Background
Although the cornea is the primary refractive tissue performing 70–80% of refraction of the eye, the major function of the lens is in accommodation and to partly help in the refraction. The lens accommodative function gradually diminishes with age, and is almost completely lost at age of > 50 years. The lens transparency plays an important role in focusing light on to the retina, but this role is gradually lost as it develops age-related opacity. Several unique factors maintain lens transparency for up to > 60 year of our life time. These include: cellular homeostasis among only two types of cells (epithelial and fiber cells) [
1], an orderly terminal differentiation of epithelial to fiber cells with precise organelles loss [
2], the unique interactions among crystallins [
3], with almost no protein turnover [
4], the specialized lens metabolism [
5], specific interactions among α-crystallin and membrane [
6], the precise maintenance of intracellular and extracellular ionic concentrations [
7], the low levels of cellular water and oxygen in the lens inner cortex and nuclear regions [
8], and a unique membrane lipid composition [
9]. Alterations among some of these lens unique factors play direct or indirect roles in pathogenesis of cataracts (e.g., pediatric- and age-related cataracts). However, additional cataract-causative factors are also identified, which include mutations in crystallins [
10], oxidative insults of crystallins, the loss of redox balance of glutathione [
11], extensive truncations of α-, β-, and γ-crystallins [
12‐
20], a variety of post-translational modifications with deamidation as being the most abundant [
21‐
25], and the loss of membrane integrity [
7,
26,
27]. These factors individually or in combination also cause lens opacity through altered lens cellular structures and contents, ionic imbalance, increased water and oxygen levels, loss of natural interactions among crystallins, and crystallins’ unfolding, degradation and cross-linking.
Our focus in this study is the potential roles of deamidation of Asn
101 of αA crystallin to Asp that introduces negative charges and shown to alter their hydrophobicity, tertiary structures, crystallin-crystallin interactions, and leads to aggregation and cross-linking [
21‐
27]. In this study, the deamidation of Asn
101 to Asp in in a mouse model was studied to determine phenotypic and molecular changes within the lens due to deamidation of a single nucleotide change in CRYAA crystallin gene. This site was chosen because our past study showed that only deamidation of Asn localized at specific sites in crystallins (e.g., deamidation of N101 but not of N123 residues in αA-crystallin [
24], and of N146 but not of N78 of αB-crystallin) exhibited the above-described deamidation-induced effects [
25]. To show the potential effects of deamidation in vivo, we have generated mouse models by inserting the human lens αA-N101D transgene in CRYαA
N101D mice, and human lens wild-type αA-transgene in CRYαA
WT mice (to act as a control). The CRYαA
N101D mice developed cortical cataract at about 7-months of age relative to CRYαA
WT mice [
28,
29]. This model showed for the first time that in vivo
expression of the deamidated αAN101D caused cortical lens opacity, which was due to the disruption of fiber cell structural integrity and protein insolubilization as aggregation [
28]. The comparative RNA sequencing and Ingenuity Pathway Analyses (IPA) of lenses from 2- and 4-months old CRYαA
N101D- and CRYαA
WT mice showed that the genes belonging to cellular assembly and organization, cell cycle and apoptosis networks were altered in αA
N101D lenses [
29]. This was accompanied with several cellular defects in αA
N101D lenses that included defective terminal differentiation (increased proliferation and decreased differentiation) of epithelial cells to fiber cells, and reduced fiber cells denucleation and expressions of Rho A and Na, K-ATPase (the major lens membrane-bound molecular transporter) [
29]. The findings also suggested the potential role of lens intracellular ionic imbalance as the major reason for the development of cataract [
29]. The above findings suggested that the altered intracellular ionic imbalance could be due to potential loss of membrane integrity that caused cortical opacity at about 7-months of age in the CRYαA
N101D mouse model. Therefore, the focus of the present study was to determine whether an increased membrane-association of αA
N101D potentially compromises membrane integrity, and causes an ionic imbalance and leads to cataract development.
Methods
Ethics statement
All animal experiments were performed per protocols approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Alabama at Birmingham (Protocol no. 130208393). Mice were housed in a pathogen-free environment at the facility of the University of Alabama at Birmingham.
