Characterization of a Novel Gja8 (Cx50) Mutation in a New Cataract Rat Model

Abstract

Purpose: To describe a novel spontaneous cataract inbred strain isolated from large-scale breeding SD rats, identify the responsible gene mutation, and understand how this mutation affects lens function.

Methods: Exome sequencing of 12 cataract-associated genes was performed in the affected and healthy relatives. Sequences of rat wild-type or mutant gap junction protein alpha 8 gene (Gja8) were transfected into cells. The expression level of protein was assayed by Western blot analysis. Subcellular localization of connexin 50 (Cx50) was analyzed in confocal fluorescent images. Wound-healing, 5-ethynyl-2ʹ-deoxyuridine incorporation, and attachment assay were performed to characterize the cell migration, proliferation and adhesion.

Results: The abnormality was found to be inheritable in an autosomal semi-dominant pattern through different mating patterns. We found a G to T transversion at codon 655 in Gja8, leading to a substitution of valine by phenylalanine (p.V219F). Gja8^V219F/+ heterozygotes expressed nuclear cataract while Gja8^V219F/V219F homozygotes manifested microphthalmia in addition to cataract. Histology revealed fiber disorders and loss of organelle-free zone in the mutant lens. Cx50^V219F altered its location in HeLa cells and inhibited the proliferation, migration and adhesion abilities of HLEB3 cells. The mutation also reduced the expression of focal adhesion kinase and its phosphorylation.

Conclusions: The c.655G>T mutation (p.V219F) is a novel mutation in Gja8, inducing semi-dominant nuclear cataracts in a new spontaneous cataract rat model. The p.V219F mutation altered Cx50 distribution, inhibited lens epithelial cell proliferation, migration, and adhesion, and disrupted fiber cell differentiation. As a consequence, the nuclear cataract and small lens formed.

The congenital cataract fulfills an important role in vision impairment among infants and children. About 22.3% of congenital cataracts are hereditary¹ whereas most of them inherit in a dominant autosomal trait. Although patients with cataracts can be treated by the lensectomy and lens implantation, there remains controversy over the surgical timing and methods of operation. Also, the complications after surgery are common and complicated, including glaucoma² and posterior capsule opacification.³ To further explore the underlying mechanism of inherited congenital cataract, several congenital animal models have been raised and used in the research.^4,5 

Connexin 50 (Cx50), encoded by gap junction protein alpha 8 gene (Gja8), is a kind of transmembrane protein widely distributed in the plasma membrane of lens epithelial cells and fiber cells, forming channels between cells to regulate the intercellular communication. The dysfunction of Cx50 has been reported to cause diverse cataract phenotypes, microphthalmia, anophthalmia in human,⁶ as well as in animals.^7,8 

Here we breed a new congenital cataract rat strain and uncover a novel mutation in Cx50, p.V219F, which is located in the fourth transmembrane region (TM4) of Cx50. This mutation induces microphthalmia and nuclear cataract which inherit in a semi-dominant autosomal fashion. The V219F mutation is the first mutation found in the TM4 region of murine Cx50, which expands the catalogue of Gja8 variants in murine models. 

Methods

Animals

In the very beginning, five rats (three males and two females) with spontaneous cataract were observed in the colony of Sprague Dawley (SD) rats raised in our laboratory. We maintained this mutant strain through over 30 generations of brother-sister mating and named this new spontaneous cataract rat model “cataractous SD rats discovered by Tian” (TSC rat). 

The Fisher 344 (F344) rats (approval no. SCXK2018-0006) and Brown Norway (BN) rats (approval no. SCXK2018-0006) were purchased from Sino-British SIPPR/BK Laboratory Animal Company Limited (Shanghai, China). The SD rats were from the laboratory animal room of Jinling Hospital (approval no. SCXK2017-0018). 

Rats were maintained in a barrier-maintained animal facility and were housed in standard, autoclavable cages maintained at 22°C ± 2°C and 55% ± 5% relative humidity. The food and filtered water were available at all time. All experimental procedures were followed in accordance with protocols approved by the Ethics Committee of Jinling Hospital (approval no. 2020JLHTXYDWLS-009) and adhered to the statement of the Association for Research in Vision and Ophthalmology for the use of animals in ophthalmic and vision research. 

