hNGF Peptides Elicit the NGF-TrkA Signalling Pathway in Cholinergic Neurons and Retain Full Neurotrophic Activity in the DRG Assay

In the last decade, Nerve Growth Factor (NGF)-based clinical approaches have lacked specific and efficient Tyrosine Kinase A (TrkA) agonists for brain delivery. Nowadays, the characterization of novel small peptidomimetic is taking centre stage in preclinical studies, in order to overcome the main size-related limitation in brain delivery of NGF holoprotein for Central Nervous System (CNS) pathologies. Here we investigated the NGF mimetic properties of the human NGF 1–14 sequence (hNGF1–14) and its derivatives, by resorting to primary cholinergic and dorsal root ganglia (DRG) neurons. Briefly, we observed that: 1) hNGF1–14 peptides engage the NGF pathway through TrkA phosphorylation at tyrosine 490 (Y490), and activation of ShcC/PI3K and Plc-γ/MAPK signalling, promoting AKT-dependent survival and CREB-driven neuronal activity, as seen by levels of the immediate early gene c-Fos, of the cholinergic marker Choline Acetyltransferase (ChAT), and of Brain Derived Neurotrophic Factor (BDNF); 2) their NGF mimetic activity is lost upon selective TrkA inhibition by means of GW441756; 3) hNGF1–14 peptides are able to sustain DRG survival and differentiation in absence of NGF. Furthermore, the acetylated derivative Ac-hNGF1–14 demonstrated an optimal NGF mimetic activity in both neuronal paradigms and an electrophysiological profile similar to NGF in cholinergic neurons. Cumulatively, the findings here reported pinpoint the hNGF1–14 peptide, and in particular its acetylated derivative, as novel, specific and low molecular weight TrkA specific agonists in both CNS and PNS primary neurons.


Dorsal Root Ganglion Dissociated Culture
(DRG dissociated primary neurons were prepared from neonatal (2 days, P2) Wistar rats (Charles River, Wilmington, MA, USA) [41,42]. All animal procedures were approved by the Ethics Committee of the CERC. Animals were handled in compliance with the national (D.Lgs26/2014) and European Union legislation guidelines for animal welfare (2010/63/EU). All efforts were made to minimize the number of animals used and suffering.

Western Blotting
Cultured neurons were digested in a RIPA buffer with "complete protease and phosphatase inhibitory cocktail" (Roche, Basel, Switzerland) and centrifuged (10,000 rpm, 20 ). The supernatants were collected, and the amount of total protein was determined by Quick Start Bradford Dye Reagent. Each sample (40 µg) was separated by SDS-PAGE in precast 4-12% Bis-Tris Plus gels (Bolt, Invitrogen), transferred to nitrocellulose membranes (0.45 µM, GE Healthcare, Chicago, IL, USA), and incubated for 1 h at RT with 10% non-fat dry milk in TBS-T (10 mM Tris, pH 7.5, 100 mM NaCl, and 0.1% . The overnight incubation with primary antibody (4 • C) was followed by incubation with the appropriate HRP-conjugated secondary antibody (1:2000, 1 h, RT, ThermoFisher Scientific, Waltham, MA, USA) and the ECL substrate (32106; ThermoFisher Scientific, Waltham, MA, USA). The films were digitalized using a professional scanner (HP4050) and quantified by gel densitometry using the ImageJ software (NIH, Bethesda, MD, USA). Measurements were standardized between the experimental groups using the same calibration system and a fixed threshold over the background. To measure the effect of NGF/peptides on the activation of the pathway, the phosphoprotein level was normalized over the total protein level, with a further normalization over β-actin to correct for sample loading bias [43]. Data are expressed as percentage optical density relative to control group, and presented as mean ± SEM.

