Axonal protection by combination of ripasudil and brimonidine with upregulation of p-AMPK in TNF-induced optic nerve degeneration

Rho kinase (ROCK) inhibitors are commercially available in some countries as a glaucoma eyedrop. ROCK inhibitors reduce intraocular pressure (IOP) by expanding empty spaces in the juxtacanalicular area and increasing retinal blood flow [1]. For example, the ROCK inhibitor netarsudil acts on the conventional outflow pathway and reduces IOP in primary open-angle glaucoma (POAG) and ocular hypertension by improving trabecular outflow facility and decreasing episcleral venous pressure [2]. Another ROCK inhibitor, ripasudil hydrochloride hydrate, also reduces IOP by increasing outflow through modulation of trabecular meshwork cell behavior and regulation of permeability in Schlemm’s canal endothelial cells [3].

The glaucoma eyedrop brimonidine tartrate is an α2-adrenoreceptor agonist that reduces IOP and is widely used in the treatment of glaucoma. It was demonstrated that topical administration of brimonidine prevented retinal ganglion cell (RGC) loss in a retinal ischemia model [4]. Moreover, it was shown that brimonidine protected RGCs in a chronic ocular hypertensive rat model [5]. It was also reported that pretreatment of optic nerve axons with brimonidine resulted in a greater survival of optic axons in an ischemic optic neuropathy model [6].

We previously reported the axonal-protective effects of ripasudil in a tumor necrosis factor (TNF)-induced axonal damage model [7] and demonstrated the axonal-protective effects of brimonidine in the TNF-induced axonal damage model [8]. There have been several recent reports showing a close relationship between TNF and glaucoma [9, 10]. Those associations are supported by previous studies showing that increased expression of TNF and TNF receptors were observed in RGCs and the optic nerve head of patients with glaucoma [11] and that TNF-mediated cell death is involved in the neurodegeneration process of glaucoma [12]. The TNF intravitreal injection model showed that axonal damage precedes the reduction of RGCs [13]. Although one possible mechanism is that brimonidine suppresses increased TNF levels, thereby resulting in RGC protection in the ischemia reperfusion model [14], the mechanism of axonal protection of ripasudil and brimonidine remains to be elucidated. In the current study, we attempted to determine whether simultaneous injection of ripasudil and brimonidine modulates axonal loss in TNF-induced optic nerve degeneration and tested the association with AMPK activation.

The present study showed substantial protective effects of combined ripasudil and brimonidine administration against axonal damage caused by TNF. In our previous study, either 20 pmol of ripasudil alone or 2 pmol of brimonidine alone showed mild protective effects that were not statistically significant [7, 8]. Thus, we used the current doses to examine the additive effects of the two-drug combination in which the doses did not have significant (only marginal) effects. However, contrary to expectations, the present study showed significant axonal protection in both the ripasudil-alone and brimonidine-alone groups. This result may be related to the different methods that the number of samples in this study was greater compared with those in previous experiments. In the present study, the ripasudil and brimonidine combination group showed a significant protective effect compared with the results in the single-agent alone groups. Therefore, it is possible that ripasudil and brimonidine may exert additive protective effects. However, the caution may be needed to interpret the protective effect, because functional effects on the retina and optic nerve evaluation using the electrophysiology, pupil light reflex testing or optokinetic responses were not performed in the present study. Thus, it is unclear whether any observed protection of axons will have real functional consequences in terms of the functional preservation.

Several studies using different ROCK inhibitors focused on axonal regeneration after optic nerve injury. These ROCK inhibitors which have axonal regeneration effects include Y-27632, Y-39983, netarsudil, and ripasudil [16‐19]. Among them, it was shown that ripasudil suppressed the phosphorylation of cofilin, a depolymerizing protein [19]. It was also shown that brimonidine exerted axonal regeneration through Erk phosphorylation [20]. Although axonal regeneration is important, we believe axonal protection is helpful in halting glaucoma progression [21]. A previous study showed that oral ripasudil administration delayed RGC death [22]. Several ROCK inhibitors were shown to have protective effects on RGCs in different damage models [16, 17, 23]. Our previous study demonstrated that ripasudil exhibited axon protection by promoting intra-axonal autophagy [7]. On the other hand, a previous study by Lambert et al. [24] found that continuous treatment with subcutaneous brimonidine injections significantly improved the survival of RGCs exposed to elevated intraocular pressure. We also found that brimonidine suppressed the increase in p62/SQSTM1 in axonal damage caused by TNF [8]. Since increased p62 means impairment of autophagy flux, it is possible that brimonidine promotes autophagic flux, thereby leading to axonal protection [8].

In the present study, the p-AMPK level tended to decrease in the TNF group, although the difference was not statistically significant. A recent study has shown that the expression of p-AMPK was downregulated in mouse neuroblastoma N2a cells after TNF treatment [25], suggesting dephosphorylation of AMPK by TNF. However, we can not exclude the possibility that TNF may downregulate AMPK protein [26]. Our immunoblot analysis also showed that the AMPK level tended to decrease in the TNF group without statistical significant.

Fasudil, a ROCK inhibitor, activates AMPK in skeletal muscle cells [27]. A recent study has demonstrated that ripasudil upregulates p-AMPK in bovine corneal endothelial cells [28]. We have recently found that the ROCK inhibitor netarsudil increased p-AMPK in the optic nerve with upregulation of autophagy [29]. Very recent studies have also shown a close relationship between ROCK inhibition and AMPK activation in several different cells [30, 31], suggesting that ROCK inhibitors may act as upstream effectors of AMPK. On the other hand, the α2-adrenoreceptor agonist dexmedetomidine protected cardiovascular endothelial cells via AMPK-autophagy [32]. In addition, it was shown that dexmedetomidine exerted neuroprotection via AMPK-autophagy in SH-SY5Y-APP cells [33]. These findings are in line with the present findings that the combination of ripasudil and brimonidine activates AMPK in the optic nerve. Treatment with the AMPK activator A769662 increased Thr-172 phosphorylation of AMPK, resulting in stimulated PGC-1α-directed mitochondrial biogenesis and autophagy induction [34]. Treatment with A769662 exerted axonal protection associated with AMPK activation and autophagy induction in TNF-induced optic nerve degeneration [29]. A very recent study showed that A769662 protected UVA-induced retinal pigmented epithelial cells [35]. Taken together, these findings suggest that the combination of ripasudil and brimonidine may have axonal-protective effects with possible involvement of AMPK activation.