Inheritance Assessment

Pairs of a TSC rat and a wild-type F344 rat were mated to produce F1 progeny. All the F1 progeny manifested a distinct ocular phenotype, compared with the TSC rats. The F1 pups were then backcrossed to their wild-type F344 parents to produce backcross rats. The F1 offspring were also crossed reciprocally to produce F2 intercross rats. 

We crossed a TSC female rat with a male wild-type BN rat and their offspring were mated with their parents to produce backcross rats. The tails of the mutant rats and wild-type SD rats were saved for exome sequencing of candidate cataract-associated genes. 

Statistical Analysis

Statistical analyses were performed by SPSS 18.0 and GraphPad Prism 7.0. All representative experimental data are presented as the mean value ± standard deviation (SD). Normal distribution was assessed with Shapiro-Wilk normality test. Two group comparisons were calculated by two-tailed t tests and the rank data were analyzed by Mann-Whitney test. The analysis of variance test was used to compare the eyeball weights of three phenotypes while Dunnett t test was used to compare two of the three groups. A P value < 0.05 was considered statistically significant. The detailed experimental procedures are described in the Supplementary File S1 and Supplementary Table S1. 

Results

Inheritance and Phenotypes of the TSC Rats

All offspring obtained from the TSC full sib mating developed pulverulent nuclear cataract and microphthalmia, with a sex ratio of about 1:1. The cross between the TSC rats and wild-type F344 rats produced F1 offspring (mating 1), among which we observed a distinct ocular phenotype. All the F1 progenies displayed dense nuclear cataract (three litters, n = 20). When F1 females and males were crossed reciprocally (mating 2), their offspring were at a ratio of about 1:2:1 (11 litters, ratio 30:72:33) for pulverulent nuclear cataract with microphthalmia, nuclear cataract, and normal phenotype, respectively. The cross between the pups with normal phenotype (mating 4) produced offspring were all phenotypically normal with respect to the eyes (two litters, n = 20). All pups from the TSC full sib mating (mating 3) exhibited pulverulent nuclear cataract with microphthalmia (three litters, n = 48). The F1 rats were then backcrossed to the wild-type F344 rats (mating 5) and the offspring developed either a nuclear cataract or normal phenotype in 1:1 ratio (four litters, ratio 19:24). The findings have been documented in detail in Table. Data support the hypothesis that the TSC rats act in an autosomal semi-dominant pattern, following the rules of monogenic Mendelian inheritance. We propose that pulverulent nuclear cataract and microphthalmia occur in a homozygous state whereas dense nuclear cataract occur in heterozygotes. No additional apparent external defects were uncovered, except the eyes. 

Table.

 

Incidence Ratio of Each Phenotype in the Offspring Produced by Different Mating

 

Connexin 50 Mutation V219F

A total of 12 congenital cataract-related genes were selected for exome sequencing. It turned out that cataractous rats (n = 3) had three missense mutations in Gja8 compared with the normal ones (n = 3). No further pathogenic mutations were found in other 11 selected genes (Epha2, Foxe3, Gja1, Gja3, Hsf4, Lim2, Nhs, Pax6, Pitx3, Sox2, Tmem114). The sequencing results of SD healthy controls (n = 23) excluded the other two point mutations, and a G to T transversion was finally identified pathogenic in codon 655 (c.655G > T). This mutation results in a valine to phenylalanine change at position 219 (p.V219F). To confirm the mutation, 22 cataractous rats with microphthalmia and 22 cataractous rats were included in the further Gja8 coding sequence analysis. It turned out that rats with pulverulent nuclear cataract and microphthalmia were homozygotes for V219F. Animals having nuclear cataract were heterozygotes for V219F. The unaffected subjects and normal controls did not have the mutation (Fig. 1B). The results allowed us to categorize the phenotype of the animals into normal phenotype, cataract and cataract with microphthalmia, and to assign them genotypes Gja8^+/+, Gja8^V219F/+ and Gja8^V219F/V219F, respectively. 

