Cisplatin induced alterations in nociceptor developmental trajectory elicits a TrkA dependent platinum-based chemotherapy induced neuropathic pain

51 Cisplatin-based chemotherapy is a common treatment for paediatric cancer. Unfortunately, 52 cisplatin treatment causes neuropathic pain, a highly prevalent adverse health related complication 53 in adult childhood cancer survivors. Due to minimal understanding of this condition, there are 54 currently no condition tailored analgesics available. Here we investigated an alteration in 55 nociceptor maturation that results in neuronal sensitisation and manifestation of cisplatin induced 56 survivorship pain in a TrkA dependent manner. Cisplatin was administered (i.p. 0.1mg/kg Postnatal 57 day 14 and 16) to neonatal male and female Wistar rats and nociceptive behavioural assays were 58 performed. In vitro studies utilised isolated neonatal dorsal root ganglia sensory neurons treated 59 with cisplatin (5μg/ml) to elucidate impact upon nociceptor activation and neurite growth, in 60 combination with TrkA inhibition (GW441756 10nM and 100nM). Cisplatin treated male and female 61 neonatal Wistar rats developed a delayed but lasting mechanical and heat hypersensitivity. 62 Cisplatin administration led to increased TrkA expression in dorsal root ganglia sensory neurons. 63 Nerve growth factor (NGF) induced TrkA activation led to sensory neuritogenesis and nociceptor 64 sensitisation, which could be prevented through pharmacological TrkA inhibition (GW441756


Introduction
Advances in paediatric cancer diagnosis and treatment, has led to an increase in patient life expectancy (Yeh et al., 2020).Unfortunately, despite these clinical successes, platinum-based chemotherapy (e.g.cisplatin) impacts the quality of a patients survivorship (Karlson et al., 2020;Moke et al., 2021) due to the development of adverse healthrelated complications, with cisplatin causing nephropathy (Jimenez-Triana et al., 2015), ototoxicity (Brock et al., 2018) and sensory neuropathy (Ness et al., 2013(Ness et al., , 2017)).In adult survivors of childhood cancer, treatment induced neuropathic pain is common (Lu et al., 2011;Alberts et al., 2018) and increasingly prevalent with age (Phillips et al., 2015), many years post diagnosis and cessation of treatment (Lu et al., 2011;Ness et al., 2013).Presence of pain is a risk to preventing ongoing cancer treatment.Furthermore, the development of neuropathic pain states is accompanied by diminished quality of life, with dependence upon analgesia and decline in mental health (Lu et al., 2011).Despite increased societal and economic burden through ongoing healthcare dependence and reductions in employment (Lu et al., 2011), there are currently few available analgesics that provide pain relief arising following platinum-based cancer treatment early in life.
Highly traumatic/noxious insults early in life lead to alterations in the ability of the sensory nervous system to process sensory information resulting in chronic pain (Schwaller and Fitzgerald, 2014).The nociceptive system is still developing during infancy (Fitzgerald, 2005), with the developmental trajectory of the nociceptor susceptible to early life 'stressors' (Schwaller and Fitzgerald, 2014).Fundamental molecular mechanisms during early life greatly influences the developmental lineage of nociceptors (Marmigere and Ernfors, 2007) and controls the hardwiring of the 'pain circuit' between the peripheral and central somatosensory nervous systems (Chen et al., 2006).However, alterations in these molecular developmental signals alters the timecourse of nociceptor maturation.In relation to platinum-based chemotherapy, there is limited understanding of the sensory neuronal maladaptations that occur in response to a chemical insult.Peripheral tissues are susceptible to platinum-based agents, with cisplatin accumulating in the peripheral somatosensory nervous system post administration (McDonald et al., 2005;Ta et al., 2006).Initial exploration has led to the development of pre-clinical experimental rodent models of adult childhood cancer survivorship pain following treatment with chemotherapy during infancy (Schappacher et al., 2017(Schappacher et al., , 2019)), highlighting alterations in the maturation of nociceptors post cisplatin treatment (Hathway et al., 2017).Despite this developing platform for investigating this clinical scenario, it remains that there is minimal understanding of the impact platinumbased chemotherapy has upon the developing somatosensory nervous system.
Our previous work (Hathway et al., 2017) demonstrated increased Tropomyosin Receptor Kinase A (TrkA) abundance in dorsal root ganglia (DRG) sensory neurons.However, there have been few investigations exploring the putative mechanisms that cause chemotherapy-induced long-lasting pain in adult childhood cancer survivors.This study explores how cisplatin impacts upon the nociceptor maturation, resulting in the manifestation of a TrkA dependent long lasting pain state, whilst also highlighting the necessity in elucidating the mechanisms behind paediatric chemotherapy-induced long-lasting pain, enabling the design of novel therapeutic strategies.

