October 2008
Volume 49, Issue 10
Free
Retinal Cell Biology  |   October 2008
Progressive Morphological and Functional Defects in Retinas from α1 Integrin-Null Mice
Author Affiliations
  • You-Wei Peng
    From the Boys Town National Research Hospital, Omaha, Nebraska; and
  • Marisa Zallocchi
    From the Boys Town National Research Hospital, Omaha, Nebraska; and
  • Daniel T. Meehan
    From the Boys Town National Research Hospital, Omaha, Nebraska; and
  • Duane Delimont
    From the Boys Town National Research Hospital, Omaha, Nebraska; and
  • Bo Chang
    The Jackson Laboratories, Bar Harbor, Maine.
  • Norman Hawes
    The Jackson Laboratories, Bar Harbor, Maine.
  • Weimin Wang
    From the Boys Town National Research Hospital, Omaha, Nebraska; and
  • Dominic Cosgrove
    From the Boys Town National Research Hospital, Omaha, Nebraska; and
Investigative Ophthalmology & Visual Science October 2008, Vol.49, 4647-4654. doi:10.1167/iovs.08-2011
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      You-Wei Peng, Marisa Zallocchi, Daniel T. Meehan, Duane Delimont, Bo Chang, Norman Hawes, Weimin Wang, Dominic Cosgrove; Progressive Morphological and Functional Defects in Retinas from α1 Integrin-Null Mice. Invest. Ophthalmol. Vis. Sci. 2008;49(10):4647-4654. doi: 10.1167/iovs.08-2011.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. The role of integrin/cell matrix interactions between the RPE and the basement membrane in retinal maintenance and function is not well characterized. In this study the functional importance of α1β1 integrin for retinal pigment epithelial cell homeostasis and retinal health was assessed by comparing α1 integrin knockout mice with strain- and age-matched wild-type mice.

methods. Immunolocalization and Western blot analysis of retinas and ARPE19 cells were performed to examine the expression of α1β1 integrin in the RPE. Retinal abnormality was assessed by funduscopy, histology, and transmission electron microscopy. Progressive retinal damage was quantified by direct counting of rod photoreceptors. Light-induced translocation of arrestin and α-transducin was documented by immunohistochemical analysis of retinal cryosections.

results. Integrin α1β1 localizes to the basal aspect of retinal pigment epithelial cells colocalizing with the basal lamina of the RPE. Integrin α1-null mice have delayed-onset progressive retinal degeneration associated with thickening of the basement membrane, dysmorphology of basal processes, synaptic malformations, and funduscopic abnormalities. Integrin α1-null mice display marked delays in transducin translocation compared with dark-adapted wild-type mice after exposure to light.

conclusions. Collectively, these data suggest an essential role for α1β1 integrin/basement membrane interactions in the RPE in basement membrane metabolism and translocation of transducin in photoreceptors. This is the first report describing evidence supporting an essential role for integrin/basement membrane interaction in the RPE. Further, this report demonstrates a direct link between integrin α1β1 function in retinal pigment epithelial and molecular defects in photoreceptor cell function before retinal abnormality is apparent.