Figure 1.

 

The distinct phenotypes and genotypes of the TSC rats. (A) The photos of rat eyes and lenses obtained from 6-week-6-day-old littermates. The Gja8^+/+ rats were with transparent lenses. The Gja8^V219F/+ heterozygotes exhibited typical nuclear opacity in the lens center. Opaque center with ashes-like opacity around was observed in small lenses from Gja8^V219F/V219F rats. The cataractous lenses were smaller than the wild type ones, with a major difference between homozygous lenses and wild-type ones. Scale bar: 1 mm. (B) Partial nucleotide sequence of Gja8 from affected and unaffected individuals. The sequence in rats with cataract and smaller sized eyeballs showed a heterozygous G→T transversion (indicated by the arrow), resulting in a substitution of valine by phenylalanine at amino acid residue 219. The sequence in animals with pulverulent nuclear cataract and microphthalmia showed a homozygous G→T transversion (indicated by the arrow), resulting in a substitution of valine by phenylalanine at amino acid residue 219. Unaffected members and the healthy controls lacked this nucleotide change. (C) The partial alignments of Gja8 sequence with the corresponding segments in diverse species. The 219th valine is highly conserved in connexin 50 proteins, indicated by the arrow.

The Ocular Phenotype of Gja8^V219F/V219F Rats

Gja8^V219F/V219F homozygotes were featured with pulverulent nuclear cataract and extreme reduction in eye size. The eyeballs obtained from Gja8^V219F/V219F animals aged between 6w1d and 6w3d (n = 30, mean ± SD = 0.04473 g ± 0.00353 g) weighed approximately a half of the Gja8^+/+ members (n = 25, mean ± SD = 0.09508 g ± 0.00403 g) of the same age (P < 0.001) (Fig. 2B). Moreover, lenses dissected from six-week-six-day-old homozygous rats were smaller than those obtained from Gja8^+/+ littermates of the same age in appearance (Fig. 1A). 

Figure 2.

 

Eyes of F1 intercross rats aged from six weeks one days to six weeks three days. (A) Whole eyeballs provided by Gja8^V219F/V219F homozygotes, Gja8^V219F/+ heterozygotes, and Gja8^+/+ unaffected members are shown. The cataractous eyeballs were smaller than the wild-type eyes. The difference was obvious between the Gja8^V219F/V219F eyes and Gja8^+/+ eyes. Scale bar: 1 mm. (B) The average weight of left and right eyes is shown for each genotype. All values are represented as the mean ± SD. There are no significant differences between males and females with the same genotype (P = 0.935, 0.302 and 0.560 for Gja8^V219F/V219F, Gja8^V219F/+, and Gja8^+/+, respectively) whereas the differences between any two of the three groups are significant (P < 0.001 among all pairs). Statistical differences between genders of the same genotype were calculated by two-tailed t tests. Analysis of variance testing was used to compare the statistical significance in three phenotypes while Dunnett t test was used to compare two of the three groups. ***P < 0.001; ns, not significant. P < 0.05 was considered statistically significant.

The opacity occupied nearly the whole lens when at eye opening which was about 14 days of age. The opacification seemed stable in the lenses of six-week-six-day-old Gja8^V219F/V219F homozygotes (Fig. 1A), compared with the 14-day-old subjects (not shown). 

The morphological differences between lenses of Gja8^V219F/V219F homozygotes and unaffected animals at the age of six weeks were obvious in hematoxylin and eosin stained histological sections. The disorderly organization of periphery lens fibers was identified in the anterior subcapsular region of lens (Fig. 4I). Some fiber cells became rounded disoriented and large, especially at the subcapsular anterior pole, which is indicated by asterisk in Figure 4I. Aberrant vacuoles-like structures were distributed among the periphery lens fibers, marked by arrowhead in Figure 4I. The number of roundish fiber cells increased and the abnormalities worsened towards the nucleus of the lens (Fig. 4K). The inner fiber cells exhibited completely disorganized and lost regular cell shape in the nucleus of lens (Fig. 4L). The differentiation of new lens fibers and the distribution of cell nuclei seemed relatively normal in the bow region compared with wild-type lenses (Fig. 4J). Instead of Gja8^+/+ lens, 6w Gja8^V219F/V219F lens lost its organelle-free zone with stained nuclei accumulated in the central region, when sections were stained by toluidine blue (Figs. 4M, 4N). 

