A 5-bp Insertion in Mip Causes Recessive Congenital Cataract in KFRS4/Kyo Rats

We discovered a new cataract mutation, kfrs4, in the Kyoto Fancy Rat Stock (KFRS) background. Within 1 month of birth, all kfrs4/kfrs4 homozygotes developed cataracts, with severe opacity in the nuclei of the lens. In contrast, no opacity was observed in the kfrs4/+ heterozygotes. We continued to observe these rats until they reached 1 year of age and found that cataractogenesis did not occur in kfrs4/+ rats. To define the histological defects in the lenses of kfrs4 rats, sections of the eyes of these rats were prepared. Although the lenses of kfrs4/kfrs4 homozygotes showed severely disorganised fibres and vacuolation, the lenses of kfrs4/+ heterozygotes appeared normal and similar to those of wild-type rats. We used positional cloning to identify the kfrs4 mutation. The mutation was mapped to an approximately 9.7-Mb region on chromosome 7, which contains the Mip gene. This gene is responsible for a dominant form of cataract in humans and mice. Sequence analysis of the mutant-derived Mip gene identified a 5-bp insertion. This insertion is predicted to inactivate the MIP protein, as it produces a frameshift that results in the synthesis of 6 novel amino acid residues and a truncated protein that lacks 136 amino acids in the C-terminal region, and no MIP immunoreactivity was observed in the lens fibre cells of kfrs4/kfrs4 homozygous rats using an antibody that recognises the C- and N-terminus of MIP. In addition, the kfrs4/+ heterozygotes showed reduced expression of Mip mRNA and MIP protein and the kfrs4/kfrs4 homozygotes showed no expression in the lens. These results indicate that the kfrs4 mutation conveys a loss-of-function, which leads to functional inactivation though the degradation of Mip mRNA by an mRNA decay mechanism. Therefore, the kfrs4 rat represents the first characterised rat model with a recessive mutation in the Mip gene.


Introduction
Kyoto Fancy Rat Stock (KFRS) strains are inbred strains derived from fancy rats to collect new rat mutations and increase the value of the rat model system. The founder rats are six fancy rats imported to Kyoto University from a fancy rat colony in the USA and six inbred lines (KFRS2/Kyo, KFRS3A/Kyo, KFRS3B/Kyo, KFRS4/Kyo, KFRS5A/Kyo, and KFRS6/ Kyo), including two sublines that were produced by brother-sister mating after the initial cross with a laboratory strain, TM/Kyo or PVG/Seac. The KFRS strains are a potential source of novel rat mutations because mutations, such as those affecting coat and eye colour, occur frequently in fancy rats. Indeed, we have identified 16 mutations that affect coat colour, eye colour, and hair pattern in the KFRS strains [1]. In addition, fancy rat colonies are thought to have been maintained relatively independently of laboratory rats [2]. This characteristic suggests that fancy rats have a unique genetic background that is more similar to that of rat strains that were recently derived from wild rats than to that of laboratory rats; therefore, KFRS strains are likely to become a new powerful tool for forward genetic studies of various pathogenic phenotypes among human populations and for providing valuable biological information regarding human disease.
We found a recessive kfrs4 mutation in a KFRS4/Kyo strain that exhibits bilateral congenital cataract with progressive severe degeneration of the lens fibre cells. Using a positional cloning approach, we discovered a mutation in the major intrinsic protein of eye lens fibre gene (Mip, also known as aquaporin 0 or Aqp0), which is the most abundant membrane protein in the lens fibre cells, constituting more than 60% of the total membrane protein content of these cells [3,4]. Spontaneous mutations in Mip (seven in humans and four in mice) have been associated with congenital cataract [5][6][7][8][9][10][11][12][13][14]. Mip mutant mice exhibit cataract as a result of disrupted lens differentiation [5,7,8,15,16]. These pathological observations suggest that MIP has essential roles in the establish-ment and maintenance of a uniform lens fibre structure and in fibre organisation.
All of the characterised Mip mutations are associated with cataract as the dominant phenotype, which could be explained by a specific dominant negative effect of the Mip mutant allele [17][18][19]. However, the cataract phenotype in KFRS4/Kyo rats is inherited in a recessive fashion, in contrast to known Mip mutations in humans and mice. In this study, we performed genetic, phenotypic and expression analyses of the KFRS4/Kyo rats. Our results suggest that this mutant should be classified as the first identified recessive mutant allele of Mip.

