Short Communication : Piezo1 plays a role in optic nerve head astrocyte reactivity

Abstract

Piezo1 (also known as Fam38A) is a mechanosensing channel required for mechanotransduction in various cell types. In astrocytes, Piezo1 activation is associated with the pathogenesis of central nervous system neurodegeneration. Expression of Piezo1 has been detected in mouse optic nerve head astrocytes, however, the functional role of Piezo1 has not been identified. In this study, we investigated the role of Piezo1 in optic nerve head astrocyte reactivity. Primary mouse optic nerve head astrocytes were cultured and subject to mechanical stretch. The expression level of Piezo1 was determined by quantitative PCR and immunocytochemistry staining. Astrocytes were further treated with Yoda1, a specific Piezo1 agonist. The intracellular calcium concentration and expression of F-actin and fibronectin were determined in Yoda1 treated cells. We found that mechanical stretch activated Piezo1 in optic nerve head astrocytes. Yoda1 induced robust Ca²⁺ responses in a dose dependent manner. In addition, Yoda1 treated cells showed a redistribution of F-actin cytoskeleton and an increased expression of fibronectin which indicated astrocyte reactivity. Our results suggest that Piezo1 responses to mechanical stimulation and plays a role in astrocyte reactivity. This study provides new insights into optic nerve head astrocyte mechanotransduction.

Astrocytes are the major glial cells in the optic nerve head (ONH). They play an important role in maintaining the microenvironment of the ONH and provide both structural and physiological support to optic nerve axons when they exit the globe. Astrocytes are mechanosensitive (Beckel et al. 2014; Ostrow, Langan, and Sachs 2000). Mechanically activated calcium currents have been described in astrocytes and associated with astrocyte reactivity (Ostrow, Langan, and Sachs 2000). In the ONH, astrocytes are exposed to substantial levels of mechanical stress/strain related to intraocular pressure (Sigal and Ethier 2009). Thus, mechanotransduction, the mechanism linking the intraocular pressure and cell signaling, has an important role in ONH astrocyte physiology and pathophysiology.

Piezo proteins have been identified as the first mammalian “professional” mechanosensitive ion channels (Kim et al. 2012; Coste et al. 2012). They respond to membrane tension and become activated independent of other cellular components (Syeda et al. 2016). Piezo1 (also known as Fam38A), the founding member of the family, is primarily expressed in non-sensory tissues exposed to fluid pressure and flow and functions in endothelial cell types under both static and shear stress conditions (Coste et al. 2010; Cahalan et al. 2015). Piezo2 (Fam38B) is required for rapidly-adapting mechanically-activated currents in somatosensory neurons, suggesting a potential role in touch and pain sensation (Coste et al. 2010).

Piezo1 activation has been associated with astrocyte reactivity in central nervous system diseases. In Alzheimer’s disease, Piezo1 expression is detected in activated astrocytes surrounding amyloid-β plaques (Satoh et al. 2006). Piezo1 also plays a role in neuroinflammation in the brain. Astrocytes upregulate Piezo1 expression following lipopolysaccharide stimulation and reduce the release of inflammatory cytokines and chemokines (Velasco-Estevez et al. 2020). Piezo channels have been detected in ONH astrocytes (Choi, Sun, and Jakobs 2015). However, the role of Piezo1 in ONH astrocyte activity has not been identified.

In this study, we cultured primary mouse ONH astrocytes, investigated response of Piezo1 to mechanical stimulation and astrocyte activity following Piezo1 activation. We found that mechanical stretch activated Piezo1, and Piezo1 activation induced astrocyte reactivity. Our results confirm the expression of Piezo1 in ONH astrocytes and indicate a functional role of Piezo1 in ONH astrocyte mechanotransduction and reactivity.

All animal procedures were performed in accordance with the National Institutes of Health guide for the care and use of Laboratory animals and the ARVO statement. All experimental protocols were approved by the Institutional Animal Care and Use Committee of University of North Texas Health Science Center.