The retinal pigment epithelium (RPE) is composed of a tight sheetlike monolayer of highly polarized epithelial cells. At the apical surface, the RPE interfaces with the photoreceptor outer segments adhering through the interphotoreceptor cell matrix. The RPE is critical for the maintenance, proper function, and health of photoreceptors. One of the well-characterized functions of the RPE occurs at this apical interface, where shed disks are phagocytosed as part of daily retinal maintenance. 1 Disruption of this phagocytic process results in retinal dystrophy in rodents and humans. 2 3 The shed outer segment (OS) disks are bound by apically localized αvβ5 integrin, activating focal adhesion kinase. 4 5 Internalization of the bound OS disks requires the scavenger receptor CD36 and the phosphorylation of Mer tyrosine kinase. 6 7  
Although much is known about the function of αvβ5 integrin in the apical RPE, little is known about integrins that localize to the basal aspect of the RPE, where the basal processes contact the basal lamina of the RPE. It is clear from related studies in other organ systems that integrin binding to matrix proteins in basement membranes modulates cell signaling pathways that play key roles in cell behavior and gene expression programs. 8 Recent studies suggest that retinal pigment epithelial cells adhere to the basement membrane through the interaction of integrin α3β1 and α6β1 with laminins −111, −115, −511, and −521. 9 In proliferative vitreoretinopathy, exposure of RPE to cytokines such as TNF-α may induce integrin α1 and α5 expression, altering the adhesion, migration, and proliferation of retinal pigment epithelial cells. 10 An essential role for integrin function at the basal aspect of the RPE in vivo has not been demonstrated. 
Integrin α1β1 is a collagen-binding integrin well characterized for its roles in regulating adhesion, migration, and proliferation. 11 Integrin α1-null mice were produced more than 10 years ago. The animals were reportedly without obvious phenotype; however, fibroblasts from these mice did show deficits in adhesion to collagen matrix. 12 Since then, integrin α1β1 has been shown to influence the progression of a number of different diseases, including diabetes, Alport syndrome, inflammatory bowel disease, and arthritis. 13 14 15 16 Although the specific mechanisms whereby α1β1 integrin blockade influences disease progression in these models remains unclear, it is known that neutralization of α1β1 integrin influences cell signaling, matrix remodeling, and regulation of matrix metalloproteinases. 
It is also known that α1β1 integrin is expressed on cultured retinal pigment epithelial cells. 10 In retinal pigment epithelial cell culture systems, it has been demonstrated that α1β1 integrin–mediated activation of mitogen-activated protein kinase (MAPK) can influence matrix remodeling in collagen gel contraction assays. 17 We surmised that α1β1 integrin might play important functional roles in retinal pigment epithelial cells in vivo. Here we explore this possibility through the structural and functional characterization of the neuroretina and the RPE of integrin α1-null mice compared with wild-type littermates. 
Materials and Methods
Mice
α1 Integrin knockout mice were a gift from Humphrey Gardner (Novartis, Cambridge, MA). 12 All mice were on the 129 Sv background (129S4/SvJae, white-bellied agouti) and were bred in-house. All animal experiments were conducted in compliance with an approved Institutional Animal Care and Use Committee animal protocol and in agreement with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 
Immunofluorescence Microscopy for α1β1 Integrin
Mouse retinas were frozen in aqueous embedding compound, sectioned at 4 μm, and fixed by immersion in cold acetone for 10 minutes. Slides were air dried, rehydrated, and reacted with a hamster monoclonal antibody specific for α1β1 integrin heterodimer (a gift from Biogen-Idec, Cambridge, MA), rabbit polyclonal antibodies against β1 integrin (Chemicon, Temecula, CA), or anti-CD31 antibodies (Cedar Lane Ltd., Hornby, ON, Canada). The antibodies were diluted at 1:100 into 7% nonfat dry milk in PBS. After several washes, sections were reacted with FITC-conjugated anti-rabbit or anti-hamster secondary antibodies (Vector Laboratories, Burlingame, CA) or Alexa 568–conjugated anti-hamster antibodies (for dual immunofluorescence), washed, and coverslipped. The fluorescent images were recorded with a fluorescence microscope (Axioplan 10; Carl Zeiss, Oberkochen, Germany) interfaced with a digital camera (SPOTFLEX-CE) supported by Image Pro-Plus software (Media Cybernetics, Bethesda, MD). Preimmune rabbit IgG or isotype-matched nonreactive hamster monoclonal antibodies were used as controls, and neither showed any immunoreactivity (data not shown). 
Funduscopic Analysis
Clinical retinal evaluation was performed as described previously. 18 Pupils were dilated with atropine and evaluated by indirect ophthalmoscopy using a 78-D lens. Photographs were taken using a small animal fundus camera (Genesis; Kowa, Torrance, CA). 
Cell Culture
ARPE-19 cells were purchased from the American Type Culture Collection (Manassas, VA) and were grown in Dulbecco modified Eagle medium/Ham F-12 (1:1) supplemented with 1% fetal calf serum, penicillin/streptomycin antibiotics (Gibco, Grand Island, NY), and 1% bovine retinal extract. Retinal extracts were prepared from freshly isolated bovine retinas using the method described by Hu and Bok. 19 Cells were differentiated by culturing for approximately 3 months under low serum (1% FCS) at confluence. Prolonged culture combined with the inclusion of retinal extract promoted tight junction formation with ZO-1 expression at cell junctions and a cobblestone morphology consistent with earlier reports. 20  
Western Blot Analysis
Neural retinas were dissected from the retinal pigment epithelial choroid. The lens was removed from the eye, and the optic nerve was cut to release the neural retina. Retinal extracts were prepared from each compartment with RIPA buffer. Protein was quantified using the Bradford microassay (Pierce, Rockford, IL). Ten micrograms protein for β1 and 40 μg protein for α1 were analyzed. Samples were incubated for 40 minutes at 55°C in the presence of sample buffer without reducing agents and then were loaded in a 7% acrylamide gel. The transfer was at 35 V, 4°C, overnight. Membranes were blocked by incubation for 4 hours at 4°C in blocking solution containing 10% nonfat dry milk in PBS. For the integrin β1 subunit, the first antibody (1/100, rabbit anti-integrin β1 polyclonal antibody, AB1952; Chemicon) was prepared in blocking solution overnight at 4°C, and the second antibody (1/20,000, anti-rabbit A9169; Sigma, St. Louis, MO) was prepared in blocking solution for 1 hour at room temperature. For the integrin α1 subunit, the first antibody (1/1000, hamster anti-mouse H318; Biogen-Idec) was prepared in blocking solution overnight at 4°C, and the second antibody (1/500, biotinylated goat anti-hamster) was prepared in blocking solution for 1 hour at room temperature. Blots were developed using streptavidin-HRP conjugate (1/500; Bio-Rad, Hercules, CA) in blocking solution for 1 hour at room temperature. For ARPE-19 cells, 10 μg protein for β1 integrin and 40 μg protein for α1 integrin were analyzed from cells grown for 3 months after confluence in DMEM/F12 (1:1), 1% FCS, and 1% bovine retina extract. Samples were run on 7% PAGE gels under reducing conditions for β1 integrin and nonreducing conditions for α1 integrin. Blots were blocked overnight at 4°C using 10% nonfat dry milk in PBS. Incubation was conducted with a rabbit polyclonal anti-β1 integrin antibody (AB1952; Chemicon) in blocking solution at 1/100 overnight and was developed with horseradish peroxidase (HRP)– conjugated anti-rabbit antibodies in blocking solution at 1/7500 for 1 hour at room temperature. For anti-α1 integrin, blots were incubated with a mouse monoclonal anti-α1 integrin antibody (MAB1973; Chemicon) in 5% nonfat dry milk at 1/100 dilution overnight and were developed using anti-mouse HRP-conjugated antibodies diluted in 5% nonfat dry milk at 1/1000 for 1 hour at room temperature. 
Light/Dark Adaptation
Procedures for handling of animals followed the guidelines of the National Institutes of Health. For dark adaptation, the animals were kept in cages in a light-proof darkroom without any detectable light. For light adaptation, the animals were kept in a darkroom for 8 hours of dark adaptation, and then they were kept in transparent cages under various intensities of diffuse white fluorescent light (1500 lux intensity at cage level) for 1 hour. Four hours of dark adaptation and 1 hour of light adaptation at this light intensity is adequate for proteins to be translocated in rod photoreceptors in wild-type mice. 21  
Immunohistochemistry for Arrestin and Transducin
Eyes were quickly removed from animals humanely killed under deep anesthesia. After removal of the anterior segments, the posterior eyecups were fixed in 4% paraformaldehyde in 100 mM sodium phosphate buffer (PB; pH 7.3) at 4°C. The tissue was transferred sequentially into 5% and 30% sucrose in PB, each at 4°C overnight. Retinal sections were cut at 4-μm thickness with a cryostat (Microm, Eden Prairie, MN) and were mounted on gelatin-coated slides. Retinal sections were incubated with 5% normal goat serum (Vector Laboratories) in PBS for 1 hour at room temperature and were incubated with primary antibodies overnight at 4°C (anti-arrestin antibody [Sigma] 1:500; anti-transducin α [Cytosignal, Irvine, CA] 1:1000), followed by three washes in PBS of 15 minutes each. The sections were then incubated with either Alexa 594–conjugated anti-mouse immunoglobulin antibody (Invitrogen, Carlsbad, CA) 1:250 or Alexa 488–conjugated anti-rabbit immunoglobulin antibody (Invitrogen) 1:250 for 2 hours at room temperature. Staining reactions were terminated by three washes with PBS (5 minutes each), and the slides were coverslipped with 50% glycerol in PBS for viewing under a fluorescence microscope (Axioplan 10; Carl Zeiss). Images were recorded, as described, for integrin immunostaining. All incubation and wash buffers contained Triton X-100 (0.3%). 
Electron Microscopy
Eyecups were fixed in 3% glutaraldehyde and 4% paraformaldehyde in 0.1 M cacodylate buffer (pH 7.3) at 4°C overnight, followed by postfixation in 1% osmium tetroxide in 0.1 M cacodylate buffer for 1 hour. After washing again in cacodylate buffer, these sections were dehydrated in a graded series of alcohol and propylene oxide and were stained with 1% uranyl acetate for 1 hour, infiltrated, and embedded in resin (EmBed 812; EMS, Fort Washington, PA). Ultrathin sections (70 nm) of the retinal cross-sections were mounted on copper grids and then stained with uranyl acetate and bismuth. The specimens were examined under a microscope (JEM-1010; JEOL, Tokyo, Japan). Digitized images were acquired using a charge-coupled device camera (Orca; Hamamatsu Photonics, Bridgewater, NJ) with AMT Advantage 12-HR software (version 5.4.2.239; Advanced Microscopy Techniques, Danvers, MA). 
