Monogenic gene therapy for glaucoma and optic nerve injury

The prevalence of glaucoma, the second leading cause of global blindness, is increasing due to aging populations. In glaucoma, degeneration of the optic nerve and retinal ganglion cells (RGCs) causes visual field defects and eventual blindness. Elevated intraocular pressure (IOP) stands out as the best-known factor affecting glaucoma. However, there exists a subtype of glaucoma, known as normal tension glaucoma, that is not associated with high IOP. A recent study has identified various factors involved in glaucoma pathogenesis, including altered retinal blood flow, glutamate neurotoxicity, oxidative stress, and others (Shinozaki et al., 2024). Furthermore, glaucoma patients may exhibit reduced amounts of neurotrophic factors, such as brain-derived neurotrophic factor (BDNF) or ciliary neurotrophic factor compared with age-matched controls. Studies indicate that intraocular injections of BDNF can rescue RGCs in a mouse model of optic nerve crush (ONC) through activation of its high-affinity receptor Tropomyosin receptor kinase B (TrkB). However, the transient nature of ligand-dependent activation poses limitations on the efficacy of this treatment. We have developed several systems, described in the following sections, to address and overcome these limitations.

We recently developed a constitutively active form of TrkB, consisting of a farnesylated form of the intracellular domain of TrkB (F-iTrkB) (Nishijima et al., 2023). Farnesylation localizes iTrkB to cell membranes independently of BDNF binding. In the conventional BDNF–TrkB pathway, when BDNF binds to full-length TrkB, TrkB monomers dimerize and transphosphorylate each other, activating downstream signaling pathways including PI3K–AKT and Ras–ERK pathways (Figure 1). In contrast, we find that constitutively active F-iTrkB does not form dimers. Instead, F-iTrkB is transphosphorylated through transient interactions, robustly stimulating downstream signaling even in the absence of BDNF (Figure 1). Moreover, the relatively compact size of F-iTrkB allows us to use the strongest version of the CAG promoter and remain within the packaging capacity limit of 4.7 kb for the adeno-associated virus (AAV) vector. AAV-mediated delivery of F-iTrkB into the eyes alters many gene expressions in RGCs, and induces both RGC protection and robust optic nerve regeneration (> 4.5 mm), with some axons reaching the optic chiasm, in a mouse model of ONC. Administration of AAV-F-iTrkA also facilitated optic nerve regeneration to a similar extent as AAV-F-iTrkB. Additionally, AAV-F-iTrkB prevented RGC dendrite retraction and loss of synapses between RGC and bipolar cells. Consistent with these findings, retinal functions, as assessed by multifocal electroretinography, were preserved by AAV-F-iTrkB (Nishijima et al., 2023). There is also a different gene therapy construct, TrkB-2A-mBDNF, designed to sustain BDNF-TrkB signaling. AAV-mediated delivery of TrkB-2A-mBDNF demonstrated neuroprotective efficacy in a mouse ONC model (Khatib et al., 2021). While a direct comparison with the effects of F-iTrkB is not feasible, each system represents a unique approach to addressing optic nerve injuries.

Figure 1.

A schematic model of sustained activation of TrkB signaling in the absence of BDNF.

AAV-mediated delivery of F-iTrkB induced sustained activation of the downstream signaling pathways in retinal ganglion cells in the absence of BDNF. Intraocular injection of AAV-F-iTrkB induced retinal ganglion cell protection and robust optic nerve regeneration in mouse models of glaucoma and optic nerve trauma. AAV: Adeno-associated virus; BDNF: brain-derived neurotrophic factor; P: phosphorylation; TrkB: tropomyosin receptor kinase B. Created with Microsoft PowerPoint.

PI3K–AKT and Ras–ERK signaling pathways downstream from TrkB are inhibited by PTEN and NF1 respectively. Prior investigations have demonstrated that PTEN deletion promotes optic nerve regeneration following ONC (Sun et al., 2011). In our study, we observed that our AAV-F-iTrkB treatment elicited more significant RGC axon regeneration compared to PTEN deletion. This suggests that F-iTrkB exerts a more potent effect than unleashing endogenous PI3K signaling by eliminating its suppressor, PTEN. Conversely, NF1 did not influence RGC axon regeneration (Nishijima et al., 2023), indicating that NF1 does not play a major role as a Ras-GAP in RGCs. Additionally, we explored the impact of gene therapy utilizing an active form of K-Ras, the most prevalent mutation in human cancer patients. Our findings revealed that AAV-mediated transduction of K-Ras in the retina demonstrated a neuroprotective effect and also facilitated robust, long-term RGC axon regeneration (< 3.0 mm) in vivo (Kiyota et al., 2023). RGC protection and axon regeneration were more robust by AAV-F-iTrkB compared with AAV-K-Ras. Intriguingly, no apparent oncogenic effects of AAV-K-Ras were observed in mature RGCs, despite K-Ras being a well-known contributor to various cancers. To further ensure safety, potential therapies combining the AAV vector with an inducible system (such as Tet on/off or optogenetics) might be employed to enable transient K-Ras expression specifically during treatment.

