Figure 1.
Retinal ischemia leads to RGC apoptotic cell death.
A. Quantitation of FG labeled surviving RGCs at 1, and 5 days post-RIRI. Changes in RGC density were expressed as percentage RGC survival in the ischemic eye, normalized to the control contra-lateral eye from the same animal (n = 8 were analyzed for each time point). Student’s t-test. **p<0.01, *p<0.05. C. TUNEL staining in control (CTRL), and ischemic (IR) retina at 6 hours, 1, and 5 day post-injury. Note that larger magnification inset shows that most of the TUNEL positive cells (purple) in the GCL 1 day post-RIRI are RGC as demonstrated by co-localization with DiA retrograde label (green). Cellular nuclei are stained with DAPI (blue). Scale bars: 100 µm. GCL = ganglion cell layer, INL = inner nuclear layer, and ONL = outer nuclear layer. (n = 4 animals for each time point).
Figure 2.
Laminin degradation is associated with increased MMP-9 expression and activity in the retina 1 day after RIRI.
A. Immunofluorescence with a pan-laminin antibody in retinal sections from control (CTRL), and ischemic (RIRI) eyes shows laminin expression (red) in the inner limiting membrane (ILM), ganglion cell layer (GCL), and inner nuclear layer (INL) (a, b, and lower magnification insets). White arrows in a, and corresponding lower magnification inset point toward sites of laminin degradation in the ECM of RGC, in the ILM, around the RGC cell body, and INL. Note thinning of laminin in the ILM, and loss of laminin expression in the RGC and INL after RIRI. (b, and lower magnification inset). Immunofluorescence with anti-MMP-9 antibodies shows increased expression of MMP-9 (red, yellow arrows; c, d), and in situ zymography demonstrates increased gelatinolytic activity (green) (e, f) in the ganglion cell layer (GCL) in ischemic eyes (n = 12). B. Western blotting of retinal extracts of control (−), and ischemic (+) rat eyes 1 day post-RIRI (n = 3). C. Densitometric analysis of bands corresponding to laminin β1 and γ1 chains in control and ischemic eyes normalized to β-actin. Error bars, SD. Student’s t test. *p<0.05. Scale bars: 100 µm.
Figure 3.
Integrin survival signaling is down-regulated in RGC in the retina 1 day after ischemic injury. A.
(i ) Western blot of retinal extracts of ischemic, RIRI (+), and control (−) rat eyes 1 day post-injury (n = 3). A representative example of at least 5 independent experiments is presented. (ii) Densitometric analysis of western blotting with all samples normalized to β-actin. Error bars, SD. Student’s t test. *p<0.05. B. β1 integrin expression is decreased in RGC 1 day post-RIRI. Immunohistochemistry with β1 integrin antibodies (red) in frozen retinal sections from control (a, b, c), and ischemic retinas (d, e, f), in which RGCs were retrogradely labeled with FG (green; a, d) shows reduced β1 integrin expression in ischemic eyes in comparison with control eyes (red; b, e) specifically in RGC (yellow; merged; c, f). Nuclei are stained with DAPI (blue; c, f). (n = 12). GCL = ganglion cell layer, INL = inner nuclear layer. Scale bars: 10 µm. C. FAK activation is reduced in RGC 1 day after ischemia. Immunohistochemistry with FAK (purple), and P-FAK [pY397] (red) antibodies in frozen retinal sections of control (a, b, c), and ischemic retinas (d, e, f) at 1 day post-RIRI, in which RGCs were retrogradely labeled with FG (yellow; a, d) shows significantly decreased expression of the activated, phospho [pY397] FAK in ischemic eyes in comparison with control eyes (red; c, f), while FAK expression remains unchanged (purple; b, e). White arrows identify FG labeled RGCs in the control retina (CTRL) expressing high levels of both FAK, and P-FAK [pY397] (a, b, c). Blue arrows point toward RGCs in the ischemic retina (RIRI) that have markedly decreased expression of P-FAK [pY397], while FAK expression is the same (d, e, f). (n = 12). GCL = ganglion cell layer, INL = inner nuclear layer. Scale bars: 10 µm.
Figure 4.
Laminin, or β1 integrin activating antibody, HUTS-21, promote integrin signaling in RGCs in vitro.
P5 RGCs were cultured on laminin (a–d), or PDL (e–l) for 48 hours. HUTS-21 (i, k) or isotype control (j, l) antibodies (1 µg/ml) were added to RGCs cultured on PDL 12 hours after plating. Immunohistochemistry was performed with β1 integrin (a, e, i, j), FAK (b, f), [pY397]-FAK (c, g, k, l), or IgG control antibodies (d, h) and analyzed by confocal microscopy. 100–200 cells per condition were analyzed. A representative experiment is shown. A total of N = 5 experiments were performed. Scale bars: 10 µm.
Figure 5.
The agonist antibody HUTS-21 mimics laminin’s pro-survival effect in RGC.