Counting Total Photoreceptor Numbers
The number of nuclei in the ONL of retinal sections at different time points was calculated. At each time point, nuclei were counted from retinal sections in central areas 2 mm eccentric from the optic nerve head site, which can be recognized under a light microscope. The central one-third of the entire length of retinal cross-section traversing the entire retina width passing through the optic nerve head from the superior ora serrata to the inferior ora serrata was defined as the central region. Vertical and horizontal sections across this site were examined. Serial retinal sections (4–5 μm each section for 12 sections) were taken. Mean number of nuclei in the ONL in these sections was counted. Only well-oriented sections with straight rod outer segments that did not have oblique orientations were used. Data obtained from integrin α1-null mice were compared with those from the wild-type retina of the same eccentricities and ages. Three animals for each age and genotype were used; data were analyzed statistically. 
Statistical Analysis
Results are expressed as mean ± SD. Statistical significance was determined using the Student’s t-test with Bonferroni correction. P < 0.05 was considered significant. 
Results
Immunolocalization of integrin α1β1 in wild-type adult mouse retinas was performed using antibodies that recognize the heterodimer. 22 Figures 1A and 1Bshows bright immunostaining in a linear pattern at the basal aspect of the RPE, consistent with the polarized distribution of integrin α1β1 previously described. 23 As expected, β1 integrin-specific antibodies show positive immunostaining for the same structures but are more widely distributed in the vessels of the choroid plexus (Fig. 1C)because this antibody reacts to all β1-containing integrin heterodimers. Immunostaining for the α1β1 integrin and the integrin β1 subunit is also observed in the capillaries of the neural retina, as shown by dual immunofluorescence immunostaining using anti-α1β1 integrin and anti-CD31 antibodies (Figs. 1E 1F 1G) . No immunostaining for α1β1 integrin was observed in retinas from integrin α1-null mice (Fig. 1D) , demonstrating the specificity of the antibody. 
To confirm the presence of α1β1 integrin in mouse retinas biochemically, Western blot analysis were performed. Protein extracts were produced from the neural retina and the remaining retinal pigment epithelial choroid, which contains the RPE and the choroid. Extracts were fractionated on PAGE gels and blotted onto nylon membranes, and duplicate blots were probed using antibodies specific for either α1 or β1 integrin subunits. The results in Figure 2show that wild-type mice express α1 and β1 integrin subunits predominantly in the retinal pigment epithelial choroid, which is consistent with findings of our immunolocalization studies. Expression in the neural retina was expected based on immunostaining data (Fig. 1)but was not observed, presumably because of the limited sensitivity of the Western blot analysis. For integrin α1-null mice, no α1 band demonstrated the specificity of the antibody on Western blots. The β1 subunit was still present, which was expected because it can form heterodimers with other α chains. Lane 5 on the two blots in Figure 2shows that differentiated ARPE-19 cells express the α1 and β1 integrin subunits, demonstrating that retinal pigment epithelial cells express α1β1 integrin heterodimers. 
Clinical, Structural, and Ultrastructural Defects in Integrin α1-Null Retinas
Because integrin α1β1 appears to localize to the basal aspect of the RPE, we presumed it may be important for retinal function. The gene knockout mouse for integrin α1 does not exhibit any life-threatening phenotypic characteristics 12 ; therefore, we used this mouse model. Several integrin α1 knockout mice as well as age- and strain-matched controls were examined by funduscopic photography. Spots consistent with retinal degenerative disease were observed in retinas of knockout mice but never in strain- or age-matched controls. Spots were small and infrequent in animals younger than 7 to 8 months of age. Shown in Figure 3Aare representative funduscopic photographs of 10-month-old wild-type and integrin α1-null (α1−/−) mice. The spots were prevalent in the peripheral retina (shown) but were more difficult to find in the central retina and appeared to worsen with age. Figure 3Bshows a typical cross-section of the retina from a 15-month-old integrin α1-null mouse and a strain- and age-matched wild-type mouse. The integrin α1-null mouse exhibits thinning of the outer nuclear layer in the retina relative to wild-type mice. Shortening of the outer segments of the photoreceptors is also evident in the retinas of the integrin α1-null mice relative to wild-type mice. This is often associated with retinal degeneration. 24 25 Given that integrin α1β1 is expressed at the basal aspect of the RPE (Fig. 1)and that this collagen-binding integrin is known to influence matrix remodeling in other experimental systems, 11 15 we surmised that we may observe defects in the basal lamina of the RPE in diseased retinas. Several 12-month-old integrin α1-null mice and age- and strain-matched wild-type mice were analyzed. From each mouse, one eye was prepared for histology and the other for transmission electron microscopy. Once pathologic findings were confirmed by histologic staining and counting of the outer nuclear layer, the other eye was processed for thin sectioning. Figure 3Cshows the typical ultrastructure of the basement membrane just beneath the basal processes of the RPE in wild-type mice and integrin α1-null (α1−/−) mice. We chose an image from a strain-matched, 24-month-old wild-type mouse to demonstrate that the normal architecture of the basement membrane and the basal processes of the RPE are maintained even in much older mice, showing a well-defined and uniform lamina densa flanked by lamina rara interna and externa, associated with a confluence of retinal pigment epithelial basal processes. In contrast, the basement membrane of the integrin α1-null mouse was markedly thickened, and the basal processes were retracted and disorganized (arrow). As noted in Figure 1E 1F 1G , the capillaries of the neuroretina were also immunopositive for α1β1 integrin. We did not observe any significant differences in capillary basement membrane ultrastructure when comparing wild-type and integrin α1-null mice (data not shown). 
To better characterize the progressive retinal degeneration in the α1-null mice, we analyzed retinal sections encompassing the full length of the retina. The total number of rods was quantified as described in Materials and Methods. The process was repeated for three wild-type mice and three integrin α1-null mice at 3, 6, 9, 12, and 15 months of age, and the data were statistically analyzed. The results in Figure 4show that loss of rod photoreceptors was first evident in the integrin α1-null retinas compared with wild-type mice when the animals were 12 months of age and progressively worsened through 15 months of age. At earlier time points analyzed, there was no significant difference in rod numbers. 
Synaptic defects are often observed in photoreceptor synaptic terminals during photoreceptor degeneration. 26 27 28 29 We compared the synaptic ultrastructure of integrin α1-null mice with that of age- and strain-matched wild-type mice to determine whether synaptic abnormalities were present. Figure 5shows synaptic defects where bipolar cell dendrites in photoreceptor synaptic terminals were missing for rod and cone photoreceptors in integrin α1-null mice older than 12 months of age. In 12-month-old wild-type mouse retinas, synaptic complexes with invaginating horizontal processes and bipolar cell dendrites can be easily and frequently detected (more than 85%) under an electron microscope (Figs. 5A 5C) . In α1-null mice, however, few intact invaginating ribbon synapses with characteristic synaptic bipolar cell dendrites were present in rod and cone terminals. More than 75% of cone synaptic terminals have only floating ribbons (arrows) without any postsynaptic component (Fig. 5B) . The flat contact synapses (small arrows) on the basal side, however, appeared intact, suggesting the synaptic contacts between cones and OFF-bipolar cells remained normal. For each calculated cone terminal, the orientation of the cone terminal was determined by the clear visible basal side, with flat contact synapses to minimize the possibility of problems in the sectioning angle. Similarly, more than 80% of rod spherules in the integrin α1-null mice displayed only a dyad structure or a double ribbon without invaginating bipolar cell dendrites (Fig. 5D) . All these results suggest the photoreceptors in integrin α1-null mice may lose the invaginating bipolar cell dendrites. 
Defects in Light-Induced Translocation of Transducin α-Subunit but not Arrestin in Integrin α1-Null Retinas
After the dark-adapted animals were exposed to light, the transducin α-subunit rapidly translocates from the outer segments to the inner segments of rod photoreceptors, and arrestin translocates from the inner segments to the outer segments. This light-driven massive protein translocation is believed to be an important aspect of tuning photoreceptor sensitivity to various light intensities. 30 Recent studies have suggested that RPE may play important roles in this translocation. In RPE65 knockout mice, which lack the retinoid isomerase needed to convert all-trans retinyl ester to 11-cis retinol for rhodopsin activation, the translocation of the transducin α-subunit and arrestin is blocked. 31 We surmised that defects in α1β1 integrin function in retinal pigment epithelial cells may influence protein translocation. We compared protein translocation in integrin α1-null mice with that of age-matched wild-type mice. Four-month-old mice (before any signs of retinal degeneration are observed) were dark adapted overnight and exposed to 1500 lux illumination for 1 hour, and the eyes were immediately fixed. Retinal sections were immunostained with antibodies specific for the arrestin or transducin α-subunit. Figure 6shows that light-induced translocation of arrestin is apparently normal in the integrin α1-null mice (Figs. 6C 6D 6G 6H) . Conversely, translocation of the transducin α-subunit is markedly delayed in integrin α1-null mice compared with wild-type mice (Figs. 6A 6B 6E 6F) . In integrin α1-null mice, after dark adaptation for 8 hours, transducin α was found primarily in the rod outer segments (Fig. 6E) , as in the wild-type mouse (Fig. 6A) . After 1 hour of light adaptation (1500 lux), transducin α in the wild-type mouse was translocated to the inner segments (Fig. 6B) . In integrin α1-null mice, however, after 1 hour of light adaptation of the same intensity, the strongest immunostaining of transducin α was still in the outer segments of the rod photoreceptors (Fig. 6F) , indicating the light-induced translocation of transducin α was delayed. The defect in transducin α-subunit translocation was apparent before photoreceptor cell degeneration was observed. Interestingly, the delay in transducin translocation was more pronounced in the peripheral retina than in the central retina (data not shown). Delayed translocation of transducin α-subunit was observed in integrin α1-null mice at 2, 6, 8, 12, and 15 months of age, and the degree of this delay compared with that of age- and strain-matched wild-type mice was not qualitatively different across these age groups (not shown), suggesting that the protein translocation defect in these animals did not worsen as a function of progressive retinal degeneration. Transducin translocation in the rod photoreceptors of integrin α-null mice was not completely blocked. After longer light adaptation (more than 4 hours), most of the transducin was translocated to the inner parts of rod photoreceptors, as in wild-type mice (data not shown). 
