I. Role of a class of neurotoxic sphingolipids, the deoxysphingolipids, in taxane-induced peripheral neuropathy.
Taxanes are a class of widely-used chemotherapy drugs; however, more than 50% of patients can experience peripheral neuropathy as a side effect, resulting in dose reduction or treatment termination. The molecular mechanism of peripheral neuropathy is not understood, and there are currently no prevention or treatment strategies available. In a pilot study with breast cancer patients receiving standard paclitaxel chemotherapy, we tested whether a class of neurotoxic lipids, deoxysphingolipids, were involved in the development of peripheral neuropathy. Deoxysphingolipids are produced when the first enzyme of the sphingolipid biosynthetic pathway, serine palmitoyltransferase, utilizes L-alanine instead of L-serine as its amino acid substrate Deoxysphingolipids were previously shown to be elevated in the plasma of those with hereditary sensory and autonomic neuropathy type I. Importantly, they have additionally been shown to be neurotoxic in vitro by us and others to and in human when the deoxysphingolipid, deoxysphinganine (ES-285), was tested as an anticancer drug. Indeed, ES-285 phase I clinical trials were terminated because patients developed severe, and in some cases fatal, neuropathy. Our study with breast cancer patients receiving standard paclitaxel chemotherapy revealed that the incidence and severity of neuropathy was associated with plasma deoxysphingolipid levels. Currently, we are investigating the molecular mechanisms of deoxysphingolipid neurotoxicity in vitro and in mouse models of taxane-induced peripheral neuropathy. Our goal is to develop strategies to address deoxysphingolipid neurotoxicity.

II. Role of de novo ceramide synthesis in neuronal homeostasis.
In mammals, ceramide biosynthesis is catalyzed by six isoforms of ceramide synthase (CerS). Each CerS isoform has a preference for a different set of CoA-activated fatty acids according to the length of their acyl chain. A sphingoid base, together with a CoA-activated fatty acid, are the two substrates for CerS enzymes.
Data from our two unique mouse models provides compelling evidence that deficiency in ceramide biosynthesis can cause neurodegeneration. Our first mouse model is based on a point mutation in the CerS1 isoform, resulting in its catalytic inactivation, the elevation of its sphingoid base substrates, the reduction of its product (18 carbon chain ceramide), and neurodegeneration. CerS1 is the main neuronal ceramide synthase isoform. The second mouse model is based on ectopic expression in the CerS1-deficient neurons of a different ceramide synthase isoform, CerS2. In this model, the CerS2 expression reduces the levels of sphingoid bases in the brain to wild type levels while rescuing the neurodegeneration phenotype. Of note, CerS1 and CerS2 isoforms share the same sphingoid base substrates, but use different fatty acid acyl-CoAs, which produces ceramides with different fatty acyl moieties.
Using these two unique mouse models, we are investigating whether high sphingoid base levels in the brain can impair the homeostasis of membrane organelles, causing neurodegeneration. Our project aims at providing important insights for understanding the pathophysiological roles of sphingolipid metabolites, such as sphingoid bases in the nervous system. It is our goal to then use this knowledge for developing sphingolipid metabolism-based strategies to reduce neurodegeneration.