Brimonidine has been shown to be effective at inhibiting the development of myopia in animal models. Carr et al. found that brimonidine at 20nmol/20µL (approximately 0.442ug/µL) and 200nmol/20µL (approximately 4.422ug/µL) of injected intravitreally was effective in slowing the progression of form-deprivation myopia in chicks [5]. Our groups previously reported similar results, in that that intravitreal injection of 4µg/µL brimonidine was effective in slowing the progression of form-deprivation myopia in guinea pigs [8]. The results of the present experiment, specifically that 4ug/µL intravitreally is effective, are generally consistent with the results of these two studies. In contrast to Liu et al. who found that brimonidine eye drops inhibited the progression of lens-induced myopia (LIM) in guinea pigs, in the present study we found that brimonidine eye drops did not slow the progression of form-deprivation myopia (FDM) in guinea pigs [4]. Differences in outcomes of these studies might be due to differences in (e.g.) strains of guinea pigs, ways of delivering eye drops, instrumentation, and techniques for measuring SE and AL; but the most likely explanation lies in the different ways of inducing myopia, as some research findings have suggested different mechanisms for LIM and FDM [15, 16].

We also found that subconjunctival injection of 40ug/µL and intravitreal injection of 2ug/µL brimonidine were effective in suppressing myopia development, while intravitreal injections of higher concentrations (20ug/µL and 40ug/µL) were ineffective. It might be due to the off-target binding of high concentrations drugs. We speculate that brimonidine– like many other ligands - is not specific to a single kind of α2-adrenoceptor or other receptors but becomes active at other receptor(s) at higher concentrations; similarly, biphasic responses to drugs over a wide range of concentrations have been reported for the effects of dopaminergic agents on naturally occurring myopia in albino guinea pigs [17]. In the same way, the three mAChR antagonists— atropine, himbacine, and MT3-bind to human α2A-adrenoceptors when administered at or above concentrations of 45µmol/L, 17 µmol/L, and 15 nmol/L, respectively, in HEK293T cells [7]. The subconjunctival injection of 40ug/µL of brimonidine could reach the same concentration as the vitreous injection of 2ug/µL and 4ug/µL, in the retina or other target tissues, and thus inhibit the effect of FDM. We have been unable to find any published studies on the pharmacokinetics and cell or tissue targets of brimonidine after subconjunctival or intravitreal injections, and further studies will be needed to understand the bases of such anomalous responses to the drug.

Research on the correlation between IOP and myopia has been a hot topic of study in recent years; however, the exact relationship between IOP and myopia is controversial. An early human study found that after excluding factors such as amblyopia, strabismus, prematurity, age, and family history of myopia, myopia was still strongly correlated with IOP [18]. Several similar studies have shown that IOP was significantly higher in the high myopia group than in the control group and that it was significantly correlated with ocular axial length [19,20,21]. However, contrary to the findings in those reports, two other studies found no statistically significant differences in IOP between control and myopia groups or between groups with different degrees of myopia [22, 23].

Comparing our multiple group data by ANOVA revealed that the IOPs in groups D4, E1, and E2 were significantly lower than those in group B, from day 14 onwards (Fig. 5). Correlation analysis of data for groups B, D4, E1, and E2 showed that all correlations were positive (Fig. 6), with IOP and AL tending to increase in the simple form deprivation group (group B) and to decrease in the brimonidine-responsive group (D4, E1, and E2), compared to group B. But the left eyes (untreated eye) of these 3 groups were essentially unchanged compared to group B (Fig. 7); these findings in the present study are consistent with the results of previous studies [8]. IOP was highly correlated with AL in group E1 (r = 0.925, P < 0.001), and less highly but still significantly correlated with AL in the B, D4, and E2 groups (r = 0.899, P < 0.05 in group A; r = 0.773, P < 0.05 in group D4; and r = 0.724, P < 0.05 in group E2). It remains unclear whether there is a causal relationship between IOP and AL, and it has been suggested [24] that lowering IOP inhibits the activation of scleral fibroblasts, thereby reducing scleral remodelling, and that a decrease in scleral dilatation force retards the balloon-like expansion of the scleral coat. It has also been suggested that lowering IOP leads to increased choroidal blood perfusion, which reduces scleral hypoxia and is accompanied by decreases in scleral remodelling. However, there is no conclusive evidence that high IOP causes myopia, and on the contrary, there is convincing evidence that scleral enlargement (stretching or active growth), rather than increased IOP, is most important. For example, in chick as well as several mammalian models it has been shown that hemiretinal form deprivation or hyperopic defocus produced localized axial elongation– specifically, the form-deprived or imposed-defocus region became larger, whereas the untreated regions of the eye wall did not [25,26,27]. Since pressure in a fluid is exerted equally in all directions, and if high IOP causes myopia by affecting scleral compliance, then the vitreous chamber should enlarge uniformly, not locally. Based on the results of the current experiment, with regard to group B (Monocular FD alone), the guinea pigs in this group did not receive brimonidine treatment, it is observed that group B began to show a decrease in IOP on day 15 (Fig. 5), while the spherical equivalent (SE) began to show myopia on day 10 (Fig. 3). The change in IOP lagged behind the changes in SE; therefore, the elevated IOP could simply be a consequence of myopia production.

