Arginine 54 and Tyrosine 118 Residues of αA-Crystallin Are Crucial for Lens Formation and Transparency

Abstract

purpose. To identify new mouse models for studying roles of αΑ-crystallin in vivo and to investigate why and how different mutations of the αΑ-crystallin gene lead to dominant or recessive cataracts.

methods. Using mouse genetic approaches and slit lamp screening, we identified two mouse cataractous mutant lines. Causative genes were mapped by a genome-wide linkage analysis. DNA sequencing verified missense mutations of αA-crystallin gene in both mutant lines. Histology, imaging of green fluorescent protein (GFP)–positive lenses, and protein 2-DE gel were used to determine the morphologic and biochemical properties of mutant lenses.

results. Two new αA-crystallin gene mutations were identified, αA-R54C (αA-Cys) and αA-Y118D, which cause recessive whole cataracts and dominant nuclear cataracts, respectively. In homozygous αA-Cys mutant mice, lens epithelial and fiber cells lost their characteristic cellular features and developed disrupted subcellular structures, such as actin filaments and mitochondria. The nuclear cataract caused by αA-Y118D mutation was associated with increased water-insoluble crystallins (α, β, and γ classes). These results suggest that the Arg⁵⁴ residue in the N-terminal region is crucial for αA-crystallin to perform its roles in lens epithelial and fiber cells during development, whereas the Y118D mutation in the central α-crystallin domain impairs αA-crystallin’s ability to maintain the solubility of crystallin proteins in the lens.

conclusions. This work demonstrates that different regions of αA-crystallin mediate distinct functions in vivo. These two mutant mouse lines provide useful animal models for further investigating the multiple roles of αA-crystallin in the lens.

The α-crystallins, consisting of the αA and αB subunits, are members of the small heat shock protein (sHSP) family. ^{1 2 3} They are structural components contributing to a transparent lens, and they are hypothesized to act as molecular chaperones to maintain lens transparency by preventing abnormal protein aggregation. ⁴  

αA-crystallin is a 173-amino-acid polypeptide encoded by the CryaA gene on mouse chromosome 17 and human chromosome 21. A splicing isoform (αA_ins) that contains additional 23-amino-acid residues between residues 63 and 64 of the αA-crystallin protein is also expressed in mice. ⁵ αB-crystallin is a 175-amino-acid polypeptide encoded by the CryaB gene on mouse chromosome 9 and human chromosome 11. The αA and αB subunits share approximately 60% amino acid sequence identity and account for 20% to 30% of the lens total proteins. ⁶ They exist as heteromers that can undergo subunit exchange. ⁷ Due to the polydispersed size distribution of both native and recombinant α-crystallins, the crystallization of α-crystallin proteins has been unsuccessful so far. Neither the three-dimensional (3-D) structure of α-crystallin proteins nor the topology of subunit assembly is known. Based on the 3-D structures of other members of the sHSP family, αA-crystallin can be divided into three regions: a variable N-terminal region (residues 1-63), the central α-crystallin domain made of β-strands (residues 64-139), and the extended C-terminal region (residues 140-173). ^{4 8 9}  

Genetic studies have reported that mutations of αA- or αB-crystallins cause dominant or recessive cataracts in both humans and mice. In αA-crystallin, two human missense mutations (αA-R49C and αA-R116C) and one mouse missense mutation (αA-V124E) lead to the development of dominant nuclear cataracts, ^{10 11 12} whereas one human nonsense mutation (αA-W9X) and two mouse mutations (αA-R54H in lop18 mice and αA^−/− knockout) cause recessive cataracts. ^{13 14 15} In αB-crystallin, a frame-shift mutation and a missense mutation (αB-R120G) cause dominant cataracts in humans. ^{16 17} These genetic studies demonstrate the importance of αA and αB-crystallins in the lens, but they do not provide explanations for why different mutations of αA- or αB-crystallin genes lead to a variety of cataract phenotypes. 