Ethical approval
All experimental procedures involving Animals were carried out in consultation with local Animal Welfare and Ethics Review Board (University of Nottingham and Nottingham Trent University) and in accordance with UK Home office animals (Scientific procedures) Act 1986 and ARRIVE guidelines.Animals had ad libitum access to standard chow and were housed under 12:12 h light:dark conditions.Wistar rats were housed in environmentally enriched cages with the mother until weaning (postnatal day 22), at which point rodents were single sex housed (typically housed as 5 rodents per cage) and were randomly allocated to each cage as well as treatment.

Induction of Cisplatin induced neuropathic pain and administration of pharmacological agents
Males and female Wistar rats (Charles River, total 108 rats were used; Vehicle (cisplatin) = 33, Cisplatin = 32, Cisplatin + TrkA antagonist = 12, Vehicle (NGF) = 13, NGF=14, NGF+TrkA antagonist = 14) were utilised in all outlined studies.Animals were randomly allocated, with individual rodents from each experimental group allocated to each litter.Neonatal Wistar rats were administered either vehicle (phosphate buffered saline; PBS) or cisplatin (Sigma-Aldrich; 0.1 mg/kg) via intraperitoneal injection delivered on 2 occasions; postnatal (p) day 14 and 16.All rodents were weaned no later than 22 days and group housed according to sex.Body weight was monitored regularly throughout the study. 1 μM Nerve Growth Factor 2.5 s (NGF, Alomone) was administered via a 20 μl subcutaneous injection under recovery anaesthesia (~2% isoflurane in O 2 ).In some instances, NGF was co-administered alongside TrkA antagonist GW441756 as a subcutaneous injection (100 nM) or as a single intraperitoneal injection or intraperitoneal injection on 4 consecutive days starting at P21 (2 mg/kg (Coelho et al., 2015)).Animals were identified via tail markers and/or ear notch.Disruption to the housing cage was minimised to prevent disruption to the mother pre weaning.

Nociceptive Behaviour
All rodents were habituated to the testing environments before nociceptive behavioural experimentation (Drake et al., 2021;Da Vitoria Lobo et al., 2022).This involved rodents becoming acclimatised to the experimenter (via handling) and the testing environment via placing the rodents in the testing enclosures for habituation periods prior to testing sessions (10 min).Removal from the litter and mother pre weaning was minimised.Mechanical withdrawal thresholds were scored following application of von Frey (vF) hairs to the plantar surface of the hindpaw.This was calculated following application of differing vF filaments of increasing force a total of five times.Force response curves were generated and mechanical withdrawal thresholds were determined as the mechanically applied force to elicit 50 % of nociceptive withdrawals.The Hargreaves test was performed to determine heat nociceptive withdrawal latency (Hargreaves et al., 1988).A radiant heat source was applied to the plantar surface of the hindpaw.The duration (latency) between onset of stimulus to the rat withdrawing their paws was recorded as the withdrawal latency (manual cut off was set at 20 s).This was measured 3 times and a mean latency was calculated for both hindpaws.

Primary dorsal root ganglia sensory neuronal cell culture
Prior to tissue collection, plates were coated with poly-L-lysine (0.01 % Sigma Aldrich) overnight at 37 • C. Wells were subsequently washed with PBS and left to dry for 1 h.Each well was further coated with 1 μg/ mL laminin (Sigma-Aldrich) incubating at 37 • C for 1hr.Postnatal day 7 C57BL/6J mice were euthanised via cervical dislocation, all DRGs were collected, pooled together (from male and female) and placed in media.DRG neuronal cell culture media consisted of Ham's/F 2 nutrient mix with 0.5 mL of 3 % bovine serum albumin, 1 mL of penicillin/streptomycin, and N-2 supplement.DRGs were incubated in 0.1 % collagenase type IX for 1 h at 37 • C; 5 % CO 2 .DRGs were mechanically dissociated via trituration.The cell suspension was added to a 15 % BSA filtration column (1 mL Ham's F12 media + 1 mL 30 % BSA) and centrifuged at 200g for 8 min.Cell pellets were resuspended in media and a cell count was performed using a C-Chip haemocytometer.DRG neurons were plated out as 2000 cells per well and incubated at 37  Tocris).Cisplatin was applied at 5 μg/ml in all in vitro experimentation (as previously described (Vencappa et al., 2015)) for 24hrs before media was refreshed with new media without cisplatin.7 days later experiments were treated with NGF2.5S and TrkA antagonist GW441756 for 24hrs as outlined above.

