The effects of GPX-1 knockout on membrane transport and intracellular homeostasis in the lens

Abstract

Glutathione peroxidase-1 (GPX-1) is an enzyme that protects the lens against H2O2-mediated oxidative damage. The purpose of the present study was to determine the effects of GPX-1 knockout (KO) on lens transport and intracellular homeostasis. To investigate these lenses we used (1) whole lens impedance studies to measure membrane conductance, resting voltage and fiber cell gap junction coupling conductance; (2) osmotic swelling of fiber cell membrane vesicles to determine water permeability; and (3) injection of Fura2 and Na+-binding benzofuran isophthalate (SBFI) into fiber cells to measure [Ca2+]i and [Na+]i, respectively, in intact lenses. These approaches were used to compare wild-type (WT) and GPX-1 KO lenses from mice around 2 months of age. There were no significant differences in clarity, size, resting voltage, membrane conductance or fiber cell membrane water permeability between WT and GPX-1 KO lenses. However, in GPX-1 KO lenses, coupling conductance was 72% of normal in the outer shell of differentiating fibers and 45% of normal in the inner core of mature fibers. Quantitative Western blots showed that GPX-1 KO lenses had about 50% as much labeled Cx46 and Cx50 protein as WT, whereas they had equivalent labeled AQP0 protein as WT. Both Ca2+ and Na+ accumulated significantly in the core of GPX-1 KO lenses. In summary, the major effect on lens transport of GPX-1 KO was a reduction in gap junction coupling conductance. This reduction affected the lens normal circulation by causing [Na+]i and [Ca2+]i to increase, which could increase cataract susceptibility in GPX-1 KO lenses.

Fig. 1. The structure of the lens, its internal circulatory system and the distribution of functional gap junction connexins

Fig. 2

The series resistance (R_S, KΩ) due to gap junction coupling between the point of recording r cm from the lens center and the surface of the lens at r = a cm. The data are graphed as a function of the fractional distance from the lens center, r/a. Insets show expanded views of R_S in the DF. The resistance is clearly lower in WT than GPX-1 KO lenses, indicating less gap junction coupling conductance in the KO lenses. The best fit values of coupling conductance (S/cm² of cell-to-cell contact) used to generate the smooth curves were WT G_DF = 0.96 and G_MF = 0.44; GPX-1 KO G_DF = 0.60 and G_MF = 0.26

Fig. 3

Gating properties of gap junction channels in 2-month-old lenses. To test pH gating properties, the lens was superfused with Tyrode solution that had been bubbled with 100% CO₂ for about 10 min. a Gating of gap junction channels in DF of WT lenses. There was a large increase in R_S when pH dropped, indicating R_DF had increased significantly. b Gating properties of gap junction channels in DF of GPX-1 KO lenses. The data show a similar pattern of change in R_S as seen in WT lenses, suggesting no differences in DF gating. C Gating properties of gap junction channels in MF of 2-month-old WT lenses. When the point of voltage recording is in the MF, the increase in R_S depends on the series connection of R_DF and R_MF. The increase is relatively small, consistent with the increase of R_DF and not R_MF. These data imply that MF gap junction channels are not pH-sensitive. d Gating properties of gap junction channels in MF of 2-month-old GPX-1 KO lenses. The data show a similar pattern of change in R_S as seen in WT lenses, suggesting no differences in MF gating

Fig. 4

Fiber cell membrane water permeability. a Typical swelling assay of a fiber cell membrane vesicle formed from isolated DF. This example was from a GPX-1 KO lens. Dashed line shows our estimate of the initial rate of volume change, which was used to calculate the water permeability, p_m. Both swelling and shrinking assays were used as the bath osmolarity was switched between 324 and 487 mM. b Average results showing the p_m of DF from GPX-1 KO and WT lenses. There is no statistically significant difference

Fig. 5

Western blots of Cx46, Cx50 and AQP0 in WT and GPX-1 KO lenses. Membrane proteins were isolated from the lenses of WT and GPX-1 KO mice as described in “Materials and Methods.” a Western blot of Cx46 using an antibody to the C terminus, which is cleaved at the DF-MF transition; hence, these data represent DF only. b Western blot of Cx46 using an antibody to the inner loop, which is not cleaved; hence, these data represent Cx46 in the DF and MF. The MF contains a number of cleavage products of Cx46, and these products are present in either WT or KO lenses; thus, they represent normal cleavage that appears to be independent of the oxidative damage. c Western blot of Cx50 using an antibody to its C terminus, which is cleaved at the DF-MF transition; hence, these data represent DF only. d Western blot of AQP0 using an antibody to its C terminus. Though there is some cleavage of AQP0's C terminus with age and internalization within the lens, it is not as abrupt as for the connexins; hence, these data represent uncleaved AQP0 in both DF and MF. e Quantitative comparison of the antibody staining in GPX-1 KO relative to WT lenses. Both Cx46 and Cx50 were significantly reduced (P < 0.05) in KO lenses, while there was no significant difference in detectable AQP0 (P = 0.36). Data are the mean ± SD of three or four blots for each antibody tested

Fig. 6

Comparison of the distribution of [Ca²⁺]_i in 2-month-old WT and GPX-1 KO lenses. Black circles represent [Ca²⁺]_i data, which were pooled from 12 different 2-month-old WT lenses, each contributing two to five locations. The data follow the black smooth curve, which was generated from the electrodiffusion model given in Eq. 11. The radial gradient suggests a flux of calcium flowing from the lens center to surface. Triangles are data from 16 different 2-month-old GPX-1 KO lenses. Overall, the MF [Ca²⁺]_i in 2-month-old GPX-1 KO lenses is higher than that in 2-month-old WT lenses at all depths into the lens but particularly in central fiber cells. However, the KO data were well fit by the electrodiffusion model (dashed line), indicating a functional circulation

Fig. 7

Comparison of the distribution [Na+]_i in 2-month-old WT and GPX-1 KO lenses. Black circles represent [Na⁺]_i data from 12 different 2-month-old WT lenses. The data follow the black smooth curve generated from the electrodiffusion model given by Eq. 11. The data suggest a flux of Na⁺ from the center to the surface. Triangles are data from nine different 2-month-old GPX-1 KO lenses. Overall, the [Na⁺]_i in 2-month-old GPX-1 KO lenses is higher than that in 2-month-old WT lenses at all depths into the lens, but the data still follow a smooth diffusion curve (dashed line), suggesting the circulation is still functional