Previous studies of αA and αB knockouts (null mutations) have revealed that αB is dispensable, whereas αA is essential for lens formation and transparency in mice. The αA^−/− knockout mice have microphthalmia and small lenses with nuclear cataracts. Light and transmission electron microscopy studies reveal the presence of inclusion bodies that contain αB-crystallins in αA^−/− lenses. ¹³ In contrast, αB^−/− knockout mice have relatively normal lenses without cataracts. ¹⁸ Moreover, double-knockout αA^−/− αB^−/− mice have abnormal fiber cells without lens sutures. ¹⁹ Therefore, αA and αB are important for both lens formation and transparency, but the in vivo functions of αA- and αB-crystallins cannot be further investigated by using these knockout mice, because these proteins are no longer present in null mutant mice. 

Here, we report our findings of two new αA-crystallin mutations, αA-R54C (αA-Cys) and αA-Y118D, which cause a severe recessive cataract and a dominant nuclear cataract, respectively. In our study, the αA-Cys mutation altered the properties of lens epithelial and fiber cells by disrupting intracellular structures, such as actin filaments and mitochondria, whereas the αA-Y118D mutation increased the level of water-insoluble crystallin proteins but maintained relatively normal lens histology. This work provides in vivo evidence for the multiple functions of α-crystallins during lens development. These two mouse lines provide important animal models for further investigating distinct functions of αA-crystallin mediated by its N-terminal domain and the central α-crystallin domain in vivo. 

Materials and Methods

Mouse Mutations, Genomic Linkage Analysis, and Causative Gene Identification

Mouse care and breeding were performed according to an Animal Care and Use Committee (ACUC)–approved animal protocol (UC Berkeley) and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The L1N mouse line was identified from a screen of ENU-mutagenized mice. Mouse pupils were dilated by a 1:1 mixture of 2.5% phenylephrine hydrochloride and 1% atropine sulfate, and lens clarity was examined by a slit lamp. Mouse breeding and genomic linkage analysis were performed as described previously. ²⁰ The L1N mutant mice, in the C57BL/6J (B6) strain background, were mated with wild-type C3H/HeN mice to produce affected G1 hybrid mice that were further crossed with wild-type C3H/HeN mice to produce second-generation (G2) mice. The G2 mice were phenotyped, and their genomic DNAs were used for a genome-wide linkage analysis. Based on the chromosomal location, causative gene candidates were identified from the mouse genome database (MGD) at the National Center for Biotechnology Information (NCBI) website. 

The nm3365 mouse line was a spontaneous recessive mutation segregated from the FVB/N-Tg (Trp53R172L)4491Jmr/J mice. A male homozygous founder had microphthalmia with dense cataracts and was mated with a wild-type BALA/cJ female to produce G1 female mice. All G1 heterozygous mice had normal lenses without cataracts. The G1 females were then backcrossed with the male homozygous founder to produce G2 mice, half of which had cataracts that had developed by weaning age. Affected G2 mice were intercrossed to produce the nm3365 mouse colony. 

Total lens RNAs were isolated from the lenses of homozygous mutant mice (TRIzol Reagent; Invitrogen Life Technologies, Gaithersburg, MD). Total RNA (2 μg) was used to generate cDNA with an RT-PCR kit (Superscript First-Strand Synthesis System; Invitrogen Life Technologies, Gaithersburg, MD). The coding region of αA-crystallin was amplified with DNA polymerase (Platinum Pfx; Invitrogen Life Technologies) and sequenced to detect mutations. A pair of primers, 5′ TCCTGATCTGACTCACTGCC 3′ and 5′ AGGCAGACTCTTTGCTGTGG 3′, was used for PCR amplification of the entire coding region of the αA-crystallin gene and for determining its DNA sequence. RT-PCR and sequencing analysis were performed on at least three different lens RNA samples from both L1N and nm3365 mutant mice, to confirm mutations in the αA-crystallin gene. 

GFP Fluorescence Imaging of Lens Epithelial Cells

Both nm3365 and L1N mice were bred with GFP transgenic mice to generate GFP-positive mutant mice. ²¹ The GFP-positive mice were screened by using a UV light. Fresh GFP-positive lenses were dissected from enucleated eyeballs and were immediately imaged for GFP-positive epithelial cells by confocal microscope (Leica, Deerfield, IL). 