Immunocytochemistry for DRG neuronal growth assays
Coverslips were washed in PBS, prior to incubation in 1 % PFA for 15 min at room temperature.Neurons were permeabilised with 0.2 % Triton X-100 in PBS for 15 min and subsequently incubated in blocking solution (5 % bovine serum albumin, 0.2 % Triton-X) for 30 min at room temperature.Coverslips were incubated in primary antibody (Table 1) for identification of neurons (Rabbit anti-Beta III Tubulin; 1:500, Abcam) overnight at 4 O C. Cells were subsequently washed in PBS and secondary antibodies (Alexa Fluor 555-conjugated donkey anti-rabbit) were added for 2hrs at room temperature.Coverslips were washed and mounted onto slides using Vector Shield (H-1000, Vector Labs).

DRG sensory neuronal calcium assay
The primary DRG sensory neuronal calcium assay was performed as previously described (Austin et al., 2017).Each well (of 96 well plate) containing dorsal root ganglia neurons were incubated in 40 μl cell culture media and 40 μL of Fluo-4 direct (Thermofisher) for 1 h at 37 • C; 5 % CO 2 .Baseline intracellular calcium recordings were performed at 37 O C and acquired via an Infinite M Plex plate reader.Following baseline recordings 20 μl of TRPV1 agonist capsaicin (Sigma-Aldrich) prepared at a stock concentration of 5 μM was added.This equates to a final capsaicin concentration as 1 μM.Capsaicin evoked intracellular calcium was recorded every 10 s for 150 s.Capsaicin evoked intracellular calcium responses as a fold change over baseline.

QPCR
RNA was extracted using TRIzol reagent (Thermofisher) from a sensory neuronal cell line, 50B11, which were cultured as described and incubated in either vehicle or cisplatin for 24hrs (Chen et al., 2007).RNA was reverse transcribed to cDNA using PrimeScript TM RT (TaKaRa).Quantitative PCR was performed (LightCycler 480 SYBR Green I Mastermix; Roche) using primers as outlined in Table 3.

Statistical analysis
All data are represented as mean ± SEM unless stated.Sample sizes were based upon our previous experience and appropriate sample size calculations.No animals were excluded from this study.Data were acquired and quantified using Microsoft Excel 2010, Image J (https://i magej.nih.gov/ij/)(Schneider et al., 2012) and Graphpad Prism 8.For primary DRG sensory neuronal cell cultures 5 mice were used per isolation to provide sufficient neuronal number to perform the assay per plate.Experimental number consisted of a minimum of 3 independent experimental plates, with a minimum of 8 technical repeats performed within each experimental plate.Intracellular Calcium responses were obtained and background fluorescence was removed.Capsaicin evoked calcium responses were then normalised and determined as fold change by dividing the fluorescence value obtained prior to capsaicin as baseline.Capsaicin evoked intracellular calcium responses were analysed using either a one way to make comparisons between area under curve or two-way ANOVA to determine the treatment effect over time.Immunofluorescence images of dorsal root ganglia (from L5 DRG extracted) and plantar skin − 5 random non-sequential sections (Z stacks) were collected and processed per animal from 5 rodents per experimental group.Only ipsilateral or right hindpaw (excluding the footpad) were processed for IENF immunohistochemistry. 5 random non-sequential sections (Z stacks) were collected and processed per animal, with average number of IENF counted per ROI and normalised to the known length of the dermal/epidermal border.This was determined per animal.In addition, sequential sections were collected on sequential slides, no sequential slides were used for immunohistochemistry. Additionally, a minimum of 6 sections of either DRGs cut to 8 μm or Plantar skin 20 μm were collected per slide, with a minimum 108 μm and 360 μm between selected sections.Mean values were compared across experimental groups utilising a one way ANOVA with post Bonferroni post hoc test.Percentage of neurons were determined as a number of neurons of interest (Runx1 positive) against total neuron number (NeuN positive).Comparisons were made using Unpaired T Tests.Membrane TrkA intensity was measured by using ImageJ to plot a plot profile.Using the line tool a line was drawn across the DRG neuronal membrane to produce a region of interest.Immunofluorescence intensity was measured and intensity was plotted across the length of this region of interest.A Two Way ANOVA comparison was performed.Neurite outgrowth was measured using ImageJ using free hand tool to measure length of neurites and number of neurites per neuronal cell body.Western blots, densitometry analysis was performed using ImageJ plugin and comparisons made using Unpaired T test or One way ANOVA with post Bonferroni post hoc test.Nociceptive behavioural