The Ocular Phenotype of Gja8^V219F/+ Rats

A reduction in the weight of eyeballs was observed in Gja8^V219F/+ heterozygotes aged from six weeks one day to six weeks three days (n = 72, mean ± SD = 0.08035 g ± 0.00413 g), compared with the Gja8^+/+ eyes (n = 25, mean ± SD = 0.09508 g ± 0.00403 g) (P < 0.001) (Fig. 2B). The size of the six-week–six-day Gja8^V219F/+ lens seemed smaller than that of Gja8^+/+ lens in appearance (Fig. 1A). 

The lens opacification did not exhibit until about three weeks after birth. The snowflake-like opacity occurred in the center of lens and the periphery was slightly less translucent compared with the wild-type lenses. The snowflake-like opacity progressed over about two weeks and the opacity of lens periphery region deepened. A sharp ring-shaped lens periphery became noticeable at about postnatal 5 weeks while the fetal nucleus appeared evidently whitish. The central lenticular opacity had been growing in region and increasing in density over the following one week. At age of six weeks, the nuclear region exhibited completely opaque with a clear ring-shaped periphery and a zonular opacity. The whole detailed progression of lens opacity is described in Figure 3. 

Figure 3.

 

Progression of lens opacity in Gja8^V219F/+ heterozygotes. The eyes of rats were photographed and the intact lenses were pictured at different time points. The opacity occurred initially around three weeks after birth. The central opacity worsened with increasing age and finally led to a totally opaque center before six weeks. White arrow indicates the clear ring-shaped zone in cataractous lenses. Scale bar: 1 mm.

The abnormality of 6w Gja8^V219F/+ rat lens was less significant than the Gja8^V219F/V219F ones at the same age. Several fiber cells became rounded and large-shaped beneath the epithelium in the anterior region, marked by asterisk in Figure 4E. The organization of inward migrating cells remained regular in the equatorial zone (Fig. 4F). Although the cell shape and arrangement were less disordered for the inner fiber cells in the center of heterozygous lens than in homozygotes, they remained less uniform in shape and size, less closely in arrangement when compared with wild-type lens (Fig. 4H). 

Figure 4.

 

Histological sections of lenses obtained from 6-week-old littermates. Images A-L were stained by hematoxylin and eosin. Images M and N were stained by toluidine blue. Histological images of lenses were from Gja8^+/+ rats (A-D, M), Gja8^V219F/+ rats (E-H) and Gja8^V219F/V219F rats (I-L, N), respectively. The positions of the images are indicated in image O. (A, E, I) cover the anterior subcapsular region of the lens. (B, F, J) show the equatorial region while (C, G, K) are close to the nucleus of lens. (D, H, L, M, N) are in the lens nucleus. Asterisk indicates abnormal rounded lens fiber cells. Arrowhead marks the vacuoles distributed among fiber cells. Scale bar: 50 µm.

The Altered Distribution of Cx50^V219F in HeLa Cells

Cx50 form gap junction channels on the plasma membrane between adjacent lens fiber cells and modulate diffusion of molecules.⁹ Both Cx50^WT-Flag and Cx50^V219F-Flag were stably expressed in transfected HeLa cells (Fig. 5A) and the expression level was not significantly different between Cx50^WT-Flag and Cx50^V219F-Flag (P = 0.700) (Fig. 5B). Immunofluorescent staining showed that Cx50^WT-Flag was localized to the cytoplasm and plasma membrane (arrows in Fig. 5C). Yet Cx50^V219F-Flag mainly accumulated in the cytoplasm (Fig. 5C). Fluorescent signals of plasma membranes between adjacent cells in Cx50^V219F group were hardly detected, which suggested that the V219F mutation might affect Cx50 trafficking to the plasma membrane. 