Immunofluorescence Labelling and Microscopy
Primary cultures were fixed for 20 in PBS containing 4% paraformaldehyde, permeabilised with PBS plus 0.3% Tween, and quenched by ammonium chloride (50 mM,30 ,RT). Aspecific staining by the secondary antibody was blocked by incubation with normal donkey serum (10%, 1 h, RT). The overnight incubation (4 • C) with primary antibodies was followed by the appropriate fluorescent secondary antibodies (1:2000, 1 h, RT).
Immunofluorescence images were acquired with an epifluorescent microscope (Leica CTR5500; Leica Microsystems, Mannheim, Germany) equipped with a CCD camera (Leica) or by confocal microscopy with the laser scanning confocal microscope TCS SP5 (Leica Microsystems) using a 40× (NA = 1.25), and a 63× (NA = 1.4) oil-immersion lens. A UV Diode laser operating at 405 nm, an argon laser at 488 nm, a HeNe laser at 543 nm were used as excitation sources.
NeuN widefield mosaic images were acquired with an inverted microscope (DMI6000 Leica Microsystems) equipped with a sCMOS camera (Zyla, Andor, Oxford Instruments, Oxon, UK) and a white light lamp. Image acquisition was controlled with Leica Application Suite (LAS X software, version 3.7.0.20979, Leica Microsystems, Mannheim, Germany). Mosaics were generated by merging several individual frames, using an automated spatial overlap of 15% with LAS X software (Leica MicroSystems, Mannheim, Germany).

Neuronal Counts
Immunofluorescent images were collected for direct comparison among experimental groups using fixed settings. The number of positive cells over the total number of DAPI stained nuclei per field (field area = 0.366 µm 2 ; 20× objective) was measured. The analysis was performed after calibrating for particle pixels size (50-400 pixels) and applying a fixed threshold over the background. Nuclei were counted both by manual and automated counting methods (ImageJ) with comparable results. Data are expressed as percentage to control group (CTR) and presented as mean ± SEM.

Analysis of DRG Bioassay
DRG neurons were fixed and permeabilized in 4% paraformaldehyde with 0.2% Triton X-100 in 0.1M Tris-HCl (pH = 7.4) for 5 , and immunolabelled for MAP2, as described above. Neurons with neurites connected to the soma, longer than the mean neuronal soma diameter, typically around 25 µM, and with an identifiable axonal arborisation were quantified for a number of parameters. In detail, the number of neurons, total number of neurites, total neurite length (sum of lengths of all neuritis in the field), maximum neurite length (length of longest neurite), and total number of branch points were measured and compared by automated cell count by means of ImageJ.
A threshold over the background was fixed by applying a contrast-based mask, followed by the calibration for particle pixels size (50-400 pixels) to exclude aspecific pins. Number of cell bodies was automatically counted by Image J, while neurites were drawn and quantified by means of the NeuronJ plugin (ImageJ).

Electrophysiology
Whole cell patch-clamp recordings were performed from primary neurons in cholinergic cultures (E17, DIV10) to study the excitatory neurotransmission in different culture conditions. The recording pipettes were pulled from borosilicate glass with an outer diameter of 1.2 mm and had open tip resistances of 3-5 MΩ. The internal solution for filling pipettes consisted of 140 mM CsCl, 1 mM EGTA, 10 mM HEPES and 6 mM d-glucose (pH 7.4 with CsOH). The standard extracellular solution consisted of 130 mM NaCl, 3 mM KCl, 2 mM MgCl 2 , 1.5 mM CaCl 2 , 10 mM HEPES, 6 mM d-glucose, and 10 mM tetraethylammonium (TEA) Cl (pH 7.4 with NaOH).
To isolate the miniature Excitatory Post Synaptic Currents (mEPSCs), 0.5 µM tetrodotoxin (TTX), 5 µM strychnine, 100 µM picrotoxin were added in the bath solution, in order to block voltage-dependent Na+ currents, glycine and GABAA receptors, respectively. Recordings were carried out for 5 from each neuron and the last 2 of each recording were analysed.
Experiments were performed at 22-24 • C, RT. Recordings were made using a MultiClamp 700B amplifier (Molecular Devices, San Jose, CA, USA). pCLAMP 9.2 software was utilized for the data acquisition system (Molecular Devices, San Jose, CA, USA). After the formation of a high-resistance seal (>1 GΩ), the capacitance and the resistance of the electrodes were compensated electronically. The whole cell capacitance was assessed online using the Membrane test function of pClamp9.2. Current signals were sampled at 100 kHz and filtered at 3 kHz.
The MTT assay is based on reduction of MTT to an insoluble formazan product by viable and metabolically active cells. DIV10 neurons, previously seeded on 96-wells microplates, were treated for 1 h with hNGF1-14 peptides at different concentrations (2, 5, 10, 50, 100 µM), or untreated (CTR) DIV10 neurons were incubated with MTT (0.5 mg/mL, 20 , 37 • C) in Hank's balanced salt solution (Life Technologies). The formazan crystals were dissolved in dimethyl sulphoxide (DMSO), and the absorbance at 570 nm wavelength was measured using a Bio-Rad microplate reader (Bio-Rad, Hercules, CA, USA). The amount of MTT conversion was evaluated as a percentage of the absorbance measured in treated cells relative to the absorbance of control cells.
Alternatively, the Trypan Blue test allows the count of living cells able of cytoplasmic dye exclusion. Briefly, DIV10 neuron seeded on glass coverslips in 24-wells plates and treated for 1 h with hNGF1-14 peptides at different concentrations (2, 5, 10, 50, 100 µM), or untreated (CTR) DIV10 neurons were incubated with Trypan Blue (4%, 15 , RT), and then cleared, and mounted on glass slides. The number of Trypan blue positive dying cells was counted, converted into percentage of surviving cells and reported in the graphs (Supplementary Figure S1).