We cultured primary optic nerve head (ONH) astrocytes from adult C57BL/6J mice as previously described (Liu et al. 2018). Briefly, the ONH was dissected and seeded into 60 cm cell culture dishes (BD Falcon, Franklin Lakes, NJ, USA). After 5–7 days in culture, cells migrating out of ONH explants were subcultured in defined astrocyte culture medium (Am-a medium, Sciencell Research Laboratories, Carlsbad, CA, USA) at 37°C in a humidified atmosphere with 5% CO₂. Cells were characterized with astrocyte markers including glial fibrillary acidic protein (GFAP), neural cell adhesion molecule (NCAM) and S100β. Cells positive for astrocyte markers were used in experiments.

The 3-D mechanical stretch stimulation has been widely used in biomechanical studies due to its physiological relevance (Moraes et al. 2013; Stucki et al. 2015). Compared to 2-D mechanical stretch, the 3-D stretch model closely mimics the biomechanics of the ONH (Sigal and Ethier 2009). A mechanical stretch device (Fig. 1), consisting of a pneumatic chamber (2.0 cm in diameter and 0.5 cm in height) and a cell culture chamber (2.0 cm in diameter and 1 cm in height), made of polydimethylsiloxane (PDMS, Sylgard 184, Dow Corning, Midland, MI, USA) was used in this study. These chambers were prepared by casting a mixture of PDMS resin and curing agent at a ratio of 10:1.05 (w/w) on the 3-D printed molds, and cured at 75°C for 2 hr. The pneumatic chamber was punched a hole, where a stainless-steel tubing (0.05” OD, 0.009” wall thickness, 2.5 cm long) was inserted and secured by applying PDMS prepolymer as the glue. The PDMS membrane was prepared by spin-coating the PDMS pre-polymer at 2500 rpm on a silicon wafer for 1 min. After cured at 75 °C for 1 hr, the PDMS membrane was sandwiched between the cell culture and pneumatic chambers by applying microtransfer assembly technique (Yang et al. 2011), i.e. using a PDMS pre-polymer adhesive layer which was formed via spin-coating at 2500 rpm for 10 min. The membrane was sterilized by using isopropyl alcohol and UV exposure, each 30 min, and coated with with 10 μg/ml poly-D lysine (Sigma-Aldrich, St. Louis, MO, USA) and 100 μg/ml laminin (Sigma-Aldrich). Primary mouse ONH astrocytes were seeded on the membrane. The 3-D mechanical stretch was applied for 24 hours at a speed of 100 μm/s and distance of 100 μm to induce 10% strain (Sigal et al. 2007) by perfusing/withdrawing air into/out of the pneumatic chamber using a PHD 2000 remote syringe pump (Harvard Apparatus, Holliston, MA).

Figure 1. Piezo1 channel responded to mechanical stretch stimulation.

(A) Illustration of 3-D printed PDMS chamber. The ONH Astrocytes were seeded on poly-D-lysine and laminin coated PDMS membrane in the cell culture chamber. Mechanical stretch was controlled via the PDMS membrane by pumping air in/out of the pneumatic chamber. (B) Illustration of side view of stretched PDMS membrane. (C) Mechanical stretch increased Piezo1 expression. The ONH astrocytes were mechanically stretched for 24 hours and immunostained with specific antibody against Piezo1 (green). DAPI nuclear counterstaining is shown in blue. Top image: cells not stretched as control. Bottom image: cells 24 hours after mechanical stretch. The insert in the bottom image shows cytoplasmic increase of Piezo1. (D) Percentage of Piezo1 positive cells in non-stretched cells and cells 24 hours after stretch. (E) Relative Piezo1 mRNA expression in non-stretched cells and cells 2 hours after stretch. Mean ± SD, n=3–4. *: p<0.05, ***: p<0.005. Unpaired t-test. Scale bar: 50 μm.

Total RNA was isolated using RNeasy mini Kit (Qiagen, Valencia CA, USA). Reverse transcription and quantitative PCR were performed using the iScript™ cDNA synthesis kit and the SSoAdvanced™ SYBR Green master mix (Bio-Rad Laboratories, Hercules CA, USA), respectively, following manufacturer’s instructions. Mouse Piezo1 was amplified using primers 5’-GCAGTGGCAGTGAGGAGATT-3’ (forward) and 5’-GATATGCAGGCGCCTATCCA-3’ (reverse). Relative mRNA abundance was calculated by the ΔΔ cycle threshold (Ct) method. Gapdh (Forward: 5’-AGAACATCATCCCTG CATCC-3’, Reverse 5’-AGCCGTATTCATTGTCATACC-3’) was used as the endogenous control.

