Axon pathology in mouse models of Huntington's disease

Axon or synapse dysfunction parallels or precedes symptom onset in many neurodegenerative disorders. In some of these conditions, not only do axon and synapse loss determine the course of pathology, but protection of those neuronal compartments is mandatory to alleviate the disease. Whether this is also the case in Huntington’s disease (HD), a devastating neurodegenerative disorder characterised by progressive deterioration of both physical and mental abilities and inevitable early death, remains unclear. Present therapeutic strategies do not address protection of axons and synapses, which may help explain why there is no effective treatment currently in use. Moreover, an accurate characterisation of the development of axon pathology relative to neuronal loss and to the deposition of mutant-Huntingtin (mHTT) aggregates, neuropathological hallmark of HD, is lacking. In the present thesis I have carried out a detailed study aiming to investigate axon degeneration in the R6/2 transgenic (Tg) and the HdhQ140 knock-in (KI) mice, two HD models, and to assess whether this occurs early in and contributes to the course of pathology. I tested the hypothesis that axon degeneration precedes or at least parallels degeneration of other neuronal compartments in these mice. To characterise axon pathology and its spatio-temporal relationship to aggregate formation, neuronal loss and symptom onset, I crossed R6/2 and HdhQ140 mice with YFP-H transgenic mice that express the yellow fluorescent protein (YFP) in a subset of neurons. Neuronal pathways labelled by YFP in this model include some reported to be affected in HD. In these mice individual fluorescent neurons can be tracked over long distance and axons can be traced back to their cell bodies. Using a powerful axon imaging method that was developed and successfully applied to study Alzheimer’s disease, I was able to place axon degeneration accurately in the sequence of pathological events and develop methods to quantify it as readout for future therapeutic studies from our group and elsewhere. I found that the morphology of axons was strikingly abnormal in some brain areas in HdhQ140 homozygous mice (HdhQ140/Q140) where large axonal swellings were detected at 6 months and at 12 months of age in stria terminalis and striatum. In these mice, the number of axonal swellings increased age-dependently and was significantly higher than that found in wild-type littermates. However, I did not detect degeneration in cell bodies, dendrites or synapses suggesting that axon pathology is the main feature of the disease in this model. To better characterise the KI model, I also performed a battery of behavioural tests to assess motor and cognitive impairment during disease progression. I used tests of locomotor activity, motor coordination and balance and sensorimotor gating to measure motor function and tests of spatial working memory and anxiety-like behaviour to assess cognitive and behavioural symptoms, respectively. A longitudinal study from 1 to 12 months was carried out to detect pathological changes from early stage and relate them to swelling formation. In all tests, I found a strong reduction in locomotor activity in HdhQ140 mice compared to the controls although balance and coordination seemed not to be impaired as rotarod performance was unaltered. Alterations were also detected in prepulse inhibition, suggesting sensorimotor defects occur in these mice, while no abnormal cognitive or psychiatric behaviour was detected in the time-frame of the study. Behavioural symptoms, as well as abnormal morphological changes found in axons, worsened with time and major impairments were found at the latest time-point, 12 months of age. Finally, I asked whether alterations in the NAD biosynthetic pathway could underline the signs of axon pathology detected in HdhQ140 homozygous mice, as it has recently emerged that this pathway regulates axon survival and axon and synapse degeneration in many neurodegenerative disorders. To this purpose, I looked at possible alterations in the level of nucleotides (NMN, NAD) and in the activity of key enzymes in this pathway (NMNAT, NAMPT). I also tested the hypothesis that mHTT interacts with NMNAT enzymes and with the Wallerian degeneration slow protein (WLDs), an NMNAT fusion protein, and interferes/impairs their normal function. As WLDs delays axon degeneration in acute and neurodegeneration models, future works may address beneficial role of WLDs in HD/WldS crossed mice. Despite no detected alterations in nucleotide levels or enzymatic activity in the KI mice compared to the controls, colocalisation was found between mHTT and WLDs and between mHTT and NMNAT2, an important axon survival factor, suggesting a possible interaction between these proteins which could play a role in HD neurodegeneration