Statistical Analysis
The graphs were generated using PRISM (GraphPad Software, Inc., San Diego, CA, USA), and the data reported as the mean ± standard error of the mean (SEM). One-way ANOVA followed by Student's t-test or Tukey-Kramer post-hoc was used to analyze the data, depending on the number of variables and groups (Statview-SAS, Cary, NC, USA). A p value <0.05 was considered statistically significant.
To analyse miniature Excitatory Post Synaptic Currents (mEPSCs), the 6.0.7 version of Mini Analysis Program (Synaptosoft Inc., Decatur, GA, USA) was used. mEPSCs were manually detected using a 8 pA threshold crossing algorithm. Frequency, event amplitude, kinetic characteristics (rise and decay time), and event area were compared in the different experimental conditions. Data were expressed as mean ± SEM. Fitting and statistical analysis were performed using SPSS 17.0.0 for Windows (SPSS Inc., Chicago, IL, USA) and Origin 7.0 (Microcal Software, Northampton, MA, USA). Statistical tests were performed by using One-Way ANOVA followed by Bonferroni post-hoc correction. Statistical significance was taken as p < 0.05.

hNGF1-14 Peptides Activate both TrkA-Shc and TrkA-PLC-γ Signalling Pathways in Cholinergic Neurons
In order to address hNGF1-14 peptides as NGF signalling agonists, we resorted to a well-established and characterized in vitro method of cholinergic neurons culture [41,42,44]. We found that the hNGF1-14 peptides exhibited NGF-like properties at micromolar concentrations, activating TrkA-related signal transduction and mimicking NGF neurotrophic effects, whereas the scrambled sequence peptide showed any NGF mimetic activity and was comparable to control, and confirmed hNGF1-14 sequence specificity.
To assess cell viability following hNGF1-14 peptides incubation, number of surviving neurons was evaluated after escalating doses (2, 5, 10, 50, 100 µM; Supplementary Figure S1 Overall, these findings pinpoint that hNGF1-14 at 10 µM concentration exerts the maximum efficacy in terms of activation of the NGF receptor TrkA without affecting neuronal viability. For these reasons, 10 µM was the working concentration of the hNGF1-14 peptides in this study.

hNGF1-14 Peptides Administration Increases the Number of Cholinergic Neurons Expressing pAKT and Nuclear pCREB
Then, we assessed whether hNGF1-14 peptides are able not only to activate key signalling molecules of the NGF-TrkA pathway, but also to engage an increased number of cholinergic neurons in the NGF-like pro-survival neurotrophic response. To this purpose, we incubated cholinergic neurons for 20 either with 10 µM scrambled hNGF1-14, or 100 ng/mL (3.84 nM) NGF, or 10 µM hNGF1-14, or 10 µM Ac-hNGF1-14, or 10 µM hNGF1-15 dimer, and number of pCREB (Figure 2A,B) and pAKT (Figure 2A,C) immunolabelled neurons were counted and compared.
Since cholinergic cultures are not pure neuronal cultures pCREB and MAP2 double immunofluorescent labellings were performed, in order to exclude any glial contribution to the neurotrophic effect. As shown in Supplementary Figure S3, a great majority of pCREB stained cells (green colour) also express MAP2 (red colour), pinpointing that CREB activation upon NGF or peptides incubation mainly occurs in neurons.
As control for the peptide sequence specificity and for TrkA-dependent NGF effect, the electrophysiological activity of