Materials
Unless stated otherwise, the molecular biology-grade chemicals were purchased from Millipore-Sigma (St. Louis, MO, USA) or Fisher (Atlanta, GA, USA) companies. The Rabbit polyclonal anti-human aquaporin-0 (AQP0) antibody was purchased from Alpha Diagnostics (San Antonio, TX, USA). Additional commercial sources of various chemicals and antibodies used in the study are described throughout the text.
Generation of transgenic mice
The mouse model that expresses a human αA-crystallin gene in which Asn-101 was replaced with Asp is referred to as αA
N101D-transgenic mouse model. This model has been considered to be “deamidated” in this study, and the mice expressing αAN101D-transgene is referred here as CRYαA
N101D mice. Both mouse models (human lens αA
N101D transgenic- and human wild-type αA-transgenic mouse models were generated in Dr. Om Srivastava’s laboratory [
28]. Independent transgenic (Tg) mouse lines were established from transgenic founders using C57BL/6 mice (Harlan Laboratories, Indianapolis, IN). αAN101D protein expression constituted about 14 and 14.2% of the total αA in the lens WS-and WI-proteins of the αA
N101D transgenic mice, respectively [
28]. The mouse lenses were extracted after the mice were euthanized using the CO
2 procedure as per approved method by the Institutional Animal Care and Use Committee (IACUC) of the University of Alabama at Birmingham (Protocol no.130208393). Adult (2–3 months) wild type mice (C57BL6) were obtained from the university breeding colony. Animals were kept under a 12/12 h light–dark cycle and had ad libitum access to food and water. We have used three mice from each group of CRYαA
WT mice control and αA
N101D mice in all the experiments described below.
Isolation of water soluble (WS)- and water insoluble (WI)-proteins from mouse lenses
The WS- and WI-protein fractions from lenses of desired ages of CRYαA
WT- and CRYαA
N101D mice were prepared as previously described by us [
28]. All procedures were performed at 5 °C unless specified otherwise. The lenses were removed under a dissecting microscope and placed in 5 °C-cold buffer A (5 mM Tris-HCl, 1 mM EDTA, 1 mM DTT, pH 7.8, and protease inhibitor cocktail [Roche Life Science, Indianapolis, IN]), and centrifuged at 14,000 x g for 15 min at 5 °C to separate the WS- and WI- protein fractions. The supernatant (WS-protein fraction) was collected, and next the pellet (WI-protein fraction) was resuspended in buffer A, centrifuged as above. The recovery of WS- and WI-protein fractions was repeated twice after centrifugation, and the WS supernatants after each centrifugation steps were pooled. The final WI-protein pellet was solubilized in 5 mM Tris-HCl, pH 7.5, containing 8 M urea, 5 mM EDTA, and 5 mM EGTA. The 8 M urea concentration was diluted to 4 M urea with buffer A prior to centrifugation as above. The protein concentrations in these fractions were determined by using a kit (Pierce Biotechnology-Thermo Fisher) using bovine serum albumin as a standard.
Membrane isolation from mouse lenses
The membranes from lenses of 1- and 6-month-old CRYαA
WT and CRYαA
N101D mice were prepared as described previously [
30,
31]. Lenses of identical ages from both types of mice were homogenized in buffer B (0.05 M Tris-HCl, pH 8.0 containing 5 mM EDTA, 1 mM DTT, 150 mM NaCl, and protease inhibitor cocktail [Roche, Indianapolis, IN]), and the preparations were centrifuged at 100,000 x g for 30 min using Beckman TL 100 centrifuge with a TLA 100.3 rotor. The supernatant was collected, and the pellets were washed twice with the above buffer B and centrifuged as above. This was followed by three additional washes with buffer B containing 8 M urea and centrifugation as above after each wash. Next, the pellet was washed twice with water and centrifuged as above. The pellet was then washed with 0.1 M cold (5 °C) NaOH [
30,
31]. A final wash of pellet was with water and centrifugation as above to recover the purified lens membrane preparations as pellets.