For immunocytochemistry staining, astrocytes were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA) in phosphate-buffered- saline (PBS) for 30 minutes at 4°C. After pretreatment with 0.5% Trixon-X (Acros Organics, NJ, USA) in PBS for 30 minutes, the cells were blocked with 10% Goat serum (ThermoFisher Scientific, Eugene, OR, USA) in PBS containing 0.1% Trixon-X for 2 hours at room temperature (RT). Primary antibodies against fibronectin (Abcam, Cambridge, MA, USA) or Piezo1 (Proteintech, Rosemont, IL, USA) were incubated overnight at 4°C, and Alexa Fluor 488 conjugated secondary antibody (ThermoFisher Scientific) against rabbit IgG was incubated for 1 hour at RT. Non-primary control immunostaining was performed using PBS instead of primary antibodies. Cells were also incubated with Alexa Fluor 568 conjugated phalloidin (ThermoFisher Scientific) for 30 minutes at RT to stain F-actin. Images were captured using a Nikon Ti-U Microscope with the Nuance Multispectral imaging system. To quantify fluorescence intensity and Piezo1 positive cells, images were taken at 9 locations (1 in the center, 2 from mid peripheral and peripheral regions of each coverslip) of each coverslip or PDMS membrane with approximately 250 cells analyzed per image. Fluorescence quantification was performed using ImageJ software and normalized to cell numbers in same visual field. Four coverslips were evaluated in a masked manner per experimental condition.

Calcium imaging was performed as previously described (Park et al. 2015). Briefly, astrocytes were pre-incubated in Krebs-Ringer buffer solution (115 mM NaCl, 2.5 mM CaCl2, 1.2 mM MgCl2, 24 mM NaHCO3, 5 mM KCl, 25 mM HEPES, and 5 mM glucose, pH 7.4) containing 3 uM of fura-2-AM (ThermoFisher Scientific) for 20 minutes at 37° C. The 340/380 nm excitation ratio for fura-2-AM was measured following stimulation of Piezo1 with Yoda1 (100, 300 and 500 nM) or vehicle control DMSO for a period of 20 minutes using a Nikon Eclipse TE200–5 microscope and the NIS-Element AR3.2 software (Nikon Instruments, Melville, NY, USA). The intracellular Ca²⁺ concentration was calculated by the Grynkiewicz equation (Grynkiewicz, Poenie, and Tsien 1985). One-way ANOVA followed by Tukey post-hoc test was performed to analyze intragroup differences. Data are presented as mean ± S.D., and a P<0.05 was considered statistically significant.

Piezo1 and Piezo2 are the two members in the vertebrate Piezo family. They induce distinct mechanically activated currents and play different roles in mechanotransduction (Coste et al. 2010). Both Piezo channels have been detected in mouse ONH astrocytes (Choi, Sun, and Jakobs 2015). To determine the activation of Piezo channels in response to mechanical stimulation, we stretched ONH astrocytes using a PDMS chamber system (Fig. 1A & 1B) and performed quantitative PCR and immunocytochemistry staining. After 2 hours stretch, there was a significant increase of Piezo1 in stretched cells (Fig. 1E). After 24 hours, compared to non-stretched control (4.9 ± 2.1 %), stretched cells showed a significantly increased expression of Piezo1 in the cytoplasm (Fig. 1C) and increased number of cells (39.9 ± 5.3 %) showing positive staining (Fig. 1D). Piezo2 was detectable in non-stretched cells but did not show significant changes after 24 hours stretch (data not shown). The activation of Piezo channels in response to elevated intraocular pressure has been assessed in mouse model of glaucoma. There is an increased mRNA expression level of Piezo1 in the ONH of glaucomatous mice. Piezo2 did not show significant changes in ocular hypertensive mice (Choi, Sun, and Jakobs 2015). In our study, the stretch stimulation applied on astrocytes mimicked the strain placed on cells due to a mild elevation of intraocular pressure (Sigal et al. 2007). Consistent with previous publication, our results show an increased expression of Piezo1 but not Piezo2, which suggests a role of Piezo1 in ONH astrocyte mechanotransduction.