Lens Phenotypes and Morphologic Analysis

Freshly dissected lenses were imaged under a dissecting microscope (MZ16 Leica) with a digital camera. Thin sections were prepared for morphologic analysis by light and electron microscopy. Briefly, mouse eyeballs or lenses were immersed in a fixative solution containing 2.5% glutaraldehyde and 2% formaldehyde in 0.1 M cacodylate buffer (pH 7.2) for 5 days at room temperature, postfixed in 1% aqueous OsO₄, stained en bloc with 2% aqueous uranyl acetate, and dehydrated through graded acetone. Samples were embedded in Epon 12-Araldyte 502 resin (Ted Pella, Redding, CA). The Sections (1 μm) were collected on glass slides and stained with toluidine blue for histologic imaging. Bright-field images were acquired with a light microscope (Axiovert 200; Carl Zeiss Meditec, Inc., Dublin, CA) with a digital camera. For TEM analysis, ultrathin sections (60 nm) were prepared and stained with Sato’s triple lead solution before examination under an electron microscope (JEM-1200EX II; JEOL, Tokyo, Japan). 

Two-Dimensional Electrophoresis of Lens Proteins

Fresh lenses were dissected from eyeballs and quickly weighed, to determine their wet weight. These lenses were homogenized in a phosphate buffer (pH 7) and prepared for the soluble fractions and the insoluble pellets as a procedure described previously. ²² Two-dimensional electrophoresis (2-DE) was performed (PROTE-AN-IEF Cell with 11-cm IPG strips [pH 3–10]; Bio-Rad, Hercules, CA), and 8% to 16% linear gradient precise gels were used for the 2-D analysis. The gels were stained with Coomassie blue. 

Results

Role of Two New αA-crystallin Mutations in Recessive or Dominant Cataracts

The nm3365 mouse line is a spontaneous recessive mutation. Heterozygous (nm3365/ ⁺) mutant mice had normal lenses, whereas homozygous (nm3365/nm3365) mutant mice had microphthalmia and very small lenses with whole cataracts (Fig. 1A) . The average lens wet weight of homozygous mice was 60% less than that of wild-type mice at weaning age, but it was difficult to obtain an accurate weight measurement for homozygous lenses due to their severe phenotype. 

The L1N mouse line was identified from ENU-mutagenized mice by slit-lamp examination. Homozygous (L1N/L1N) mutant mice showed much denser nuclear cataracts than heterozygous mutant (L1N/ ⁺) mice at weaning age (Fig. 1A) . The average wet weights of wild-type, heterozygous and homozygous mutant lenses of 3-week-old mice were 3.55 ± 0.07 (n = 8), 3.31 ± 0.22 (n = 9), and 2.87 ± 0.04 (n = 11) mg. Thus, heterozygous lenses weighed approximately 7% less than wild-type lenses, and homozygous lenses weighed approximately 19% less than wild-type lenses. 

Genome wide linkage analyses mapped both mutations to the linkage marker D17Mit175, which is in the proximity of the CryaA (αA-crystallin) gene on chromosome 17 (Fig. 1B) . Because αA-crystallin mutations were linked to cataracts, we performed sequence analyses using cDNAs synthesized from homozygous nm3365 and homozygous L1N lens RNAs. DNA sequencing of nm3365 cDNA samples revealed a single nucleotide change (C-to-T) at position 160 in exon 1 of the CryaA gene, which resulted in the substitution of the arginine residue at codon 54 by a cysteine (αA-R54C; Fig. 2A ). Of interest, the nm3365 mutation occurred at the same Arg⁵⁴ as the lop18 mutation, which is αA-R54H mutation due to a G-to-A change at position 161. ¹⁴ Sequencing data revealed that L1N was also a missense mutation (T-to-G) of the CryaA gene that resulted in the replacement of the tyrosine residue at codon 118 by an aspartic acid (αA-Y118D; Fig. 2B ). 

