Neurite outgrowth in normal and injured primary sensory neurons reveals different regulation by nerve growth factor (NGF) and artemin

Highlights

• We compare artemin and nerve growth factor regulation of in vitro neurite outgrowth.

• Both facilitate neurite initiation in adult rat dorsal root ganglia sensory neurons.

• Artemin has weak efficacy for neurite elongation.

• Somatic and visceral sensory neuron respond differently to in vivo nerve injury.

• Regulation of neurite outgrowth by artemin is more selective than NGF.

Abstract

Neurotrophic factors have been intensively studied as potential therapeutic agents for promoting neural regeneration and functional recovery after nerve injury. Artemin is a member of the glial cell line-derived neurotrophic factor (GDNF) family of ligands (GFLs) that forms a signalling complex with GFRα3 and the tyrosine kinase Ret. Systemic administration of artemin in rodents is reported to facilitate regeneration of primary sensory neurons following axotomy, improve recovery of sensory function, and reduce sensory hypersensitivity that is a cause of pain. However, the biological mechanisms that underlie these effects are mostly unknown. This study has investigated the biological significance of the colocalisation of GFRα3 with TrkA (neurotrophin receptor for nerve growth factor [NGF]) in the peptidergic type of unmyelinated (C-fibre) sensory neurons in rat dorsal root ganglia (DRG). In vitro neurite outgrowth assays were used to study the effects of artemin and NGF by comparing DRG neurons that were previously uninjured, or were axotomised in vivo by transecting a visceral or somatic peripheral nerve. We found that artemin could facilitate neurite initiation but in comparison to NGF had low efficacy for facilitating neurite elongation and branching. This low efficacy was not increased when a preconditioning in vivo nerve injury was used to induce a pro-regenerative state. Neurite initiation was unaffected by artemin when PI3 kinase and Src family kinase signalling were blocked, but NGF had a reduced effect.

Introduction

Injuries to the peripheral axons of primary afferent neurons are a cause of sensory deficits and other neurological symptoms, such as chronic pain, that require clinical treatment (Baron et al., 2010, von Hehn et al., 2012). Common causes include accidental trauma, surgery, drug therapy (e.g. anti-neoplastic cancer drugs) and metabolic dysfunction (e.g. diabetes). Several obstacles need to be addressed to develop new clinically useful therapies for peripheral sensory nerve damage (Christie and Zochodne, 2013, Scheib and Höke, 2013). The most basic requirement is to promote the regeneration of injured axons without excessive aberrant outgrowth. However, new axons must not only be correctly targeted but also need to recover their function. Recent evidence suggests that function may only return if re-innervation occurs within a critical period, after which functional recovery can fail (Ma et al., 2011). This has identified a need to accelerate regeneration so that re-innervation can be achieved within this optimal time window. Finally, new therapies must not themselves cause dysfunction of uninjured neurons in either the sensory or motor systems.

Neurotrophic factors have been extensively researched as candidate treatments for peripheral nerve injury. Two major classes of these endogenous regulators target unmyelinated (C-fibre) primary afferent neurons. Different types of C-fibre sensory neurons in dorsal root ganglia (DRG) are delineated by expression of receptors for neurotrophins and glial cell line-derived neurotrophic factor (GDNF) family ligands (GFLs) (Ernfors and Lallemend, 2012). During development, these receptors are crucial determinants of survival, peripheral innervation of appropriate targets, and functional characteristics sculpted by the expression of ion channels, receptors and molecular properties of the neurons.

Most C-fibre sensory neurons express TrkA and/or Ret (Ernfors and Lallemend, 2012, Molliver et al., 1995, Molliver et al., 1997). TrkA is a high affinity receptor for the prototypic neurotrophin NGF (nerve growth factor), whereas Ret serves as a signalling co-receptor for the GFLs (Airaksinen and Saarma, 2002). NGF has been studied intensively in the context of peripheral nerve regeneration but it has properties that could lead to side effects when used as a therapeutic. Of particular significance is the function of NGF in sensitisation of nociceptive C-fibre sensory neurons, as this is a cause of acute and chronic pain (Lewin et al., 1993, Pezet and McMahon, 2006). This property is clinically significant as demonstrated by clinical trials evaluating NGF-sequestering antibodies as analgesic therapies (Sanga et al., 2013, Wang et al., 2014a). The search for alternatives to NGF has identified artemin, a member of the GFL family, as another prototypic drug candidate for promoting sensory regeneration (Gardell et al., 2003, Wang et al., 2008, Wang et al., 2014b). Mammalian primary afferent neurons are functionally regulated by three GFLs—GDNF, neurturin and artemin (Ernsberger, 2008). Each has a ligand binding co-receptor (GFRα1, GFRα2 and GFRα3, respectively) that is required for signalling via Ret (Airaksinen and Saarma, 2002). In situ hybridization and immunohistochemical studies that have localised co-localised the GFRα subtypes in DRG neurons suggest that GDNF and neurturin target non-peptidergic C-fibre sensory neurons, which do not express TrkA and are insensitive to NGF (Bennett et al., 1998, Bennett et al., 2000). By contrast, GFRα3 localises to some peptidergic sensory neurons, which suggests there is a major type of C-fibre DRG neuron that is co-regulated by both NGF and artemin (Ernsberger, 2008, Forrest and Keast, 2008, Kalous et al., 2009, Keast et al., 2010, Orozco et al., 2001).

In vivo administration of artemin is reported to have beneficial effects in rodent nerve injury models—with injuries of both peripherally and centrally projecting sensory axons having been studied (Gardell et al., 2003, Wang et al., 2008, Wang et al., 2014b). The positive effects of artemin include enhanced regeneration, recovery of sensory functions, and reversal of hyperalgesia used to measure pain. However, relatively little is known about the mechanisms underlying artemin-induced regeneration of injured sensory axons. To begin addressing this, we have used well-established and extensively characterised in vitro assays to compare the effects of NGF and artemin on neurite outgrowth in adult rat DRG sensory neurons. As the artemin receptor GFRα3 and the NGF receptors are co-expressed primarily in peptidergic unmyelinated (C-fibre) sensory neurons, we have focused our study on this neuron type. Finally, to determine the effects of artemin and NGF on sensory neurons previously injured by axotomy, we have compared the impact of nerve transection on visceral (bladder projecting) and somatic (sciatic nerve projecting) sensory neurons. While many studies have examined the impact of injury on somatic afferents, relatively little is known about how visceral afferents respond to injury or if injury affects their responsiveness to neurotrophic factors. We chose bladder afferents to investigate this issue as they include many that express receptors for both artemin and NGF (Forrest and Keast, 2008, Forrest et al., 2013). In each case, the injury was performed unilaterally to enable direct comparison with the contralateral side of the same animal.