P5 purified RGCs were cultured on either laminin, or PDL for 48 hours. HUTS-21, or isotype control antibodies (1 µg/ml) were added to RGCs cultured on PDL 12 hours after plating. RGCs were pretreated with the Src/FAK inhibitor PP2 (25 µM). Live-cell imaging with an Axiovert Zeiss 200 M microscope was used to acquire micrographs of RGC in culture immediately after HUTS-21 addition, t = 0, and at the end of the experiment, 48 hours after plating, t = 48 hours. A. To evaluate RGC survival 100–500 live RGC/experimental condition were counted manually in the acquired micrographs to determine the percentage of live RGC at t = 48 hours when compared with t = 0. (N = 6 experiments were performed). B. Average total neurite length per RGC was evaluated by manually measuring RGC neurites in photo-micrographs acquired by live-cell imaging. (N = 6 experiments). Error bars, SD. Student’s t test. **p<0.01; *p<0.05.
Figure 6.
FAK regulates RGC survival, and neurite growth.
P3 RGCs were transfected by electroporation with FAK siRNA and non-targeting control siRNA and plated on laminin for 72 hours. A. The knockdown efficiency of the FAK siRNA was verified by RT-PCR with specific FAK primers after 24, and 72 hours in culture. (N = 3 experiments). Error bars, SD. Student’s t test. **p<0.01; *p<.0.05. B. Double immunolabeling with FAK (red), and TUJ1 (neuronal class III β-tubulin) (green) antibodies in RGC 72 hours post-transfection showed almost complete depletion of FAK protein expression in RGC transfected with FAK siRNA (a, white arrows; b), whereas FAK expression is not affected in RGC transfected with non-targeting control siRNA (c, yellow arrowheads; d). Boxed insets in a, and b represent higher magnification images (63X) of a representative cell in each condition; n≥100 RGC per condition were analyzed. A representative experiment is shown. At least N = 3 independent experiments were performed. Scale bar: 10 µm. FAK siRNA knock-down resulted in (C) decreased RGC survival, and (D) reduced average total neurite length per RGC when compared with non-targeting siRNA control 72 hours post-transfection; n≥100 RGC per condition were analyzed for both survival and neurite length. N = 3 experiments were performed. Error bars, SD. Student’s t test. *p<0.05.
Figure 7.
The β1 agonist antibody HUTS-21 enhances RGC survival and rescues β1 integrin-FAK signaling in vivo after RIRI.
HUTS-21, or isotype control (CTRL) rat antibodies (0.5 mg/ml), or PBS were administered intravitreally (5 µl) in the rat eye twice, 30 minutes, and 2 days after RIRI. A. Immunostaining with a secondary anti-rat fluorescent antibody demonstrating localization of HUTS-21 antibody into the retina 24 hours after RIRI (n = 3 animals were analyzed). Note that HUTS-21 binds to activated β1 receptors on RGC cell body, dendrites, and axons (a, white arrowheads). B. Representative photo-micrographs of retinal flat-mounts of control rats (a), non-treated (RIRI) (b), and intravitreally treated with HUTS-21 (c), IgG control (IgG Ctrl) (d), and vehicle PBS (e) ischemic rats 5 days post-injury (n = 4–6 animals for each group). C. RGC survival was determined by counting FG retrogradely labeled RGCs in flat-mounted retinas 5 days post-RIRI (n = 4–6 animals for each group). Error bars, SD. Student’s t test. **p<0.01; *p<0.05. D. Western blot analysis of retinal extracts with β1 integrin, and P-FAK antibodies, 1 day post-RIRI (n = 4). A representative experiment is shown. N = 3 experiments.
Figure 8.
Schematic diagram of laminin-integrin survival signaling in RGC.
RGC adhesion to laminin induces integrin activation and initiates integrin signaling. Integrin β1 undergoes a conformational change to an “active” high affinity bound state [83]. Alternatively, in the absence of laminin, or in the context of MMP-9 mediated laminin degradation observed after RIRI in vivo, a β1 integrin stimulatory antibody can bind and maintain integrin β1 in the “active”, signaling conformation at the RGC’s membrane. Activated β1 integrin initiates signal transduction by recruiting the tyrosine protein kinase, FAK. In response to integrin engagement, FAK auto-phosphorylates at Tyr 397 which represents the initial, major step in FAK activation. FAK P-Tyr 397 creates a binding site for Src family kinases. Src recruitment, and activation leads to phosphorylation of multiple amino acids including Tyr 576, 577 required for FAK’s full catalytic activity [37]. PP2 inhibits Src, therefore FAK’s complete activation required for downstream survival signaling. Initial Tyr 397 phosphorylation is also important for recruitment of PI3K. FAK promotes cell survival downstream of laminin and integrin β1 by enhancing PI3K mediated activation of Akt by phosphorylation at Ser 473. Activated Akt is associated with survival [35], [37], and inhibition of apoptosis by increasing expression of anti-apoptotic protein bclxL.