This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.Müller glia (MG) are the primary and specialised glial cells in the adult vertebrate retina. In certain non-mammalian species, such as the teleost fish, they have a remarkable ability to re-enter the cell cycle and dedifferentiate into retinal progenitor cells (RPCs) that can proliferate and differentiate into lost neurons (1). However, in adult mammalian retina, the proliferative response of MG to retinal injury is extremely limited (2). In response to injury in rodent retina, the majority of MG undergo reactive gliosis and hypertrophy, but a few rapidly re-enter the cell cycle, as inferred by increased cyclin gene expression (3, 4). Although this proliferative response does not result in regeneration, it suggests that the regenerative potential of MG can be stimulated.
The Hippo pathway plays a pivotal role in organ size control, tissue regeneration, and stem cell maintenance (5-7). The core of this pathway consists of a kinase signalling cascade in which MST1/2 kinases (mammalian sterile 20-like kinase 1 and 2) phosphorylate and activate LATS1/2 kinases (large tumour suppressor kinases 1 and 2). The LATS1/2 kinases directly phosphorylate the transcriptional coactivators, YAP (Yes-associated protein) and TAZ (transcriptional coactivator with PDZ-binding motif), leading to their interaction with 14-3-3 that causes the cytoplasmic retention of YAP/TAZ and their subsequent proteasomal degradation (8). When the Hippo signalling pathway is inactive, YAP is dephosphorylated and translocated into the nucleus to interact with the TEAD transcription factor. This activation of YAP/TAZ results in promotion of the expression of genes involved in cell proliferation, differentiation, and survival (9). Recent studies have identified the Hippo pathway as an endogenous blocking mechanism that blocks mammalian MG cell cycle re-entry and dedifferentiation to a progenitor cell state (10).
Photobiomodulation (PBM) therapy is a technique based on the use of low-level laser light or light emitting diodes (LED) with red to near-infrared spectral range of 600-1,100 nm wavelength, and has been investigated over the past decade as an alternative approach to treating retinal diseases and disorders, including AMD (11, 12). The studies in rodent models of retinal degeneration have shown that PBM therapy with 670 nm light reduces photoreceptor cell death (11, 13). Furthermore, clinical studies have demonstrated the neuroprotective effect of PBM in retinal degeneration (12). The primary chromophore for red or near-infrared photons is generally considered to be cytochrome c oxidase, which is unit IV of the mitochondrial respiratory chain (14). The photon absorption leads to the secondary effects, including increase in ATP production, increase in nitric oxide production, and modulation of calcium levels (15, 16). PBM can produce a brief burst of ROS in normal cells, but when used in oxidative stressed cells and tissues, ROS levels are reduced (16, 17). PBM has been proposed to induce the activation of numerous intracellular signalling pathways, and alter the transcriptional factors involved in cell proliferation, survival, and differentiation (12, 15, 18). However, the molecular and cellular mechanisms of PBM underlying the regenerative capacity of MG remain unknown. Here, we reveal PBM therapy as a promising tool to stimulate MG cell cycle re-entry and dedifferentiation into progenitor-like state in both the injured and uninjured retina. We also describe an unreported mechanism whereby PBM inhibits Hippo pathway, leading to YAP activation in MG.
To investigate whether PBM affects the Hippo pathway, we treated mouse retinas with 670 nm LED light (Supplementrary Fig. 1), and immunostained the retinal sections with anti-YAP antibody to determine the expression of YAP protein. In the untreated retina, YAP protein was specifically expressed in cells that were labelled with glutamine synthetase (GS), indicating MG identity (Fig. 1A). In addition, the expression of YAP protein was observed in both MG cytoplasm (arrows) and nucleus (dashed line boxes). Interestingly, treatment of 670 nm light markedly increased the nuclear-localised active YAP protein expression in the inner nuclear layer (INL). To further confirm the effect of 670 nm light on YAP activity, we evaluated YAP phosphorylation in the retinas treated with 670 nm light by western blot analysis using an antibody specific to YAP phosphorylated at serine 127. Total YAP protein levels progressively increased between 6 and 24 h, and were followed by a decline by 48 h after light treatment (Fig. 1B). However, treatment of 670 nm light decreased YAP phosphorylation, as seen by quantitative analysis of the ratio of phosphor-YAP to total YAP protein levels (Fig. 1C). Moreover, we found that 460 and 520 nm light did not affect the nuclear expression of YAP protein, as seen by co-immunostaining analysis for YAP and MG nuclear marker SOX9 (Fig. 1D). These results indicate that YAP is activated only under specific light wavelength.