Correlations of AL with IOP in baseline group B and treatment groups D4, E1 and E2

Changes in intraocular pressure over the course of 3 weeks in left eye (untreated eye)

Most studies suggest that the control of myopia progression is primarily mediated through the action of mAChRs [28, 29], However, recent studies have shown that atropine is unlikely to inhibit myopia progression by acting upon mAChRs, and more likely does so via adrenergic receptors [30]. For example, McBrien et al. found in myopia experiments with chicks and tree shrews that retinal acetylcholine (ACh) levels in myopic animals were not significantly different from those in controls [31]. In addition, many other antagonists of mAChRs, unlike atropine, did not inhibit the development of myopia [32]. Thomson et al. also found that muscarinic, nicotinic, and non-specific cholinergic agonists inhibited FDM development, leading them to question whether atropine inhibits myopia via cholinergic antagonism [33]. These findings suggest that mAChRs may not affect the development of myopia. In addition, Näreoja found that certain muscarinic toxins not only interfere with binding of acetylcholine to its receptors, but also have moderate to high affinity for adrenergic receptors [34]. It has even been found that atropine interacts with α-adrenoceptors in addition to muscarinic receptors [35]. Carr’s group also found that muscarinic antagonists block signaling via α2A-adrenoceptors at concentrations comparable to those used to inhibit chick myopia in vivo [7]. Currently, some experimental animal studies have found that α2-adrenoceptor agonists can inhibit the development of myopia in animals (including chicks and guinea pigs) [4, 5, 8]. The results of these studies suggest that α2-adrenergic receptors may be the real target of atropine’s action in inhibiting myopia development. However, while brimonidine is equally effective as atropine in inhibiting form-deprivation myopia in chicks [5, 32], the required concentration still greatly exceeds the effective concentrations found in receptor binding and activity assays [36, 37]. Another study published by our experimental group [8] also found that the expression of adrenergic signaling-related genes in the retina of form-deprived guinea pig eyes injected with brimonidine was not significantly different from that of the control group. This might suggest that the mechanism by which brimonidine slows myopia progression is not mediated by adrenergic signaling-related pathways. The specific mechanism by which brimonidine slows myopia progression still needs to be investigated further.

Although the guinea pig eye is similar to the human eye in many ways, if we want to study it further in monkeys or humans, we need to know the differences between the guinea pig eye and the primate eye. For example, the guinea pig retina is avascular and under physiological conditions the oxygen content of the retina and choroid is much lower than in other animals with vascularised retinas [38]. Pharmacokinetic experiments are required to determine the efficiency of retinal absorption rates. Compared to primates, guinea pigs are more hyperopic at birth and undergo emmetropization during post-natal development. Emmetropization in guinea pigs is rapid during the first 3 weeks of age and then slows down. During the first five weeks after birth. the increase in axial length in guinea pigs is primarily determined by the thickening of the crystalline lens. However, the axial growth of the eye in primates is mainly determined by lengthening of the vitreous chamber, followed by deepening of the anterior chamber and the lens thickening [39]. This suggests that the timing of brimonidine administration in relation to these developmental processes may require further discussion.

Despite these results showing the inhibitory effect of brimonidine on myopia, this study has limitations. Firstly, this experiment did not investigate the mechanisms and pathways involved. The mechanism leading to the occurrence of myopia has been widely investigated, and factors such as retinal dopamine secretion [40], scleral extracellular matrix [41, 42], and scleral hypoxia [43] have been implicated, but none of them can fully explain the mechanism of myopia in this study. Secondly, due to budgetary constraints, the experimental animals used in this study were guinea pigs. Although guinea pigs, like humans, are mammals, they differ to some extent from humans in terms of eye structure and the aetiology of myopia. This is not conducive to further clinical trials of this drug. If future researchers are interested in further studies, it is suggested that primates, which are closer to humans, should be used as research subjects.

In conclusion, our study suggests brimonidine at appropriate doses significantly reduced the development of FD myopia in guinea pigs. The IOP may change with FD myopia. There is a positive correlation between IOP and AL, with IOP increasing as AL increases. Brimonidine is a highly promising drug for future myopia treatment.