Several studies have employed a combination of two or more factors to achieve long-distance optic nerve regeneration in mice. For instance, PTEN deletion with the manipulation of SOCS3, Lin28, and ATF3 resulted in a greater length and intensity of axon regeneration, with some regenerating axons even passing the optic chiasm (Sun et al., 2011). However, our findings reveal that intraocular injection of AAV-F-iTrkB or AAV-K-Ras alone into wild-type mice induces robust axon regeneration, with some axons reaching optic chiasm after ONC. Furthermore, intraocular injection of AAV-F-iTrkB promotes axon regeneration at the injury site, restores connections to brain targets, and leads to a partial recovery of visual behavior in an optic tract transection model (Nishijima et al., 2023). These data indicate that overexpression of F-iTrkB or K-Ras may be effective in stimulating robust axon regeneration as a monogenic treatment.

We also assessed the effects of AAV-F-iTrkB in GLAST knockout mice, an animal model of normal tension glaucoma, and found that intraocular injections significantly enhanced the survival of RGCs (Harada et al., 2007). In addition, AAV-F-iTrkB protected RGCs in high IOP model mice, without affecting IOP (Nishijima et al., 2023). These results suggest that gene therapy using F-iTrkB may be useful in preventing disease progression in both normal and high IOP glaucoma (Harada et al., 2019). Collectively, these findings demonstrate that AAV-F-iTrkB treatment can protect neurons from disease and injury and mediate robust axon regeneration.

Besides the gene therapies mentioned above, TrkB agonists are demonstrating promising results in promoting RGC survival in acute and chronic models of glaucoma (Lambuk et al., 2022). Further, we reported that the overexpression of dedicator of cytokinesis 3 (DOCK3), an atypical member of the guanine nucleotide exchange factors, plays a crucial role in optic nerve regeneration (Namekata et al., 2010). Recently, we screened 462,169 low-molecular-weight compounds and identified DOCK3 activators that stimulate neuroprotection and axon regeneration (Namekata et al., 2023). It will be interesting to compare the effects of TrkB agonists with TrkB gene therapies to develop more efficient and safer therapeutic strategies in the future.

In addition to RGCs, recent studies have demonstrated the involvement of retinal astrocytes in the pathogenesis of glaucoma (Shinozaki et al., 2022; Cameron et al., 2024). The loss of ATP-binding cassette transporter A1, a common genetic risk factor for primary open-angle glaucoma, in mouse astrocytes induced normal tension glaucoma-like phenotypes (Shinozaki et al., 2022). Moreover, proliferating neuroprotective astrocytes promote RGC survival by inhibiting microglial activation and preventing downstream neurotoxic astrocyte differentiation (Cameron et al., 2024). These results suggest that retinal astrocytes are potential therapeutic targets for glaucoma, and gene therapies targeting both RGCs and astrocytes may yield synergistic effects in treating glaucoma.

Many novel and promising therapies have been reported to stimulate neuroprotection and axon regeneration for glaucoma and optic nerve injury although we mainly introduced our new method of gene therapy using F-iTrkB in this perspective. Longitudinal studies are essential to determine the long-term efficacy and durability of the F-iTrkB therapy. In addition, we will examine the potential off-target effects on non-RGCs to ensure the therapy’s safety. Finally, further studies are required to validate our results in larger animal models or develop additional evidence supporting the translatability of the therapy to human eyes. In anyway, a novel therapy distinct from lowering IOP is urgently needed to restore visual function in glaucoma patients.

This work was supported by Japan Society for the Promotion of Science (JSPS) KAKENHI Grants-in-Aid for Scientific Research (JP22K09804 to CH; JP21K09688 to XG; JP19KK0229, JP21H04786, JP21H02819 and JP21K18279 to TH), the Shiseido Female Researcher Science Grant (to XG), Mitsubishi Foundation and Takeda Science Foundation (to TH).