Rats
We used KFRS4/Kyo, BN/CrlCrlj, WIAR/lar and DOB/Oda rats as the wild-type rats in all of the experiments. KFRS4/Kyo and DOB/Oda rats were supplied by The National BioResource Project for the Rat in Japan (NBRP Rat: http://www.anim.med. kyoto-u.ac.jp/nbr/Default.aspx), Kyoto University (Kyoto, Japan). The BN/CrlCrlj and WIAR/Iar rats were purchased from Charles River Japan (Yokohama, Japan) and from the Institute for Animal Reproduction (Kasumigaura, Japan), respectively. All of the procedures involving animals met the guidelines described in the Proper Conduct of Animal Experiments, as defined by the Science Council of Japan, and were approved by the Animal Care and Use Committee of the Tokyo Metropolitan Institute of Medical Science, Tokyo University of Agriculture and Kyoto University.

Phenotypic Analysis
The rat pupils were dilated using Mydrin-P (Santen Pharmaceutical, Osaka, Japan), and both eyes were observed after 5 min. The diagnosis of cataracts was performed by macroscopic examination, as described previously [20].
Both the right and left eyeballs from embryonic and postnatal wild-type, kfrs4/+ heterozygous and kfrs4/kfrs4 homozygous rats were examined for histological analysis. The rats were sacrificed, and both eyeballs were enucleated and fixed in Superfix (Kurabo, Tokyo, Japan) overnight at room temperature. After fixation, specimens were transferred to methanol, dehydrated, embedded in paraffin, and sectioned (5 mm). After removing the paraffin, the sections were stained with haematoxylin and eosin and observed under a Leica DM2500 light microscope.

Genetic Mapping
Genetic mapping of the kfrs4 mutant locus was performed by intercrossing progeny derived from the mating of (KFRS4/Kyo 6 DOB/Oda) F 1 6 KFRS4/Kyo. The backcrossed progeny with a mutant phenotype were easily identified by the overt lens opacity induced by mydriatic instillation. DNA samples from 58 offspring, including 31 cataract-presenting rats of a KFRS4/Kyo and DOB/ Oda cross, were genotyped using 108 polymorphic microsatellite markers selected from The NBRP Rat (Table S1) and six microsatellite markers (Table S2) developed from the rat genomic sequence (Ensembl: http://asia.ensembl.org/Rattus_norvegicus/ Info/Index). Genotyping was carried out using PCR (Table S2) and 4% agarose gel electrophoresis. The map position was refined using the Map Manager computer program [21].

Mutation Analysis
A genomic fragment covering the four coding exons of Mip was amplified from genomic DNA isolated from wild-type rats (DOB/ Oda, BN/CrlCrlj and WIAR/Iar), kfrs4/+ heterozygous rats (F 1 rats from a kfrs4/kfrs4 and wild-type cross), and kfrs4/kfrs4 homozygous rats. The primers Mip_F and Mip_R were used for amplification, and the following primers were used for sequencing: Mip_F1, Mip_F2, Mip_F3, Mip_F4, Mip_R1, and MIP_R2 (Table S2). The PCR products were purified using the QIAquick Gel Extraction Kit (Qiagen, Valencia, CA), sequenced using a BigDye Terminator kit (Life Technologies, Grand Island, NY) and analysed using an Applied Biosystems 31306l Genetic Analyzer.
The kfrs4 allele was genotyped using PCR to amplify genomic DNA and fluorescently labelled using Mip_del_F and Mip_del_R (Table S2)