Purification of recombinant WTαA- and αA-N101D-crystallins, their conjugation with Alexa Fluor 350 and membrane binding
The WTαA- and αA-N101D mutant proteins were expressed in
E.coli and purified by a Ni-affinity column chromatographic method as previously described by us [
28]. Each protein was labeled with Alexa-350 using a protein labeling kit as suggested by the manufacturer (Molecular Probes, Carlsbad, CA). The binding of Alexa 350-conjugated WT αA- and αA-N101D mutant proteins to mouse lens membrane (isolated from C57BL non-transgenic mice) was determined as previously described [
32,
33]. During the binding assay, the purified lens membrane (containing 2.5 mg protein; isolated from 1 to 3-month old non-transgenic C57 mice) was incubated with increasing but identical concentrations of either Alexa-labelled WT αA- or αA-
N101D proteins at 37
οC for 6 h. Next, the incubated preparations were centrifuged at 14,000 X g and the supernatant and pellet (membrane fraction) recovered. After washing the membrane fraction with water and centrifugation as above, the relative levels of fluorescence of membranes incubated with WT αA- and αA-
N101D mutant proteins were determined using Perkin Elmer Multiplate Reader (Model Victor1420–04).
Determination of intracellular Ca2+ in epithelial cells in culture from lenses of CRYαAWT - and CRYαAN101D mice
To culture epithelial cells, six 5-months old lenses from CRYαAWT - and CRYαAN101D mice were excised and incubated with 0.25% trypsin at 37ο C for 2.5 h in an incubator with 5% CO2-humidified air. Next, the lens cells in trypsin solution were centrifuged at 1200 rpm for 3 min, and trypsin (in the supernatant) was discarded. The lens epithelial cells (recovered as pellet) were suspended in medium 199 (Thermo Fisher Scientific, Grand Island, NY) containing 10% fetal calf serum (Hyclone, Logan, Utah) and 1% antibiotic-antimycotic solution (Thermo Fisher Scientific, Grand Island, NY) in 12-well plates (Corning, Franklin Lakes, NJ). After 24 h, the unattached cells were discarded by washing with the above medium. The old medium was replaced with fresh medium after every 48 h, and the cells were allowed to grow for 7 to 10 days until confluent. Next, the confluent cells were trypsinized and seeded in 12-well plates for intercellular Ca2+ determination and were allowed to grow for 24 h. The cells were washed with medium 199 without phenol red, incubated in calcium orange dye (Thermo Fisher Scientific, Grand Island, NY) at a final concentration of 4 μM for 30 min at room temperature as instructed in the manufacturer’s protocol. After 30 min, the cells were washed with the above medium, and Ca2+ indicator was examined under a microscope (Leica DMI 4000B) using a Texas Red filter.
Western blot and Immunohistochemical analyses
The WS- and WI-proteins and membrane fractions isolated from lenses were analyzed for their immunoreactivity with anti-aquaporin-0 antibody (to visualize the membrane intrinsic protein), and Mouse anti-His monoclonal antibody ([Novagen, Madison, WI], to visualize WTαA and αA
N101D) during Western blot analysis. The SDS-PAGE analysis was carried out as described by Laemmli [
34].
The confocal immunohistochemical analysis of lens axial sections of WTαA and αA
N101D was carried out as previously described by us [
28]. The analysis was performed at the High-Resolution Imaging core facility of the University of Alabama at Birmingham.