Discussion
Epithelial cells adhere to basement membranes through cell matrix integrin interactions, and, in a broad array of cell culture and organ systems, these interactions have been shown to influence cytoskeletal organization and cell signaling. 32 These cell matrix integrin interactions play pivotal roles in the development and maintenance of a highly differentiated state, and disruption of these interactions has profound effects on tissue development and function. Despite the proliferation of this evidence in other systems, no studies have addressed whether integrin interactions with basement membrane proteins in the retinal pigment epithelial compartment influence retinal pigment epithelial function or retinal health. In this study, we show that α1β1 integrin localizes to the basal aspect of the RPE, where it likely binds to matrix molecules in the basal lamina of the RPE. Expression of α1β1 integrin in cultured ARPE19 cells confirms retinal pigment epithelial expression of this integrin. α1β1 Integrin is a collagen-binding integrin that displays distinct recognition for various collagen subtypes. 33 One well-established function of α1β1 integrin/collagen interaction is to activate signaling cascades that influence MMP regulation. 34 35 36 37 Basement membrane thickening in the basal lamina of the RPE in integrin α1-null mice suggests an imbalance in synthesis or turnover of basement membrane proteins. This basement membrane thickening results in an apparent loss of focal contacts between the basal processes of the RPE and the basement membrane. These morphologic changes are progressive and become apparent when the animals approach 10 months of age, which coincides with the onset of our observed reduction in the rod number. This accumulation of electron-dense material is reminiscent of basal laminar deposit (BLD) associated with the aging macula in humans and with certain mouse models. 38 39 BLD, which is largely composed of extracellular matrix, 39 accumulates between the RPE and a structurally identifiable basement membrane. In our mice the basement membrane was not identifiable separately from the thickened material, suggesting it may be distinct from BLD. 
Synaptic malformations were observed at synaptic termini for rods and cones, suggesting both photoreceptor cell types are influenced by α1β1 integrin signaling in the RPE. Recent studies on different RP animal models indicate that the absence of invaginating bipolar cell dendrites from photoreceptor terminals may be an early sign of photoreceptor death. 28 29 Thus, rods and cones show synaptic malformations in integrin α1-null mice, suggesting that degeneration involves both photoreceptor cell types. 
Collectively, these data strongly suggest that basement membrane/integrin interactions at the retinal pigment epithelial interface with the basal lamina of the RPE play an important role in regulating retinal pigment epithelial cell function. Absence of α1β1 integrin/cell matrix interactions at this interface may influence retinal pigment epithelial cell morphology, basement membrane homeostasis, and photoreceptor health, and it culminates in photoreceptor degeneration. It should be noted that α1β1 integrin also localizes to the capillary endothelium; thus, some of the pathologic changes in α1-null retinas might have emanated from this compartment. We feel this was unlikely, however, because we observed no structural or ultrastructural abnormalities in the capillaries of α1-null mice compared with wild-type mice. 
Eighteen integrin α-subunits and eight β-subunits can heterodimerize in various combinations to produce 24 known integrin heterodimers. 8 Integrins bind a broad spectrum of ligands with the common feature that the integrin binding site within the ligand generally contains the arginine-glycine-aspartic acid sequence. The study of integrin function in the RPE has been focused largely on the role of αvβ5 integrin in phagocytosis of shed disks at the apical surfaces of the cells, 4 40 a process that interestingly appears regulated through the activation of focal adhesion kinase signaling. 5  
Early work on integrins at the basal aspect of retinal pigment epithelial cells showed that β1 integrins promoted the attachment of cultured cells to provisional matrices, including the basal lamina of the RPE. 10 41 Although early studies to define polarized expression of different integrins on retinal pigment epithelial cells were inconclusive, 42 more recently, a specific subset of basally localized integrins has been described that may function in retinal pigment epithelial attachment to matrix. 23 One study defined integrin α3 and α6 containing heterodimers bind to laminin 511, 9 suggesting that these integrin/cell matrices may also be important for retinal pigment epithelial cell function. 
The potential role for α1β1 integrin in collagen matrix remodeling has been explored for a variety of tissues and cell types, including retinal pigment epithelial cells. The experimental platform most widely used involves culturing cells in a collagen gel and assessing the function of specific integrins in facilitating contraction of these gels, which is considered a measure of collagen matrix remodeling. Through this approach, integrin α1β1 was first implicated in fibroblast matrix remodeling using neutralizing antibodies against the α1 integrin subunit. 34 These observations correlate well with later studies showing collagen 1 dysregulation, matrix accumulation, and MMP-13 upregulation in the skin of integrin α1-null mice. 43 Similarly, integrin α1β1 influences collagen gel contraction in cultured mesangial cells from the renal glomerulus through the activation of the MAPK signaling pathway. 44 45 Recent in vivo data bear out these observations, showing that integrin α1β1 does influence MAPK-mediated regulation of MMP-2, -9, and -14 and that dysregulation of this pathway can influence progressive glomerular pathogenesis in various disease models. 37  
Interestingly, recent studies have shown that the collagen gel contraction of ARPE-19 cells is inhibited by neutralizing antibodies against α1 integrin 17 and that this effect is mediated through focal adhesion kinase and downstream activation of MAPK. 46 In light of these findings and the work discussed here, it is likely that α1β1 integrin in retinal pigment epithelial cells influences basement membrane homeostasis, at least in part, by altering the turnover of extracellular matrix through influences on MMP expression. These influences may in part account for the basement membrane thickening and dysmorphology of retinal pigment epithelial basal processes observed in our study. 
Light-driven massive protein translocation is a prominent feature in photoreceptors. It was discovered more than 20 years ago and was reconfirmed more recently. 21 47 48 49 Its functional roles in photoreceptors, however, remain unclear, and the mechanism of such massive protein translocation is under investigation. 28 50 In healthy eyes, when light stimulation reaches a certain threshold intensity, 51 transducin is moved out of the outer segments to reduce the activation of the phototransduction pathway, and arrestin is transported to the outer segments to inactivate rhodopsin. 52 One of the functions of protein translocation, therefore, may be to prevent constant activation of saturated rods under strong light intensity. 50 Diseases with defects in photoreceptor protein translocation show retinal degeneration. 53 54 55 It is not known how aberrant protein translocation leads to photoreceptor degeneration. Defects in protein translocation may delay the inactivation of the pathway under light adaptation. Under these circumstances, long periods of high-intensity light adaptation may induce the degeneration of photoreceptors. 53 56  
The influence of the RPE on protein translocation in photoreceptors is unclear. RPE65 knockout mice show defects in translocation of transducin and arrestin, 57 suggesting the RPE may have a dominant influence on the translocation mechanism. Current interpretation is that the absence of retinoid replacement in RPE65 knockout retinas blocks the activation of rhodopsin, which is critical for triggering protein translocation. Our results suggest that the RPE may play more sophisticated roles in affecting photoreceptor protein translocation. One interpretation is that integrin α1β1 signaling regulates the velocity of transducin translocation in rods. Translocation defects in integrin α1β1-null mice cannot be explained by the absence of rhodopsin activation because arrestin translocation is not affected. Our results, therefore, also suggest that the translocation of arrestin and transducin may be regulated by different mechanisms. 
In summary, here we provide the first example suggesting an essential role for integrin/matrix interaction at the basal aspect of the RPE in retinal health. Given the complexity of laminins in the basal lamina of the RPE and the presence of additional integrins at the basal aspect of the RPE, it appears likely that integrin adhesion and signaling at the basement membrane interface play important roles in maintaining retinal pigment epithelial architecture and function. Although this is not surprising considering what we know from related studies in other tissues and organs, it underscores the need for further research aimed at understanding this niche in the RPE. 
 