Figure 5.

 

Impaired expression of Cx50^V219F protein in cellular membranes. (A) Cx50^WT-Flag and Cx50^V219F-Flag protein were stably expressed in transfected HeLa cells. HeLa cells transfected with vector showed no Cx50-Flag expression. (B) The relative Cx50-Flag amount compared to GAPDH. Protein expression of Cx50^V219F-Flag was not significantly different from that of Cx50^WT-Flag (P = 0.700). Statistical differences were calculated by Mann-Whitney test. Each bar represents the quantification (mean ± SD) of Western blots from three independent experiments. ns, not significant. P < 0.05 was considered statistically significant. (C) Immunofluorescent imaging of Cx50^WT-Flag and Cx50^V219F-Flag in HeLa cells. Cells were immunostained with anti-Flag monoclonal antibody (red). DAPI shows nuclear DNA staining (blue). The merged panels show the superposition of the Flag and DAPI fluorescence signals. The arrows show junction plaques formed at areas of cell-cell contact and the arrowheads indicate areas of cell to cell apposition. Scale bar: 20 µm.

Impaired Cx50^V219F Cell Proliferation, Adhesion and Migration

To understand the underlying mechanism for disrupted lens growth of Gja8^V219F/V219F lenses, 5-ethynyl-2ʹ-deoxyuridine (EdU) labeling assay was performed in HLEB3 cells expressing Cx50^V219F or Cx50^WT (Figs. 6A, 6B). Compared with the Cx50^WT cells, a pronounced decrease in proliferation activity (lower ratio of EdU⁺ DAPI⁺ cells /DAPI⁺ cells) occurred in the Cx50^V219F cells (P = 0.003). 

Figure 6.

 

The impaired functions of HLEB3 cells with Cx50^V219F expression. (A) Representative pictures of EdU-labeled cells. Scale bar: 250 µm. (B) The ratio of EdU-labeled cell counts to DAPI-labeled cell counts for cells transfected with Cx50^WT, Cx50^V219F or vector. The ratio of Cx50^V219F group is less than that of Cx50^WT group (P = 0.003). (C) The number of remaining cells on Matrigel-coated plates was evaluated by CCK8 assay. The adhesion pattern of Cx50^V219F group is lower than Cx50^WT (P < 0.001). (D) Cells were scraped to create wounds of similar sizes. Representative images were pictured at timepoints 0, 12 and 24 h after cell scratches. Scale bar, 250 µm. E. The wound closure ratios were calculated at each timepoint by ImageJ. The ratio of Cx50^V219F group is lower than Cx50^WT group (P < 0.001 for 12 and 24 hours, respectively). Statistical differences were calculated by two-tailed t tests. Each bar represents the quantification (mean ± SD) of three independent experiments. **P < 0.01; ***P < 0.001. P < 0.05 was considered statistically significant.

Previous reports have indicated that Cx50 with p.P88L mutation interrupted migration process of lens epithelial cells and mutations in the second extracellular domain of Cx50 decreased adhesive properties of lens fibroblast cells.^10,11 To examine whether Cx50 V219F mutation could interfere with lens epithelial cell adhesion and migration, cell attachment assay (Fig. 6C) and wound-healing assay (Figs. 6D, 6E) were performed in HLEB3 cells transfected with Cx50^V219F, Cx50^WT or vector. Cell scratch wound assay showed that the expression of Cx50^V219F caused a significantly slower rate of wound closure than wildtype (P < 0.001 for 12 hours and 24 hours). The attached Cx50^V219F HLEB3 cell count showed a statistically significant decrease (compared with Cx50^WT group, P < 0.001). 