Localization by Immunohistochemical-transmission Electron microscopic method
The analysis was performed at the High-Resolution Imaging core facility of the University of Alabama at Birmingham. His-tagged αAWT- and αAN101D-crystallins were localized in lens cells by an Aurion immunogold method and the reagents used were Aurion Conventional Immunogold reagents (Electron Microscopy Science [PA]). Lenses of desired ages were fixed in phosphate-buffered saline, pH 7.4 containing 4% paraformaldehyde and 0.05% glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA) for 2 h at room temperature, and then overnight at 4 °C. The fixed lenses were washed with water (Millipore, Billerica, MA). Samples were dehydrated by ascending ethanol gradient series followed by infiltration overnight at 4 °C with absolute ethanol: London Resin (LR) white (1:1). Next, the samples were incubated overnight with pure LR white resin on a rotating platform. The lenses were removed and transferred to gelatin capsules containing fresh LR white and allowed to polymerize for 24 h at 45–50 °C. Ultra-thin (silver gold to light gold) LR white lens sections were collected on nickel mesh grids. The color of sections was silver-gold to light gold, and based on their color, the thickness was estimated to between 70 and 80 nm. For immunogold-labeling, the protocol as described in Electron Microscopy Sciences (Hatfield, PA) was precisely followed. To inactivate aldehyde groups present after aldehyde fixation, the samples on grids were incubated on 0.05 M glycine in PBS buffer for 10–20 min. Next, the grids were transferred onto drops of the matching Aurion blocking solution for 15 min, and then were washed for 15 min in incubation solution (PBS containing 0.1% bovine serum albumin and 15 mM NaN3, pH 7.3). This was followed by a 2X wash in incubation buffer, each time for 5 min. The grids were incubated with two primary antibodies (Mouse anti-His monoclonal antibody and Rabbit anti-aquaporin-0 polyclonal antibody for 1 h. In controls, the primary antibodies were omitted. The grids were then washed 6X (5 min each time) with the incubation solution and transferred to following secondary antibody conjugates {(goat anti-Rabbit EM grade conjugate 25 nm diameter) and (goat anti- mouse EM grade conjugate 10 nm diameter)} and were incubated for 30 min to 2 h. The grids were washed on drops of incubation solution for 6X (5 min each time). The grids were washed twice with PBS for 5 min, post-fixed in 2% glutaraldehyde in PBS for 5 min, and finally washed with distilled water and contrasted according to standard procedures. Lens sections were imaged using an FEI 120kv Spirit TEM (FEI-Thermo Fisher), and images were collected using an AMT (AMT-Woburn, MA) digital camera.
RNA extraction and reverse transcription-quantitative polymerase chain reaction (RT-PCR [qPCR])
RNA was extracted with Trizol reagent (Invitrogen) from cultured lens epithelial cells from CRYαAN101D and CRYαAWT mice, and all the samples were analyzed in triplicates. Real-time PCR quantifications were performed using the BIO-RAD iCycler iQ system (Bio-Rad, Hercules, CA), using a 96-well reaction plate for a total volume of 25 μL. RNA was extracted as described above. Primers were designed using Primer3 for the following genes:
Atp1a2 Forward-5’CGGGAGCCATAAGGGTTTGT 3′, and Atp1a2 Reverse- 5’GCACTGACTTGGCTGTTGTG 3′.
The ACTB gene was used for normalization. The reaction mixture included 12.5 μL of Real- Time SYBR Green PCR master mix, 2.5 μL of reverse transcription product, 1 μL of forward and reverse primer and 8 μL of DNase/RNase free water. The reaction mixtures were initially heated to 95 °C for 10 min to activate the polymerase, followed by 40 cycles, which consisted of a denaturation step at 95 °C for 15 s, annealing at 55 °C for 60 s and an elongation step at 72 °C. The qRT-PCR data were analyzed by the comparative ΔCt method.
Discussion
Several past studies have shown in vitro effects of deamidation of crystallins on their structural properties including those in αA-, and αB-crystallins [
21‐
25]. It has been reported that deamidation of Asn and Gln was the major modification identified in several human cataractous and aged lenses and these totaled 66% of the modification in the water-soluble and water-insoluble protein fractions that was analyzed by 2D LC/MS [
23]. The mass spectrometric analysis found that there is negligible (less than 1%) deamidation at αA
N101 site in both aged and cataractous human lenses [
38]. These studies suggested that because of low levels of deamidation of αA- at N101 to D in normal and cataractous lenses, the αAN101D might not play a significant role in cataract development. However, additional studies suggest otherwise. For example, our in vitro studies showed significant altered structural and functional properties of αA-crystallin on deamidation of N101 residue but not of N123 residue [
24]. We also showed that the WS-protein fraction from 50 to 70 year-old human donors contained αA- fragments with deamidation of N101 to D [
39]. This finding is significant because recent studies have also shown an increasing role of crystallin fragments in cataract development [
40,
41]. In the present study, the cortical cataract development in mice on the introduction of αA-N101D transgene further show significance of deamidation of this site and altered changes in the lens. However, the exact in vivo molecular mechanism of αAN101D-induced crystallin’s aggregation is yet to be fully understood.
