Comparative Proteomics Analysis of Growth-Primed Adult Dorsal Root Ganglia Reveals Key Molecular Mediators for Peripheral Nerve Regeneration

Abstract Injuries to peripheral nerves are frequent, yet no drug therapies are available for effective nerve repair. The slow growth rate of axons and inadequate access to growth factors challenge natural repair of nerves. A better understanding of the molecules that can promote the rate of axon growth may reveal therapeutic opportunities. Molecular profiling of injured neurons at early intervals of injury, when regeneration is at the maximum, has been the gold standard for exploring growth promoters. A complementary in vitro regenerative priming model was recently shown to induce enhanced outgrowth in adult sensory neurons. In this work, we exploited the in vitro priming model to reveal novel candidates for adult nerve regeneration. We performed a whole-tissue proteomics analysis of the in vitro primed dorsal root ganglia (DRGs) from adult SD rats and compared their molecular profile with that of the in vivo primed, and control DRGs. The proteomics data generated are available via ProteomeXchange with identifier PXD031927. From the follow-up analysis, Bioinformatics interventions, and literature curation, we identified several molecules that were differentially expressed in the primed DRGs with a potential to modulate adult nerve regrowth. We then validated the growth promoting roles of mesencephalic astrocyte-derived neurotrophic factor (MANF), one of the hits we identified, in adult rat sensory neurons. Overall, in this study, we explored two growth priming paradigm and shortlisted several candidates, and validated MANF, as potential targets for adult nerve regeneration. We also demonstrate that the in vitro priming model is a valid tool for adult nerve regeneration studies.


Introduction
Peripheral nerve injuries affect millions of people worldwide (Grinsell and Keating, 2014;López-Cebral et al., 2017). Severe injuries interrupt nerve continuity leading to long-term functional disabilities in affected individuals. Surgical reconstruction of injured nerves is a treatment choice for severe injuries. However, it does not often restore full functions because of the poor regeneration of axons (Bruyns et al., 2003;Bergmeister et al., 2020). For example, axons regenerate at a poor rate of 1-2 mm/d, which combined with insufficient access to growth factors, limits target reinnervation and functional recovery (Gordon, 2020). Approaches that can improve the rate of axon growth and availability of growth factors may improve nerve regeneration and functional recovery.
Several preclinical studies demonstrated that growth factors such as the nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF) and glial-cell derived neurotrophic factor (GDNF) promote axon regeneration (Karchewski et al., 2002;Gordon, 2009). However, none of them translated into effective nerve repair therapies because of their poor efficacy or adverse outcomes in the clinical settings. This demands identification of novel molecular candidates with superior efficacies. A novel in vitro model, known as in vitro priming model, has been recently developed to study neuron outgrowth (Krishnan et al., 2021). This new model involves growth priming of sensory neurons by incubating the whole dorsal root ganglia (DRG) in suspension in a standard culture media in vitro overnight. It was demonstrated that the outgrowth capacity of the in vitro primed neurons is as good as the in vivo primed neurons, which are generated by a preconditioning transection injury to sciatic nerve (SN) in adult rats (Krishnan et al., 2021). However, the molecular profile of in vitro primed DRGs was not known. In this work, we characterized the in vitro primed DRGs using a whole-tissue proteomics approach and compared their molecular signature with that of the in vivo primed and nonprimed (control) DRGs. We found differential expression of several growth regulatory molecules in the in vitro and in vivo primed DRGs compared with control, revealing potential molecular targets for nerve regeneration and repair. Through this approach, we found a previously unrecognized growth promoting role for the mesencephalic astrocyte-derived neurotrophic factor (MANF) in the peripheral nervous system (PNS), along with the detection of its specific receptor, neuroplastin, in PNS tissues.