Thus, we named the nm3365, lop18, and L1N mutations as αA-Cys, αA-His, and αA-Y118D, respectively. Both Arg⁵⁴ and Tyr¹¹⁸ are conserved residues that are located either in the N-terminal domain or in the central α-crystallin domain of the αA-crystallin protein (Fig. 2C) . 

Effect of the αA-Cys Mutation on Lens Epithelium and Fiber Cells

Histologic sections showed no obvious changes in lens epithelial and fiber cells of αA(Y118D/Y118D) lenses when compared with wild-type αA^+/+ lenses at weaning age (Fig. 3A) . Although some intracellular vacuoles were observed in the epithelial cells of αA(Y118D/Y118D) lenses, similar vacuoles were also seen in the epithelial cells of wild-type lenses (data not shown). Thus, histologic data did not provide useful evidence to explain why or how the αA-Y118D mutation causes smaller lenses. 

In contrast, histologic sections of postnatal day 2 (P2) αA(Cys/Cys) lenses showed severely altered epithelial cells and degenerated fiber cells (Fig. 3B) . Anterior epithelial cells were either detached from or aggregated underneath the lens capsule, and posterior fiber cells were only partially elongated. A large lumen in the anterior half of the lens was confirmed by histology and phalloidin staining. These data suggest that aberrant growth of epithelial cells and insufficient elongation of fiber cells lead to the small size of αA(Cys/Cys) lenses (Fig. 1A) . 

We also examined the abnormality of epithelial cells in fresh αA(Cys/Cys) lenses by using a transgenic mouse line expressing green fluorescent protein (GFP) under an actin promoter. ²¹ A mosaic expression pattern of GFP-positive and GFP-negative lens epithelial cells was observed in enucleated lenses from this transgenic mouse line (Fig. 3C , left), which was similar to another GFP-transgenic mouse line. ²³ The mosaic GFP pattern allows direct observation of epithelial cell morphology in fresh, enucleated lenses. We have, therefore, generated both αA(Y118D/Y118D) and αA(Cys/Cys) mutant mice that contain the GFP transgene. Direct observation of the GFP-positive lens epithelial cells further confirmed that epithelial cells in αA(Y118D/Y118D) lenses had normal morphology, but epithelial cells in αA(Cys/Cys) lenses were severely altered (Fig. 3C , middle, right). Irregularly shaped cells and vesicle-like cellular structures were observed in the epithelial layer of αA(Cys/Cys) lenses. Thus, the characteristic features of lens epithelium as a polarized monolayer of cuboidal epithelial cells were disrupted in αA(Cys/Cys) lenses. 

These results suggest that αA-Cys mutant proteins probably interrupt some basic subcellular structures to alter characteristic properties of both epithelial cells and fiber cells during early lens development, whereas αA-Y118D mutant proteins probably perturb lens growth and transparency. 

Interrupted Actin Filaments and Degenerated Mitochondria in αA(Cys/Cys) Embryonic Lenses

Actin filaments are necessary for maintaining epithelial cell polarity, cell shape, cell migration, and other cellular events. Actin filament assembly is also suggested to be regulated or protected by α-crystallins. Frozen sections of αA(Cys/Cys) embryonic lenses at different stages were stained by fluorescent-labeled phalloidin. Severely disrupted actin filaments were observed in the contact regions between anterior epithelial cells and posterior fiber cells of embryonic day 16.5 (E16.5) αA(Cys/Cys) lenses (Fig. 4A) . Moreover, F-actin staining signals were disorganized in epithelial and fiber cells of αA(Cys/Cys) lenses when compared to wild-type αA^+/+ lenses. 

TEM data showed that degenerated mitochondria (Fig. 4B , arrows) with severe loss of cristae and matrix or vacuoles were present in peripheral fiber cells of E16.5 αA(Cys/Cys) lenses, whereas normal mitochondria (arrowheads) were present in the cells of E16.5 wild-type αA^+/+ lenses. Thus, disrupted F-actin and degenerated mitochondria are probably two major defects that abolish the characteristic features of epithelial and fiber cells in αA(Cys/Cys) lenses. 