Previously we showed that the three recombinant deamidated αA-mutants (N101D, N123D, and N101D/N123D) exhibited reduced levels of chaperone activity, alterations in secondary and tertiary structures, and larger aggregates relative to WT-αA-crystallin [
24,
25]. Among the above three mutants, the maximally affected and altered properties were observed in the recombinant αAN101D mutant [
25]. Additionally, our recent results show that in vitro
, the deamidated αA-, and αB-crystallins facilitated greater interaction with βA3-crystallin, leading to the formation of larger aggregates, which might contribute to the lens cataractogenic mechanism [
42]. As a further extension of our previous studies [
28,
29], in some studies the 7-month old lenses were chosen because of the development of cortical cataract at about 7-month of age in the αA-
N101D mice relative to αA
wt mice. In other experiments, lenses from 5-month old of both types of mice were used to determine the progression of phenotypic changes in lenses to determine their significance in cataractogenic mechanism.
The present study show that the introduction of αAN101D trans-gene in a mouse model resulted in the following major in vivo effects in lenses of CRYαA
N101D- relative to CRYαA
WT mice: (A) An age-related difference in protein profiles with an increasing association αA
N101D with WI-protein fraction suggesting its insolubilization after 4-months of age. (B) The WS-HMW protein fraction showed a higher level of proteins with a greater M
r. (C) Mass spectrometric analysis showed preferential insolubilization of αA-, αB-, γD- and γE-crystallins, and nestin, which remained insoluble even in 8 M urea. (D) The tight association of αAN101D with membranes relative to WTαA, which could not be fully dissociated with 8 M urea treatment. (E) In vitro, αA
N101 a showed greater affinity and binding to lens membranes relative WTαA. (F) The greater number of immunogold-labeled αA
N101 relative WTαA binding to membrane along with relatively greater swelling of lens membranes, suggesting the potential water uptake due to intracellular ionic imbalance, and (G) The ionic imbalance was suggested by the greater Ca
2+ uptake and 75% reduction in mRNA levels of Na, K-ATPase in the epithelial cells cultured from CRYαA
N101D lenses relative to those from CRYαA
WT lenses. Our mass spectrometry analysis showed that retinal dehydrogenase was absent in the N101D mice. It has been shown earlier that
Aldh1a1(−/−) knock-out mice developed lens opacification later in life [
43]. Retinal dehydrogenase 1 may protect the lens against cataract formation by detoxifying aldehyde products on lipid peroxidation in both cornea and lens. It has been shown that antimalarial drug chloroquine which binds and inhibit retinal dehydrogenase 1 [
44] induce cataract in rats [
45]. Together, these findings suggest altered membrane integrity (possibly due to greater levels of αA
N1010D binding to membrane than WTαA) resulting in intracellular ionic imbalance in CRYαA
N101D lenses, which could play a major role in the cortical cataract development.
Among the lens crystallins, only α-crystallin show an association with the membrane in both normal and cataractous lenses [
6,
46‐
50]. Lens membranes contain both a high-affinity saturable and low-affinity non-saturable α-crystallin-binding sites [
46,
50‐
52]. Alpha-crystallin binding to native membranes was enhanced on stripping of extrinsic proteins from the lens membrane surface to expose lipid moieties [
32,
33], which contradicted a previous report that the crystallin mostly interacts with lens membrane MP26 protein [
53]. Even after stripping extrinsic membrane proteins by alkali-urea treatment, the full-length αA-, and αB-crystallins remained associated with membranes of both bovine and human lenses [
6]. Additionally, αB-crystallin showed three-fold higher binding to lens membrane relative to αA-crystallin, and their binding was affected by the residual membrane-associated proteins, suggesting that their binding behaviors were affected by an intrinsic lens peptides [
6]. A large-scale association of proteins with cell membranes in the lens nucleus (mostly in the barrier region) occurs after middle age in human lenses [
48], and such association was enhanced by mild thermal stress [
49]. The in vitro studies further supported this because the binding capacity of α-crystallin from older lenses to lipids increased with age and decreased in diabetic donors who were treated with insulin [
50]. This implied that under diabetic conditions, abnormal binding of α-crystallin to lens membrane occurred. Such information in the literature about membrane binding of native vs. post-translationally modified crystallins including the deamidated αAN101D species is presently lacking. Therefore, the results of the present study showing relatively increased binding of αAN101D relative to WTαA are highly significant.