Piezo1 can be selectively activated by a specific agonist Yoda1 (Syeda et al. 2015), which acts as a molecular wedge, facilitating force-induced conformational changes and lowering the channel’s mechanical threshold for activation (Botello-Smith et al. 2019). We treated astrocytes with various concentrations of Yoda1 (100 nM, 300 nM and 500 nM) and assessed the calcium response. As expected, Yoda1 induced robust Ca²⁺ influx in a dose dependent manner (Fig. 2A–B). The [Ca²⁺]_I induced by vehicle control 0.005% DMSO was 198.1 ± 145.5 nM. With Yoda1 treatment, the [Ca²⁺]_I was 1552.1 ± 620.3, 4696.4 ± 623.6, and 5996.4 ± 213.0 nM as the concentration of Yoda1 increased. These results further confirm the existence of a functional Piezo1 channel in mouse ONH astrocytes.

Figure 2. Activation of Piezo1 induced ONH astrocyte reactivity.

Chemical activation of Piezo1 in ONH astrocytes. (A) Calcium traces of cells treated with vehicle control 0.005% DMSO, 100 nM, 300 nM and 500 nM Yoda1. (B). Bar graph shows [Ca^2+]i after DMSO or Yoda1 treatment. Yoda1 increased [Ca^2+]i in a dose-dependent manner. Mean ± SD, n=5–7. **: p<0.01, ****: p<0.001 compared to all other groups. One-way ANOVA followed by Tukey’s test. (C-J) Chemical activation of Piezo1 activated ONH astrocytes. ONH astrocytes were treated with vehicle control DMSO (C, E and G) or 300 nM Yoda1 (D, F, and H) for 72 hours and stained with phalloidin (E-H), specific antibodies against fibronectin (C-D). DAPI nuclear counterstaining is shown in blue. (I-J) Bar graphs show relative fluorescence intensity in DMSO and Yoda1 treated cells. Cells treated with Yoda1 show drastic shapeshifting, redistributed F-Actin cytoskeleton and increased expression levels of fibronectin and F-Actin. Mean ± SD, n=4. *: p<0.05, ***: p<0.005. Unpaired t-test. Scale bar: 50 μm.

Astrocytes remain quiescent and become activated after trauma or other insults. Activation of astrocytes involves various morphological and functional cellular responses (Sun, Qu, and Jakobs 2013; Ho, Lambert, and Calkins 2014). To evaluate the functional roles of Piezo1 in mouse ONH astrocytes, we further treated ONH astrocytes with Yoda1 and assessed astrocyte activity following Piezo1 activation. Based on the Ca²⁺ dose response results, we treated astrocytes with 300 nM Yoda1, and stained ONH astrocytes with phalloidin to evaluate the F-actin cytoskeleton changes induced by Pizeo1 activation. We found that Yoda1 induced drastic astrocyte shapeshifting between polygonal and stellate morphologies and F-Actin cytoskeleton redistribution from submembrane to across the cytoplasm (Fig. 2G–H) which is an important morphological characteristic of activated astrocytes (Tehrani et al. 2014; Tehrani et al. 2019). We further assessed expression levels of fibronectin and F-Actin in treated ONH astrocytes. Compared to vehicle control group, Yoda1 treated astrocytes showed significant increases of fibronectin and F-Actin (Figure 2C–F, 2I–J). Astrocytes are a major source of extracellular matrix components in the ONH. Reactive astrocytes increase extracellular matrix production and are heavily involved ONH extracellular matrix remodeling (Schneider and Fuchshofer 2016). The results showing an increased fibronectin production further support a reactive phenotype of astrocytes following Piezo1 activation.

In summary, the present study provides evidence that Piezo1 channel responses to mechanical stretch in ONH astrocytes. Activation of Piezo1 channel induces astrocyte reactivity. This study provides new insights into mechanisms linking astrocyte mechanotransduction and reactivity.