In vitro and in vivo priming of DRGs
In vitro and in vivo priming of L4-L6 DRGs were done as described previously (Krishnan et al., 2021). For the in vitro priming, intact DRGs from four-to six-week-old healthy male SD rats were incubated in standard primary neuron culture media [DMEM/F12 (catalog #11330032; ThermoFisher Scientific) 1 N2 supplement (catalog #17502048; ThermoFisher Scientific) 1 100 ng/ml NGF (catalog #13257-019; ThermoFisher Scientific) 1 Cytiva Hyclone antibiotic/antimycotic solution (catalog #SV3007901; Fisher Scientific)] at 37°C and 5% CO 2 conditions for 24 h. In vivo priming was done by performing axotomy in adult male SD rats. Briefly, a small incision was made at the mid-thigh level and muscle layers were separated using a fine scissor to expose sciatic nerve (SN). A blunt transection was then performed onto the SN for complete axotomy. The muscle layers and the skin were then sutured back, and the animals allowed to survive for 3 d before the isolation of DRGs. All animal protocols were approved by the Institutional Animal Research Ethics Board Committee.

Whole-tissue proteomics studies
Total proteins from DRGs were isolated using RIPA lysis buffer (catalog #PI89900; Fisher Scientific) containing Halt protease and phosphatase inhibitor cocktail (catalog #PI78441; Fisher Scientific). Protein concentration was determined by Bradford assay. Proteins were denatured with trifluoroethanol (catalog #BP622-100; Fisher Scientific) and digested in-solution following a previously published protocol (Nair et al., 2021). Briefly, denatured proteins were reduced with dithiothreitol (catalog #194821; MP Biomedicals), alkylated with iodoacetamide (catalog #AC122270050; Fisher Scientific), and then treated with cold acetone (catalog #A18-4; Fisher Scientific) to eliminate interfering compounds (e.g., salts and detergents) that can prevent digestion. Proteins were digested overnight at 37°C in a buffer-containing trypsin (catalog #V5111; Promega Corporation; 50 ng/ml trypsin in hydrochloric acid)/ammonium bicarbonate) at a 40:1 protein:trypsin ratio. To ensure complete digestion of proteins into peptides, samples were further incubated the following morning for 2 h at 37°C in the same ratio trypsin buffer. Digested peptides were dried in speedvac (Labconco) and stored at À80°C until mass spectrometric analysis. Peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) using an Agilent 6550 iFunnel quadrupole time-of-flight mass spectrometer equipped with an Agilent 1260 series liquid chromatography instrument and a Chip Cube LC-MS interface (Agilent Technologies) in data dependent acquisition mode. Chromatographic peptide separation was accomplished using a high-capacity high performance LC Polaris chip (Polaris C18-A, 180 Å, 3-mm stationary phase; Agilent Technologies), consisting of a 360-nl enrichment column and a 75 mm Â 150 mm analytical column. Peptide samples were first loaded onto the enrichment column with 0.1% formic acid in water (ThermoFisher Scientific) at a flow rate of 2.0 ml/min and then transferred onto the analytical column for gradient elution separation. The linear gradient system consisted of solvent A (0.1% formic acid in water) and solvent B [0.1% formic acid in acetonitrile (catalog #A9554; Fisher Scientific)]. The linear gradient program was 3-25% solvent B for 105 min followed by 25-90% solvent B for 20 min at a flow rate of 0.3 ml/min. Positive-ion electrospray MS data were acquired using a capillary voltage set at 1900 V, the ion fragmentor set at 360 V, and the drying/collision gas (nitrogen) set at 225°C with a flow rate of 12.0 l/min. Spectra were collected over a mass range of 250-1700 mass/charge (m/z) at a scan rate of 8 spectra/s. Tandem mass spectrometry data were collected over a range of 100-1700 m/z and a set isolation width of 1.3 atomic mass units. The top 20 most intense precursor ions for each MS scan were selected for MS/ MS with active exclusion for 0.25 min.
Tandem mass spectra were searched against the SwissProt rat (Rattus norvegicus; UniProt May 2020 release date) nonredundant database using Spectrum Mill (Agilent Technologies) as the protein database search engine. Search parameters included a fragment mass error of 50 ppm, a parent mass error of 20 ppm, trypsin cleavage specificity, and carbamidomethyl as a fixed modification of cysteine. Carbamylated lysine, acetyl lysine, oxidized methionine, pyroglutamic acid, deamidated asparagine and phosphorylated serine, threonine, and tyrosine were used as variable modifications in various stages during the database search. In addition, semi-trypsin nonspecific C terminus and semi-trypsin nonspecific N terminus were also used during the search to enhance protein identification. Validation of the Spectrum Mill search was performed at peptide and protein levels (1% false discovery rate; FDR). Three samples each representing the control, in vitro primed and in vivo primed DRGs were used for the analysis. Differential expression of proteins in the in vitro and in vivo primed DRGs was determined relative to control based on a fold change (FC) of !2 in spectral intensities. Mass Profiler Professional (MPP, version 15.0, Agilent Technologies) software was used for the statistical analysis (one-way ANOVA). A cutoff value of p , 0.05 and the Benjamini and Hochberg FDR set at ,1% were used to obtain statistically significant results. Each of the significant proteins has more than two peptides identified. The mass spectrometry proteomics data are deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD031927 (Perez-Riverol et al., 2022).