Figure 1.
 
Immunofluorescence localization of integrin α1β1 at the basolateral aspect of the RPE. An α1β1 integrin-specific antibody shows bright immunostaining at the basal aspect of the RPE (arrow) in the mouse retina (A). Bright-field image of this same stained section (B) clearly identifies the signal just basal to the pigmented epithelium. (C) Immunostaining of wild-type mouse retina for the integrin β1 subunit, which also localizes to the basal aspect of the RPE (arrow). (D) Absence of staining for integrin α1 in retinal sections from the integrin α1-null mouse, demonstrating the specificity of the antibody. Specific immunostaining is also observed in the capillaries of the neuroretina (arrowheads), as confirmed by dual immunofluorescence immunostaining with anti-CD31 antibodies: anti-α1β1 integrin (E), anti-CD31 (F), and merged images (G). PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bars, 50 μm.
Figure 1.
 
Immunofluorescence localization of integrin α1β1 at the basolateral aspect of the RPE. An α1β1 integrin-specific antibody shows bright immunostaining at the basal aspect of the RPE (arrow) in the mouse retina (A). Bright-field image of this same stained section (B) clearly identifies the signal just basal to the pigmented epithelium. (C) Immunostaining of wild-type mouse retina for the integrin β1 subunit, which also localizes to the basal aspect of the RPE (arrow). (D) Absence of staining for integrin α1 in retinal sections from the integrin α1-null mouse, demonstrating the specificity of the antibody. Specific immunostaining is also observed in the capillaries of the neuroretina (arrowheads), as confirmed by dual immunofluorescence immunostaining with anti-CD31 antibodies: anti-α1β1 integrin (E), anti-CD31 (F), and merged images (G). PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bars, 50 μm.
Figure 2.
 
Western blot analysis of α1β1 integrin in mouse retinas and ARPE 19 cells. Detergent extracts of neuroretina (lanes 1 and 3) and retinal pigment epithelial choroid (lanes 2 and 4) after removal of the neuroretina from wild-type mice (lanes 1 and 2) and integrin α1-null mice (lanes 3 and 4) were analyzed by Western blots probed with antibodies specific for either integrin α1 (upper) or β1 (lower) subunits. Western blots (lane 5) of ARPE19 cell extracts probed with antibodies against the human α1 integrin subunit (upper) or the human β1 integrin subunit (lower).
Figure 2.
 
Western blot analysis of α1β1 integrin in mouse retinas and ARPE 19 cells. Detergent extracts of neuroretina (lanes 1 and 3) and retinal pigment epithelial choroid (lanes 2 and 4) after removal of the neuroretina from wild-type mice (lanes 1 and 2) and integrin α1-null mice (lanes 3 and 4) were analyzed by Western blots probed with antibodies specific for either integrin α1 (upper) or β1 (lower) subunits. Western blots (lane 5) of ARPE19 cell extracts probed with antibodies against the human α1 integrin subunit (upper) or the human β1 integrin subunit (lower).
Figure 3.
 
Integrin α1-knockout mice show clinical pathology of the retina associated with thinning of the outer nuclear layer and thickening of the basal lamina of the RPE. (A) Typical funduscopic images from 10-month-old wild-type mice and integrin α1-null (α1KO) mice reveal spots in the peripheral retina of a α1-null mouse characteristic of retinal degeneration. (B) Histologic examination of 15-month-old wild-type and integrin α1-null mice. The retina of the integrin α1-null mouse showed significantly reduced numbers of nuclei in the outer nuclear layer (ONL) when compared with the wild-type retina of the same age. The outer segments (OS) of photoreceptors appear shortened in integrin α1-null mice compared with wild-type mice. (C) Ultrastructural analysis of the basal lamina of the RPE in the 24-month-old wild-type mouse and the 12-month-old integrin α1-null mouse. Basement membranes in the basal lamina of the RPE show retraction of the basal processes of the RPE associated with irregular thickening (arrow) of the basal lamina of the RPE for the integrin α1-null mice. This basement membrane thickening was not evident in integrin α1-null mice before observed reductions in ONL thickness (between 8 and 10 months).
Figure 3.
 
Integrin α1-knockout mice show clinical pathology of the retina associated with thinning of the outer nuclear layer and thickening of the basal lamina of the RPE. (A) Typical funduscopic images from 10-month-old wild-type mice and integrin α1-null (α1KO) mice reveal spots in the peripheral retina of a α1-null mouse characteristic of retinal degeneration. (B) Histologic examination of 15-month-old wild-type and integrin α1-null mice. The retina of the integrin α1-null mouse showed significantly reduced numbers of nuclei in the outer nuclear layer (ONL) when compared with the wild-type retina of the same age. The outer segments (OS) of photoreceptors appear shortened in integrin α1-null mice compared with wild-type mice. (C) Ultrastructural analysis of the basal lamina of the RPE in the 24-month-old wild-type mouse and the 12-month-old integrin α1-null mouse. Basement membranes in the basal lamina of the RPE show retraction of the basal processes of the RPE associated with irregular thickening (arrow) of the basal lamina of the RPE for the integrin α1-null mice. This basement membrane thickening was not evident in integrin α1-null mice before observed reductions in ONL thickness (between 8 and 10 months).
Figure 4.
 
Kinetics of rod loss as a function of age in integrin α1-null mouse retinas compared with wild-type mouse retinas. Data points represent quantitative measures of outer nuclei across the entire retina for wild-type mice compared with integrin α1-null mice at the indicated ages. *Statistically significant differences between wild-type and integrin α1-null groups (P < 0.05).
Figure 4.
 
Kinetics of rod loss as a function of age in integrin α1-null mouse retinas compared with wild-type mouse retinas. Data points represent quantitative measures of outer nuclei across the entire retina for wild-type mice compared with integrin α1-null mice at the indicated ages. *Statistically significant differences between wild-type and integrin α1-null groups (P < 0.05).
Figure 5.
 
Synaptic malformations are observed in rod and cone synaptic terminals in integrin α1-null mice. (A) Retinal section from a 12-month-old wild-type mouse under transmission electron microscopy (TEM) showing the cone synaptic terminal. CP, cone pedicle. In this normal cone photoreceptor synaptic terminal, invaginating synapses (recognized by the synaptic ribbons, as indicated by the arrows) with postsynaptic horizontal cell processes (H) and ON-cone bipolar cell dendrites (B) are located superficially on the basal side of the terminal. (B) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the cone synaptic terminal (CP). Most invaginating synapses (recognized by the synaptic ribbons, as indicated by large arrows) do not have postsynaptic bipolar cell dendrites and are located deep inside the cone synaptic terminal. The basal side can be recognized by conventional flat contact synapses (small arrows). (C) Retinal section from a 12-month-old wild-type mouse under TEM illustrating the outer plexiform layer. The rod synaptic (RS) termini display the normal triad structure. (D) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the outer plexiform layer. Most RS photoreceptor terminals do not have a normal triad structure. Rather, they have double ribbons (arrows) or dyads. The CP does not have any invaginating synapses. Scale bar, 500 nm.
Figure 5.
 
Synaptic malformations are observed in rod and cone synaptic terminals in integrin α1-null mice. (A) Retinal section from a 12-month-old wild-type mouse under transmission electron microscopy (TEM) showing the cone synaptic terminal. CP, cone pedicle. In this normal cone photoreceptor synaptic terminal, invaginating synapses (recognized by the synaptic ribbons, as indicated by the arrows) with postsynaptic horizontal cell processes (H) and ON-cone bipolar cell dendrites (B) are located superficially on the basal side of the terminal. (B) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the cone synaptic terminal (CP). Most invaginating synapses (recognized by the synaptic ribbons, as indicated by large arrows) do not have postsynaptic bipolar cell dendrites and are located deep inside the cone synaptic terminal. The basal side can be recognized by conventional flat contact synapses (small arrows). (C) Retinal section from a 12-month-old wild-type mouse under TEM illustrating the outer plexiform layer. The rod synaptic (RS) termini display the normal triad structure. (D) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the outer plexiform layer. Most RS photoreceptor terminals do not have a normal triad structure. Rather, they have double ribbons (arrows) or dyads. The CP does not have any invaginating synapses. Scale bar, 500 nm.
Figure 6.
 
