Neurobiology of Disease (Apr 2016)
Interplay between TDP-43 and docosahexaenoic acid-related processes in amyotrophic lateral sclerosis
Abstract
Background: Docosahexaenoic acid (DHA), a key lipid in nervous system homeostasis, is depleted in the spinal cord of sporadic amyotrophic lateral sclerosis (sALS) patients. However, the basis for such loss was unknown. Methods: DHA synthetic machinery was evaluated in spinal cord samples from ALS patients and controls by immunohistochemistry and western blot. Further, lipid composition was measured in organotypic spinal cord cultures by gas chromatography and liquid chromatography coupled to mass spectrometry. In these samples, mitochondrial respiratory functions were measured by high resolution respirometry. Finally, Neuro2-A and stem cell-derived human neurons were used for evaluating mechanistic relationships between TDP-43 aggregation, oxidative stress and cellular changes in DHA-related proteins. Results: ALS is associated to changes in the spinal cord distribution of DHA synthesis enzymatic machinery comparing ten ALS cases and eight controls. We found increased levels of desaturases (ca 95% increase, p < 0.001), but decreased amounts of DHA-related β-oxidation enzymes in ALS samples (40% decrease, p < 0.05). Further, drebrin, a DHA-dependent synaptic protein, is depleted in spinal cord samples from ALS patients (around 40% loss, p < 0.05). In contrast, chronic excitotoxicity in spinal cord increases DHA acid amount, with both enhanced concentrations of neuroprotective docosahexaenoic acid-derived resolvin D, and higher lipid peroxidation-derived molecules such as 8-iso-prostaglandin-F2-α (8-iso-PGF2α) levels. Since α-tocopherol improved mitochondrial respiratory function and motor neuron survival in these conditions, it is suggested that oxidative stress could boost motor neuron loss. Cell culture and metabolic flux experiments, showing enhanced expression of desaturases (FADS2) and β-oxidation enzymes after H2O2 challenge suggest that DHA production can be an initial response to oxidative stress, driven by TDP-43 aggregation and drebrin loss. Interestingly, these changes were dependent on cell type used, since human neurons exhibited losses of FADS2 and drebrin after oxidative stress. These features (drebrin loss and FADS2 alterations) were also produced by transfection by aggregation prone C-terminal fragments of TDP-43. Conclusions: sALS is associated with tissue-specific DHA-dependent synthetic machinery alteration. Furthermore, excitotoxicity sinergizes with oxidative stress to increase DHA levels, which could act as a response over stress, involving the expression of DHA synthetic enzymes. Later on, this allostatic overload could exacerbate cell stress by contributing to TDP-43 aggregation. This, at its turn, could blunt this protective response, overall leading to DHA depletion and neuronal dysfunction.
