Cognitive decline and dementia are among the most common nonmotor changes in neurodegenerative disorders such as PD and DLB and are often associated with intraneuronal a-syn deposits which lead to synaptic dysfunction and degeneration 
. Elevations in a-syn protein levels contribute to the pathogenesis observed in PD, as gene duplications/triplications of the a-syn gene have been shown to cause familial PD or parkinsonism 
. In the present study, we set out to study the relationship between UCH-L1 and a-syn as these proteins are implicated in PD and have been previously shown to interact 
. We examined hippocampal tissues to assess potential UCH-L1-dependent alterations that may occur in this brain region due its inactivation. In doing so, we investigated mechanisms that could potentially lead to neuronal dysfunction and loss and ultimately cognitive decline and dementia. Here we show that inhibition of UCH-L1 activity in non tg mice leads to an increase in a-syn protein expression levels. Similarly, inhibition of UCH-L1 activity in primary hippocampal neurons expressing endogenous levels of a-syn led to an increase in the size and number of presynaptic a-syn clusters. In contrast, blocking UCH-L1 activity in a-syn transgenic mice or neurons that over express a-syn resulted in a decrease in a-syn levels that correlated with its enhanced clearance from synapses. Furthermore, the observed alterations in a-syn levels were not due to changes in its mRNA expression levels or neuron cell density but correlated with dysregulation of the autophagy pathway in a-syn transgenic mice. Our data point to differential effects of UCH-L1 inhibition in the context of normal and pathologic conditions, in that blocking UCH-L1 activity leads to opposing effects on a-syn distribution and protein levels in cells that express endogenous a-syn levels vs. those that over express the protein.
The precise nature of, and mechanisms underlying, how UCH-L1 might modulate a-syn are not clear. One possibility might involve a direct interaction. a-syn is a primarily pra-synaptic protein whilst UCH-L1 is localized both in the pre and post-synaptic regions 
. Interestingly, in a previous study 
we have shown that the effects of LDN were preferentially on post-synaptic proteins such as PSD-95. However, as we show here, presynaptic proteins such as a-syn might also be affected.
A recent study examined synuclein levels in ubiquitin carboxyl-terminal hydrolase L1-deficient gad mouse and reported β- and a-synuclein immunoreactive spheroids in the gracile nucleus in these mice 
. In contrast, alpha-synuclein immunoreactivity was barely detectable in spheroids. However, it is worth noting that they did not investigate a-syn levels in the synapses which where we found re-distribution of a-syn rather than in axonal swellings. In addition, other factors that might explain the differences between our results and theirs are compensatory mechanisms in the context of a developmental UCH-L1 KO compared to a functional transient effect of a pharmacological inhibitor such as LDN.
In our previous study, we demonstrated that pharmacological suppression of UCH-L1 activity led to reduction in mono-ubiquitin levels, which in turn suppressed ubiquitin-dependent protein degradation by the proteasome 
. Consequently, UCH-L1-dependent perturbation of proteasome activity affected the stability of important synaptic scaffold proteins, PSD-95 and Shank, which are known targets for ubiquitination and degradation by the UPS 
. Here, we show that reduction in UCH-L1 activity, and subsequent decreased mono-ubiquitin levels, under non-pathogenic conditions, results in accumulation of a-syn, which is degraded by the UPS 
. These observations are in line with our previous study and demonstrate that alterations in ubiquitin homeostasis could affect ubiquitin-dependent degradation of certain proteins, including a-syn, by the UPS.