RT-PCR
Total RNA was isolated from the eyes of 7-week-old wild-type and kfrs4/kfrs4 rats using TRIzol (Life Technologies) and the TRIzol Plus Purification Kit (Life Technologies) according to the manufacturers' protocols. The Superscript VILO cDNA synthesis kit (Life Technologies) was used to generate cDNA using 2 mg of DNase-pretreated total RNA. We also prepared cDNA from wildtype, kfrs4/+, and kfrs4/kfrs4 rats at 7 and 10 weeks for quantitative RT-PCR (qRT-PCR). Primers for Mip and Gapdh were purchased from Qiagen (Valencia, CA). A total of 9 specific transcripts (Cryaa, Crygb, Crygd, Casp6, Lim2, Bfsp1, Bfsp2, Gja3, and Gja8) for lens fibre cells were used as the initial controls and to verify the quality of the fibre cells (Table S2). Subsequently, qRT-PCR was performed as previously described [20].

Antibodies
An anti-MIP rabbit polyclonal antibody (MIP-Cter) that targets a 17-amino acid (aa) peptide in the C-terminal cytoplasmic domain of MIP was previously characterised [8,22] and commercially acquired from Alpha Diagnostic International (San Antonio, TX). We also generated an anti-MIP rabbit polyclonal antibody (MIP-Nter) to a peptide (C+TPPAVRGNLALNT) in the N-terminal region of the MIP peptide from aa 108 to 120 (NM_001105719). We used Ncadherin (CDH2) and b-catenin (CTNNB1) as markers of the lens. Both proteins were evenly expressed in lens fibre cells [23][24][25][26]. Mouse monoclonal anti-CDH2 and anti-CTNNB1 antibodies were purchased from BD Biosciences (San Jose, CA). For a secondary antibody, Horseradish peroxidase (HRP)-conjugated donkey anti-rabbit and HRP-sheep anti-mouse IgG antibodies (GE Healthcare Life Science, Piscataway, NJ) were used for Western blotting. Alexa Fluor-conjugated secondary antibodies were obtained from Invitrogen.

Immunoblotting
Membrane proteins from the eyes of wild-type, kfrs4/+, and kfrs4/kfrs4 rats at 7 and 8 weeks of age were purified as described by Sidjanin et al. [7]. Samples containing approximately 500 ng of protein were separated on 15% SDSpolyacrylamide gels, and the separated proteins were transferred onto a Hybond-P PVDF membrane (GE Healthcare Life Science). MIP protein bands were detected using the anti-MIP-Cter (1:1,000) and -Nter (1:5,000) antibodies, followed by HRP-anti-rabbit (1:20,000) IgG secondary antibodies. ECL Prime Western blotting detection regents (GE Healthcare Life Science) were used for the enhanced chemiluminescent detection of specifically bound antibody, and the membrane was exposed to autoradiography film (GE Healthcare Life Science). The membrane was then stripped and blotted with anti-CDH2 (1:2,000) and CTNNB1 (1:500) antibodies, followed by HRPanti-mouse IgG secondary antibody (1:20,000) as a protein loading control. For quantitative Western blot analyses, the chemiluminescently labelled bands were quantified using the ImageJ software (http://rsb.info.nih.gov/ij). The expression level in wild-type rats was assigned an arbitrary value of 1, and the differences were evaluated using one-way ANOVA with a post hoc Bonferroni multiple comparison test.

Immunohistochemistry
Both the right and left eyes from wild-type (9, 11 and 14 weeks of age), kfrs4/+ (8, 9, 11, 13 and 15 weeks of age), and kfrs4/kfrs4 (9 and 11 weeks of age) rats were used for immunostaining. The procedure for immunohistochemistry of lens paraffin sections was described previously [20], and the sections were stained with anti-MIP-Cter (1:100), anti-MIP-Nter (1:100), anti-CDH2 (1:2,000) and anti-CTNNB1 (1:500). Fluorescence images were obtained using a Zeiss LSM780 confocal microscope and processed using Adobe Photoshop software. For each immunofluorescence labelling and lens fibre cell quantification experiment, we used three images from three different section preparations. The fluorescence intensity of MIP was analysed using the Image J software to analyse confocal images that were taken under identical conditions and adjusted using the intensity of CDH2 and CTNNB1 as a control. The expression level in wild-type rats was assigned an arbitrary value of 1, and differences were evaluated using a twotailed t-test.