Table 1
List of Primary Antibodies for Immunofluorescence.

Results
A rodent model of cisplatin induced neuropathic pain led to a pronounced sensory polyneuropathy (Fig. 1).Neonatal administration of cisplatin at P14 and P16 led to reductions in mechanical nociceptive withdrawal threshold (Fig. 1A) and decreased heat nociceptive withdrawal latency (Fig. 1B) in both hindpaws when compared to age matched sham control rodents.In addition, comparisons of nociceptive behavioural phenotypes were conserved across male and female, with both mechanical (Fig. 1C) and heat (Fig. 1D) hyperalgesia present in cisplatin experimental groups for both sexes.Tropomyosin receptor kinase A (TrkA) was expressed in lumbar dorsal root ganglia (DRG) sensory neurons taken from vehicle and cisplatin treated groups (Fig. 1E).When evaluating TrkA immunoreactivity, immunofluorescence intensity was measured across the sensory neuronal membrane to indicate the expression abundance of TrkA (Fig. 1F).TrkA expression was increased in the cisplatin DRG sensory neurons (P45) when compared to vehicle treated age matched control rodents (Fig. 1G).To determine the impact of nociceptor developmental trajectory, markers of sensory neuronal development and maturation were investigated following cisplatin treatment.A sensory neuronal cell line, 50B11, was treated with either cisplatin or vehicle for 24hrs led to elevated expression of TrkA (Fig. 2A; representative westerns blots for TrkA and RUNX1; Fig. 2B TrkA) and reductions in runt related transcription factor 1 (RUNX1; Fig. 2C).Reductions were similarly observed in runx1 mRNA levels in cisplatin treated 50B11 neuronal cell line (Fig. 2D).Furthermore, immunoreactivity of RUNX1 in lumbar DRG sensory neurons extracted from vehicle and cisplatin animals demonstrated a reduction in the number of neurons expressing RUNX1 at P16 (Fig. 2E; representative images of RUNX1 immunoreactivity in DRG sensory neurons labelled with NeuN at P16 and P45; Fig. 2F).Body weight does not differ following cisplatin administration to vehicle treated age matched vehicle controls (Fig. 2G).Cisplatin treatment induces mechanical (Fig. 2H) and heat hypersensitivity (Fig. 2I), which can be prevented following 4 consecutive days of intraperitoneal administration of TrkA antagonist from day 21.
DRG sensory neurons treated with NGF for 24 h led to increased sensory neuritogenesis when compared to vehicle treated neurons (Fig. 3A=average length, 3B=number of neurites, 3C=maximum neurite length & 3D=total neurite length per neuron).This NGF induced sensory neurite growth was inhibited by a TrkA antagonist (GW441756) in a dose dependent manner (Fig. 3A, 3B, 3C & 3D).Representative images of DRG sensory neurons following 24 h treatment with either vehicle (Fig. 3E), NGF+Veh (Fig. 3F), NGF+10 nM GW441756 (Fig. 3G) and NGF+100 nM GW441756 (Fig. 3H) demonstrating suppression of NGF induced neurite outgrowth following TrkA inhibition.NGF led to sensitisation of nociceptor DRG sensory neuron activity through elevating intracellular calcium fluorescence following stimulation with TRPV1 agonist Capsaicin (Fig. 4A).Co-treatment with TrkA antagonist GW441756 led to reductions in NGF induced sensitisation of TRPV1 activity (Fig. 4A), with inhibition occurring in a dose dependent manner (Fig. 4B).NGF led to increased phosphorylation of TrkA, which was inhibited with the TrkA antagonist, GW441756 (representative TrkA and pTrkA western blots; Fig. 4C, D & E).Subcutaneous administration in the plantar surface of the hindpaw of NGF co-treated with TrkA antagonist GW441756 prevented NGF induced mechanical (Fig. 4F) and heat (Fig. 4G) hypersensitivity in the ipsilateral hindpaw.Furthermore, alterations in the contralateral hindpaw were observed with reductions in withdrawal thresholds to mechanical stimulation (Fig. 4H), but no change in relation to heat (Fig. 4I) stimuli.In addition, intraperitoneal delivery of GW441756 prevented peripherally administered NGF induced mechanical (Fig. 5A) and heat (Fig. 5B) hypersensitivity in the ipsilateral hindpaw.In the contralateral hindpaw, mechanical (Fig. 5C) nociceptive behavioural phenotypes were reduced in NGF group, though no change was observed in response heat stimuli (Fig. 5D).TrkA is expressed upon small diameter nociceptor DRG neurons (CGRP positive neurons).Peripheral delivery of NGF into the plantar surface of the hindpaw led to increased CGRP IENF in the ipsilateral plantar skin (representative CGRP positive IENF in plantar skin Fig. 6A) versus vehicle treated group (Fig. 6B&C).NGF delivered inconjunction with TrkA antagonist, GW441756, delivered via intraperitoneal injection, prevented NGF induced CGRP aberrant IENF growth in the plantar skin (Fig. 6B&C).Furthermore, GW441756 prevented NGF induced aberrant growth of pan neuronal PGP9.5 IENF (Fig. 6D-F).
DRG sensory neurons were treated with cisplatin for 24hrs.Subsequently, neurons were washed with fresh media that included the presence of NGF and were kept for a further 7 days (Work flow of induction of cisplatin induced nociceptor sensitisation in vitro Fig. 7A).TRPV1 activity was measured in nociceptors in response to increasing concentrations of the TRPV1 agonist, capsaicin led to a pronounced increase in intracellular calcium (Fig. 7B & C).As previously shown (Hulse et al., 2014;Bestall et al., 2018), 1 μM capsaicin demonstrated a significant increase in intracellular calcium in DRG neurons and was used for all subsequent experiments.DRG neurons treated with cisplatin demonstrated a pronounced increase in TRPV1 activity versus vehicle treated DRG sensory neurons Fig. 7D & E).Additionally, inhibition of TrkA activity through treatment with GW 441756 led to suppression of cisplatin induced TRPV1 enhanced activity, in a dose dependent manner (Fig. 7F-G).Intraperitoneal administration of GW441756 in a rodent model of neonatal cisplatin induced neuropathic pain led to the inhibition of mechanical (Fig. 8A) and heat (Fig. 8B) hyperalgesia.Additionally, intraperitoneal administration of GW441756 ameliorated cisplatin induced mechanical hyperalgesia in both male (Fig. 8C) and female (Fig. 8D) rodents.Rodents in all experimental groups did not display any alterations in body weight (Fig. 8E).Previous investigation has identified that in the neonatal model of cisplatin induced survivorship pain (Hathway et al., 2017), pain is accompanied by aberrant

