We report strain differences in the responsiveness to transduction of mouse eyes with mutant human myocilin. These data show two important, tractable and genetically distinct differences in the response to a known human glaucoma causing gene. First, intravitreal injection of Ad5.MYOC.Y437H induced ocular hypertension in A/J, BALB/cJ, and C57BL/6J mice but had minimal effect on C3H/HeJ mice. Second, intravitreal injection of Ad5.MYOC.Y437H induced optic nerve damage in A/J mice but had no significant damaging effect on optic nerves of BALB/cJ and C57BL/6J mice, even though all 3 strains had ocular hypertension phenotypes. As expected, given the lack of ocular hypertension, the C3H/HeJ mice had no optic nerve damage. These data provide an excellent resource to study the onset and progression of glaucoma phenotypes in mice, to elucidate genetic pathways involved in IOP regulation and optic nerve damage through genetic mapping experiments, as well as provide a novel model system that can be applied to any mouse strain.
The genetic and molecular pathways involved in IOP regulation are not well known. In addition, changes in aqueous humor dynamics and IOP are major risk factors for developing glaucoma. IOP measurement is also the primary clinical tool for monitoring the potential onset and development of other glaucoma phenotypes. Here we show strain differences in IOP elevation in response to human mutant myocilin. The exact mechanism of how mutant myocilin affects IOP has not been completely elucidated. Wild-type myocilin is secreted from TM and is expressed in other eye tissues (Jacobson et al., 2001
; Nguyen et al., 1998
), but its function is not known. Previously it has been shown that mutations in human myocilin induce exposure of a cryptic peroxisomal targeting sequence, whose interaction with the PTS1R is necessary for IOP elevation in mice (Shepard et al., 2007
). Based on these data it has been hypothesized that instead of being secreted, mutant myocilin is misfolded and shuttled to peroxisomes, leading to cellular and endoplasmic reticulum stress and resulting in deleterious TM function. The human mutant myocilin adenovirus used in our studies presented here is presumably acting in a similar fashion, resulting in peroxisome dysfunction. However, ocular hypertension was only induced in three of the strains tested. C3H/HeJ mice did not develop a pronounced mutant myocilin induced ocular hypertension even though the human myocilin transgene was expressed in the TM (). Interestingly, C3H/HeJ mice have altered levels of enzymes associated with peroxisome proliferation as well as polymorphisms in PPARG (peroxisome proliferator activated receptor γ) (Ackert-Bicknell et al., 2008
; Butler et al., 1988
). Therefore, it is possible that C3H/HeJ mice may have altered peroxisome function affecting how the TM cells process mutant myocilin, resulting in maintenance of IOP versus the increased IOP seen in other mouse strains. Further experiments are needed to test this hypothesis. By demonstrating dramatic differences in genetically distinct mouse strains, we can now utilize the power of mouse genetics to begin to tease out the genes and pathways responsible for these differences.
We have shown that A/J mice are susceptible to mutant myocilin induced ocular hypertension and optic nerve damage, whereas BALB/cJ and C57BL/6J mice displayed only the ocular hypertension phenotype. These data suggest that there are strain differences in the ability to progress from increased IOP to axonal damage in the optic nerve. Previously, glaucoma related strain differences have been identified between the DBA/2J and C57BL/6J mouse strains (Anderson et al., 2006
). However, in that particular case when the disease causing genes were transferred from the DBA/2J background to the C57BL/6J background, both the IOP and optic nerve damage phenotypes were lost. Here, we were able to separate the IOP phenotype from the optic nerve damage phenotype. Interestingly, the BALB/cJ strain was not sensitive to optic nerve damage or RGC loss in our model, but this strain is most susceptible to RGC loss in the optic nerve crush model (Li et al., 2007
). Expansion of the strain survey and/or backcross studies could identify the modifier genes associated with this response, which will lead to a better understanding of the molecular mechanisms responsible for glaucomatous optic nerve and RGC damage. This model would also be ideally suited to look at the roles of individual genes in glaucomatous damage and optic neuropathy neuroprotection. Thus, our model may provide an invaluable resource to begin to dissect the genetic and molecular components involved in the progression from increased IOP to axon and RGC damage.
It has been shown in other models of glaucoma that damage to the optic nerve precedes RGC damage and visual center damage in the brain (Howell et al., 2007
; Libby et al., 2005c
; Schlamp et al., 2006
; Soto et al., 2011
). Here, we recapitulate those results in our model. The A/J mice were susceptible to both ocular hypertension and optic nerve damage. However, there was no evidence of RGC damage or damage to the SC after 8-weeks post-injection of the human mutant myocilin adenovirus. This suggests that the initial insult is first occurring at the optic nerve. Further time course analysis is necessary to evaluate the temporal sequence of RGC and SC damage after the initial optic nerve insult.
Recently, there have been several mouse models of glaucoma developed using mutant myocilin transgenic mice (Senatorov et al., 2006
; Zhou et al., 2008
; Zode et al., 2011
). These model systems are beneficial in that they have TM damage with open irideocorneal angles, elevated IOP, RGC damage, and optic nerve damage. Zode et al have also demonstrated that therapeutic intervention can protect against glaucomatous damage in their transgenic mouse (Zode et al., 2011
). The model system we used in this study also has several benefits compared to many of the glaucoma mouse models developed to date. Using human mutant myocilin adenovirus is unique in that it is inducible and can be applied to any mouse strain, without having to extensively breed animals to transfer mutant genes from one background to another. The insult also mimics human glaucoma like the transgenic mice, by using a similar pathogenic pathway (mutant myocilin) to raise IOP. The response within each mouse strain was also homogeneous, yielding similar phenotypes in all animals. IOP elevation was achieved within a week post-injection of the adenovirus and optic nerve damage was shown by 8 weeks, providing a reasonable and cost effective time frame. Since we have identified strain differences in response to human mutant myocilin, we can use the model to exploit the power of mouse genetics to molecularly dissect pathogenic mechanisms.