Increased Water-Insoluble Crystallins in αA(Y118D/Y118D) Lenses

Because the histology of αA-Y118D mutant lenses was relatively normal, we hypothesized that αA-Y118D mutant proteins fail to maintain the solubility of other lens proteins in mature fiber cells and cause nuclear cataracts. To test this hypothesis, we determined the changes of crystallin proteins between wild-type αA^+/+ and αA(Y118D/Y118D) lenses by comparing the 2-DE protein profiles of wild-type αA^+/+ and homozygous αA(Y118D/Y118D) lens samples. 

The 2-DE data of αA^+/+ and αA(Y118D/Y118D) water-soluble proteins were almost identical (Fig. 5A 5C)except one protein spot. The αA_ins spot of αA(Y118D/Y118D) lenses shifted left due to the lower PI of the αA_ins-Y118D protein and probably overlapped with the βA1 spot (Figs. 5A 5C , arrows). 2-DE data of water-insoluble proteins showed that intact αA, αB, and several β-crystallin isoforms, but not γ-crystallin isoforms, were major components from wild-type αA^+/+ lenses (Fig. 5B) . However, almost all crystallin proteins were present in the water-insoluble proteins of αA(Y118D/Y118D) lenses (Fig. 5D) . In addition, cleaved or phosphorylated forms of αA-Y118D proteins, several β-crystallin isoforms and a substantial amount of different γ-crystallin isoforms were also present. Thus, these results indicate that increased water-insoluble crystallins are correlated with the nuclear cataract of αA-Y118D mutant lenses. 

Discussion

This work identified and characterized a recessive mutant mouse line with whole cataracts and a dominant mutant mouse line with nuclear cataracts, caused by two new αA-crystallin gene mutations, αA-Cys and αA-Y118D. The recessive cataracts in αA(Cys/Cys) mutant mice are associated with disruptions of essential subcellular structures, such as F-actin and mitochondria, in embryonic lens fiber cells. These data indicate that α-crystallins are involved in maintaining the characteristic features of epithelial cells and in regulating appropriate formation of differentiating fiber cells during lens development. Although the mechanism for how αA-Cys mutant proteins disrupt these subcellular structures is not clear, this study provides direct evidence for a new hypothesis that the Arg⁵⁴ residue in the N-terminal domain is essential for the function of αA-crystallin in lens epithelial and fiber cells during development. The nuclear cataracts of αA(Y118D/Y118D) mutant mice are associated with increased water-insoluble crystallins including α-, β-, and γ-isoforms. Therefore, αA-Y118D mutation reduces αA-crystallin’s ability to maintain the solubility of other crystallin proteins in the lens and that the nuclear cataract probably results from the light-scattering of abnormal aggregates of crystallin proteins in mature fiber cells. 

Role of Insoluble and Modified Crystallins on Cataracts in αA-Y118D Mutant Mice

The αA-Y118D mutant protein adds a negatively charged residue (D), whereas both αA-R116C and αB-R120G mutant proteins lose a positively charged residue in the α-crystallin domain. Thus, the net charge change in these three mutant proteins is similar. The Arg¹¹⁶of αA-crystallin corresponds to the Arg¹²⁰ of αB-crystallin. Both αA-R116C and αB-R120G mutations lead to dominant cataracts in humans. ^{16 17} Biochemical studies have shown that αA-R116C and αB-R120G mutant subunits alter protein structure and decrease their chaperone-like function in vitro. ^{24 25} Other studies have identified several subunit–subunit interaction sites of αB-crystallin in vitro. ^{26 27} Mutant αB-R120G subunits also increase their binding capacity to desmin filaments, causing protein aggregation in desmin-related myopathy. ²⁸ A recent study reports that αB-R120G mutant proteins are unstable and more susceptible to proteolytic degradation, have enhanced interaction with their target proteins, and increase the size of their protein aggregates. ²⁹ Studies of biochemical properties of αA-Y118D mutant proteins may be very informative regarding the molecular basis of cataracts caused by the αA-Y118D mutation in vivo. 