Discussion
With increased life expectancy following childhood cancer diagnosis, importance has now been placed upon the quality of survivorship.Long lasting neuropathic pain in these individuals has been highlighted to arise due to cisplatin treatment.However, understanding the mechanisms that underlie cisplatin induced survivorship pain is lacking, which is exemplified by the lack of effective analgesia.In this study we have identified a modulation of nociceptor maturation following exposure to cisplatin early in life, a mechanism that underpins TrkA dependent nociceptor sensitisation and adult childhood cancer survivorship pain.

Cisplatin early in life leads to neuropathic pain in adulthood in females and males
Advances in childhood cancer research has led to increased understanding and early diagnosis of cancer, with large proportions of paediatric patients surviving 10 years post diagnosis.However, due to the principle cytotoxic nature of chemotherapeutic agents that compromises a multitude of physiological systems, a number of adverse health related complications develop.Consequently, majority of these patients treated with chemotherapy live with therapy associated side-effects for many years after treatment has stopped (Geenen et al., 2007).Adult childhood cancer survivors highlight pain as a side-effect of treatment (Lu et al., 2011;Ness et al., 2013;Kandula et al., 2018), with platinum-based chemotherapy induced sensory neurodegeneration leading to long-term alterations in sensory perception in adult childhood cancer survivors (Solheim et al., 2019;Kandula et al., 2020).Our work here using a rodent model of childhood cancer survivorship pain, demonstrates early life exposure to cisplatin results in a delayed but lasting pain that presents and remains into adulthood.This is typical of the human clinical scenario, whereby pain manifests many years post diagnosis and after the treatment has ended (>7yrs) (Glendenning et al., 2010).In addition, neuropathic pain presents typically during adolescence, becoming increasingly prevalent with age (Lu et al., 2011;Ness et al., 2013;Khan et al., 2014;Phillips et al., 2015).Furthermore, the cisplatin induced symmetrical sensory polyneuropathy presented here is equally prevalent in male and female rodents.However, it is noted that in adult rodent models of chemotherapy induced sensory neuropathy, the underlying mechanisms that induce pain can differ when considering sex (Luo et al., 2019;Legakis et al., 2020;Saika et al., 2020).Clinically, there is minimal evidence in adult survivors of childhood cancer to strongly support a sex specific neuropathy, though some evidence highlights neuropathic pain predominating in female patients (Khan et al., 2014).However, as presented here, when considering neonatal exposure to chemotherapy in rodent models there are no differences in nociceptive behavioural phenotypes when investigating vinca alkaloids treatment (Schappacher et al., 2017), which supports the overall narrative of the field.

Impact of cisplatin induced alterations in nociceptor development and induction of neuropathic pain
The diagnosis of neuropathic pain not only arises from patients pain experiences but also include disturbances in the integrity of the peripheral sensory nervous system.The incidence of sensory neurodegeneration in adults (Ferdousi et al., 2015) and adult survivors of childhood cancer (Kandula et al., 2020) is high following exposure to platinum based (including cisplatin) chemotherapeutic agents.This is  typified by the presence of axonal atrophy and sensory disturbances of patients, which are depicted through disturbances in axonal conduction properties and reductions in IENF profiles in skin (Argyriou et al., 2005;Ferdousi et al., 2015).These hallmarks of platinum-based chemotherapy induced neuropathic pain are also present in rodent models (Cata et al., 2008;Joseph and Levine, 2009;Park et al., 2014;Vencappa et al., 2015), as well as similar experimental outcomes in vitro sensory neuronal studies demonstrating reduced sensory neurite outgrowth (Meijer et al., 1999;Ta et al., 2006;Podratz et al., 2011).This sensory neuronal remodelling is accompanied by alterations in a number of key regenerative signalling cascades that underlie the regression of the sensory nerve.However, this sensory neuronal remodelling is not permanent.The sensory architecture and accompanying pain experience recovers close to, if not, back