Bioinformatics and functional enrichment analysis
The protein expression data as observed by the FC values were used to identify common and unique proteins differentially expressed in both in vivo and in vitro samples. The STRING database and the Gene Ontology (GO) terms were used to retrieve the biological processes for all differentially expressed proteins (Ashburner et al., 2000;Szklarczyk et al., 2021). In addition, for the commonly upregulated proteins (proteins that were commonly upregulated in both in vivo and in vitro samples compared with control), due their small number, we performed AmiGO2 gene ontology search for retrieving the biological functions (Carbon et al., 2009). Plots were generated using the ggplot library available in the R statistical package.

Neurite outgrowth analysis
Neurite outgrowth analysis was performed as described previously (Krishnan et al., 2018b). Briefly, adult DRG neurons at the thoracic and lumbar levels were isolated from six-week-old male SD rats. The DRGs were then incubated in 0.1% collagenase (catalog #17104019; ThermoFisher Scientific) in L15 (catalog #11415064; ThermoFisher Scientific) for 90 min at 37°C for enzymatic digestion. The individual cells were then mechanically dissociated from the DRGs by repeated pipetting, followed by the cells pelleted by centrifugation at 800 rpm for 6 min. The cell suspension was then laid over 15% bovine serum albumin (BSA; catalog #SH3057402; Fisher Scientific) and centrifuged at 800 rpm for 6 min to remove the debris that concentrated at the middle layer of the supernatant. The pellet was then washed with L15 and suspended in DMEM/F12 media. Equal number of cells were then seeded on a Nunc Lab-Tek 4 well Chamber Slide (catalog #177437; ThermoFisher Scientific) coated with 0.01% poly-l-lysine solution (catalog #A-005-C; Millipore Sigma) and 10 mg/ml laminin (catalog #23017015; ThermoFisher Scientific). The neurons were cultured in standard primary neuron culture media described above.
To examine the growth promoting effect of MANF, 100 ng/ ml recombinant MANF (catalog #3748-MN; Biotechne) was supplemented to the culture for 48 h. The cultures were then fixed using 4% paraformaldehyde and stained with NF200 (mouse monoclonal, catalog #N0142, Millipore Sigma, 1:200 dilution) followed by incubation with the secondary antibody (goat anti-mouse-Alexa Fluor-488; catalog #A-11001; ThermoFisher Scientific; 1:100 dilution). The neurite outgrowth images were captured using the inverted fluorescence microscope (Zeiss Axio Observer) and the outgrowth parameters were measured using the WIS-Neuromath software (Rishal et al., 2013).