hNGF1-14 Peptides Sustain the Survival and Promote Neurite Outgrowth of Dissociated Dorsal Root Ganglia Neurons
Furthermore, to extend the biological characterization of hNGF1-14 peptides, we also studied their action in the classical DRG assay, a model system of choice to validate new NGF agonists for drug discovery [44]. Indeed, NGF is a critical survival, phenotypic and functional regulatory factor for sensory neurons during developmental and early postnatal period [47][48][49]. A continuous NGF supply is needed for dissociated DRG neurons survival and differentiation in vitro [50]. To assess the neurotrophic bioactivity of peptides, post-natal day 2 (P2) DRG neurons were cultured for 5 days in low-serum medium with hNGF1-14 peptides (10 µM) or NGF (100 ng/mL; 3.84 nM). DRG neurons cultured with NGF or hNGF1-14 peptides and MAP2 stained neurons were analysed ( Figure 6).

Discussion
The aim of this study was to evaluate in vitro efficacy of hNGF1-14 and its derivatives as potential drugs for in vivo neuroprotection, of interest for novel therapeutic approaches in neurodegenerative pathologies. Given the limitation of immortalized cell lines studies, to test NGF mimetic activity of the hNGF1-14 sequence we resorted to two types of well-known NGF-target primary neurons, as cellular paradigm of choice.
In this study, we investigated and compared the hNGF1-14, its monomeric derivative peptide Ac-hNGF1-14, and the hNGF1-15 dimer, using NGF as positive and a scrambled peptide as negative controls. Here, we report that the Ac-hNGF1-14, hNGF1-14 and, to some extent, the hNGF1-15 dimer trigger a TrkA-dependent NGF-like signalling cascade in cholinergic neurons (Supplementary Figure S1E) and show survival and neuritogenic properties in dissociated DRG neurons.
To deeply assess their ability to stimulate NGF/TrkA signalling and NGF typical bioactivity, we resorted to two elective cellular models of NGF-target neurons of the central (CNS) and peripheral (PNS) nervous systems, the cholinergic and DRG primary neurons, respectively. The cultured cholinergic neurons do not need NGF to survive, thus being a good paradigm to test the short-term effect of an acute NGF stimulus, e.g., signalling pathway and neuronal activity. Opposite, since DRG neurons require NGF/TrkA continuous supply to survive and differentiate, they represent the model of choice for in vitro survival assays with putative NGF mimetics.
Here we show that the monomeric (hNGF1-14) acetylated monomeric (Ac-hNGF1-14) and dimeric hNGF1-15 peptides present NGF typical signalling properties, being able to stimulate its specific receptor TrkA, its early adaptors ShcC and PLC-γ, as well as its downstream PI3K and MAPK signalling molecules, finally leading to activation of survival promoting AKT pathways (Figure 1). Taken together, these findings suggest that the TrkA pathway initiated by hNGF1-14 peptides lead to convergence of Shc and PLC-γ neurotrophic signalling on key downstream molecules and cellular endpoints in cholinergic neurons, resembling the NGF prototypical mechanism of action.
Albeit glial cells are a minor component of embryonic septal cultures (less than 5% total cells), [45] and low density plating used in this study is sufficient per se to minimize glial survival and growth, a glial contribution to hNGF1-14 peptides-driven NGF pathway activation cannot be excluded. To address this question, we performed double pCREB/MAP2 immunofluorescence labelling. As reported in Supplementary Figure S3, we found that all pCREB cells are also MAP2 positive in control conditions as well as upon NGF/hNGF1-14 peptides incubation, indicating that pCREB expressions upon NGF/peptides stimuli occur mainly in neurons (~85% of cells). Our data are in line with previous works indicating that NGF specifically affects cholinergic neurons in culture [45]. Overall, glial targeting by hNGF1-14 peptides in cholinergic cultures may be rule out.
As a result of CREB activation, a crucial hub in neurotrophic responses [55], the hNGF1-14 peptides showed the ability to increase neuronal activity, measured by nuclear c-Fos level ( Figure 3A,B). The c-Fos is an early immediate gene, induced by neuronal depolarization. For this reason, it is commonly considered a marker for neuronal activity both in vitro and in vivo [56,57]. The observed increase of mature BDNF ( Figure 3C,D) and ChAT ( Figure 3C,E) levels indicated a coherent stimulation of the synaptic neurotrophin and of Ach metabolism. Noteworthy, CREB has been suggested to have a pivotal role in NGF-driven c-Fos promotion [58,59], leading to subsequent BDNF [60], and ChAT [61] expressions in neurons.
Further, our findings are in line with the previously reported increase of BDNF mRNA level and BDNF release upon hNGF1-14 peptides incubation of PC12 cells [28,36]. Of note, reduced ChAT expression deficits occur in several age-related CNS pathologies, like AD [62] and PD [63] and in neurodevelopmental disorders [64] even in absence of neuronal death. Moreover, BDNF is recognized as a key neurotrophin modulating forebrain LTP and synaptic plasticity [65], and its expression is induced by NGF in target neurons [66]. Thus, the use of hNGF1-14 peptides to sustain BDNF and ChAT levels is of interest for several human diseases associated to cholinergic disturbances, like AD, PD, DS and cancer.
