAR3 bind tightly to each other through their c-terminal domains, thus potentially connecting their associated signaling networks (Wallez et al. 2014). This complex signaling regulates epithelial and mesenchymal cell adhesion, motility and responses to growth factors (Makkinje et al. 2012). BCAR3 is expressed strongly in lens epithelial cells found around the equator, known as the transitional zone (McAvoy et al. 1999; Near et al. 2009). In this region, the differentiation of lens epithelial cells into lens fiber cells involves the loss of cell nuclei and cell organelles (Bassnett and Mataic 1997). Since the first abnormality observed was lens fiber vacuolization and swelling in the lens equator, we speculated that alteration in BCAR3 function disrupts normal differentiation of lens epithelial cells to lens fiber cells. Subsequently, abnormal lens fiber cells were less able to interact normally with lens fiber cells, lens epithelial cells and the posterior lens capsule. Finally, the posterior lens capsule was ruptured and the lens nucleus was extruded (Fig. 6).

Fig. 6
figure 6

Possible mechanism of lens extrusion cataract formation

Full size image

It is reported that BCAR3 knockout mouse exhibits the pathology of lens extrusion. This mouse presents with vacuolization and liquefaction of the lens cortex at 3 days after birth. Lens rupture with extrusion of cortical fiber material is detected at 24–33 days after birth. Other phenotypic alterations observed in adult BCAR3 knockout mouse included abnormally deep anterior chambers and anterior synechiaes covering the trabecular meshwork, ectropion uveae and mild to moderate retinal ganglion cell loss. Morphologic abnormalities are restricted to the eye (Near et al. 2009).

Similarities and differences exist between this spontaneous mutation mouse and BCAR3 knockout mice. Morphological characteristics of the two strains are very similar. For example, first vacuolization was observed in the lens cortex and successively the lens ruptured at the posterior pole. Also, abnormalities were limited to ocular disease. However, the two strains differ greatly in the age of onset. BCAR3 knockout mice presented with vacuolated lens fiber 3 days after birth, and lens rupture with extrusion of cortical fiber material was detected at 24–33 days after birth (Near et al. 2009). In contrast, in CF#1/lr mice, the lens fiber was vacuolated at 3 weeks after birth and lens rupture with extrusion of cortical fiber material was detected from 5 weeks after birth. Moreover, visual observation of CF#1/lr mice revealed that all mice developed cataract by 14 weeks of age. In CF#1/lr mice, BCAR3 lacks the GEF-like domain and retains the SH2 domain (Fig. 5a). It is reported that BCAR3 deletion constructs lacking the GEF-like domain have demonstrated a loss of BCAR3-induced cell motility (Cai et al. 2003; Schrecengost et al. 2007). Protein tyrosine phosphatase α (PTPα) regulates integrin signaling, focal adhesion formation and migration. BCAR3 forms a molecular bridge between phospho-PTPα and p130Cas with the SH2 domain of BCAR3 binding to phosphoTyr789 on PTPα and the GEF-like domain of BCAR3 associating with the C-terminal region of p130Cas (Mace et al. 2011; Sun et al. 2012). Interestingly, however, not all BCAR3-associated signaling requires formation of a BCAR3-p130Cas complex. BCAR3-mediated p130Cas phosphorylation requires the amino-terminal BCAR3 SH2 domain, but occurs in the absence of the GEF-like domain, indicating its independence of BCAR3-p130Cas complex formation (Vanden Borre et al. 2011). We have identified that the delay in onset time of CF#1/lr mice might have been caused by disruption of many BCAR3-associated signaling and remaining the few BCAR3-associated signaling. In addition, BCAR3 knockout mice are reported to be on 129 background strains, whereas CF#1/lr mice are on the CF#1 strain; thus, the apparent difference may be due to differences in genetic background. If future studies demonstrate that the function of BCAR3 is not completely knocked out, the usefulness of this model will be considered enhanced.

In conclusion, we report here the first spontaneous mutation in the Bcar3 gene leading to lens rupture. The CF#1/lr mouse represents an exciting tool for studying the molecular biology of cataractogenesis. Moreover, CF#1/lr mice represent a useful new cataract model and will enable investigation into the function of the BCAR3 protein.