Isolation and Phenotypic Characterisation of the Spontaneous kfrs4 Mutant
While characterising rats from the KFRS4/Kyo strain, we observed that all of the rats developed severe bilateral lens opacity throughout the eye within 1 month of birth ( Figure 1A). To classify the cataract type caused by the kfrs4 mutation, lenses from wildtype and KFRS4/Kyo rats were dissected and observed under dark-field microscopy ( Figure 1B-E). The opacity of the KFRS4/ Kyo rat lenses was observed in the anterior nuclear regions at P0 ( Figure 1C). The cataract progressed to a nuclear cataract and then to an all-over opacity that included the cortical fibres at 7 weeks of age ( Figure 1D). The F 1 progeny from the crossbreeding of DOB/Oda, BN/CrlCrlj, and WIAR/lar rats with KFRS4/Kyo rats showed normal lens phenotypes (Table 1). Crossbreeding between the (KFRS4/Kyo 6 DOB/Oda) F 1 rats and KFRS4/ Kyo rats produced 31 offspring with cataract and 27 offspring without cataract. Moreover, mating (KFRS4/Kyo 6 WIAR/lar) F 1 rats with kfrs4 rats produced 23 offspring with cataract and 29 offspring without cataract. These results indicate that the kfrs4 mutation is recessive ( Figure 1E, Table 1).
To define the histological defects in the lenses of kfrs4 mutants, sagittal sections of the eye were prepared from wild-type, kfrs4/+, and kfrs4/kfrs4 rats at various embryonic and postnatal stages (Figure 2). At E15.5, the phenotypes of the lens in kfrs4/kfrs4 homozygotes showed a pattern of development similar to that of kfrs4/+ heterozygotes (Figure 2A, B). However, there were signs of disorganisation in the fibre cells in the nuclear region of the kfrs4/ kfrs4 homozygous lens; the nuclei of the fibre cells remained in this region to a significant degree in comparison to the kfrs4/+ heterozygotes ( Figure 2C, D). Degeneration of the lens fibre was pronounced in both the anterior and posterior regions in the kfrs4/ kfrs4 homozygotes at P0, with lens fibre swelling ( Figure 2E, F). Over the next few weeks, the fibre cells became progressively more disorganised ( Figure 2G, H), and irregularly shaped cells and huge vacuoles were present by 9 weeks of age ( Figure 2I). In contrast, the lenses of kfrs4/+ heterozygotes exhibited compactly packed and uniform fibre cells at 19 weeks of age, similar to the lenses of wild-type rats ( Figure 2J, K).  (Table 1) were genotyped using 114 microsatellite markers (Table S1) on rat chromosome 1-20 to determine the location of the kfrs4 mutation. Using this linkage analysis, we mapped the kfrs4 mutation to an interval of approximately 9.7-Mb between markers Cwf19l2 and D7Wox43 on chromosome 7 ( Figure 3A). This region contains more than 350 protein coding genes, including the kfrs4 candidate region [Ensembl]. Given its genomic locus and its involvement in a similar mutant pathology in humans and mice, Mip was the strongest candidate gene for the kfrs4 mutation. The kfrs4 phenotype was non-recombinant with a microsatellite marker in the 39 region of Mip in 58 [(KFRS4/Kyo 6 DOB/Oda) 6KFRS4/Kyo] N 2 progeny ( Figure 3A).

