Pathway analysis identifies altered mitochondrial metabolism, neurotransmission, structural pathways and complement cascade in retina/RPE/ choroid in chick model of form-deprivation myopia

Abstract

Purpose: RNA sequencing analysis has demonstrated bidirectional changes in metabolism, structural and immune pathways during early induction of defocus induced myopia. Thus, the aim of this study was to investigate whether similar gene pathways are also related to the more excessive axial growth, ultrastructural and elemental microanalytic changes seen during the induction and recovery from form-deprivation myopia (FDM) in chicks and predicted by the RIDE model of myopia.

Methods: Archived genomic transcriptome data from the first three days of induction of monocularly occluded form deprived myopia (FDMI) in chicks was obtained from the GEO database (accession # GSE6543) while data from chicks monocularly occluded for 10 days and then given up to 24 h of normal visual recovery (FDMR) were collected. Gene set enrichment analysis (GSEA) software was used to determine enriched pathways during the induction (FDMI) and recovery (FDMR) from FD. Curated gene-sets were obtained from open access sources.

Results: Clusters of significant changes in mitochondrial energy metabolism, neurotransmission, ion channel transport, G protein coupled receptor signalling, complement cascades and neuron structure and growth were identified during the 10 days of induction of profound myopia and were found to correlate well with change in axial dimensions. Bile acid and bile salt metabolism pathways (cholesterol/lipid metabolism and sodium channel activation) were significantly upregulated during the first 24 h of recovery from 10 days of FDM.

Conclusions: The gene pathways altered during induction of FDM are similar to those reported in defocus induced myopia and are established indicators of oxidative stress, osmoregulatory and associated structural changes. These findings are also consistent with the choroidal thinning, axial elongation and hyperosmotic ion distribution patterns across the retina and choroid previously reported in FDM and predicted by RIDE.

Keywords: Bile acid metabolism; Gene set enrichment analysis; Mitochondrial energy metabolism; Myopia; Neurotransmission.

Figure 1. MRI images of chick eye following 10 days of FDM induction and 3 days recovery.

(A) Monocular form-deprivation of the right eye (RE) for 10 days demonstrates abnormal ocular growth, excess vitreal volume, and thinned choroid of RE compared to its fellow left eye (LE). (B) 72 h post-occlusion recovery (i.e., 72 h normal visual experience) resulted in vitreous volume decrease and choroidal expansion. Previous studies (Liang et al., 2004) have shown ∼300% increase in choroidal thickness in RE compared to LE after three days recovery from FD. Note: Images (same magnification) in (B) slightly more dorsal than in (A). Image credit: G Egan.

Figure 2. Ocular Biometrics for FDMI and FDMR.

Mean (±SE) measures of refractive status, axial length (AL) and vitreous chamber depth (VCD). To complement the data reported by McGlinn et al. (2007), (A) refraction and (B) AL & VCD were collected during 6 h and 72 h of normal development and 6 h and 72 h following seven-days induction of myopia. Refraction, AL & VCD measures for 24 h recovery after prolonged (i.e., 10 days) form deprivation is shown in (C) and (D). Both refractive state and axial length changes were highly correlated (r = .78) during occluder wear (E). Note: Measures for anterior chamber and lens thickness are included in Fig. S1.

Figure 3. Enrichment map for highly clustered pathways in normal eye development.

Gene set enrichment analysis revealed 61 biological pathways that can be functionally grouped into 10 clusters using a co-efficient of similarity altered during the 10 days of normal eye development in retina/RPE/choroid. Note: Each node represents a biological pathway from File S1. The colour of each node emphasises the direction of expression and normalised enrichment score (NES). Node size is relative to the number of genes in the pathway. Thickness of the connections (green) between each node reflects the degree of similarity between each gene set. Twenty-six pathways did not meet the clustering similarity coefficient of 0.5 and hence are not shown here. Note cluster names: GAG, glycosaminoglycan; GPCR, g-protein coupled receptors.

Figure 4. Enrichment map for highly clustered pathways in form deprivation induction and recovery.