Measurement of MANF in culture supernatants
We performed LC-MS/MS analysis in four individual sensory neuron cultures grown for 14 h to determine whether MANF is released by adult sensory neurons in culture. At first, total protein from culture supernatants was isolated using acetone precipitation protocol. Briefly, one part of the culture supernatant was added to 10 parts of acetone (1:10 ratio) and the mixture stored at À80°C overnight. The resulting supernatant was removed the next day, and 1:10 acetone was again added to the pellet and incubated at À80°C for another 1 h. The final supernatant was then removed and the pellet air dried. The sample pellet was dissolved in PBS before in-solution digestion, which was performed as described for the whole-tissue proteomics studies. Targeted MS analysis on tryptic peptides that resulted from in silico digestion of MANF was performed. Database search of the acquired spectra, for the identification of MANF, was done as described above.

In vitro and in vivo priming induce common molecular changes in adult DRGs
We detected a total of 1475 proteins in adult DRGs after processing the mass spectra data against the rat database (Extended Data Fig. 1-1). Principal component analysis (PCA) and heatmap showed a similar protein profile between the biological replicas (n = 3) for each group, indicating reproducibility in the results (Fig. 1).
Fold change analysis revealed that 63 and 64 proteins were upregulated in the in vivo and in vitro primed DRGs, respectively, compared with control (Extended Data Fig.  1-1). A single protein (coatomer subunit g 2) was shown inverse differential expression (up and down) between the in vivo and in vitro models. Strikingly, 13 proteins were commonly upregulated in both the models indicating some shared molecular features in these models (Table  1). Similarly, 118 and 94 proteins were downregulated after in vivo and in vitro priming, respectively, compared with control (Extended Data Fig. 1-1). Out of these, 64 proteins were commonly downregulated in both models (Table 2). Overall, of the 263 differentially expressed proteins (p 0.05) observed combinedly in the two models, 77 (;29%) proteins showed a similar change (up or down) between the models, indicating that in vitro priming partially reproduces the molecular features that occur in in vivo primed DRGs.
In vivo and in vitro primed adult DRGs upregulate common proteins involved with metabolism, neuron survival, and outgrowth We found that 13 proteins were commonly upregulated in the in vitro and in vivo primed DRGs (Table 1). UniProt knowledgebase indicated that these proteins are associated with cell metabolism, transport, cytoskeleton organization and dynamics, stress regulation, and neuron projection and survival functions (Table 1). Increased metabolism, stress related signaling cascade, neurite outgrowth and structural reorganization are critical events for efficient nerve regeneration. Hence, it is interesting to note that in vitro priming also induces such molecular changes in DRGs in similar ways as observed after in vivo priming. Because of the small number of proteins in this pool, we used AmiGO2 gene ontology search for their functional enrichment analysis (Extended Data Fig.  2-1). Following, our focused analysis identified the candidates that enrich nerve growth-related biological processes, such as, actin filament binding, growth factor activity, GTPase activity, neuron projection development, cell polarity establishment, response to endoplasmic reticulum stress, and signal transduction (Fig. 2). Strikingly, we noted that MANF and Rap-1b enrich at least three biological processes related to nerve growth (Fig. 2). MANF and Rap-1b were previously shown to promote the survival, polarization and outgrowth of the CNS neurons (Schwamborn and Püschel, 2004;Tseng et al., 2017;Eesmaa et al., 2021). However, their growth modulatory roles in the PNS neurons are not well established. We also found that plastin-3 is commonly upregulated in the primed DRGs. Plastin-3 was earlier shown to promote CNS axon growth and survival (Alrafiah et al., 2018). Overall, we found that in vitro priming induces at least some critical molecular changes in adult DRGs similar to what seen after in vivo priming, suggesting that in vitro primed DRG may be an additional suitable model for exploring potential molecular cues involved with adult PNS regeneration.
In vivo and in vitro primed adult DRGs downregulate common proteins that modulate biological processes related to nerve regeneration We found that 64 common proteins were downregulated in the two primed DRG models compared with control. The list of commonly downregulated proteins and their molecular functions identified from the UniProt knowledgebase are given in Table 2. We also performed functional enrichment analysis of these proteins using the STRING database and found them clustering at several biological processes, including cellular and metabolic processes, biological regulation, structure development, and cellular component organization, etc. (Fig. 3). Our focused analysis then identified proteins that clustered onto the biological processes related to nerve regeneration, such as, anatomic structure development, nervous system development, cell communication, cytoskeleton organization, signaling, neurogenesis, microtubule-based process, neuron differentiation, kinase activity and negative regulation of cell death. This focused analysis revealed 39 potential proteins for future investigation (Fig. 4). Some of these proteins, such as,   cyclin-dependent kinase 5 (cdk5), neural cell adhesion molecule 1 (Ncam1), and macrophage migration inhibitory factor (Mif) were previously shown to have growth regulatory roles in the PNS, while drebrin (Dbn1) and contactin-1 (Cntn1) were known to regulate CNS neurons (Nishio et al., 2002;Thornton et al., 2008;Tanabe et al., 2014;Koganezawa et al., 2017;Y.A. Chen et al., 2018;Hwang and Namgung, 2020). Considering the growth promoting roles of these proteins, it is likely that their supplementation may be therapeutic for nerve regeneration. Capturing these common molecular changes in the in vitro and in vivo primed DRGs indeed substantiates the validity of the in vitro priming model for future studies. Interestingly, the common downregulation of the transport proteins, the kinesin-like protein KIF1A (Kif1a), a-soluble NSF attachment protein (napa) and plasma membrane calcium-transporting ATPase 3 (Atp2b3) suggest that these proteins may not be critical for the communication of injury signals or transport of growth-related molecules across injured nerve tips and neuronal soma during regenerative reprogramming of adult neurons. Similarly, common downregulation of the actin modulators such as filamin-C and brefeldin A-inhibited guanine nucleotide-exchange protein 1 (Arfgef1) suggests that similar modes of cytoskeletal changes occur in neurons, and may be satellite glial cells, in the in vitro and in vivo primed DRGs. Although we found downregulation of a few growth promoters in the two models, additional molecular changes unique to these individual models may compensate the loss of these proteins for promoting outgrowth in the primed neurons. The unique protein changes observed in the two priming models are presented below.