Since MG responds to retinal injury, we sought to determine whether 670 nm light affects the YAP activity in degenerative retina. We induced retinal degeneration with NaIO3, whose systemic administration is known to selectively induce the injury of retinal pigment epithelium, causing photoreceptor cell death (19). We treated retinas with 670 nm light three days after NaIO3 injection, at the onset of the reactive gliosis of MG and photoreceptor cell death (Supplementary Fig. 2A, B). A marked increase in levels of phosphorylated YAP protein was observed in the retinas of NaIO3-injected mice, compared to PBS-injected mice (Fig. 1E). However, treatment of 670 nm light significantly reduced the YAP phosphorylation in the retinas of NaIO3-injected mice, as assessed by quantifying the ratio of phospho-YAP to total YAP protein (Fig. 1F). Taken together, these data demonstrate that treatment of 670 nm light facilitates the MG to overcome the inhibition of YAP activity by retinal injury.
To determine if 670 nm light directly stimulated MG to activate YAP, we employed an in vitro culture system using MG isolated from mouse retina. MG cultures were treated with 670 nm light, and further cultured for 24 h. The number of nuclear YAP and SOX9-double-positive cells was significantly increased in cultures treated with 670 nm light, compared to control (Fig. 2A), as determined by quantitative immunostaining analysis using anti-YAP and anti-SOX9 antibodies. Western blot analysis of these cultures also showed an elevated level of YAP protein and a decreased level of phosphorylated YAP in cultures treated with 670 nm light, compared to control (Fig. 2B, C). In addition, 670 nm light decreased the levels of phosphorylated LATS1 protein, as compared with that in control (Fig. 2B, C).
Previous studies have reported that a decrease in ATP production activates AMP-activated protein kinase (AMPK), which inhibits YAP activity (20). We observed a significant increase in intracellular ATP levels (Supplementary Fig. 3) and a decrease in AMPK phosphorylation (Fig. 2D) in MG cultures treated with 670 nm light, prompting us to investigate whether 670 nm light promotes AMPK-mediated regulation of the Hippo pathway. To test this, we exposed MG cultures to AICAR, a direct activator of AMPK, or DMSO, and then treated them with 670 nm light. Western blot analysis from these cells showed that the treatment of AICAR increased the level of phosphorylated LATS1 and YAP proteins, as compared with that of DMSO (Fig. 2E, F). Taken together, these results demonstrate that 670 nm light stimulates MG to inhibit AMPK-mediated Hippo pathway, leading to YAP activation.
We next investigated whether 670 nm light can stimulate MG to acquire the regenerative capacity, and whether this process requires YAP. We treated MG expressing YAP shRNA or control shRNA with 670 nm light, and assessed proliferative activity using the EdU incorporation assay. We observed a significant increase in the number of EdU-positive cells in MG expressing control shRNA when treated with light, compared to control. However, this effect was not observed in MG expressing YAP shRNA (Fig. 3A, B). This result indicates that YAP is necessary for the proliferative response of MG to 670 nm light. We also found a loss of glial marker GFAP in MG expressing control shRNA when treated with 670 nm light, while YAP shRNA-expressing MG showed robust GFAP expression (Fig. 3C). Western blot analysis of MG treated with 670 nm light further confirmed that YAP-depleted MG did not show a reduction in the expression level of GFAP protein (Fig. 3D). Additionally, these cells exhibited a decrease in the expression level of the cell cycle marker Cyclin D1, which was increased by 670 nm light treatment (Fig. 3D). To further assess the effect of 670 nm light on the dedifferentiation of MG, we analysed the expression of the progenitor markers. Quantitative real-time PCR analysis showed upregulated mRNA levels of RPC markers, including those of RAX, Ascl1, Pax6, HEY1, and HES5, in MG treated with 670 nm light (Fig. 3E). However, knockdown of YAP suppressed the upregulation of RPC markers. These results suggest that 670 nm light can activate the regenerative potential of MG, and demonstrate that YAP is required for this cellular function of PBM with 670 nm.
