PPARγ modulates refractive development and form deprivation myopia in Guinea pigs

Highlights

• PPARγ agonism and antagonism have opposing effect on refractive development and FDM.

• PPARγ agonism reverses FD-induced declines of choroidal thickness and choroidal blood perfusion.

• PPARγ agonism reverses FD-induced increases of scleral Hif-1α expression.

• PPARγ agonism reverses FD-induced declines scleral collagen type 1 expression.

Abstract

Form deprivation myopia (FDM) is characterized by loss of choroidal thickness (ChT), reduced choroidal blood perfusion (ChBP), and consequently scleral hypoxia. In some tissues, changes in levels of peroxisome proliferator-activated receptor γ (PPARγ) expression modulate hypoxia-induced pathological responses. We determined if PPARγ modulates FDM through changes in ChT, ChBP, scleral hypoxia-inducible transcription factor (HIF-1α) that in turn regulate scleral collagen type 1 (COL1) expression levels in guinea pigs. Myopia was induced by occluding one eye, while the fellow eye served as control. They received daily peribulbar injections of either the PPARγ antagonist GW9662, or the GW1929 agonist, with or without ocular occlusion for 4 weeks. Ocular refraction and biometric parameters were estimated at baseline, 2 and 4 weeks post-treatment. ChT and ChBP were measured at the 2- and 4-week time points. Western blot analysis determined the expression levels of scleral HIF-1α and COL1. GW9662 induced a myopic shift in unoccluded eyes. Conversely, GW1929 inhibited FDM progression without affecting the refraction in unoccluded eyes. GW9662 reduced both ChT and ChBP in unoccluded eyes, while GW1929 inhibited their declines in occluded eyes. Scleral HIF-1α expression rose in GW9662-treated unoccluded eyes whereas GW1929 reduced HIF-1α upregulation in occluded eyes. GW9662 downregulated scleral COL1 expression in unoccluded eyes, while GW1929 reduced their decreases in occluded eyes. Therefore, PPARγ modulates collagen expression levels and FDM through an inverse relationship between changes in PPARγ and HIF-1α expression levels.

Introduction

Globally, myopia has become the second most common ocular disorder, which in severe cases can lead to blindness (Bourne et al., 2013). By 2050, it is predicted that the number of people afflicted with myopia and high myopia (≥6.00 diopters [D]) will increase to 49.8% and 9.8% of the world population, respectively(Holden et al., 2016). In Asian countries, myopia prevalence is already approaching crisis levels in some cohorts (Chen et al., 2018; Koh et al., 2014; Lam et al., 2012). Even though myopia is a non-life-threatening disorder, the visual pathology associated with severe cases can lead to cataract, glaucoma, retinal detachment, and myopic macular degeneration (Iwase et al., 2006; Yip et al., 2011). Therefore, a pressing need exists to clarify the underlying pathogenic mechanisms and identify novel targets for selectively improving therapeutic management of this condition.

Mammalian myopia develops from excessive ocular elongation and scleral extracellular matrix (ECM) remodeling that render the sclera more extensible, particularly at the posterior pole (Funata and Tokoro, 1990; McBrien et al., 2001; Phillips et al., 2000). These ECM changes are largely due to declines in the content of collagen I (COL1) that account for about 90% of the dry weight of the mammalian sclera (Rada et al., 2006; Yang et al., 2009).

We recently showed that the development of form deprivation myopia (FDM) is associated with upregulation of the scleral hypoxia-inducible transcription factor (HIF-1α) (Wu et al., 2018). Upon cessation of form deprivation (FD), HIF-1α reverts back to levels similar to those in the fellow unoccluded eyes (Wu et al., 2018). This association between changes in HIF-1α expression levels and myopia progression supports the notion that this transcription factor contributes to the control of genetic events inducing ECM remodeling and myopia progression. Other recent studies provided additional support for the conclusion that hypoxia triggers myopia progression. For example, during FDM and lens-induced myopia, hypoxia likely develops due to declines in both choroidal thickness (ChT) and choroidal blood perfusion (ChBP) whereas cessation of FD reversed these choroidal changes in guinea pigs (Zhang et al., 2019). These changes are not species specific because an inverse association between declines in choroidal blood flow and increased myopia progression was also identified in chicks after 14 days of FD (Shih et al., 1993). These findings are consistent with other reports that showed declines in choroidal blood flow occur in highly myopic humans (Dimitrova et al., 2002; Shih et al., 1991; Yang and Koh, 2015). In order to improve therapeutic management of myopia, it is relevant to identify targets controlling scleral hypoxia whose modulation can selectively inhibit responses underlying this sight compromising disease.

Peroxisome proliferator-activated receptor (PPAR) subtypes are ligand-activated transcription factors belonging to the nuclear hormone receptor superfamily with many diverse regulatory functions (Lakatos et al., 2007). The PPAR family includes three subtypes, PPARα, PPARβ, and PPARγ. PPARγ is widely viewed as an essential regulator of lipid metabolic processes (Wang et al., 2018) that may also contribute to myopia (Cordain et al., 2002; Lim et al., 2010). There is recent mounting evidence suggesting that hypoxia inhibits PPARγ expression and activation in different tissues. For example, hypoxic activation of ERK1/2 and NF-κB stimulates Nox4-mediated H₂O₂ generation, which in turn reduces PPARγ expression and activity (Lu et al., 2013). PPARγ has been shown to be inhibited by hypoxic stress during adipocyte differentiation (Yun et al., 2002). On the other hand, PPARγ upregulation inhibits hypoxia-induced proliferation of arterial smooth muscle cells (Yang et al., 2018). Such effects by changes in PPARγ expression are thought to stem from its interactions with inhibits signaling pathway mediators at upstream sites in a signaling pathway mediating control of responses to hypoxia. However, a few other studies reported an opposite effect of hypoxia on PPARγ expression. For example, long-term hypoxia induced instead increases in PPARγ mRNA in fetal sheep adipose tissue (Myers et al., 2008), and it stimulated differentiation of C2C12 and G8 murine myogenic cell lines (Itoigawa et al., 2010). Also, hypoxia caused PPARγ mRNA and protein levels to rise above those in normoxic HepG2 cells and promote carcinogenesis. (Zhao et al., 2014).

While studies carried out in other tissues highlight a possible relationship between the effects of hypoxia on PPARγ expression levels, their interaction during myopia development is unknown. We aimed to determine if PPARγ agonism and antagonism affect parameters modulating scleral hypoxia and collagen expression levels, which are reflective of changes in ECM remodeling during myopia progression.