Interestingly enough, the peptidomimetics property of hNGF1-14 peptides was abolished by previous incubation with the TrkA specific inhibitor GW441756 (15 µM, 90 ). As compared to other general tyrosine kinase receptors inhibitors used so far, like K252a and synthetic derivatives (e.g., CEP), the GW441756 is a potent TrkA inhibitor (15 µM, IC 50 = 2 nM) with more than 100-fold selectivity over other kinases, thus allowing the highly specific targeting of the NGF/TrkA pathway. The lack of hNGF1-14 effect in presence of the inhibitor strongly suggests that an active TrkA receptor is required for its CREB/c-Fos related neurotrophic action in cultured cholinergic neurons.
This study was mainly focused on the ability of hNGF1-14 peptides to activate the NGF specific TrkA receptor. Other receptors expressed by cholinergic neurons, like p75 common neurotrophic receptor and the specific BDNF receptor TrkB, were not addressed. However, the crystal structure analysis of the NGF-TrkA-IgGL2 complex clearly showed that N-terminal residues are dispensable for NGF interaction with the p75 receptor [26], while p75-NGF binding occurs via the L1, L3 and L4 loops [16], and the D1 domain in presence of TrkA [22,26]. Also, it has been previously observed that hNGF1-14 is able to promote activation upon endocytosis of TrkA, but not of p75NTR in neuronal cells [28,36]. Cumulatively, and given the observed neurotrophic outcome of hNGF1-14 peptides in TrkA-expressing cholinergic neurons, the induction of the p75-driven signalling by 10 µM hNGF1-14 peptides may be excluded. Whether the peptides are able to activate a p75NTR homodimers apoptotic signalling [67] in absence of TrkA remains to be elucidated.
As for hNGF1-14 peptides interplay with the BDNF pathway, crystallization studies of the peptides complexed with the TrkB-D5 domain are not yet available. Based on current knowledge, human BDNF lacks the two histidines critical for TrkA binding at its N-terminus, and it was reported not to bind TrkA-D5, further suggesting neurotrophins specificity at their N-terminal region [32,33]. In line, cross-interactions of hNGF1-14 peptides with other neurotrophic pathways are seemingly unlike in these cellular models. Nonetheless, this critical issue deserves further scrutiny in preclinical testings of hNGF peptides.
Increased NGF availability has been reported to significantly facilitate the induction of hippocampal long-term potentiation (LTP) [68], by augmenting the frequency of mEPSCs in the cholinergic system [46,69]. Thus, the electrophysiological functions of Ac-hNGF1-14, the peptide with the highest in vitro activity, were assessed. We found that 10 µM Ac-hNGF1-14 induces a facilitation of presynaptic excitatory neurotransmitter release (Figure 5), and exerts a potent and specific stimulatory presynaptic action on cholinergic nerve terminals [46,69] similarly to 3.84 nM NGF holoprotein [45,70]. According to the literature, NGF-induced neurotransmitter release potentiation was found to occur upon a significant mEPSC frequency elevation ( Figure 5B), and a comparable effect was found following incubation with the Ac-hNGF1-14 peptide ( Figure 5B) in cholinergic cultures. As control for the peptide sequence specificity and for TrkA-dependent NGF effect, the electrophysiological activity of scrambled hNGF1-14 (Scr; 10 µM, 30') and of NGF (NGF; 100 ng/mL; 3.84 nM, 30') in presence of the TrkA inhibitor GW441756 (GW441756 + NGF; 10 µM, 45') were analysed in another set of experiments. Any statistical significance was found for mEPSC parameters examined, like frequency, amplitude, rise time, decay time, and area (Supplementary Figure S4A,B).
Since TrkA activation is known to occur by NGF dimers binding, dimeric or cyclised peptidomimetics have been generally preferred to facilitate TrkA conformational rearrangement and subsequent autophosphorylation [71,72]. Nonetheless, major limitations have been often reported for dimeric compounds because of their partial TrkA agonism or even antagonism [73].
Interestingly enough, peptidic as well as non-peptidic monomeric compounds, like D3, have been shown to induce TrkA dimerization and subsequent DRG differentiation [12]. The exact molecular mechanisms underlying this activation need to be elucidated. However, peptides clustering over the receptor accounting for its dimerization, has been proposed at least in vitro [28].
Taking all these findings into account, preferring primary neurons to test hNGF1-14 peptides allowed full disclosure of their NGF-mimetic potential. At variance with studies performed on PC12 cell lines [27,36], micromolar concentrations of hNGF1-14 peptides were able to sustain neuronal survival and differentiation in dissociated DRG neurons (Figure 6), and ERK/MAPK activation in cholinergic neurons ( Figure 1F), at an extent comparable with the NGF optimal dose. A possible reason for the discrepancies may relies on the distinct signalling machineries, activation thresholds and even differentiative status characteristics of PC12 derived neuron-like cells and primary neurons. In particular, the need of PC12 for a sustained TrkA activation because of the very low ERK/MAPK basal level during their first differentiation week may account for the suboptimal hNGF1-14 effect observed in PC12 cells [51]. In line with this, the higher ERK level in differentiated PC12 [51] allowed its activation by hNGF1-14 peptide-driven phosphorylation [36], as well as neuroprotection upon NGF withdrawal [27].
Overall, the findings here reported establish hNGF1-14 sequence as a linear monomeric TrkA agonist with full NGF-like bioactivity on NGF-target neurons. Further assessment of their efficacy as neuroprotective drugs in animal models of human CNS neurodegenerative pathologies is underway.