To further explore the impact of 670 nm light on the regenerative potential of MG in response to retinal damage, we first assessed GFAP and Vimentin expression as the markers of reactive gliosis. Because 670 nm light activated YAP transiently, we treated retina with light every other day starting from day 3 after NaIO3-injection to sustain the YAP activity (Supplementary Fig. 4A, B). At day 8 after injection, immunostaining analysis showed that GFAP and Vimentin expression were increased in all retinal layers of NaIO3-injected mice, compared to PBS-injected mice, confirming MG reactive gliosis (Fig. 4A). As expected, 670 nm light suppressed GFAP and Vimentin expression in NaIO3-injected mice. These data indicate that the treatment of 670 nm light inhibits MG reactive gliosis. The reactive gliosis of MG responding to retinal injury is associated with the activation of cell cycle-related regulators, including Cyclin D1 (3). Therefore, we next assessed Cyclin D1 expression to determine whether 670 nm light affected the re-entry of MG into the cell cycle. Cyclin D1 expression was observed in MG of NaIO3-injected mice, but not in PBS-injected mice, as seen by immunostaining analysis using anti-Cyclin D1 antibody (Fig. 4B). In NaIO3-injected mice, 600 nm light treatment significantly increased the number of Cyclin D1/GS double-positive MG, as compared to that in untreated mice (Fig. 4C). These data demonstrate that 670 nm light stimulates the ability of MG to re-enter the cell cycle and dedifferentiate in both injured and uninjured retinas.
To determine whether YAP is required for the effect of 670 nm light in the regenerative response of MG to retinal injury, lentivirus expressing YAP shRNA or control shRNA was intravitreally injected into the retinas of NaIO3-injected mice. Immunostaining analysis showed a marked reduction of nuclear YAP expression in MG of retinas transduced with lentivirus expressing YAP shRNA (Supplementary Fig. 5). YAP knockdown resulted in a significant decrease in the number of Cyclin D1/GS-double-positive MG in retina treated with 670 nm light, compared to control shRNA (Fig. 4E, F). We also observed increased expression of GFAP and Vimentin in YAP knockdown retina treated with light, compared to control retina (Fig. 4D). Western blot analysis further confirmed that Cyclin D1 expression level decreased in YAP-deficient retina, while GFAP expression level increased (Fig. 4G). Moreover, quantitative real-time PCR analysis from these retinas showed that YAP knockdown inhibited the upregulation of mRNA levels of RPC markers in retina treated with 670 nm light (Fig. 4H). Taken together, these data demonstrate that the depletion of YAP abolishes the effect of 670 nm light on MG cell cycle re-entry and dedifferentiation in the injured retina.
PBM therapy has been shown to attenuate the retinal disease in experimental and clinical studies. However, the molecular and cellular mechanism of PBM underlying the regenerative potential of MG is largely unexplored. Our results reveal a previously unreported capability of 670 nm light in the regulation of the Hippo pathway. We discovered that 670 nm light activates YAP, a major effector of the Hippo pathway, in MG, thereby triggering the cell cycle re-entry and dedifferentiation of MG into a progenitor-like state. In addition, we demonstrate that in the absence of retinal injury, 670 nm light treatment is sufficient to activate MG spontaneously.
The low intensity light in the red or near-infrared spectra range increases ROS production and ATP synthesis. As signalling molecules, ROS is involved in various extrinsic and intrinsic signalling pathways. A recent study showed that ROS production increases the expression of YAP protein, and stimulates the proliferation of neural progenitor cells (21). When we assessed YAP activity in MG treated with a ROS inhibitor NAC, no significant change in YAP phosphorylation was observed between 670 nm light-treated and untreated cells (data not shown). A previous study also reported that the low level of ATP activates AMP-activated kinase (AMPK)-mediated Hippo pathway and inhibits YAP activity and cancer cell proliferation (20). We showed that PBM with 670nm light increases ATP production in MG. Although it remains unclear whether 670 nm light can specifically regulate the level of ATP in the AMPK-Hippo signalling cascade, our study demonstrates that 670 nm light directly stimulates MG to increase YAP activity. In primary MG culture, 670 nm light treatment led to the dephosphorylation of LATS kinase and the marked reduction of YAP phosphorylation and its nuclear localisation. Interestingly, this effect of 670 nm light was inhibited when treated with AMPK activator. Therefore, the AMPK-Hippo-YAP signalling pathway is involved in the underlying molecular mechanism of photo-stimulatory effect induced by 670 nm light in MG.