Identification of the kfrs4 Mutation
By sequencing Mip, we identified a 5-bp insertion at nucleotide position 360 in exon 1 of Mip in the (KFRS4/Kyo 6 DOB/Oda) F 1 and KFRS4/Kyo rats ( Figure 3B). This 5-bp insertion, AACAC, is present as a tandem repeat, most likely resulting from the duplication of the same 5 bp 59 of the insertion ( Figure 3B). We designed a genotyping PCR primer pair to detect a 104-bp fragment for the wild-type Mip allele and a 109-bp fragment for the kfrs4 allele. Based on these PCR amplicons, we were able to confirm the genotypes of the +/+, kfrs4/+, and kfrs4/kfrs4 rats ( Figure 3B). Moreover, the deletion was only observed in the kfrs4 mutants and was not present among a set of 21 rat strains, including other KFRS strains ( Figure S1).
The kfrs4 5-bp insertion is predicted to cause a frameshift mutation that results in truncation of the peptide chain by generating a stop codon at amino acid position 127, and this kfrs4 frameshift mutation causes a truncation of the MIP protein that removes three transmembrane domains (H4, H5 and H6), the HE hemichannels, cytoplasmic loop D, and extracellular loop E ( Figure 4A, B). To confirm this truncation of the Cterminal domain in kfrs4 rats, we performed immunoblot and immunohistochemistry analyses using a rabbit polyclonal antibody, anti-MIP-Cter, that targets a 17-aa peptide in the Cterminal cytoplasmic domain of MIP [Alpha Diagnostic International]. The 28-kDa (previously reported as the MIP band [27]) bands were abundant in extracts from the wild-type and kfrs4/+ heterozygote eyes but were not detectable in the eye extracts from kfrs4 homozygotes ( Figure 4C). Moreover, we generated an antibody, anti-MIP-Nter, to a peptide within an LC extracellular domain in the N-terminal region of the kfrs4 mutation to detect the presence of the mutant MIP protein in kfrs4 mutants. In the eye extracts of wild-type and kfrs4/+ rats, this antibody, as well as anti-MIP-Cter antibody, detected the 28-kDa band ( Figure 4C). In addition, we detected a variant of approximately 23 kDa. Although we have not yet identified whether these bands represent nonspecific antibody binding or a degradation product, these bands could not be detected in the eye extracts of kfrs4/kfrs4 homozygotes.
As reported previously [27], MIP was expressed throughout the fibre cells of the lens in wild-type rats; we also observed strong signals in the anterior fibre cells of adult rats ( Figure 4D). The MIP staining was abolished in the kfrs4 homozygous rats, confirming the truncation of the C-terminal domain in rats with the kfrs4 mutation ( Figure 4D). Immunohistochemistry using the anti-MIP- rats. Wild-type rats show normal transparency (B), but nuclear opacity is observed in the anterior (left) and side (right) views in a lens from a kfrs4/ kfrs4 homozygote at P0 (C). The kfrs4/kfrs4 homozygous rat developed a dense nuclear cataract with full penetrance at 7 weeks of age (D). At 17 weeks of age, both kfrs4/+ heterozygotes and +/+ rats showed normal lens transparency (E). doi:10.1371/journal.pone.0050737.g001 Nter antibody revealed a nearly identical staining pattern to that of anti-MIP-Cter ( Figure 4E). We found that the staining of MIP is completely ablated in the kfrs4/kfrs4 homozygotes.