The RNA sequence and IPA data of our previous study [
29] further support the findings of the present study. This study [
29] showed that the genes belonging to gene expression, cellular assembly, and organization, and cell cycle and apoptosis networks were altered, and specifically, the tight junction-signaling and Rho A signaling were among the top three canonical pathways that were affected in the CRYαA
N101D lenses relative to CRYαA
WT lenses. The present study showed an increased association of αAN101D to membrane, and this could lead to potential ionic imbalance affecting tight junction assembly and RhoA GTPase expression. This in turn causes increased proliferation and decreased of differentiation and denucleation of epithelial cells, and an accumulation of nuclei and nuclear debris in the lens anterior inner cortex and fiber cell degeneration. Some of these phenotypic changes could be cause or effects, but together could be responsible for the age-related cortical cataract development in CRYαA
N101D lenses.
To maintain ionic balance within lens cells, a permeability barrier close to the surface of the lens is responsible for the continuous sodium extrusion via Na, K-ATPase-mediated active transport [
35‐
37]. Without an active sodium extrusion, lens sodium and calcium contents are shown to increase resulting in lens swelling that leads to loss of lens transparency [
35]. Similarly, an excessive intracellular Ca
2+ levels can be detrimental to lens cells, and its increased levels play an important role in development of cortical cataract [
37]. Therefore, homeostasis of Na
+, K
+, Ca
2+ and other ions within the lens has been recognized as of fundamental importance in lens pathophysiology. These have been altered as shown in our present and our previous studies [
29]. It is also possible that the increased Ca
2+ levels could in turn lead to calpain activation and proteolysis of crystallins, which will be investigated in future.
Similar to our study, other studies have shown that an increased membrane binding of α- crystallin in the pathogenesis of different forms of cataracts. The high molecular weight complexes (HMWCs), comprised of α-crystallin and other crystallins, accumulate with aging and show a greater membrane binding capacity than native α-crystallin [
50]. Other mutants of αA-crystallin, like the αA
N101D mutant, also exhibit a greater membrane binding than corresponding wild-type species [
54]. For example, in the αAR116C-associated congenital cataracts, an increased membrane binding capacity along with changes in complex polydispersity, and the reduction of subunit exchange were considered potential factors in the cataract pathogenesis [
54]. Similarly, αA-crystallin R49C neo mutation influenced the architecture of lens fiber cell membranes and caused posterior and nuclear cataracts in mice [
55].
Interactions between proteins and the cell membrane are an integral aspect of many biological processes, which are influenced by compositions of both membrane lipids and protein structure [
56]. Reports have shown the age-related lipid compositional changes in the lens membrane, which might affect α-crystallin binding, i.e., in the nucleus of the human lenses, the levels of glycerophospholipids declined steadily by age 40 as opposed to the levels of ceramides and dihydroceramides increased approximately 100 fold during middle-age [
57,
58]. Further, it has been shown that because of the elevation of sphingolipid levels with species, age, and cataract, lipid hydrocarbon chain order, or stiffness increased. Therefore, the increased membrane stiffness caused an increase in light-scattering, reduced calcium pump activity, altered protein-lipid interactions, and perhaps slow fiber cell elongation [60]. Presently, whether similar changes occur in αA
N101D lenses are not known.
Alpha A- and αB-crystallins differently associate with the cellular membrane, i.e. αA-crystallin may interact exclusively with membrane phospholipids, and thereby unaffected by the presence of extrinsic proteins on the membrane, whereas these proteins may act as conduits for αB-crystallin to bind to the membrane [
58]. Presently, the specific binding mechanism of αAN101D to the membrane and age-related changes in lipid composition in lenses CRYαA
N101D vs. CRYαA
WT are unknown, and these are presently the focus of our investigations.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.