Our study shows that 670 nm light requires YAP to trigger MG cell cycle re-entry and dedifferentiation to RPC-like state. The 670 nm light treatment failed to increase the expression of Cyclin D1 and RPC markers (Rax, Ascl1, Pax6, Hey1, and Hes5), or to reduce the expression of GFAP in YAP-deficient MG. Consistent with our finding, a recent report showed that conditional YAP deletion in MG impeded Cyclin D1 upregulation and enhanced GFAP expression upon retinal injury (10). The regenerative response of MG critically depends on retinal damage cues. Retinal injury stimulates the release of cytokines and mitogens, including EGF, FGF, CNTF, and Wnt, which play a key role in cell proliferation and differentiation. However, in the adult mammalian retina after injury, MG does not spontaneously re-enter the cell-cycle and completely dedifferentiate, and therefore, they lack regenerative capacity. The absence of an adequate regenerative response in mammalian MG suggests that endogenous blocking mechanisms are in place to limit their proliferation and dedifferentiation. A recent study showed that loss of Hippo signalling promotes spontaneous MG proliferation in NMDA-induced retinal ganglion cell degeneration (10). We also show that YAP activity is suppressed in NaIO3-injured retina. Thus, although some MG express Cyclin D1 in response to NaIO3-induced retinal injury, it is like that the expression of these proteins is largely suppressed by activation of the Hippo pathway. Previously, in an attempt to boost the regenerative response of MG, ectopic expression of YAP5SA, a constitutively active form of YAP, and β-catenin, a key factor of the Wnt signalling pathway, were used (10, 22, 23). While these studies establish the regenerative potential of mammalian MG, they used primarily invasive surgical approaches using intraocular injection of adeno-associated virus (AAV). Here, we show that non-invasive 670 nm light treatment stimulates a sustained YAP activation of MG and enhances the expression of cell cycle and RPC markers in the injured retina. Therefore, our study suggests that PBM with 670 nm wavelength would be more useful as a therapeutic approach for driving MG to a proliferative, RPC-like state to achieve the reversal of retinal degeneration.
The detailed information is available in the Supplementary material and methods.
The authors have no conflicting interests.
Fig. 1. 670 nm light increases YAP activity in both uninjured and injured retina. (A) Immunostaining analysis for YAP and MG marker GS in retina 24 h after 670 nm light treatment. High magnification shows the nuclear localization of YAP in MG (dashed line boxes). Arrows indicate cytoplasmic YAP. Quantitative analysis for the fluorescent intensity of nuclear Yap is shown in the bottom panel. (B) Western blotting analysis of phospho-YAP and total YAP expression at 0, 6, 12, 24, and 48 h in retina after light treatment. (C) Relative YAP phosphorylation levels were quantified by the ratio between phospho-YAP and total YAP from C. (D) Immunostaining analysis and quantitative analysis for YAP and SOX9 in INL of retina 24 h after light treatment with different wavelengths. (E) Western blotting analysis of phospho-YAP and total YAP in retinas 24 h after light treatment in NaIO3-injected mice. (F) Relative YAP and YAP phosphorylation levels were quantified by the ratio of total YAP to Actin and of phospho-YAP to total YAP from E, respectively. For quantification, levels are presented as mean ± SD (n = 5 retinas for each group in A, n = 3 retinas for each group in D, and n = 3 independent samples for each group in C and F). Student’s t test (*P < 0.05, **P < 0.01, ***P < 0.001). NFL: nerve fiber layer, INL: inner nuclear cell layer, ONL: outer nuclear cell layer. Scale bars: 50 μm in (A), 25 μm in (D).