Reduced FAK Expression and Phosphorylation With Cx50^V219F Expression

Given that previous reports have shown a close relationship between focal adhesion kinase (FAK) and lens development,¹² we studied the protein expression level of FAK and its phosphorylation on Y³⁹⁷ (pFAK-Y³⁹⁷) in HLEB3 cells transfected with Cx50^V219F, Cx50^WT or vector (Fig. 7). Total FAK protein expression relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) showed a considerably decrease in Cx50^V219F HLEB3 cells, compared with the HLEB3 cells with Cx50^WT expression (P = 0.044). The expression of GAPDH-related protein pFAK-Y³⁹⁷ also declined significantly with Cx50^V219F expression in HLEB3 cells, compared with Cx50^WT group (P < 0.001). The differences in FAK phosphorylation suggest different levels of focal adhesion signaling between lens epithelial cells with Cx50^V219F or Cx50^WT expression. 

Figure 7.

 

Protein level of FAK and pFAK-Y³⁹⁷ in HLEB3 cells. (A) FAK protein level in HLEB3 cell clones stably transfected with Cx50^V219F-Flag, Cx50^WT-Flag or vector. Cx50^WT-Flag and Cx50^V219F-Flag protein were stably expressed in transfected HLEB3 cells. Expression of Cx50^V219F-Flag reduced protein expression of total FAK. (B) Protein level of pFAK-Y³⁹⁷ in HLEB3 cell clones stably transfected with Cx50^V219F-Flag, Cx50^WT-Flag or vector. Cx50^WT-Flag and Cx50^V219F-Flag protein were stably expressed in transfected HLEB3 cells. Expression of Cx50^V219F-Flag reduced protein expression of pFAK-Y³⁹⁷. (C) The relative FAK amount compared to GAPDH. Cx50^V219F reduced protein expression of total FAK, compared with Cx50^WT group (P = 0.044). (D) The V219F mutation reduced expression level of pFAK-Y³⁹⁷ (compared with Cx50^WT group, P < 0.001). Statistical differences were calculated by two-tailed t tests. Each bar represents the quantification (mean ± SD) of Western blots from three independent experiments. *P < 0.05; ***P < 0.001. P < 0.05 was considered statistically significant.

Discussion

Because lenses do not have vessels, the main ion flow depends on the gap junction channels in the plasma membrane of lens cells. Connexins form the channel between adjacent cells, modulating the intercellular communication. Cx50 distribute through the whole lens,¹³ playing a critical role in maintaining the transparency of lens. Former reports of different point mutations in murine Cx50 are summarized in Supplementary Figure S1,^5,14–20 Many of human Cx50 mutations have been reported, among which mutations in the transmembrane regions of human Cx50 are common (searched on the Cat-map website: https://cat-map.wustl.edu/). However, only one reported mutation is located in the human TM4 region (p.N220D) while no disorders of TM4 region have been found in rats or mice yet.^6,21 We identified a spontaneously new point mutation of TM4 domain in Cx50 which is novel in murine models (See in Supplementary Fig. S1). The missense mutation (G to T) in codon 655 resulted in the valine at codon 219 being replaced by a phenylalanine. The valine at codon 219 is highly conserved in different species (Fig. 1C), suggesting its importance in maintaining the function of Cx50. 

Connexins are synthesized in endoplasmic reticulum, modified in Golgi, and then transported to plasma membrane.²² The Cx50 with V219F mutation mainly localized in the cytoplasm. The abnormal localization of mutant Cx50 might alter the neighboring cells contact. Evidence has suggested that cell adhesion ability declined in the HLEB3 cells with Cx50^V219F expression. The layout and shape of Gja8^V219F/V219F lens fibers were also abnormal. The ordered arrangement of lens fiber cells may be interrupted without strong cellular contact, which was considered to correlate with lens opacity. 