The αA-Y118D mutation causes an increase of insoluble, truncated, and phosphorylated crystallins in the lens. Many different posttranslational modifications of α-crystallin proteins have been identified in normal lenses, aging lenses, cataractous lenses, and lenses treated with different stimuli. ^{30 31} These modifications can affect chaperone-like activity, size of protein aggregates, rate of subunit exchange, solubility, subcellular distribution, or other properties of the protein in vitro. However, it has been difficult to interpret the functional consequences of these modifications in vivo. ^{32 33} Phosphorylation of Ser¹²² or Ser¹⁴⁸ and C-terminal truncation of αA-crystallin proteins have been previously examined by 2-DE with mass spectrometry protein identification. ^{33 34} The P1 and P2 spots in the αA-Y118D 2-DE gel (Fig. 5D)match phosphorylated Ser¹²²or Ser¹⁴⁸ protein forms, respectively, and the cleaved spot also matches the C-terminal 1-168 truncated protein. 

Loss-of-Function Mutant Proteins in the αA-Cys and αA-His Mutant Mice

Unlike the αA-Y118D dominant mutation, both αA-Cys and αA-His are recessive mutations. It is unexpected that the αA-Cys mutation (nm3365) occurs at the same Arg⁵⁴ residue as the previously reported αA-His (lop18) mutation. Both heterozygous αA(Cys/+) and αA(His/+) mutant mice have relatively normal lenses, whereas homozygous mice have severe cataracts. At present, the molecular mechanism for how the αA-Cys (or αA-His) mutation leads to unique defects in αA(Cys/Cys) or αA(His/His) lenses is unknown. Based on the fact that these heterozygous mice have transparent lenses, we assume that the mutant proteins expressed from one allele are nontoxic to lens cells and that wild-type αA-crystallin proteins expressed from the other allele are sufficient for normal lens formation. The last assumption is supported by the fact that heterozygous αA^+/− knockout mice also have normal lenses. 

Previous studies have suggested that α-crystallins are multifunctional proteins that play roles in various cellular events and that α-crystallins interact with other crystallin proteins, plasma membrane, and cytoskeletal components, such as actin, intermediate filaments, and microtubules. ^{35 36 37 38 39 40 41 42} Studies of αA^−/− lenses suggest that αA-crystallin is needed for preventing lens epithelial cell apoptosis. ⁴³ A recent study also suggests that α-crystallins play an important role in Golgi reorganization during the cell cycle. ⁴⁴ Thus, we hypothesize that R54C (αA-Cys) and R54H (αA-His) mutations impair the N-terminal domain that is necessary for the distinct functions of αA-crystallin in lens epithelial and fiber cells. These two mutations provide a useful experimental system that is different from the αA-Y118D mutation for investigating multiple functions of αA-crystallin during lens development. 

 Supported by National Eye Institute Grants EY13849, EY03897, EY015073, and EY07758.

 Submitted for publication February 17, 2006; revised March 17, 2006; accepted May 22, 2006.

 Disclosure: C. Xia, None; H. Liu, None; B. Chang, None; C. Cheng, None; D. Cheung, None; M. Wang, None; Q. Huang, None; J. Horwitz, None; X. Gong, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Both the nm3365 recessive and the L1N dominant mutations mapped to mouse chromosome 17 in the proximity of the CryaA (αA-crystallin) gene. (A) Lens photographs of the various mouse strains (as labeled) at the age of 3 weeks. (B) Linkage analysis of the nm3365 and L1N mutations. Left: 67 meioses (numbers listed in the bottom line) from a backcross between nm3365 and CAST/EiJ were phenotyped and genotyped. The columns represent haplotypes (▪: nm3365 allele; □: CAST/EiJ allele). The mutation is linked to several markers on mouse chromosome 17. The genetic map shows that the nm3365 mutation is linked to the D17Mit175 marker at position 17.70 cM on chromosome 17. The CryaA locus is in the proximity of this marker. Right: 88 meioses (numbers listed in the bottom line) from a backcross between B6-L1N/+ and C3H/HeJ were phenotyped and genotyped. The columns represent haplotypes (▪: L1N allele; □: C3H/HeJ allele). The L1N mutation and the D17Mit175 marker were tightly linked (0/88). Scale bar: (A) 1 mm.

Figure 2.

