Retinal Detachment-Induced Müller Glial Cell Swelling Activates TRPV4 Ion Channels and Triggers Photoreceptor Death at Body Temperature

Müller glial cell swelling in an RD mouse model. A, Immunostaining of glutamine synthetase (GS; a Müller glial cell marker, represented as green) was performed in RD retinae. Nuclei were stained by Hoechst (blue). In the bright-field image, the region of detachment from the RPE is represented as RD, in contrast to the normal region (represented as control). ILM, Inner limiting membrane. Scale bar, 200 μm. B, High magnification of the region highlighted by a dashed box in A. ONL, Outer nuclear layer; INL, inner nuclear layer; IPL, inner plexiform layer. Scale bar, 30 μm. Fluorescent images were converted to binary images by image software. Significant swelling of Müller processes in RD is indicated by arrowheads.

Müller glial TRPV4 is activated by mechanical stimuli. A, Morphology of acutely dissociated Müller glial cells. White arrowheads and arrows represent soma and endfeet, respectively. Top, Immunostaining of acutely dissociated cells with glutamine synthetase (a Müller glial marker). Black arrows represent neurons. B, Schematic drawing of a Müller glial cell and electrophysiological recording by stepped positive pressure stimulation (+10 mmHg per step). C, D, Representative traces of WT or TRPV4KO cells after application of stepped positive pressure with ramp pulses (−100 to +100 mV). The holding potential was at −60 mV. E, Quantification of densities of positive pressure-evoked currents. WT, n = 10; TRPV4KO, n = 12. F, Quantification of time constant of activation periods in WT recording (n = 10). G, Quantification of time constant of inactivation periods in WT recording (n = 10).

Müller glial TRPV4 is synergistically activated by mechanical stimuli and body temperature. A, Representative traces at 25 or 37°C induced by application of minimal positive pressure (+10 mmHg) by the HSPC system (red traces) with ramp pulses (−100 to +100 mV) compared with basal traces (black traces). The holding potential was at −60 mV. B, The outward rectified the current–voltage relationship of mechanical stimulus-evoked current (red trace) corresponding to the application of 10 mmHg pressure. The black trace represents the basal current–voltage relationship. C, Quantification of densities of positive pressure-evoked currents (n = 5, *p = 0.00056, Student t test).

RD-induced Müller glial swelling activates TRPV4 at body temperature. A, B, Representative traces of Müller glia in a WT retinal explant at 37°C. The recording pipette was placed in the endfeet. The holding potential was at −60 mV. Ramp pulses (−100 to +100 mV) were applied in each 5 s interval. The black lines represent application of a TRPV4 antagonist HC (10 mm). Control is from normal retinal explants. RD is from RD-evoked retinal explants. C, Quantification of densities of TRPV4-evoked currents in WT explants (n = 5, *p = 0.0058, Student t test). D, E, Representative traces of Müller glia in TRPV4KO a retinal explant at 37°C. The recording pipette was placed in the endfeet. The holding potential was at −60 mV. Ramp pulses (−100 to +100 mV) were applied in each 5 s interval. The black lines represent application of a TRPV4 antagonist HC (10 mm). Control is from normal retinal explants. RD is from RD-evoked retinal explants. F, Quantification of densities of TRPV4-evoked currents in TRPV4KO explants (n = 5). We failed to observe any TRPV4 currents.

Müller glial TRPV4 activation accelerates photoreceptor cell death in RD. A, Genetic manipulation of the Trpv4 gene allowing conditional deletion of its sixth allele. The top row shows the wild-type locus. Arrows show the sequence targeted by the primers for the genotyping PCR. The middle row shows the targeted Trpv4 gene, with the exon 6 flanked by loxP recombination sites. The bottom row shows the disrupted gene after Cre recombination. B, PCR genotyping of blank (a; showing a nonspecific band), Trpv4^w/w (b; 188 bp fragment), Trpv4^wt/lox (c; 188 and 327 bp fragments), Trpv4^−/− (d; KO allele where the reverse primer site is not present), and Trpv4^lox/lox (e; 327 bp fragment). C, Müller glia/astrocyte-specific TRPV4 conditional KO mice (TRPV4CKO) were generated by crossing hGFAP-Cre mice with TRPV4-flox mice. Shown are representative traces of [Ca^{2 +}]_i changes in cultured Müller glial cells (blue diamond line, WT, n = 58 cells; red triangle line,TRPV4CKO, n = 46 cells). Four days after culture, [Ca²⁺]_i changes were measured by Fluo-4 AM. The data were quantified as ΔF/F0. A TRPV4 agonist (10 nm GSK) was applied during recording. At the end of each experiment, we applied ionomycin (5 μm) to identify the surviving cells. D, Representative images of TUNEL staining in RD retinal tissues (WT, TRPV4KO, and TRPV4CKO). Conventional TRPV4KO mice showed significantly less photoreceptor cell death than WT mice 24 h after retinal detachments (*p = 0.06706, n = 7 or 8 eyes). The TRPV4CKO mice showed significantly less photoreceptor cell death than WT mice (**p = 0.020863, n = 6 eyes). The TRPV4CKO results were perfectly matched with those in TRPV4KO, indicating that Müller glial TRPV4 activation accelerates photoreceptor cell death in RD. ONL, Outer nuclear layer; INL, inner nuclear layer.

