Abstract
Purpose :
Progress in the development of therapies to enhance remyelination in demyelinating diseases has been hampered by a lack of appropriate preclinical models - functional measures are often lacking or variable. We sought to develop a tractable and reproducible model of optic nerve demyelination with precise structural and functional measures.
Methods :
Oligodendrocytes of MBP-DTR 100a transgenic mice express diphtheria toxin receptor (DTR) and systemic diphtheria toxin (DT) administration induces diffuse demyelination of the central nervous system. In the present study we used retrobulbar DT injection to induce focal demyelination of the optic nerves of 3-month-old MBP-DTR 100a mice.
Dose optimisation: anaesthetised mice underwent unilateral retrobulbar DT injection with 5, 10 or 15ng/kg DT (n=7 per dose, 1 µL per injection). Tissues were harvested three weeks after injection.
Time-course study: Following baseline visual evoked potential (VEP) recording, electroretinogram (ERG) and optical coherence tomography (OCT), mice underwent retrobulbar DT injection with 15ng/kg DT or 1µL PBS. Follow-up measurements were taken at 2 (n=5 DT, 5 PBS), 4 (n=6 DT, 6 PBS), 8 (n=9 DT, 9 PBS) or 12-weeks (n=7 DT, 7 PBS). Animals were culled at each timepoint for tissue analysis.
Tissue analysis: Optic nerves were resin embedded, sectioned (1µm) and stained with toluidine blue for myelin analysis, or cryosectioned for immunofluorescence, and retinas were flat-mounted for ganglion cell counts.
Results :
3 weeks after injection with 15ng/kg DT, optic nerves showed colocalisation of activated caspase 3 & olig2, consistent with the apoptosis of oligodendroglia. Gliosis and axonal degeneration were evident.
VEP N1 latency was not delayed in DT treated nerves (compared to PBS) at 2 weeks (DT 66.6±6.9 ms, PBS 62.4±4.2ms, p=0.66). Significant delays were seen at 4 (DT 80.2±4.6ms, PBS 61.3±1.3ms, p=0.002) and 8 (DT 73.0±1.8ms, PBS 61.8±2.2ms, p=0.003) weeks. N1 latency was not delayed at 12 weeks (DT 62.7±2.0ms, PBS 59.3±2.6ms, p=0.33). Recovery of VEP latency was associated with remyelination.
No differences in positive scotopic threshold response (p=0.503) or retinal layer thickness were apparent between groups at any timepoint.
Conclusions :
We describe a reproducible model of optic nerve demyelination that can be used to study remyelination and therapies to enhance the regenerative process. Functional and structural sequelae are closely correlated.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.