Paclitaxel is a microtubule-stabilizing chemotherapeutic agent that is widely used in cancer treatment and in a number of curative and palliative regimens. Despite its beneficial effects on cancer, paclitaxel also damages healthy tissues, most prominently the peripheral sensory nervous system. The mechanisms leading to paclitaxelinduced peripheral neuropathy remain elusive, and therapies that prevent or alleviate this condition are not available. We established a zebrafish in vivo model to study the underlying mechanisms and to identify pharmacological agents that may be developed into therapeutics. Both adult and larval zebrafish displayed signs of paclitaxel neurotoxicity, including sensory axon degeneration and the loss of touch response in the distal caudal fin. Intriguingly, studies in zebrafish larvae showed that paclitaxel rapidly promotes epithelial damage and decreased mechanical stress resistance of the skin before induction of axon degeneration. Moreover, injured paclitaxel-treated zebrafish skin and scratch-wounded human keratinocytes (HEK001) display reduced healing capacity. Epithelial damage correlated with rapid accumulation of fluorescein-conjugated paclitaxel in epidermal basal keratinocytes, but not axons, and upregulation of matrix-metalloproteinase 13 (MMP-13, collagenase 3) in the skin. Pharmacological inhibition of MMP-13, in contrast, largely rescued paclitaxel-induced epithelial damage and neurotoxicity, whereas MMP-13 overexpression in zebrafish embryos rendered the skin vulnerable to injury under mechanical stress conditions. Thus, our studies provide evidence that the epidermis plays a critical role in this condition, and we provide a previously unidentified candidate for therapeutic interventions.aclitaxel is a microtubule-stabilizing chemotherapeutic agent that is widely used in the treatment of common cancers, including breast, ovarian, and lung cancer. Despite its promising anticancerous properties, paclitaxel also damages healthy tissues, most prominently peripheral axons of somatosensory neurons (reviewed in ref. 1). Paclitaxel-induced peripheral neuropathy initiates in the distal extremities and presents as neuropathic pain syndrome, temperature sensitivity, and paresthesia (tingling and numbness). Nerve biopsies from patients suggest that axon degeneration is the primary manifestation of this condition, followed by secondary demyelination and nerve fiber loss in severely affected patients (1, 2). Certain drugs have been shown in vitro and in vivo to protect against paclitaxel-induced nerve damage, including acetyl-L-carnitine, erythropoietin, alpha-lipoic acid, olesoxime, amifostine, nerve growth factor, and glutamate (reviewed in ref.3). However, so far, these agents have either not successfully passed clinical trials or merely alleviate symptoms such as pain without prevention (1). Thus, a better understanding of the underlying causes of paclitaxel-induced peripheral neuropathy is necessary and may help identify new candidate drugs with which to treat this condition.A widely...

Paclitaxel induces peripheral neuropathy as a side effect of cancer treatment. The underlying causes are unclear, but epidermal, unmyelinated axons have been shown to be the first to degenerate. We previously utilized an in vivo zebrafish model to show that the epidermal matrix-metalloproteinase 13 (MMP-13) induces degeneration of unmyelinated axons, whereas pharmacological inhibition of MMP-13 prevented axon degeneration. However, the precise functions by which MMP-13 is regulated and affects axons remained elusive. In this study, we assessed mitochondrial damage and reactive oxygen species (ROS) formation as possible inducers of MMP-13, and we analyzed MMP-13-dependent damage. We show that the small ROS, H2O2, is increased in basal keratinocytes following treatment with paclitaxel. Cytoplasmic H2O2 appears to derive, at least in part, from mitochondrial damage, leading to upregulation of MMP-13, which in turn underlies increased epidermal extracellular matrix degradation. Intriguingly, also axonal mitochondria show signs of damage, such as fusion/fission defects and vacuolation, but axons do not show increased levels of H2O2. Since MMP-13 inhibition prevents axon degeneration but does not prevent mitochondrial vacuolation, we suggest that vacuolization occurs independently of axonal damage. Finally, we show that MMP-13 dysregulation also underlies paclitaxel-induced peripheral neuropathy in mammals, indicating that epidermal mitochondrial H2O2 and its effectors could be targeted for therapeutic interventions.

We demonstrate that zebrafish are suitable to study glucose-induced neurotoxicity. Given the effects in zebrafish and mice, MMP-13 inhibition may be beneficial in the treatment of human diabetic neuropathy.

Paclitaxel is a chemotherapeutic agent that causes peripheral neuropathy (PIPN) as a side effect of cancer treatment. Severely affected patients need to terminate chemotherapy, diminishing their chance of survival. The underlying causes of PIPN are unclear, but epidermal, unmyelinated axons have been shown to be the first to degenerate. We previously utilized a zebrafish in vivo model to show that the epidermal matrix-metalloproteinase 13 induces degeneration of unmyelinated axons, whereas pharmacological inhibition of MMP-13 prevented axon degeneration. The precise functions by which MMP-13 is regulated and affects axons, however, remained elusive. In this study, we assessed mitochondrial damage and reactive oxygen species (ROS) formation as possible inducers of MMP-13, and we analyzed MMP-13-dependent damage. We show that the small ROS, H 2 O 2 , is increased in keratinocytes following treatment with paclitaxel. Epidermal mitochondrial damage appears to be a source of ROS leading to cytoplasmic H 2 O 2 elevation, upregulation of MMP-13, and increased matrix degradation. Intriguingly, although axonal mitochondria also show aberrant morphologies and are vacuolized, as shown in other neuropathies, these axonal mitochondria do not produce increased H 2 O 2 levels. We suggest that mitochondrial vacuolization occurs independently of axonal damage given that MMP-13 inhibition prevents axon degeneration, though vacuoles persist. We further show that MMP-13 dysregulation also underlies PIPN in rodent paclitaxel models, and that this function appears to be DRG neuron-extrinsic. These findings suggest that vacuolization is not a cause of paclitaxel-induced neuropathy, and that epidermal MMP-13 is a strong candidate for therapeutic interventions in cancer patients with neuropathy. 6 were obtained for von Frey (tactile), Hargreaves (thermal) and cold plate or acetone assays, as described below. Plasmids and transgenic linesUAS-mito-dsRed fish (gift from Dr. Alvaro Sagasti (UCLA)) were injected with krt4:Gal4VP16 to label skin mitochondria and CREST3:Gal4VP16_14xUAS-GFP to label sensory neurons.krt4:HyPer-cyto was previously published previously (1). tp63:HyPer was generated by amplifying HyPer-cyto from the original Evrogen plasmid with 5' BamHI and 3' NotI sites via PCR. The HyPer amplicon was inserted into tp63:AcGFP (gift of Dr. Gromoslav Smolen) after AcGFP was removed via BamH/NotI digestion. CREST3:Gal4VP16_14x-HyPer-cyto was cloned using the Gateway system (Life Technologies). pDONR vectors for CREST3 and Gal4VP16-14xUAS were a gift from Dr. Alvaro Sagasti (UCLA). HyPer-cyto was amplified via PCR with attB primers to insert as 3' element into the pDEST vector together with CREST3 and Gal4VP16-14xUAS using recombination. Transgenic isl2b:GFP fish (2) were used for axon degeneration studies. Drug administrationsZebrafish treatments: Starting at 2dpf, zebrafish were placed individually into each well of 24well plates. The control groups were typically incubated in 0.09% DMSO/Ringers solution to match the DMSO concentra...