Reduction in Mip mRNA and MIP Protein Expression by the kfrs4 Mutation
As mentioned above, the frameshift in the kfrs4 mutant produces a truncated MIP protein that is missing several domains in the C-terminal region (Figure 4). To examine the effect of the kfrs4 mutation on Mip mRNA expression, we carried out real-time qRT-PCR using RNA from wild-type, kfrs4/+ heterozygous, and kfrs/kfrs4 homozygous rats. The relative abundance of Mip transcripts in the eyes of kfrs4/+ and kfrs/kfrs4 rats was approximately 34.6 and 7.1% of wildtype levels, respectively ( Figure 5). To confirm this reduction of Mip mRNA in kfrs4/+ and kfrs/kfrs4 eyes and determine whether it caused a loss of lens fibre cells to induce cataracts, we performed qRT-PCR analysis of nine other lens-specific transcripts. Five of these transcripts, Casp6, Lim2, Bfsp1, and Gja8, did not exhibit significant expression changes among the wild-type, kfrs4/+, and kfrs/kfrs4 rats ( Figure 5). The expression of three crystallins, Cryaa, Crygd, and Crygd, exhibited similar levels between the wild-type rats and the kfrs4/+ heterozygous rats, but significant reductions were detected in the kfrs/kfrs4 homozygous rats, suggesting the possibility that the reduction of these crystallins caused structural defects in the lens fibre cells of kfrs/kfrs4 homozygous rats. Although the expression of Bfsp2 (a code lens structural protein) was lower in kfrs4/+ rats relative to wild-type, a change in expression was not detected in the kfrs/ kfrs4 rats ( Figure 5). These results may confirmed that the reduction of Mip expression in kfrs4/+ and kfrs4/kfrs4 rats is not the result of cataract formation but is instead caused by the expression of certain genes that were affected in the kfrs4 mutant. Gja3 (also known as connexin 46) showed a significantly lower expression in kfrs4/+ and kfrs/kfrs4 rats compared with wild-type ( Figure 5). Although we could not explain why the expression of Gja3 is also reduced in kfrs/kfrs4 homozygous rats, the GJA3 protein directly interacts with MIP, mediated by the C-terminal region [28]; therefore, it may be an effect of the reduction in Mip mRNA as a result of the kfrs4 mutation.
Quantitative immunoblotting and immunohistochemistry were also performed to confirm the MIP protein expression levels in wild-type and kfrs4/+ heterozygous rats. Although the difference in protein levels was not significant according to a statistical analysis, MIP was 1.59 times less abundant in kfrs4/+ eyes than in wild-type eyes ( Figure 6A). Quantitative immunohistochemical analysis of MIP in the fibre cells of the anterior lens revealed that the expression level of MIP significantly decreased in the kfrs4/+ rat, whereas the level of CDH2 did not decrease ( Figure 6B). As shown in Figure 6C, the amount of MIP in kfrs4/+ eyes was 33.1% of the level observed in wild-type eyes. Similar results were obtained using the anti-MIP-Nter antibody   [17]). The locations of the six transmembrane domains (H1, H2, H3, H4, H5, and H6), two hemichannels and CTNNB1 as control for quantitative analysis ( Figure 6D-F, Figure S2).