In contrast to our observations in non-pathogenic conditions, inhibition of UCH-L1 activity under pathogenic conditions (h-a-syn over expression) reduces a-syn protein levels, and this could potentially be beneficial for neuronal survival. Many lines of evidence have suggested that dysfunction in the autophagy pathway is common in many neurodegenerative disorders including in PD and DLB 
. Furthermore, a-syn aggregates have been shown to interfere with the autophagy mechanisms and ultimately lead to neurodegeneration in vivo
and in vitro
. Interestingly, excess a-syn and its aggregates are cleared by autophagy 
. This observation is in line with a recent study by Spencer et al., where it was shown that LV-mediated over expression of beclin 1, a major component in initial steps of the autophagy pathway, reduced abnormal accumulation of a-syn and associated neuronal deficits in a transgenic mouse model of DLB 
. The observed hippocampal accumulation of p62, whose levels are known to increase in response to inhibition in autophagy, with a concurrent decrease in cleaved LC3 II levels in our a-syn transgenic model point to a dysfunction in the autophagy pathway in these mice 
. Furthermore, Kabuta et al. have shown that UCH-L1 interacts with members of the chaperon-mediated autophagy (CMA), and that the familial PD-associated UCH-L1I93M
interacts abnormally with these members, leads to inhibition of CMA and as a result an increase in a-syn levels in cultured cells 
. Taken together, under pathological conditions (h-a-syn over expression), blocking UCH-L1 activity may enhance autophagy, which in turn will serve to degrade and reduce a-syn levels. The mechanism(s) by which inhibition of UCH-L1 by LDN affects its interaction with the members of autophagy pathway are unknown. Interestingly, our double-immunolabeling for UCH-L1 and a-syn, only in non tg mice treated with LDN, showed a redistribution in UCH-L1 staining pattern. It is unclear how this redistribution may affect the interaction of UCH-L1 with members of the autophagy pathway, and whether these alteration lead to the opposing effects on a-syn observed in non tg vs. a-syn tg mice. Our data suggest that loss of UCH-L1 function in the context of h-a-syn-induced pathologies may have neuroprotective effect in the hippocampus.
A recent study by Liu et al. demonstrated that a population of UCH-L1 is membrane-bound (UCH-L1M
), and C-terminal farnesylation of UCH-L1 promotes its association with cellular membranes 
. Over expression of wild type UCH-L1 led to an increase in the levels of UCH-L1M
expressed in cells and correlates with accumulation of a-syn and neurotoxicity 
. Interestingly, a mutation in the farnesylation sequence (CKAA→SKAA) of UCH-L1, resulting in the C220S mutant, eliminates the membrane-associated species of UCH-L1, and has no effect on a-synuclein levels. Pharmacological inhibition of UCH-L1 farnesylation, on the other hand, was shown to reduce a-syn levels possibly by promoting its degradation through lysosomal pathway, and increased cell viability 
. In addition, approximately 30% of UCH-L1 was found to be membrane-associated in cortical tissues of diseased and normal human brains (e.g. PD and AD). However, an association between UCH-L1M
(soluble UCH-L1) ratio and disease, in this case, in cortical tissues from PD patients was not detected 
. These data clearly demonstrate a link between UCH-L1 farnesylation, membrane association and a-syn expression levels. However, the effects of modulation of UCH-L1 activity (e.g. by LDN) on these conditions are unknown. It is worth noting that the C220S mutation does not affect hydrolytic activity of UCH-L1 
. Interestingly, Kabuta et al. did not report an increase in a-syn expression levels due to wild type UCH-L1 over expression using the same cell culture model system 
. Also, we have used a model system in which UCH-L1 and a-syn are abundantly expressed in the hippocampus and cultured neurons contrary to these other systems where neither proteins were endogenously expressed. Our use of a system where proteins of interest are endogenously expressed presents a more suitable system and reflective of what may occur in vivo
In conclusion, we show that suppression of UCH-L1 activity has differential effects on a-syn in neurons that express normal or excessive levels of a-syn. Under normal physiological conditions, perturbation of UCH-L1 activity has a dramatic effect on distribution and protein levels of presynaptic a-syn, which in turn could have detrimental effects on normal neuronal function. Under pathological conditions, however, loss of UCH-L1 function proves to be beneficial as it not only enhances a-syn degradation but also relieves the a-syn-mediated block of the autophagy pathway. Further studies are needed to evaluate the potential value of blocking UCH-L1 as a therapeutic target in a-synucleinopathies.