Axial elongation during 10 days of form-deprivation compared to normal unoccluded controls resulted in 130 altered pathways in retina/RPE choroid (inner node) while 24 h recovery (outer annulus) identified only one statistically significant pathway i.e., bile acid & bile salt metabolism. Pathways not statistically enriched during FDMR are shown for comparison purposes. Notably, expression profiles of FDMI and FDMR are consistent despite the fact that only the FDMR data includes choroidal tissue. Pathways highly expressed during induction (red inner node) were often suppressed during normal vision and recovery (blue outer annulus) and vice versa. Note: Each node represents a biological pathway from File S1. The colour of each node emphasises the direction of expression and normalised enrichment score (NES). Node size is relative to the number of genes in the pathway. Thickness of the connections (green) between each node reflects the degree of similarity between each gene set. There were 22 unclustered pathways in FDMI that did not meet the clustering similarity coefficient of 0.5. Note cluster names: CCC, complement and coagulation cascades; CME, clatherin-mediated endocytosis; FA, Fatty acid; GF, growth factors; GPCR, g-protein coupled receptors; MAPK, mitogen-activated protein kinases; NGF, nerve growth factor.

Figure 5. Median expression of pathways involved in mitochondrial metabolism.

Graphs of the seven mitochondrial metabolism pathways with significant expression shifts across 240 h of occluder wear relative to unoccluded controls (A) Alzheimer’s disease, (B) Huntington’s disease, (C) Parkinson’s disease, (D) Oxidative phosphorylation, (E) Respiratory electron transport, (F) Respiratory electron transport/ATP synthesis by chemiosmotic coupling and heat production by uncoupling proteins, (G) TCA cycle and respiratory electron transport.

Figure 6. Median expression of pathways involved in neurotransmission during normal ocular development and in FDMI.

Graphs of the neurotransmission-related pathways with significant expression shifts during normal ocular development (dotted lines) and FDMI (solid lines) are shown. (A–D) Four pathways were significant in both normal development and FDMI. The leading-edge subsets for these pathways identified 115 common core genes shared within these pathways during normal development and during FDMI and 27 other core genes specific to normal development and nine specific to FDMI (File S1). (E–H) Graphs indicate FDMI induced down regulation of expression shift in four additional neurotransmission-related pathways with significant expression shifts during FDMI (solid lines) only. These pathways were not significant during normal ocular development but data are shown for comparison purposes (dotted lines). (A) Neuronal system (B) Neurotransmitter release cycle (C) Neurotransmitter receptor binding & downstream transmission in the postsynaptic cell (D) Transmission across chemical synapse (E) Activation of NMDA receptor upon glutamate binding and postsynaptic events (F) CREB phosphorylation through the activation of RAS (G) Long-term potentiation (H) Post-NMDA receptor activation events.

Figure 7. Median expression of pathways involved in ion transport during normal ocular development and in FDMI.

Expression of the (A) Ion channel transport and (B) Ligand-gated ion channel transport pathways with significant expression shifts during FDMI (solid lines) compared to normal development (dotted lines) are shown. The ‘Ligand-gated ion channel transport pathway’ pathway was also significantly altered during normal development and was clustered with the neurotransmission pathway.

Figure 8. Median expression of pathways involved in the complement and coagulation cascade.

Graphs indicate greater expression shift in the complement & coagulation cascade between 72 h and 240 h of occluder wear for both (A) formation of fibrin clot/clotting cascade and (B) complement and coagulation cascades. Note that the ‘complement and coagulation cascades’ pathway was also significantly altered during normal development.

Figure 9. Median expression of pathways involved in cytochrome p450 metabolism.

Graph indicates enhanced expression in cytochrome p450 related pathways in FDMI compared to normal development. (A) Biological oxidations (B) Cytochrome P450 arranged by substrate type (C) Drug metabolism, cytochrome P450 (D) Drug metabolism of xenobiotics by cytochrome P450 (E) Phase 1 functionalization of compounds.

Figure 10. Median expression of core genes in the bile acid and bile salt metabolism pathway during FDM.

Graphs shows median change of the core genes during normal ocular development, FDMI, and FDMR. This pathway was found to be significant for FDMI (left, solid line) and FDMR (right, solid line) but not in normal development. Note: Median expression value was calculated based on core genes identified in each experimental group. This pathway was not significantly altered during normal ocular development but data are shown for comparison purposes (dotted line).