Fig 2: Lens expression of p120-catenin subfamily proteins. (A–O) Immunofluorescent labeling of equatorially cryosectioned juvenile mouse lenses (P7) using antibodies specific for p120-catenin (A–E) δ-catenin (F–J), and Arvcf (K–O). (P) Diagram of a portion of the lens indicating the plane of section and approximate location within the lens for (A–O). Expression within the lens epithelium (B,G,L), the outermost elongating lens fiber cells near the fulcrum (C,H,M), the outermost cortical lens fiber cells within 50 microns of the surface (D,I,N), and lens fiber cells 100 microns from the surface (E,J,O) is depicted. Note that p120-catenin and Arvcf but not δ-catenin are localized to the junctions of lens epithelial cells and that δ-catenin and Arvcf are much stronger than p120-catenin in lens fiber cells. Arvcf is also localized strongly to bicellular lens fiber cell junctions while δ-catenin appears restricted to lens fiber cell tricellular junctions. (Q–S) Immunofluorescent labeling of transversely cryosectioned mouse embryos at the indicated ages immunofluorescently labeled for Arvcf (red). Blue signal for all panels represents Hoechst labeling of nuclei. A Plan-Apochromat ×40/1.3 Oil objective with wide-field fluorescent microscope was utilized. Scalebars = 25 microns.

Fig 3: Arvcf is required for N-cadherin/β-catenin localization within interlocking protrusions. (A–L) Representative individual images of similarly sized, interlocking protrusions from Arvcf +/+ (A,G) and Arvcf −/− (B,H) lenses immunofluorescently labeled with N-cadherin (A–B) or β-catenin (G–H) were aligned and used to measure the fluorescent intensity of each pixel. The mean intensity for each pixel was calculated and used to generate an artificial image that represents the average signal intensity for N-cadherin (C–D) or β-catenin (I–J) in Arvcf +/+ (C,I) and Arvcf −/− (D,J) fiber cell interlocking protrusion. (E,K) represent an artificial image where the fold change at each pixel position is indicated by a heatmap. Regions where Arvcf +/+ interlocking protrusions have more signal than Arvcf −/− protrusions have warmer (redder) colors and regions where differences are less pronounced have cooler colors (bluer). (F,L) represent an artificial image indicating where statistically significant differences in pixel intensity are located. Note that the tips of interlocking protrusions have the greatest and statistically significant differences (p < 0.05). (M–P) The mean intensity and standard error of pixels from individual interlocking protrusion images along the lines drawn parallel to the distal-proximal or left-right axes from each experimental group are depicted in the graphs. Note the depressed N-cadherin and β-catenin intensity along both axes Arvcf −/− protrusions. All images were taken with a Plan-Apochromat ×63/1.40 oil objective and a Zeiss Airyscan equipped microscope was utilized.

Fig 4: Arvcf is required for lens fiber cell interlocking protrusion morphology. (A–B) Whole lens fiber cells immunofluorescently co-labeled for Arvcf (blue) and N-cadherin (green). The box in A is magnified in (B). The asterisks mark interlocking protrusions with lateral margins that are enriched for both Arvcf and N-cadherin. A Plan-Apochromatic CS2 ×100/1.4 oil objective was utilized with a Leica TCS SP8 STED equipped microscope. (C–D) Whole lens fiber cells immunofluorescently co-labeled for β-catenin (blue) and N-cadherin (green). The box in C is magnified in (D). The arrowheads mark regions of discontinuous signal of both β-catenin and N-cadherin in Arvcf −/− fiber cells (D). Also note that interlocking protrusions are not well defined by β-catenin and N-cadherin in Arvcf −/− fiber cells. A Plan-Apochromat ×63/1.40 oil objective was utilized with a Zeiss Airyscan equipped microscope. (E–P) SEM images of lens fiber cells from 1 month old (E–N) and 7 month old (M–P) Arvcf +/+ and Arvcf −/− lenses located at the indicated depths from the lens surface. Note that while N-cadherin/β-catenin do not delineate interlocking protrusion borders, they are still present in abundance but are often elongated and misshapen. Asterisks mark regions where the characteristic paddle structures appear missing. Lens fiber cells in (M–N) are shaded to view morphology. Scalebars in (E–F) = 10 microns, (G–P) = 1 micron.

Fig 5: Arvcf is required to maintain lens transparency, refraction, and biomechanical properties. (A–B) Images of the eye of 9 month-old mice with the indicated genotypes. The inset is a magnified region of the pupil. Note the presence of an anterior polar cataract is visible in (B). (C–P) Representative images of lenses placed on a dark background that were extracted from the eyes of mice with the indicated genotype. The asterisks mark regions of the superficial lens that remain transparent, the arrowheads indicate regions of cortical spoking, and the arrows indicate additional regions of cortical opacity within the initially transparent and expanding region. The number on the bottom of each image are the number of animals examined at each age. Note that when opacities were observed they were always bilateral. (Q–S) Compilation of several overlaid images taken of a single representative lens with the paths of several beams of light from a He/Ne laser aimed at several positions along the equatorial axis. Note that the approximate location of several focal lengths are visible and that the arrow denotes that Arvcf mutant lenses have a visibly longer focal length. When aimed at a region of opacity, the beam of light is blocked (S). (T) Diagram of the laser beam light path positional groups used for quantitation of the focal length measurements. Measurements from each colored region were grouped together. (U) The graph depicts the mean focal lengths and standard error for laser beam paths that were calculated from all of the measurements taken from multiple positions (indicated by the colored regions in (T)). The asterisk indicates a statistically significant difference from the Arvcf +/+ measurements p < 0.05 (Arvcf +/+ : n = 11, Arvcf +/- : n = 18, Arvcf −/− n = 6). Scalebar = 500microns. (V)) Representative stereomicroscope images of 60-day old mouse Arvcf +/+ (top) and Arvcf −/− (bottom) lenses taken through a 45° mirror unloaded (left) or loaded with 10 coverslips (right). (X) The graph depicts the mean strain in the axial (E) or equatorial dimension following the addition of 1–10 coverslips placed on the anterior pole. The error bars represent the standard error and the asterisks denote data points that are significantly different than the control (p < 0.05). Note that the Arvcf −/− lenses do not deform as readily as the control in both dimensions. (Y) Graphs of the axial (top) and equatorial (bottom) dimensions of control (left) and mutant lenses (right) before coverslips were added (pre) and 2 min following their removal (post). Note that after this time period the lenses have not returned to their pre-load dimensions. Each black line depicts the change between two time points of the same lens from a single experiment.