Integrin α1-null miceshow defects in light-induced protein translocation of the transducin α-subunit but not arrestin. Immunostaining of the transducin α-subunit on wild-type (A, B) and integrin α1-null (E, F) retinas under dark adaptation for 8 hours (A, E) and light adaptation for 1 hour (B, F). Note the failure of transducin-α subunit to completely translocate from the outer segments to the inner segments in light-adapted retinas from α1-null mice (compare B with F). On the other hand, arrestin translocation to the outer segments after 1-hour light adaptation appeared qualitatively similar to that of wild-type mice (compare G, H with C, D). OS, outer segments; IS, inner segments; PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar, 25 μm.
Figure 6.
 
Integrin α1-null miceshow defects in light-induced protein translocation of the transducin α-subunit but not arrestin. Immunostaining of the transducin α-subunit on wild-type (A, B) and integrin α1-null (E, F) retinas under dark adaptation for 8 hours (A, E) and light adaptation for 1 hour (B, F). Note the failure of transducin-α subunit to completely translocate from the outer segments to the inner segments in light-adapted retinas from α1-null mice (compare B with F). On the other hand, arrestin translocation to the outer segments after 1-hour light adaptation appeared qualitatively similar to that of wild-type mice (compare G, H with C, D). OS, outer segments; IS, inner segments; PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar, 25 μm.
The authors thank John Kennedy (Boys Town National Research Hospital, Omaha, NE) for help in figure preparation and Charlotte Lieser for secretarial help. 
StraussO. The retinal pigment epithelium in visual function. Physiol Rev. 2005;85:845–881. [CrossRef] [PubMed]
EdwardsRB, SzamierRB. Defective phagocytosis of isolated rod outer segments by RCS rat retinal pigment epithelium in culture. Science. 1977;197:1001–1003. [CrossRef] [PubMed]
GalA, ThompsonDA, WeirJ, et al. Mutations in MERTK, the human orthologue of the RCS rat retinal dystrophy gene, cause retinitis pigmentosa. Nat Genet. 2003;26:270–271.
FinnemannSC, BonilhaVL, MarmorsteinAD, Rodriguez-BoulanE. Phagocytosis of rod outer segments by retinal pigment epithelial cells requires alpha(v)beta5 integrin for binding but not for internalization. Proc Natl Acad Sci U S A. 1997;94:12932–12937. [CrossRef] [PubMed]
FinnemannSC. Focal adhesion kinase signaling promotes phagocytosis of integrin-bound photoreceptors. EMBO J. 2003;22:4143–4154. [CrossRef] [PubMed]
FinnemannSC, SilversteinRL. Differential roles of CD36 and αvβ5 integrin in photoreceptor phagocytosis by the retinal pigment epithelium. J Exp Med. 2001;194:1289–1298. [CrossRef] [PubMed]
FengW, YasumuraD, MatthesMT, LaVailMM, VollrathD. Mertk triggers uptake of photoreceptor outer segments during phagocytosis by cultured retinal pigment epithelial cells. J Biol Chem. 2002;277:17016–17022. [CrossRef] [PubMed]
TakadaY, YeX, SimonS. The integrins. Genome Biol. 2007;8:215. [CrossRef] [PubMed]
AisenbreyS, ZhangM, BacherD, YeeJ, BrunkenWJ, HunterDD. Retinal pigment epithelial cells synthesize laminins, including laminin 5, and adhere to them through α3- and α6-containing integrins. Invest Ophthalmol Vis Sci. 2006;47:5537–5544. [CrossRef] [PubMed]
JinM, HeS, WorpelV, RyanSJ, HintonDR. Promotion of adhesion and migration of RPE cells to provisional extracellular matrices by TNF-α. Invest Ophthalmol Vis Sci. 2000;41:4324–4332. [PubMed]
PozziA, ZentR. Integrins: sensors of extracellular matrix and modulators of cell function. Nephron Exp Nephrol. 2003;94:e77–e84. [CrossRef] [PubMed]
GardnerH, KreidbergJ, KotelianskiV, JaenischR. Deletion of integrin alpha 1 by homologous recombination permits normal murine development but gives rise to a specific deficit in cell adhesion. Dev Biol. 1996;175:301–313. [CrossRef] [PubMed]
CosgroveD, RodgersK, MeehanD, et al. Integrin α1β1 and transforming growth factor-β1 play distinct roles in Alport glomerular pathogenesis and serve as dual targets for metabolic therapy. Am J Pathol. 2000;157:1649–1659. [CrossRef] [PubMed]
de FougerollesAR, SpragueAG, Nickerson-NutterCL, et al. Regulation of inflammation by collagen-binding integrins α1β1 and α2β1 in models of hypersensitivity and arthritis. J Clin Invest. 2000;105:721–729. [CrossRef] [PubMed]
KrieglsteinCF, CerwinkaWH, SpragueAG. Collagen-binding integrin alpha 1 beta 1 regulates intestinal inflammation in experimental colitis. J Clin Invest. 2002;110:773–1782.
ZentR, YanX, SuY, et al. Gomerular injury is exacerbated in diabetic integrin α1-null mice. Kidney Int. 2006;70:460–470. [PubMed]
BandoH, IkunoY, HoriY, SayanagiK, TanoY. Mitogen-activated protein kinase (MAPK) and phosphatidylinositol-3 kinase (PI3K) pathways differently regulate retinal pigment epithelial cell-mediated collagen gel contraction. Exp Eye Res. 2006;82:529–537. [CrossRef] [PubMed]
HawesNL, SmithRS, ChangB, DavissonM, HeckenlivelyJR, JohnSW. Mouse fundus photography and angiography: a catalogue of normal and mutant phenotypes. Mol Vis. 1999;5:22. [PubMed]
HuJ, BokD. Technical brief: a cell culture medium that supports the differentiation of human retinal pigment epithelium into functionally polarized monolayers. Mol Vis. 2000;7:14–19.
LuoY, ZhuoY, FukuharaM, RizzoloLJ. The effects of culture conditions on heterogeneity and the apical junctional complex of the ARPE-19 cell line. Invest Ophthalmol Vis Sci. 2006;47:3644–3655. [CrossRef] [PubMed]
SokolovM, LyubarskyAL, StrisselKJ, et al. Massive light-driven translocation of transducin between the two major compartments of rod cells: a novel mechanism of light adaptation. Neuron. 2002;34:95–106. [CrossRef] [PubMed]
MendrickDL, KellyDM, duMontSS, SandstromDJ. Glomerular epithelial and mesangial cells differentially modulate the binding specificities of VLA-1 and VLA-2. Lab Invest. 1995;72:367–375. [PubMed]
ZarbinMA. Analysis of retinal pigment epithelium integrin expression and adhesion to aged submacular human Bruch’s membrane. Trans Am Ophthalmol Soc. 2003;101:499–520. [PubMed]
KedzierskiW, LloydM, BirchDG, BokD, TravisGH. Generation and analysis of transgenic mice expressing P216L-substituted rds/peripherin in rod photoreceptors. Invest Ophthalmol Vis Sci. 1997;38:498–509. [PubMed]
MachidaS, KondoM, JamisonJA, et al. P23H rhodopsin transgenic rat: correlation of retinal function with histopathology. Invest Ophthalmol Vis Sci. 2000;41:3200–3209. [PubMed]
JansenHG, SanyalS. Development and degeneration of retina in rds mutant mice: electron microscopy. J Comp Neurol. 1984;224:71–84. [CrossRef] [PubMed]
LibbyRT, LavalleeCR, BalkemaGW, BrunkenWJ, HunterDD. Disruption of laminin β2 chain production causes alterations in morphology and function in the CNS. J Neurosci. 1999;19:9399–9411. [PubMed]
PengYW, HaoY, PettersRM, WongF. Ectopic synaptogenesis in the mammalian retina caused by rod photoreceptor-specific mutations. Nat Neurosci. 2000;3:1121–1127. [CrossRef] [PubMed]
PengYW, SendaT, HaoY, MatsunoK, WongF. Ectopic synaptogenesis during retinal degeneration in the royal college of surgeons rat. Neuroscience. 2003;119:813–820. [CrossRef] [PubMed]
CalvertPD, StrisselKJ, SchiesserWE, PughEN, Jr, ArchavskyVY. Light-driven translocation of signaling proteins in vertebrate photoreceptors. Trends Cell Biol. 2006.560–568.
JinM, LiS, MoghrabiWN, SunH, TravisGH. Rpe65 is the retinoid isomerase in bovine retinal pigment epithelium. Cell. 2005;122:449–459. [CrossRef] [PubMed]
BerrierAL, YamadaKM. Cell-matrix adhesion. J Cell Physiol. 2007;213:565–573. [CrossRef] [PubMed]
NykvistP, HongminT, IvaskaJ, KapaylaJ, PihlajaniemiT, HeinoJ. Distinct recognition of collagen subtypes by a1b1 and a2b1 integrins: α1β1 mediates cell adhesion to type XIII collagen. J Biol Chem. 2000;275:8255–8261. [CrossRef] [PubMed]
LangholzO, RockelD, MauchC, et al. Collagen and collagenase gene expression in three-dimensional collagen lattices are differentially regulated by α1β1 and α2β1 integrins. J Cell Biol. 1995;131:1903–1915. [CrossRef] [PubMed]
PozziA, MobergPE, MilesLA, WagnerS, SolowayP, GardnerHA. Elevated matrix metalloprotease and angiostatin levels in integrin α1 knockout mice cause reduced tumor vascularization. Proc Natl Acad Sci U S A. 2000;97:2202–2207. [CrossRef] [PubMed]
YangC, ZeisbergM, LivelyJC, NybergP, AfdhalN, KalluriR. Integrin α1β1 and α2β1 are the key regulators of hepatocarcinoma cell invasion across the fibrotic matrix microenvironment. Cancer Res. 2003;63:8312–8317. [PubMed]
CosgroveD, MeehanDT, DelimontD, et al. Integrin α1β1 regulates matrix metalloproteinases via P38 mitogen-activated protein kinase in mesangial cells: implications for Alport syndrome. Am J Pathol. 2008;172:761–773. [CrossRef] [PubMed]
Espinosa-HeidmannDG, SallJ, HernandezEP, CousinsSW. Basal laminar deposit formation in APO B100 transgenic mice: complex interactions between dietary fat, blue light, and vitamin E. Invest Ophthalmol Vis Sci. 2004;45:260–266. [CrossRef] [PubMed]
Van der SchaftTL, MooyCM, deBruijnWC, BosmanFT, deJongPT. Immunohistochemical light and electron microscopy of basal laminar deposit. Graefes Arch Clin Exp Ophthalmol. 1994;232:40–46. [CrossRef] [PubMed]
NandrotEF, KimY, BrodieSE, HuangX, SheppardD, FinnemannSC. Loss of synchronized retinal phagocytosis and age-related blindness in mice lacking αvβ5 integrin. J Exp Med. 2004;200:1539–1545. [CrossRef] [PubMed]
HoTC, Del PrioreLV. Reattachment of cultured human retinal pigment epithelium to extracellular matrix and human Bruch’s membrane. Invest Ophthalmol Vis Sci. 1997;38:1110–1118. [PubMed]
BremRB, RobbinsSG, WilsonDJ, et al. Immunolocalization of integrins in the human retina. Invest Ophthalmol Vis Sci. 1994;35:3466–3474. [PubMed]
GardnerH, BrobergA, PozziA, LaatoM, HeinoJ. Absence of integrin α1β1 in the mouse causes loss of feedback regulation of collagen synthesis in normal and wounded dermis. J Cell Sci. 1999;112:263–272. [PubMed]
KagamiS, KondoS, LosterK, et al. α1β1 Integrin-mediated collagen matrix remodeling by rat mesangial cells is differentially regulated by transforming growth factor-β and platelet-derived growth factor-BB. J Am Soc Nephrol. 1999;10:779–789. [PubMed]
KagamiS, UrushiharaM, KitamuraA, et al. PDGF-BB enhances α1β1 integrin-mediated activation of the ERK/AP-1 pathway involved in collagen matrix remodeling by rat mesangial cells. J Cell Physiol. 2004;198:470–478. [CrossRef] [PubMed]
MoralesSA, MareninovS, PrasadP, WadehraM, BraunJ, GordonLK. Collagen gel contraction by ARPE-19 cells is mediated by a FAK-Src dependent pathway. Exp Eye Res. 2007;85:790–798. [CrossRef] [PubMed]
BrannMR, CohenLV. Diurnal expression of transducin mRNA and translocation of transducin in rods of rat retina. Science. 1987;235:585–587. [CrossRef] [PubMed]
PhilpNJ, ChangW, LongK. Light-stimulated protein movement in rod photoreceptor cells of the rat retina. FEBS Lett. 1987;225:127–132. [CrossRef] [PubMed]
WhelanJP, McGinnisJF. Light-dependent subcellular movement of photoreceptor proteins. J Neurosci Res. 1988;20:263–270. [CrossRef] [PubMed]
KalraD, ElsaesserR, GuY, VenkatachalamK. Transducin in rod photoreceptors: translocated when not terminated. J Neurosci. 2007;27:6349–6351. [CrossRef] [PubMed]
SokolovM, LyubarskyAL, StrisselKJ, et al. Massive light-driven translocation of transducin between the two major compartments of rod cells: a novel mechanism of light adaptation. Neuron. 2002;34:95–106. [CrossRef] [PubMed]
CalvertPD, StrisselKJ, SchiesserWE, PughEN, ArshavskyVY. Light-driven translocation of signaling proteins in vertebrate photoreceptors. Trends Cell Biol. 2006;11:560–568.
KongL, FengL, SolemanCE, et al. Bright cyclic light accelerates photoreceptor cell degeneration in tubby mice. Neurobiol Dis. 2006;21:468–477. [CrossRef] [PubMed]
MendezA, LemJ, SimonM, ChenJ. Light-dependent translocation of arrestin in the absence of rhodopsin phosphorylation and transducin signaling. J Neurosci. 2003;23:3124–3129. [PubMed]
Abd-El-BarrMM, SykoudisK, AndrabiS, et al. Impaired photoreceptor protein transport and synaptic transmission in a mouse model of Bardet-Biedl syndrome. Vision Res. 2007;47:3394–3407. [CrossRef] [PubMed]
ChenJ, SimonMI, MatthesMT, YasumuraD, LaVailMM. Increased susceptibility to light damage in an arrestin knockout mouse model of Oguchi disease (stationary night blindness). Invest Ophthalmol Vis Sci. 1999;40:2978–2982. [PubMed]
MendezA, LemJ, SimonM, ChenJ. Light-dependent translocation of arrestin in the absence of rhodopsin phosphorylation and transducin signaling. J Neurosci. 2003;23:3124–3129. [PubMed]
Figure 1.
 