Discussion
We present several lines of evidence demonstrating that a mutation in the Mip gene underlies congenital cataract in kfrs4 mutant rats. First, Mip is located within the candidate region identified for the kfrs4 mutation ( Figure 3A). Second, the kfrs4 mutant phenotypes were consistent with the Mip mutation genotypes in all of the rats examined ( Figure 3B, Figure S1). Third, the absence of MIP-positive bands in the Western blots of eye tissue from kfrs4 mutants and the lack of MIP-specific immunofluorescence in eye sections from kfrs4 mutants indicate the absence of the normal Mip gene product ( Figure 4B-G). The identified mutation is a 5-bp nucleotide insertion within the coding region of the amino terminus of MIP, adding to the catalogue of known Mip mutations that cause cataract in mammals.
In humans and in mice, point mutations [5,6,[10][11][12][13][14], intragenic in-frame deletions [7,8] and frameshift mutations [5,9] in Mip have been characterised to cause dominant phenotypes ( Figure  S3). The cataract phenotype in Mip Hfi/+ heterozygous mice suggested that a 76-bp deletion in Mip is a gain-of-function mutation, consistent with data showing that a normal level of wild-(HB and HE), and the extracellular (LA, LC, and LE) and intracellular (LB and LD) loops are indicated. The schematic illustrates premature truncation of the protein expressed from the Mip kfrs4 allele as a result of the mutation (B). C. Western blots of homogenates prepared from the eyes of +/+, kfrs4/+ and kfrs4/kfrs4 rats detected with anti-MIP-Cter (top) and anti-MIP-Nter (middle) antibodies. Note the absence of a specific band at ,28 kDa (arrows) in the kfrs4/kfrs4 mutant homogenates. The samples were processed for indirect immunofluorescence using an anti-CNTTB1 antibody (bottom). D, E. Immunofluorescence labelling of MIP in the lens of +/+ (left panels) and kfrs4/kfrs4 (right panels) rats at 7 weeks of age. These panels show merged images of the detection of MIP (red)+CNTTB1 (green)+DAPI (blue). The small panels within the large panels show higher magnifications of the dashed boxed areas in the large panels. ALE, anterior lens epithelium. Scale bars = 200 mm (large panels) and 50 mm (small panels). doi:10.1371/journal.pone.0050737.g004 Figure 5. Quantitative analysis of selected lens fibre specific gene expression in +/+, kfrs4/+, and kfrs4/kfrs4 rats. A. Relative levels of mRNA in the lenses of +/+ (blue bars), kfrs4/+ (orange bars), and kfrs4/kfrs4 (red bars) rats at 7 weeks of age. mRNA expression was detected using real-time quantitative RT-PCR analysis. *P,0.05, **P,0.01, ***P,0.001. doi:10.1371/journal.pone.0050737.g005 type MIP protein expression observed in Mip Hfi/+ could not protect the lens from cataract formation [7]. Another Mip mutation in mice, Mip Cat-Tohm is a 12-bp deletion in Mip that does not alter the open reading frame, but these mutant mice exhibited a more severe cataract phenotype than did Mip knockout mice [15,16] with a null mutation in Mip [8]. Interestingly, the lens fibre cells of a transgenic mouse that expressed both the wild-type and the Mip Cat-Tohm mutated Mip showed more severe degeneration at birth than did the lens cells of Mip Cat-Tohm/Cat-Tohm homozygotes, which expressed only Mip Cat-Tohm . In addition, the E134G/T138R mutant of human MIP leads to congenital cataract and results in a loss of water permeability owing to a failure in protein trafficking to the plasma membrane in homozygotes [6]; however, when the E134G/T138R mutant is co-expressed with wild-type MIP protein, the mutant protein reaches the plasma membrane but causes tetramer instability and a loss-of-function of wild-type MIP [17]. Moreover, FRET analysis of human Mip ADC2 and wild-type MIP coexpressed in mammalian cells demonstrated that this mutation is dominant because the hetero-oligomerisation of the wild-type and mutant MIP molecules traps the wild-type MIP in the endoplasmic reticulum [18]. These studies indicate that mutated versions of the MIP protein have a strong dominantnegative effect on lens transparency and that cataract in these mouse and human cases is caused by gain-of-function mutations. However, we characterised kfrs4 as a recessive mutation because kfrs4/+ heterozygous rats exhibit a lens structure that is similar to that of wild-type rats until the late stages of development (Figure 1, 2, Table 1), and the kfrs4 mutation was effectively mapped in a linkage study using backcrosses to follow the recessive phenotype ( Figure 3A). The kfrs4/+ heterozygous and kfrs4/kfrs4 homozygous rats appear to have reduced Mip mRNA expression ( Figure 5). Moreover, the mutant MIP protein in kfrs4 rats presumably lacks 136 aa from the C-terminus because MIP signals were not detected by Western blotting and immunohistochemistry when an anti-MIP-Cter antibody that recognise C-terminal region of MIP was used (Figure 4, 6). These results suggest that the frameshift caused by the kfrs4 mutation leads to functional inactivation though the rapid degradation of mRNA by nonsense-mediated mRNA decay; this observation supports the characterisation of kfrs4 as a loss-of-function mutation in Mip, which may be explained by a partial loss of MIP function via dosage effects. Indeed, Mip null mutations may be characterised as semi-dominant because heterozygous mutants present with a milder cataract phenotype than do homozygotes [22,27]. In addition, Mip Cat-Ft/+ (in which aa 203-263 at the C-terminus has been replaced with a transposon sequence) and Mip Lop/+ (A51P) heterozygous mutants also show mild phenotypes compared with the corresponding homozygous mutants [5]. Furthermore, the phenotype can be rescued in chimeric mice, which have lenses containing equal numbers of mutant and wild-type fibre cells [29].
Finally, the difference in heterozygous phenotypes and the inheritance mode in kfrs4 rats may result from the diversity of the genetic background and species differences between rats and mice compared to humans, but we lack the evidence to make this determination. In humans, individuals with MIP mutations have disorders of multiple genetic origins that exhibit marked phenotypic heterogeneity [6,[10][11][12][13][14]30]. In particular, it is known that cataracts caused by different types of mutant MIP have different phenotypes, suggesting that Mip mutations cause phenotypic heterogeneity [6]. The diverse phenotypes exhibited by the Mip gene suggest that other genetic modifiers are likely to influence the expression and function of MIP in lens development and in lens fibre formation.