to normality to pre-chemotherapy levels (Flatters and Bennett, 2006), in particular following cisplatin exposure (Paice, 2011) indicative of sensory neuronal regenerative capacity.In addition, sensory neural regeneration can be induced pharmacologically by activating key cellular repair mechanisms following nerve injury i.e. activating transcription factor 3 (ATF3) (Seijffers et al., 2007;Cheng et al., 2021).It has been previously identified that aberrant nociceptor IENF density in the plantar skin is evident in adult rodents following early life exposure to cisplatin (Hathway et al., 2017), and is widely associated with onset of chronic pain states (Jimenez-Andrade et al., 2010).This is indicative of this sensory neuronal regenerative capacity, which is aligned with elevated numbers of TrkA positive DRG sensory number post cisplatin administration (Hathway et al., 2017).Here our work identifies that this process is dependent upon influencing the molecular signals that control nociceptor subclassification during development.The fundamental aspects of our developing pain experience is the early life exposure of nociceptors to stressors that influence the molecular checkpoints that delineate nociceptor lineage (Marmigere and Ernfors, 2007).Post-birth the classification of the dorsal root ganglia (DRG) sensory neurons and projections of sensory nerve fibres are not yet hardwired.During adolescence cell fate is finalised and classified according to the expression of distinct sensory neuronal markers.This includes broad nociceptor classifications equating to peptidergic (TrkA+ve) and non-peptidergic (Ret + ve/RUNX1 + ve), with such classifications controlled by numerous transcriptional programming during early postnatal days (Marmigere and Ernfors, 2007).At early stages of postnatal development maturation of the nociceptor, in particular that of TrkA positive neurons, is dependent upon RUNX1 expression patterns (Chen et al., 2006;Marmigere and Ernfors, 2007).Studies have exemplified that loss of RUNX1 leads to elevated abundance in TrkA positive neurons during sensory neuronal development.This is accompanied by aberrant nociceptor growth in the periphery and central terminals, as well as pain hypersensitivity dependent upon TRPV1 modulation (Chen et al., 2006).This is indicative of a prominent TrkA signalling pathway, activation of which has been proven to be fundamental to nociceptor sensitisation (Obreja et al., 2018), sensory neuronal growth (Gallo et al., 1997) and survival (Fischer et al., 2001) as well as onset of pain and pathological pain states (Ashraf et al., 2016).
NGF-TrkA axis has been implicated in the onset of pain hypersensitivity following early life activation of TrkA signalling, which instigates a pronounced nociceptor recruitment to the developing pain experience during childhood (Li and Baccei, 2011).This is represented by increased recruitment of nociceptor dependent activity following tissue damage and/or chemical activation of nociceptors (Fitzgerald, 2005;Brewer and Baccei, 2020).This accelerates C fibre dependent maturation of the somatosensory system that drives the long-lasting pain experience (Brewer and Baccei, 2020).Our previous work has identified that early life exposure to cisplatin led to rodents in adulthood presenting alterations in nociceptor maturation, encompassing aberrant IENF and central projections in the dorsal horn in adulthood (Hathway et al., 2017).
Here we present that this nociceptor maturation and sensitisation is TrkA dependent and is significant contributing factor to cisplatin induced adult childhood cancer survivorship pain.However, it is worth noting the DRG neuronal population is heterogeneous.Here we have not exploited the diversity present in the DRG to fully explore nociceptor sensitisation in particular to allow understanding of the involvement of putative silent nociceptors that may underpin this pain hypersensitivity.
Here we demonstrate that long lasting pain in adult survivors of childhood cancer arises due to a platinum induced alteration in the developmental trajectory of nociceptors, leading to a TrkA dependent activation of nociceptors.With growing numbers of adult survivors of childhood cancer there is an urgent need to understand the underlying mechanisms that cause pain following chemotherapy treatment.This study highlights a novel mechanism that allows development of new analgesic strategies to target this pain.