Immunofluorescence localization of integrin α1β1 at the basolateral aspect of the RPE. An α1β1 integrin-specific antibody shows bright immunostaining at the basal aspect of the RPE (arrow) in the mouse retina (A). Bright-field image of this same stained section (B) clearly identifies the signal just basal to the pigmented epithelium. (C) Immunostaining of wild-type mouse retina for the integrin β1 subunit, which also localizes to the basal aspect of the RPE (arrow). (D) Absence of staining for integrin α1 in retinal sections from the integrin α1-null mouse, demonstrating the specificity of the antibody. Specific immunostaining is also observed in the capillaries of the neuroretina (arrowheads), as confirmed by dual immunofluorescence immunostaining with anti-CD31 antibodies: anti-α1β1 integrin (E), anti-CD31 (F), and merged images (G). PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bars, 50 μm.
Figure 1.
 
Immunofluorescence localization of integrin α1β1 at the basolateral aspect of the RPE. An α1β1 integrin-specific antibody shows bright immunostaining at the basal aspect of the RPE (arrow) in the mouse retina (A). Bright-field image of this same stained section (B) clearly identifies the signal just basal to the pigmented epithelium. (C) Immunostaining of wild-type mouse retina for the integrin β1 subunit, which also localizes to the basal aspect of the RPE (arrow). (D) Absence of staining for integrin α1 in retinal sections from the integrin α1-null mouse, demonstrating the specificity of the antibody. Specific immunostaining is also observed in the capillaries of the neuroretina (arrowheads), as confirmed by dual immunofluorescence immunostaining with anti-CD31 antibodies: anti-α1β1 integrin (E), anti-CD31 (F), and merged images (G). PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bars, 50 μm.
Figure 2.
 