A, TRPV4 agonist or control PBS was injected into the subretinal space when generating retinal detachment. The eyes with TRPV4 agonist (GSK, 1 mm) showed a significant increase in photoreceptor cell death compared with control eyes 24 h after retinal detachments (**p = 0.002622, n = 7). Shown are representative images of TUNEL staining in RD retinal tissues (control or TRPV4 agonist). The graphs show mean ± SEM. Scale bar, 50 μm. B, TRPV4 inhibitor (n = 8) or control PBS (n = 6) was injected into the subretinal space when creating retinal detachment for WT mice. The eyes with TRPV4 inhibitor (HC, 1 mm) showed significantly less photoreceptor cell death compared with control eyes 24 h after retinal detachments (**p = 0.009823). Shown are representative images of TUNEL staining in RD tissues from WT mice (control or TRPV4 inhibitor). The graphs show mean ± SEM. Scale bar, 50 μm. ONL, Outer nuclear layer; INL, inner nuclear layer.

TRPV4 activation triggers MCP-1 release from Müller glial cells and leads to photoreceptor apoptosis in the detached retina. A, Real-time PCR assays were performed using specific MCP-1 primers. The values were normalized against β-actin expression, and the bar graphs compare values with the MCP-1/β-actin expression level of control eyes in WT mice. Asterisks (vs control) and hash tags (vs WT RD) represent a significant difference at p < 0.01 (*p = 0.000116, ^#p = 0.000232, Student t test, n = 5). B, ELISAs of MCP-1 protein were performed in control and RD eyes for WT and TRPV4KO mice. The asterisk represents a significant difference at p < 0.05 (p = 0.0010509, Mann–Whitney U test, n = 6). C, ELISA assays of MCP-1 protein performed using cultured Müller glial cells from WT mice. Asterisks (vs control) and hash tags (vs GSK or Hypo) represent a significant difference at p < 0.01 (*GSK, p = 0.000247; *Hypo, p = 0.00944; ^#GSK+HC, p = 0.001683; ^#Hypo+HC, p = 0.00092; t test, n = 4). D, ELISAs of MCP-1 protein were performed using the supernatant of cultured Müller glial cells (from WT mice) under Ca²⁺-free conditions. Only the small amount of MCP-1 release was observed independent from TRPV4 activation or inhibition (GSK or HC). These results were perfectly different from the physiological Ca²⁺ condition (2 mm) shown in Figure 5C. Thus, we can conclude that TRPV4-triggered Ca²⁺ influx evokes the MCP-1 release from Müller glial cells.

TRPV4 activation occurs upstream of MCP-1 release in Müller glial cells and accelerates photoreceptor cell death in RD. Control anti-IgG antibody, neutralizing anti-MCP-1 antibody, or control PBS was injected into the subretinal space when creating retinal detachment for WT and TRPV4KO mice. A, Representative images of TUNEL staining in RD retinal tissues from WT mice (PBS, IgG, or neutralizing anti-MCP-1 antibody). The WT eyes with neutralizing anti-MCP-1 antibody (100 μg/ml) showed significantly less photoreceptor cell death compared with control eyes (PBS or IgG) 24 h after retinal detachments (*p = 0.012611 vs PBS, *p = 0.012851 vs IgG; n = 10). B, Representative images of TUNEL staining in RD retinal tissues from TRPV4KO mice (PBS, IgG, or neutralizing anti-MCP-1 antibody). The TRPV4KO eyes (n = 8) with neutralizing anti-MCP-1 antibody (100 μg/ml) did not show a significant difference in photoreceptor cell death compared with control eyes 24 h after retinal detachments. ONL, Outer nuclear layer; INL, inner nuclear layer.

RD-induced TRPV4 activation recruits macrophages in subretinal space. Immunostaining of CD11b (a macrophage marker, represented as green) was performed in RD retinae. Nuclei were stained by Hoechst (blue). Shown is quantification of CD11b-positive cell numbers in WT or TRPV4 retinae 24 h after retinal detachments (n = 8, **p = 0.00632, Student t test). Scale bar, 100 μm. ONL, Outer nuclear layer; INL, inner nuclear layer.