Funding
This work was supported by the European Foundation for the Study of Diabetes Microvascular Programme supported by Novartis to RPH (Nov 2015_2 to RPH), the EFSD/Boehringer Ingelheim European

Fig. 1 .
Fig. 1.Cisplatin induced increases in TrkA expression in mouse dorsal root ganglia in a rodent model of platinum induced neuropathic pain.Intraperitoneal injection of (0.1 mg/kg) cisplatin in Wistar rats at Postnatal (P) days 14 and 16 results in a delayed but lasting symmetrical pain hypersensitivity to [A] mechanical (**p = 0.0013) and [B] heat (*p = 0.025, **p = 0.0019, ***p = 0.0008) stimuli, when compared to age matched vehicle treated rodents.Similarly, sex comparisons between male and female rodents demonstrated a similar nociceptive behavioural phenotype, with cisplatin induced reductions in [C] mechanical withdrawal threshold (**p = 0.001) and [D] heat (*p = 0,01, **p = 0.002) withdrawal latency, when compared to age matched vehicle treated rodents.Lumbar 5 dorsal root ganglia were extracted and processed for confocal imaging of neuronal TrkA immunoreactivity.[E] Representative images of DRG neurons for vehicle and cisplatin treated animals expressing Beta III Tubulin (Green) and TrkA (red).[F] Z plots were mapped across the neuronal membrane to score the level of TrkA immunoreactivity in the neuron.[G] Cisplatin increased TrkA expression in DRG sensory neurons compared to age matched vehicle treated animals at P45 (*p = 0.01, **p = 0.007,).(* p < 0.05, **p < 0.01, ***P<0.001,Two way ANOVA with post Bonferroni test, number of animals per group = Male Vehicle = 9, Female Vehicle = 9, Male Cisplatin = 9, Female Cisplatin = 6).