Western blot analysis of α1β1 integrin in mouse retinas and ARPE 19 cells. Detergent extracts of neuroretina (lanes 1 and 3) and retinal pigment epithelial choroid (lanes 2 and 4) after removal of the neuroretina from wild-type mice (lanes 1 and 2) and integrin α1-null mice (lanes 3 and 4) were analyzed by Western blots probed with antibodies specific for either integrin α1 (upper) or β1 (lower) subunits. Western blots (lane 5) of ARPE19 cell extracts probed with antibodies against the human α1 integrin subunit (upper) or the human β1 integrin subunit (lower).
Figure 2.
 
Western blot analysis of α1β1 integrin in mouse retinas and ARPE 19 cells. Detergent extracts of neuroretina (lanes 1 and 3) and retinal pigment epithelial choroid (lanes 2 and 4) after removal of the neuroretina from wild-type mice (lanes 1 and 2) and integrin α1-null mice (lanes 3 and 4) were analyzed by Western blots probed with antibodies specific for either integrin α1 (upper) or β1 (lower) subunits. Western blots (lane 5) of ARPE19 cell extracts probed with antibodies against the human α1 integrin subunit (upper) or the human β1 integrin subunit (lower).
Figure 3.
 
Integrin α1-knockout mice show clinical pathology of the retina associated with thinning of the outer nuclear layer and thickening of the basal lamina of the RPE. (A) Typical funduscopic images from 10-month-old wild-type mice and integrin α1-null (α1KO) mice reveal spots in the peripheral retina of a α1-null mouse characteristic of retinal degeneration. (B) Histologic examination of 15-month-old wild-type and integrin α1-null mice. The retina of the integrin α1-null mouse showed significantly reduced numbers of nuclei in the outer nuclear layer (ONL) when compared with the wild-type retina of the same age. The outer segments (OS) of photoreceptors appear shortened in integrin α1-null mice compared with wild-type mice. (C) Ultrastructural analysis of the basal lamina of the RPE in the 24-month-old wild-type mouse and the 12-month-old integrin α1-null mouse. Basement membranes in the basal lamina of the RPE show retraction of the basal processes of the RPE associated with irregular thickening (arrow) of the basal lamina of the RPE for the integrin α1-null mice. This basement membrane thickening was not evident in integrin α1-null mice before observed reductions in ONL thickness (between 8 and 10 months).
Figure 3.
 
Integrin α1-knockout mice show clinical pathology of the retina associated with thinning of the outer nuclear layer and thickening of the basal lamina of the RPE. (A) Typical funduscopic images from 10-month-old wild-type mice and integrin α1-null (α1KO) mice reveal spots in the peripheral retina of a α1-null mouse characteristic of retinal degeneration. (B) Histologic examination of 15-month-old wild-type and integrin α1-null mice. The retina of the integrin α1-null mouse showed significantly reduced numbers of nuclei in the outer nuclear layer (ONL) when compared with the wild-type retina of the same age. The outer segments (OS) of photoreceptors appear shortened in integrin α1-null mice compared with wild-type mice. (C) Ultrastructural analysis of the basal lamina of the RPE in the 24-month-old wild-type mouse and the 12-month-old integrin α1-null mouse. Basement membranes in the basal lamina of the RPE show retraction of the basal processes of the RPE associated with irregular thickening (arrow) of the basal lamina of the RPE for the integrin α1-null mice. This basement membrane thickening was not evident in integrin α1-null mice before observed reductions in ONL thickness (between 8 and 10 months).
Figure 4.
 
Kinetics of rod loss as a function of age in integrin α1-null mouse retinas compared with wild-type mouse retinas. Data points represent quantitative measures of outer nuclei across the entire retina for wild-type mice compared with integrin α1-null mice at the indicated ages. *Statistically significant differences between wild-type and integrin α1-null groups (P < 0.05).
Figure 4.
 
Kinetics of rod loss as a function of age in integrin α1-null mouse retinas compared with wild-type mouse retinas. Data points represent quantitative measures of outer nuclei across the entire retina for wild-type mice compared with integrin α1-null mice at the indicated ages. *Statistically significant differences between wild-type and integrin α1-null groups (P < 0.05).
Figure 5.
 
Synaptic malformations are observed in rod and cone synaptic terminals in integrin α1-null mice. (A) Retinal section from a 12-month-old wild-type mouse under transmission electron microscopy (TEM) showing the cone synaptic terminal. CP, cone pedicle. In this normal cone photoreceptor synaptic terminal, invaginating synapses (recognized by the synaptic ribbons, as indicated by the arrows) with postsynaptic horizontal cell processes (H) and ON-cone bipolar cell dendrites (B) are located superficially on the basal side of the terminal. (B) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the cone synaptic terminal (CP). Most invaginating synapses (recognized by the synaptic ribbons, as indicated by large arrows) do not have postsynaptic bipolar cell dendrites and are located deep inside the cone synaptic terminal. The basal side can be recognized by conventional flat contact synapses (small arrows). (C) Retinal section from a 12-month-old wild-type mouse under TEM illustrating the outer plexiform layer. The rod synaptic (RS) termini display the normal triad structure. (D) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the outer plexiform layer. Most RS photoreceptor terminals do not have a normal triad structure. Rather, they have double ribbons (arrows) or dyads. The CP does not have any invaginating synapses. Scale bar, 500 nm.
Figure 5.
 
Synaptic malformations are observed in rod and cone synaptic terminals in integrin α1-null mice. (A) Retinal section from a 12-month-old wild-type mouse under transmission electron microscopy (TEM) showing the cone synaptic terminal. CP, cone pedicle. In this normal cone photoreceptor synaptic terminal, invaginating synapses (recognized by the synaptic ribbons, as indicated by the arrows) with postsynaptic horizontal cell processes (H) and ON-cone bipolar cell dendrites (B) are located superficially on the basal side of the terminal. (B) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the cone synaptic terminal (CP). Most invaginating synapses (recognized by the synaptic ribbons, as indicated by large arrows) do not have postsynaptic bipolar cell dendrites and are located deep inside the cone synaptic terminal. The basal side can be recognized by conventional flat contact synapses (small arrows). (C) Retinal section from a 12-month-old wild-type mouse under TEM illustrating the outer plexiform layer. The rod synaptic (RS) termini display the normal triad structure. (D) Retinal section of 15-month-old integrin α1-null mouse under TEM showing the outer plexiform layer. Most RS photoreceptor terminals do not have a normal triad structure. Rather, they have double ribbons (arrows) or dyads. The CP does not have any invaginating synapses. Scale bar, 500 nm.
Figure 6.
 
Integrin α1-null miceshow defects in light-induced protein translocation of the transducin α-subunit but not arrestin. Immunostaining of the transducin α-subunit on wild-type (A, B) and integrin α1-null (E, F) retinas under dark adaptation for 8 hours (A, E) and light adaptation for 1 hour (B, F). Note the failure of transducin-α subunit to completely translocate from the outer segments to the inner segments in light-adapted retinas from α1-null mice (compare B with F). On the other hand, arrestin translocation to the outer segments after 1-hour light adaptation appeared qualitatively similar to that of wild-type mice (compare G, H with C, D). OS, outer segments; IS, inner segments; PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar, 25 μm.
Figure 6.
 
Integrin α1-null miceshow defects in light-induced protein translocation of the transducin α-subunit but not arrestin. Immunostaining of the transducin α-subunit on wild-type (A, B) and integrin α1-null (E, F) retinas under dark adaptation for 8 hours (A, E) and light adaptation for 1 hour (B, F). Note the failure of transducin-α subunit to completely translocate from the outer segments to the inner segments in light-adapted retinas from α1-null mice (compare B with F). On the other hand, arrestin translocation to the outer segments after 1-hour light adaptation appeared qualitatively similar to that of wild-type mice (compare G, H with C, D). OS, outer segments; IS, inner segments; PRL, photoreceptor layer; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar, 25 μm.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×