Genotype, Age, Genetic Background, and Sex Influence Epha2-Related Cataract Development in Mice

Purpose Age-related cataract is the leading cause of blindness worldwide. Variants in the EPHA2 gene increase the disease risk, and its knockout in mice causes cataract. We investigated whether age, sex, and genetic background, risk factors for age-related cataract, and Epha2 genotype influence Epha2-related cataract development in mice. Methods Cataract development was monitored in Epha2+/+, Epha2+/−, and Epha2−/− mice (Epha2Gt(KST085)Byg) on C57BL/6J and FVB:C57BL/6J (50:50) backgrounds. Cellular architecture of lenses, endoplasmic reticulum (ER) stress, and redox state were determined using histological, molecular, and analytical techniques. Results Epha2−/− and Epha2+/− mice on C57BL/6J background developed severe cortical cataracts by 18 and 38 weeks of age, respectively, compared to development of similar cataract significantly later in Epha2−/− mice and no cataract in Epha2+/− mice in this strain on FVB background, which was previously reported. On FVB:C57BL/6J background, Epha2−/− mice developed severe cortical cataract by 38 weeks and Epha2+/− mice exhibited mild cortical cataract up to 64 weeks of age. Progression of cataract in Epha2−/− and Epha2+/− female mice on C57BL/6J and mixed background, respectively, was slower than in matched male mice. N-cadherin and β-catenin immunolabeling showed disorganized lens fiber cells and disruption of lens architecture in Epha2−/− and Epha2+/− lenses, coinciding with development of severe cataracts. EPHA2 immunolabeling showed intracellular accumulation of the mutant EPHA2-β-galactosidase fusion protein that induced a cytoprotective ER stress response and in Epha2+/− lenses was also accompanied by glutathione redox imbalance. Conclusions Both, Epha2−/− and Epha2+/− mice develop age-related cortical cataract; age as a function of Epha2 genotype, sex, and genetic background influence Epha2-related cataractogenesis in mice.


Sample preparation
To precipitate the protein, 200 μL aliquots of the lens homogenates were mixed with ice-cold 80% methanol in pre-weighed microcentrifuge tubes. The samples were centrifuged at 7400 ×g for 15 min at 4°C. The supernatants containing unbound GSH and GSSG were removed, and the protein pellets were washed twice with ice-cold 80% methanol. The protein pellets were dried in a Vacufuge 5301 (Eppendorf, NY, USA) at room temperature for 4.5 hours. The weight of each tube and pellet was recorded. Each pellet was then resuspended in 0.5 mL of a pH 7.0, 100 mM ammonium bicarbonate / formic acid buffer. The samples were then reduced by mixing 350 μL aliquots of the suspended protein samples with 10 μL of 75 mM dithiothreitol (DTT) and 140 μL of LC-MS grade water. The protein samples were shaken vigorously for one hour to allow for reduction and liberation of protein-bound glutathione. The samples were then centrifuged at 11000 ×g for 15 min at room temperature. Finally, 9 μL of supernatant was diluted with 900 μL of 0.1% formic acid (LC-MS grade) and then filtered through a 0.22 μm nylon membrane filter.

Chromatography
A Shimadzu Prominence UFLC system consisting of a degasser (DGU-30A3), two pumps (LC-20 AD XR), an autosampler (SIL-20AC XR), and a column oven (CTO-20A), were coupled to an AB Sciex 4000QTRAP LC/MS/MS system and controlled via AB Sciex Analyst Software to carry out the separation and quantification of analytes in this study. A Phenomenex HydroRP column (4 μm, 250 × 2 mm) was employed for separation under positive electrospray ionization (ESI) conditions. UFLC was performed at 40°C with binary flow at a rate of 0.3 mL/min with mobile phase A consisting of 0.1% formic acid in ultra-pure water and mobile phase B consisting of 0.1% 3 formic acid in acetonitrile. The method began with a pre-run equilibration of 8 minutes at 100% A, followed by injection of 20 μL of sample. A linear gradient increased to 100% B from 0-8 minutes, isocratic 100% B from 8-10 minutes, and finally a linear decrease to 100% A from 10-13 minutes. GSH was detected in multiple reaction monitoring (MRM) mode with Q1 m/z 307.985 and Q3 m/z 179.100.

Calibration curves
Determination of free GSH, GSSG, and total soluble GSH The GSH standard curve (Supplementary Figure S1) was constructed by plotting the known concentrations of GSH calibration standards versus the corresponding peak areas. The internal standard was employed to account for minor variations in sample volume due to evaporation since the same amount of NAC should be present in each sample. Correction of GSH peak area was done by multiplying each individual GSH peak area ( , ) by the ratio of the average NAC peak area ( ) to the NAC peak area of the individual sample ( , ) as follows: In order to validate the accuracy of the GSH standard curve for determination of GSSG following reduction, reduced standards prepared with known GSSG concentrations were plotted against the resulting peak areas to confirm a direct correlation with the GSH calibration curve (Supplementary Figure S1). GSSG was calculated by subtracting free GSH from total GSH and dividing by two.
All concentrations are expressed in nanomoles per milligram of protein as determined from the Bradford assay. The peak areas resulting from reduction of GSSG check standards matched the 4 GSH calibration curve, indicating that complete reduction of GSSG and stoichiometric yield of GSH were achieved over the full concentration range of interest.

Estimation of PSSG
Levels of PSSG in the lens samples were estimated by solving the regression equation fit to a calibration curve (Supplementary Figure S2) generated from analyzing GSSG standards of known concentration. Standards of GSSG were used since analytical-grade PSSG is not commercially available. The peak areas measured correspond to GSH that was liberated from protein by reduction with DTT. The concentration of PSSG was expressed as nmol/mg of dried tissue.  Figure S1: Calibration curve for determination of free GSH, GSSG, and total soluble GSH.

Supplementary
Standards of known concentrations of GSH and GSSG were reduced and processed as described in Supplementary Methods. The yellow diamonds represent GSH calibration standards (0-50 μM).
The black line is a linear least-squares regression fit to data from GSH standards, and the blue squares represent GSSG quality-check standards (0-25 μM). Concentrations of GSSG were multiplied by 2 since the stoichiometric yield of GSSG reduction is 2 GSH per GSSG. 7

Supplementary Figure S2: Calibration curve for estimation of [PSSG].
Glutathione disulfide (GSSG) standards of known concentration were reduced with DTT and processed in parallel with lens samples. Plotted are the known concentrations (multiplied by a factor of two) vs. the resulting chromatographic peak areas corresponding to GSH. The yellow diamonds represent peak areas resulting from reduced GSSG standards. The black line is a linear least-squares regression fit to the data. Concentrations of GSSG were multiplied by 2 since the stoichiometric yield of GSSG reduction is 2 GSH per GSSG while reduction of PSSG yields 1 GSH per PSSG. 8