 

DNA sequencing reveals an αA-Cys (R54C) point mutation in the nm3365 mutant mice and an αA-Y118D point mutation in the L1N mutant mice. (A) DNA sequencing shows a C-to-T change in the first nucleotide for codon 54, which results in the replacement of Arg⁵⁴ by a cysteine (αA-Cys) in the nm3365 mutation. (B) A nucleotide T-to-G change causes the substitution of the Tyr¹¹⁸ by an aspartic acid (αA-Y118D) in the L1N mutation. (C) Schematic drawing shows both αA-Cys (nm3365) and αA-His (lop18) mutations are located in the N-terminal domain. Lop18 is another αA recessive mutation where Arg⁵⁴ is substituted by a histidine as previously reported. The αA-Y118D mutation is located in the central α-crystallin domain of the αA-crystallin protein. The αA_ins, an alternative splicing isoform in mice, has additional 23 amino acid residues.

Figure 3.

 

Normal epithelial cells are observed in αA(Y118D/Y118D) lenses, whereas aberrant epithelial cells and degenerated fiber cells appear in αA(Cys/Cys) lenses. (A) Toluidine blue–stained lens sections show normal lens fiber cells in the bow regions of lenses from 3-week-old mice. (B) Images of toluidine blue–stained lens sections of 2-day-old mice. In αA(Cys/Cys) lenses, anterior epithelial cells are either dissociated from or aggregated under-neath the lens capsule (arrows), and posterior degenerated fiber cells are elongated about half the distance to the anterior epithelium (arrowheads, leading edge of the fiber cells). A fluorescent image of phalloidin- and 4′,6′-diamino-2-phenylindole (DAPI)-labeled αA(Cys/Cys) lens frozen sections further confirms that the anterior half of the lens is filled with liquid rather than fiber cells (right). White arrowhead: leading edge of fiber cells; small white arrow: epithelial cells. (C) GFP-fluorescent images of lens epithelial cells from freshly dissected lenses of 7-day-old GFP mice. Normal epithelial cells show a mosaic expression pattern of GFP-positive (green) and GFP-negative (black) cells in the αA^+/+ lenses. The epithelial cells of αA(Y118D/Y118D) lenses are normal, whereas the epithelial cells of αA(Cys/Cys) lenses have irregular morphology and abnormal vesicle-like cellular structures (white arrows). Scale bars: (A, B, C) 50 μ.

Figure 4.

 

Disrupted actin filaments and degenerated mitochondria in αA(Cys/Cys) embryonic lenses. (A) Fluorescent images of phalloidin- and 4′,6′-diamino-2-phenylindole (DAPI)-labeled frozen lens sections of E16.5 embryos. Disrupted actin filaments are observed in the junctions between epithelial cells and fiber cells in the αA(Cys/Cys) lens compared with the αA^+/+ lens (white arrowheads). (B) TEM data show normal mitochondria (arrowheads) in cortical fiber cells of E16.5 wild-type αA^+/+ lens. Vacuoles and degenerated mitochondria (arrows) with severe loss of cristae and matrix are observed in cortical fiber cells of the E16.5 αA(Cys/Cys) embryonic lens. Scale bars: (A) 50 μm; (B) 500 nm.

Figure 5.

 

Increased water-insoluble crystallin proteins, especially γ-crystallins, in αA(Y118D/Y118D) lenses. (A, C) 2-DE results of water-soluble proteins (15 μg proteins are loaded) of lenses of 3-week-old mice. Crystallin proteins αA, αA-Y118D, and αB are labeled. Arrows: αA_ins. (B, D) 2-DE results of water-insoluble proteins (one third of the pellet from two lenses was loaded) of lenses of 3-week-old mice. (D, arrowhead) Cleaved form of αA, P1, and P2 indicate two different phosphorylated forms of αA. The pH gradient is from 3 on the left side to 10 on the right side of the 2-DE images.

The authors thank Eric Wawrousek (National Eye Institute) for providing the αA/B-crystallin knockout mice, and Bruce Beutler and Xin Du at Scripps Research Institute for providing the ENU-induced mutant mice.