A clear lens requires the removal of subcellular organelles and complete degradation of cell nuclei in the central lens fiber cells, to create and maintain a transparent organelle-free zone.²³ Mutations that impair this differentiating progress are associated with loss of the organelle-free zone and congenital cataract.²⁴ Nuclear breakdown was disturbed in the Gja8 mutant lens, as the nuclei still remained in the Gja8^V219F/V219F lens center and cortex. FAK contributes to lens epithelial cell migration and differentiation during the whole lens development.¹² FAK is expressed and activated in the posterior germinative zone and transitional zone, where epithelial cells exit the cell cycle, elongate and differentiate.¹² Most certain cellular functions regulated by FAK act in a kinase-dependent way and autophosphorylation at Y³⁹⁷ is the initiation step.¹² Expression of pFAK-Y³⁹⁷ and FAK decreased significantly in HLEB3 cells with Cx50^V219F expression, which may inhibit cell migration and differentiation. Taken together, we propose that the dysregulation of cell differentiation and migration was involved in the cataract formation, among which FAK might play a key role. 

The Gja8^V219F/V219F rats developed smaller eyes and lenses, compared with the wild-type. The abnormal size reduction in eyeballs was also the most common presentation in other Cx50 mutant cataract murine models.^15,17,25,26 The Gja8^V219F/+ lenses were also smaller than the wild-type lenses in appearance. The reduction in lens and eyeball size suggested the disruption of ocular development caused by Cx50 mutation, indicating the important role Cx50 played in the ocular development. Lenses principally express two kinds of connexins: connexin 46 (Cx46) and Cx50. Unlike Cx46 causing cataract only, the knockout of connexin 50 caused both microphthalmia and cataract.^7,27 It has also been reported that when Cx50 was replaced by Cx46, the lens of mice remained clear while the lens growth was interrupted, which finally caused reduced size of lens and eye.²⁸ The investigators suggested that the opacity of lens was caused by the reduction of water-soluble proteins while the lens growth was affected by complex homeostatic requirements of lens fiber cells.²⁸ Previous reports suggested that Cx50 selectively mediated growth factor signal transduction in lens epithelia, which recruited cells into mitosis in the postnatal lens.²⁹ This kind of Cx50 channel property correlates with cell proliferation and one allele of Cx50 is sufficient to promote normal lens growth.^7,30.31 In our study, the Cx50^V219F was less effective in HLEB3 cell proliferation, which might result in the reduction of lens size. 

The features of congenital cataract vary, among which the most common one is nuclear cataract. According to histological results, the differentiation of secondary fiber cells in the bow region of lens displayed in a relatively normal way while the primary fiber cells were completely abnormal in organization and shape. A previous study of connexin knockout mouse models suggested that the peripheral lens fiber cells were normal in the Gja8 knockout homozygous mouse model and the lens did not show opacity but in a small size, when the mice were 3 weeks old.¹⁸ The absence of both Cx50 and Cx46 caused the cataract in mice by affecting the shape of lens inner fibers instead of the periphery, which was attributed to the block of outflow of small molecules in inner fiber cells.¹⁸ The cause of cataract in TSC rats might be multi-factorial. The Gja8^V219F/+ cataract formation might be encouraged by the swelling of inner fiber cells due to Cx50^V219F dysfunction. The opacity of Gja8^V219F/V219F lens might be determined more by the interrupted nuclear breakdown in the primary fiber cells at an early stage of lens growth process, since the complete lens opaque appeared at eye opening with no further progress. Further lens development research will allow the collection of more data to confirm the precise reason for delayed denucleation in primary lens fiber cells. Mutant connexins have also been reported to impair the lens circulation by reducing the gap junctional communications between cells, with Ca²⁺ accumulates in central lens regions, inducing the formation of cataract.⁹ 

In summary, we have generated a new congenital cataract rat strain with a novel mutation of Cx50. Although several spontaneous Cx50 mutant murine models have been applied in the cataract research, our mutation is the first point mutation identified in the 4th transmembrane domain of murine Cx50. Congenital cataract still requires relevant animal models and our mutant rat strain may serve as an ideal research subject. 

Acknowledgments

Supported by the Jiangsu Natural Science Foundation (Grant No. BK20191233), Military Laboratory Animal Fund (Grant No. SYDW [2017]10). 

Disclosure: J. Shen, None; Q. Wu, None; J. You, None; X. Zhang, None; L. Zhu, None; X. Xia, None; C. Xue, None; X. Tian, None