A Multi-Scale MRI Approach to Investigate Novel Drug Treatment Strategies in Mouse Models of Alzheimer's Disease

Several emerging therapies focus on preventing and clearing the accumulation of pathological lesions of tau in Alzheimer's disease. As such, there is high demand for early biomarkers sensitive to tau pathology in vivo. To assess the sensitivity of MRI to tau suppression, we longitudinally imaged two cohorts of rTg4510 mice, with doxycycline administered from 3.5 months (‘early’ intervention) and 4.5 months (‘late’ intervention) to coincide with the development of pathological tau products within the forebrain from 2.5 months. We investigated the exchange of endogenous mobile proteins using amide proton transfer(APT), microstructural diffusion using diffusion tensor imaging(DTI) and morphometric changes using high-resolution structural MRI. 39 rTg4510, and 18 wildtype (WT) litter-matched mice were scanned at baseline (3.5 months for the ‘early’ intervention study and 4.5 months for the ‘late’ intervention study) for APT, DTI and structural imaging using a 9.4T MRI and parameters previously described. 10 rTg4510 from each cohort were then treated with doxycycline hyclate(10mg/kg) and maintained on a doxycycline diet for the remainder of the study to supress tau-expression. The doxycycline-treated rTg4510s, untreated rTg4510s and WTs were imaged again at 5.5 and 7.5 months for APT, DTI and structural imaging. Atrophy was detectable in rTg4510s from 5.5 months (Fig.1A). Volume loss was lower in animals treated with doxycycline from 3.5 months than in those treated from 4.5 months. Both interventions produced an increase in APT towards WT values by 7.5 months (Fig.1B,E). Following early treatment (3.5 months), APT within the treated rTg4510s remained within the range of WTs until 7.5 months (Fig.1B). Here, both the treated rTg4510s and the WTs were significantly greater than untreated rTg4510s. DTI discriminated between groups: with increased diffusivity detected within the rTg4510s at 7.5 months (Fig.1C,F). Following treatment from 3.5 months, DTI illustrated excellent sensitivity to doxycycline and readily discriminated between the treated and untreated rTg4510s (Fig.1C). Multiparametric MRI results following early (A-C) and late (D-F) therapeutic intervention. Results are shown for the high tau pathology regions. Error bars represent the standard error of the mean. WT vs. rTg4510: * = p ≤ 0.05; ** = p≤0.01; *** = p ≤ 0.001; **** = p ≤ 0.001. WT vs. rTg4510(+DOX); + = p ≤ 0.05; ++++ = p ≤ 0.0001. rTg4510s. vs. rTg4510(+DOX); -- = p ≤0.01; ---- = p ≤ 0.0001. We demonstrate sensitivity of MRI biomarkers to both early and late suppression of pathological tau in an animal model of tauopathy. This longitudinal approach enables comparison between the multiparametric methods at different stages of pathology, which can guide MRI investigation of novel therapies. Clinically, this could provide valuable insights towards assessment of tau-related pathologies, including diagnosis and prognosis, and assessing the efficacy of tau-targeted therapeutics.