Molecular Genetics Of Early Eye Development

National Eye Institute

Investigators

Linked publications & trials

Abstract

Olfactomedin 1 (Olfm1), Olfm2 and Olfm3 (also known as Noelin1-3) are secreted glycoproteins expressed throughout the brain and retina in a cell type-specific manner. Olfm1-3, neuritin (Nrn1), brorin (Vwc2), and brorin-like (Vwc2l) proteins are extracellular components of the AMPA receptor (AMPAR) complex. AMPARs are the major ionotropic glutamate receptors within both the brain and the retina that play a major role in synaptic formation and communication. Changes in AMPAR composition and activity may lead to chronic neurological disorders such as Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, amyotrophic lateral sclerosis, epilepsy, and glaucoma. Our previous results suggested that changes in the AMPAR activity induced by the elimination or mutations in their extracellular component (Olfm1-3, Brorin) can lead to retinal pathologies. Therefore, it is important to elucidate the molecular mechanisms of action of these proteins and understand how they contribute to brain and eye pathologies through modifications of AMPAR physiology. Our previous observations demonstrated that Vwc2 directly interacts with Olfm2 and GluA2, one of the core subunits of AMPARs, and Vwc2 and Olfm2 mutually facilitate interaction with GluA2. An in-silico modeling suggests that the Olfm2 dimer, Vwc2, and GluA2 tetramer form a complex. The interactions between these proteins were stabilized by the presence of hydrogen bonds, salt bridges and hydrophobic interactions. The olfactomedin domain of Olfm2 is critical for interaction with both Vwc2 and GluA2, while the N-terminal part of the Olfm2 dimer in the AMPAR complex protrudes into the extracellular space facilitating AMPAR clustering. In the brain, clustering of AMPARs determines synaptic strength and is essential for learning and memory. In the retina, clustering may affect efficiency of neuronal information processing. We continued to use Olfm1 KO, Olfm2 KO, Olfm3 KO, and Olfm1-3 triple knockout (TKO) mice to study the impact which the removal of these extracellular proteins has on retinal functions. The RNA-Seq data from retinal samples of different KO and wild-type (WT) lines for the identification of differentially expressed genes. The impact of KO condition was observed prominently in the TKO retinas and was less prominent in single KO conditions. The spectrum of differentially expressed genes in Olfm-TKO retinas showed higher similarity with the Olfm1-KO retina followed by Olfm3-KO and Olfm2-KO retinas. The pathway enrichment analysis revealed the major impact on the immune regulatory processes followed by the extracellular matrix organization, and ligand-receptor binding processes. Our data suggest microglial cells, photoreceptors, and RGCs may be affected in different Olfm KO. Olfm1 and Olfm2 are strongly expressed in RGCs, with Olfm1 broadly expressed across all subtypes, Olfm2 restricted to specific subtypes, and Olfm3 expressed in C26, C39, and C40 (M1dup) RGC subtypes. Multi-electrode array recordings (in collaboration with the Dr. Jiang, Baylor College of Medicine) revealed a marked loss of light-ON RGCs in the Olfm TKO retina. RNA Scope in situ hybridization using subtype-specific probes showed a statistically significant reduction in Neurod2-positive cells (41.6%), with smaller, non-significant decreases in Foxp2- and Tasc5-positive cells compared with WT. Other RGC subtypes were unaffected. These results suggest that N-RGCs and F-RGCs expressing Neurod2 or Foxp2/Tasc5 are selectively vulnerable to the loss of olfactomedin proteins, consistent with the multi-electrode recording. To further resolve these changes, single-cell RNA sequencing of RGCs in WT and TKO mice is underway. In collaboration with Dr. Fakler (University of Freiburg, Germany), we analyzed cerebellar circuitry and motor behavior in KO, Olfm2/3 double KO, and Olfm TKO mice. These animals exhibited profound defects in motor coordination and motor learning, most severe in the TKO line. Notably, connections from mossy fibers to Purkinje cells via granule cells and parallel fibers were disrupted. These genes may play a role in human oculomotor disorders and could be implicated in Parkinson’s disease and amyotrophic lateral sclerosis. To better understand how Nrn1 affects the visual pathway, we investigated RGC function, axon projection, and visual behavior in Nrn1 KO mice. In Nrn1 KO line, pattern electroretinography revealed a ~47% reduction in P1–N2 amplitude in Nrn1 KO mice compared to WT (p<0.0001), indicating impaired RGC function. RNAscope and immunostaining confirmed RGC-specific expression of Nrn1. Despite this dysfunction, optomotor reflex testing showed no difference in spatial frequency or contrast sensitivity between KO and WT mice. This aligns with preserved axon projections to accessory optic system (AOS) targets, including the medial terminal nucleus and nucleus of the optic tract, as revealed by cholera toxin B-based anterograde tracing. While the dorsal lateral geniculate nucleus (dLGN) showed reduced signal intensity in KO mice, its axonal coverage remained intact. These preserved subcortical circuits likely support reflexive visual behavior. In contrast, KO mice exhibited significantly reduced axon projections to non-AOS targets, including a ~50% reduction in mean intensity in the superior colliculus (SC) (p<0.001), and reductions in the olivary pretectal nucleus, ventral LGN, and suprachiasmatic nucleus (SCN), supporting selective disruption of long-range visual pathways. Notably, these deficits occurred despite minimal reported Nrn1 expression in SC and SCN, suggesting that axonal abnormalities arise intrinsically from RGCs rather than target-driven degeneration. RNA sequencing further identified downregulation of genes involved in postsynaptic density and Notch signaling. Finally, intravitreal delivery of Nrn1 protein with its binding partner Olfm1 significantly enhanced RGC survival after optic nerve crush compared to vehicle. These findings demonstrated that Nrn1 is critical for RGC integrity and connectivity and may represent a promising therapeutic target in glaucomatous neurodegeneration. Olfm2 may also have another unexpected non-ocular role. In collaboration with Dr. Francisco J. Ortega’s laboratory (Institut d'Investigació Biomèdica de Girona: Girona, Catalonia, Spain) we demonstrated that OLFM2 modulates key metabolic and structural pathways, including PPAR signaling, citrate cycle, fatty acid degradation, axon guidance and focal adhesion. On the molecular level, OLFM2 deficiency in differentiated adipocytes predominantly down regulated genes involved in cell cycle. Olfm2 depletion in mice resulted in impaired adipose cell cycle gene expression, with the latter also displaying fat mass accretion and metabolic dysfunction. Collectively, these results demonstrated a critical role for OLFM2 in adipocyte biology and supported a causative link between reduced adipose OLFM2 and the pathophysiology of obesity. In summary, our results demonstrate that Olfm1, Olfm2, and Olfm3 act synergistically to maintain motor functions and RGC populations in mice. Our findings support the concept that secreted AMPAR-associated proteins act individually or in concert to regulate visual and motor functions.