Fig. 2. 670 nm light inhibits AMPK-mediated YAP phosphorylation and the Hippo pathway. (A) MG were treated with or without 670 nm light and then immunostained 24 h after light treatment with anti-YAP and anti-SOX9 antibodies. Hoechst dye served as a counterstain. Percentages of YAP/SOX9 double-positive cells is shown in the right panel. Scale bars, 100 μm. (B) Western blot analysis for expression of phospho-LATS1, LATS1, phospho-YAP, and YAP from MG cultures described in A. (C) Quantitative analysis for the ratio of phospho-YAP to total YAP and of phospho-LATS1 to LATS1 from B, respectively. (D) Western blot analysis of phospho-AMPKα and AMPKα expression from MG cultures described in A. Quantitative analysis are shown in right panel. (E) MG exposed to AICAR or DMSO were treated with or without 670 nm light and subjected to western blotting analysis 24 h after light treatment. (F) Relative LATS1 and YAP phosphorylation levels were quantified by the ratio of phosph-LATS1 to total LATS1 and of phospho-YAP to total YAP from E. For quantification, levels are given as mean ± SD (n = 3 independent samples for each group). Student’s t test (*P < 0.05, **P < 0.01).
Fig. 3. YAP mediates the cellular function of 670 nm light to change Müller glia state. (A) MG cultures expressing YAP shRNA or control shRNA were subjected to the EdU incorporation assay 24 h after 670 nm light treatment. (B) Percentage of EdU-positive cells. (C) Immunostaining of YAP and GFAP in MG cultures expressing YAP shRNA and control shRNA 24 h after 670 nm light treatment. (D) MG expressing YAP shRNA and control shRNA were treated with 670 nm light, and their cell lysates were subjected to western blot with antibodies against GFAP, Cyclin D1, and YAP. (E) Quantitative real-time PCR showed the expression of retinal progenitor markers, Rax, Ascl1, Pax6, Hey1, and Hes5 in MG culture expressing control shRNA and YAP shRNA after treatment of 670 nm light. For quantification, levels are presented as mean ± SD (n = 4 independent samples for each group in B, and n = 3 independent samples for each group in E). Student’s t test (*P < 0.05, **P < 0.01). Scale bars: 50 μm.
Fig. 4. 670 nm light suppresses Müller glia gliosis and stimulates their dedifferentiation into progenitor-like cells in the injured retina. (A-C) The retinas of NaIO3-injected and PBS-injected mice were treated with or without 670 nm light. The retinal sections were then immunostained at 5 days after light treatment with antibodies against reactive gliosis markers GFAP and Vimentin (A), and cell cycle marker Cyclin D1 and MG marker GS (B), respectively. Cyclin D1/GS-double-positive cells (arrows) were counted, and the number of Cyclin D1/GS-double-positive cells are shown in B (C). (D-F) The lentiviruses expressing YAP shRNA and control shRNA were transduced to retinas of NaIO3-injected mice using intravitreal injection, and these retinas were treated with 670 nm light. The retinal sections were immunostained at 5 days after light treatment with antibodies against GFAP and Vimentin (D), and Cyclin D1 and GS (E). Quantitative analysis of Cyclin D1/GS-double-positive cells (arrows) in the INL shown in E (F). (G) Western blot analysis for expression of GFAP, Cyclin D1 and YAP from retinas described in D-F. Quantitative analysis are shown in right panel. (H) Quantitative real-time PCR showed the expression of retinal progenitor markers, Rax, Ascl1, Pax6, Hey1, and Hes5 in 670 nm light-treated or untreated retinas expressing YAP shRNA or control shRNA in NaIO3-injected and PBS-injected mice. For quantification, levels are presented as mean ± SD (n = 5 retinas for each group in C and F, and n = 3 independent samples for each group in G and H). Student’s t test (*P < 0.05, **P < 0.01, ***P < 0.001). NFL: nerve fiber layer, IPL: inner plexiform layer, INL: inner nuclear cell layer, OPL: outer plexiform layer, ONL: outer nuclear cell layer. Error bars represent the mean ± SD of three independent experiments. Two-tailed Student’s t test (***P < 0.001). Scale bars: 50 μm.
Author ORCID Information
Jungmook Lyu
Funding Information Konyang University Korea Health Industry Development Institute |
 |