Fig. 2 .
Fig. 2. Cisplatin induces alterations in nociceptor developmental trajectory.A sensory neuronal cell line, 50B11, were treated with either vehicle or cisplatin for 24 h.Protein expression ([A] Representative western blot for TrkA and Runx1) for [B] TrkA was increased in the cisplatin group versus vehicle treated cells (p = 0.05, Mann Whitney Test, number of samples 4 per group, p = 0.0511).[C] In contrast, Runx1 protein expression was decreased in the cisplatin group versus vehicle treatment (*p = 0.02, Mann Whitney Test, number of samples 4 per vehicle group and 3 for cisplatin group).[D] Similarly, mRNA of Runx1 was downregulated in the cisplatin group compared vehicle samples (***P=0.0001,Mann Whitney Test, number of samples 4 per group).[E] Immunohistochemistry of Lumbar 5 DRG sensory neurons, Runx1 protein expression was reduced at P16, with fewer neurons expressing Runx1 in cisplatin treated animals compared to age matched vehicle treated rodents (*p = 0.007Two way ANOVA with Bonferroni test, number of animals per group = 5).[F] Representative images of Runx1 immunoreactivity in Lumbar 5 DRG sensory neurons from vehicle and cisplatin treated rodents at P16 and P45 (NeuN Red, Runx1 Green, DAPI Blue).[G] Body weights of rodents post administration of sham injection/vehicle and cisplatin treated rodents demonstrating no difference in body weight.Long term administration on 4 consecutive days via Intraperitoneal injection of the TrkA antagonist, GW441756, prevented cisplatin induced [H] mechanical (**p = 0.001, ***p = 0.0001 all comparisons to cisplatin group) and [I] heat hypersensitivity (***P=0.0004all comparisons to cisplatin group, Two way ANOVA with post Bonferroni test, number of animals per group = Vehicle = 6, Cisplatin = 4, Cisplatin + TrkA antagonist = 4, F (2, 25) = 71.13).Scale bar = 50 μm.

Fig. 4 .
Fig. 4. TrkA dependent nerve growth factor induced pain hypersensitivity and nociceptor sensitisation.[A] Intracellular calcium fluorescence was measured in dissociated DRG sensory neurons following stimulation with TRPV1 agonist, 1 μM Capsaicin.24 h treatment with nerve growth factor led to a potentiation of the capsaicin induced intracellular calcium response, which was inhibited by TrkA antagonist GW441756.[B] TrkA dependence was demonstrated via dose dependent decrease of NGF induced increases in capsaicin induced intracellular calcium responses (AUC=area under the curve, *p = 0.03, ***p = 0.0001).[C] Furthermore, GW441756 inhibited NGF induced phosphorylation of TrkA (representative blot of pTrkA and TrkA, [D] TrkA densitometry analysis and [E] pTrkA densitometry analysis, *p = 0.031).Subcutaneous injection of NGF in the plantar surface of the hindpaw led to a pronounced nociceptive behavioural hypersensitivity to [F] mechanical (***p = 0.0001) and [G] heat stimuli (*p = 0.01) in the ipsilateral hindpaw versus vehicle treated controls.In [H] contralateral hindpaws mechanical hypersensitivity was observed (*p = 0.009), [I] no alterations in relation to heat stimuli (* p < 0.05, Two way ANOVA with post Bonferroni test, number of animals per group = Vehicle = 7, NGF=8, NGF+TrkA antagonist = 8).

Fig. 7 .
Fig. 7. Cisplatin induced prolonged nociceptor sensitisation is TrkA dependent.[A] Work flow for the experimental protocol utilising dissociated DRG sensory neurons from neonatal (P7) mice that were pre-treated with cisplatin for 24hrs.Increasing concentrations of TRPV1 agonist, Capsaicin, led to a dose dependent increase in intracellular calcium fluorescence, calcium response over [B] time (***P<0.001,Two way ANOVA with post Bonferroni test) and [C] area under curve (AUC), ***P=0.001,One way ANOVA with post Bonferroni test).Cisplatin treatment induced significant increases in NGF induced potentiation of capsaicin induced intracellular calcium responses in cisplatin treated DRG sensory neurons [D] over time and [E] AUC (*P=0.035,**P<0.0098,One way ANOVA with post Bonferroni test)).In a dose dependent manner treatment with TrkA antagonist, GW441756, suppressed cisplatin induced nociceptor sensitisation [F] over time and [G] AUC (*P=0.01,**p = 0.002, ***P=0.0001,One way ANOVA with post Bonferroni test).