Conclusions
The drawbacks and limitations of previous clinical trials and NGF-based therapeutic approaches also in recent times, paved the way for the screening of novel, potent and selective NGF mimetics. Nowadays, the characterization of novel small peptidomimetic is taking centre stage in order to overcome poor oral NGF bioavailability, size-related limitations of NGF brain delivery, as well as pain related side-effects of peripheral NGF administration for spinal cord injury or diabetic neuroinflammation. The use of small peptidic molecules or non-peptidic compounds with NGF activity is instrumental to non-invasive brain treatment by means of nasal [74,75] and ocular [76,77] administrations, bypassing BBB and avoiding side effects (e.g., pain) typical of systemic NGF administration.
Author Contributions: All authors have read and agree to the published version of the manuscript. V.T. had a major role in designing and performing the experiments, analyzed the data, prepared the figures, and wrote the manuscript; E.F. and V.S. contributed to biochemical assays; E.F. contributed with bibliographic compilation; M.T.C. was in charge of primary neuronal culture; S.C. performed and analysed electrophysiological data; C.Z. supervised the E.F. analysis; P.T. and P.C. provided financial support; E.R., D.L.M. and C.S. provided hNGF peptides and contributed to study design and data discussion; D.M. provided murine NGF; P.T. and P.C. supervised the study. All authors critically revised the manuscript, approved the final version of the manuscript and agreed to be accountable for accuracy and integrity of any part of the work.

Funding:
The project was partially founded by the PRIN (MIUR) grant n • 20152EKS4Y to PT, n • 20152EKS4Y_005 to CS, and by FIRB (RBAP10L8TY_004) to PC. The founders had no role in the design of the study, the analysis, and interpretation of the data.