Oxidative stress and autophagy

Abstract

Many neurodegenerative conditions have oxidative stress burdens where levels of reactive oxygen species (ROS) exceed the antioxidant capacity of the neuron. ROS can induce wide-ranging damage in a cell and this is prevented by the activation of antioxidant responses including autophagy. Junkinase (JNK) is stimulated by ROS and mediates antioxidant responses via the activation of the transcriptional activators Fos and Jun (AP-1). In recently published work we examined Drosophila mutants with overgrown larval neuromuscular synapses, mutants that also show all the hallmarks of lysosomal storage disease (LSD). We find that we can reverse this synaptic overgrowth by reducing the oxidative stress burden, and that synaptic overgrowth is mediated by autophagy and JNK-AP-1 activity. We also examined animals defective for protection from oxidative stress and found that they too have synapse overgrowth generated by JNK-AP-1 activity. Treatment of larvae with a known ROS-generating toxin, paraquat, yielded similar synaptic responses. The observations that oxidative stress responses, potentially acting through autophagy, can generate synaptic growth suggest that ROS may be a potent regulator of synapse size and function. These findings have intriguing implications for aging neurons, neurodegenerative conditions and the interpretation of metabolic demand during learning and memory.

Normal synaptic development is a highly regulated process, controlled by growth pathways that are, to some degree, likely to be responsive to neuronal activity. A number of growth pathways have been identified using forward and reverse genetic approaches with the Drosophila larval neuromuscular junction (NMJ). The larval NMJ has many attractions as a model synapse for reasons of accessibility, size and the powerful molecular-genetic toolbox. Using this system, the JNK-AP-1 pathway has been identified as a potent pre-synaptic regulator of synapse growth and function. Mutations in the neuronally expressed E3-ubiquitin ligase highwire have revealed overgrown synapses generated by release of inhibition on a JNKKKinase, wallenda. More recent work has elegantly demonstrated that autophagy can degrade high-wire protein to release the inhibition on the JNK-AP-1 pathway. This work raises the idea that oxidative stress, acting through JNK-AP-1 and autophagy could regulate synapse growth. It has been widely demonstrated in other experimental systems that ROS acts as an activator of JNK-AP-1 signaling generating the suggestion that this signaling pathway could play an important role in regulation of synaptic growth and development. By their very nature, neurons, being post-mitotic and having an elevated energy demand, are predisposed to oxidative stress. The neuronal responses to oxidative stress, however, have been little investigated.

Previously identified Drosophila mutations in spinster (spin), were known to induce a neurodegenerative LSD-like condition. LSDs are characterized by the loss of a gene essential to lysosomal function or generation, reduced life span, lysosomal swelling, lipofuscin accumulation and, in some forms of the disease, synaptic overgrowth. LSDs also have elevated ROS levels. A marked phenotype in spin is the synaptic overgrowth of the larval NMJ. We hypothesized that an oxidative stress burden, acting through the JNK-AP-1 pathway could be generating the synaptic overgrowth in spin. We examined spin for oxidative stress and found excessive levels of peroxidated lipids and activation of a transgenic oxidative stress marker. We then asked a direct question: if we reduce oxidative stress, will synapse overgrowth be reduced? Expression of antioxidant transgenes (superoxide dismutase (SOD1), catalase and thioredoxin-reductase) in the spin mutant background reduces the synaptic overgrowth by 40%. We did not rescue overgrowth entirely: expression of SOD1 potentially generates a knock-on burden of peroxide, but each transgene gives an effective reduction in growth. Antioxidant expression also rescues a synaptic fatigue phenotype suggesting that oxidative stress impinges negatively on synapse function.

We followed our spin observations by asking: is oxidative stress sufficient to generate synaptic growth? We examined mutants defective in mechanisms for reducing oxidative stress (SOD1, SOD2). Synapses in these mutants we found to be overgrown, though not to the same level as spin. We also grew larvae in the mitochondrial poison paraquat to generate oxidative stress. These animals, though smaller in size due to the paraquat, have an elevated synapse size for their muscle surface area. These data suggested a generality for the proposition that oxidative stress can activate synapse growth.

The JNK-AP-1 pathway is a potent mediator of oxidative stress responses, and we examined and found in spin an activation of the AP-1 pathway. Blocking JNK-AP-1 signaling in spin, SOD1, and SOD2 animals blocks the synaptic overgrowth. The recent finding that autophagy can regulate synapse growth prompted an examination of this pathway as an oxidative stress response regulating synapse growth. Introducing autophagy mutations into a spin mutant background completely blocks synaptic overgrowth. Inhibiting function of JNK-AP-1 in either nerve or muscle in spin partially reduces the synaptic overgrowth in these animals suggesting a pre-and a post-synaptic contribution to growth. This suggestion is supported by functional knockdown of Atg5 function in either compartment to reduce synapse overgrowth by 50%.

The data from spin therefore differ from highwire in that there is a muscle contribution to growth in spin. This identifies a novel AP-1 and autophagy contribution to synaptic growth in the muscle. How autophagy is regulating synapse growth in spin remains unclear; what we have failed to do so far, is to image autophagy at the synapse and we await more markers that might illuminate this process. What our data can suggest is that there may be a vicious cycle in LSDs where oxidative stress induces JNK-AP-1 signaling and autophagy, generating an added burden on the endosomal system. This in turn could induce more oxidative stress.

Taken together, our findings implicate oxidative stress as an activator of the JNK-AP-1 signaling pathway and autophagy, known mediators of synaptic growth and function, and provide potential mechanistic insight (easily tested) into synaptic growth in LSD. There may be broader implications for synaptic growth. Metabolic demand generates ROS through activation of mitochondrial activity. The JNK-AP-1 pathway has been known as a mediator of learning and memory processes for many years though the transcriptional targets remain obscure. Might antioxidant activities contribute to synaptic plasticity?

Punctum to: Milton VJ, Jarrett HE, Gowers K, Chalak S, Briggs L, Robinson IM, et al. Oxidative stress induces overgrowth of the Drosophila neuromuscular junction. Proc Natl Acad Sci USA. 2011;108:17521–17526. doi: 10.1073/pnas.1014511108.