Unique molecular changes in the in vivo primed adult DRGs
Expression of 104 proteins was uniquely altered in in vivo primed DRGs with 50 and 54 proteins upregulated and downregulated, respectively, compared with control. The list of these proteins and their potential functions identified from the UniProt knowledgebase are given in Extended Data Figure 1-1. STRING databasebased functional enrichment analysis clustered them at the biological processes, including cellular and metabolic processes, response to chemical, response to stimulus, and development process, etc. (Fig. 3). Further, our specific focus on nerve growth-related biological processes, such as, anatomic structure development, microtubulebased process, nervous system development, neurogenesis, regulation of cell communication, regulation of cell death, regulation of cell motility, regulation of cytoskeleton organization, regulation of signaling, and response to stress revealed 62 potential proteins for future investigation (Fig. 5). Some of these proteins, such as, fibronectin (Fn1), signal transducer and activator of transcription 3 (Stat3) and E3 ubiquitin-protein ligase NEDD4, (Nedd4) are well known for their roles in PNS regeneration indicating the validity of our proteomics analysis (Bailey et al., 1993;Bareyre et al., 2011;Christie et al., 2012). The upregulation of the kinase, the mitogen activated protein kinase kinase kinase 8 (Map3k8), in this model suggests its potential role in  (Tuba8), and kinesin heavy chain isoform 5A (Kif5a) indicates their likely roles in communicating injury signals, and transporting growth promoting molecules across the injured nerve tips and neuronal soma, during the regenerative reprogramming of adult PNS neurons. Overall, our proteomics data substantiated that in vivo priming promotes key events such as cell metabolism, intracellular transport and trophic signaling cascades to facilitate neuron outgrowth and revealed additional molecular candidates for future exploration in adult nerve regeneration and repair models. We found that a few proteins that modulate actin dynamics were downregulated in the in vivo primed DRGs (Extended Data Fig. 1-1). These were tropomodulin-2 (Tmod2), actin-related protein 2/3 complex subunit 2 (Arpc2), destrin (Dstn), and profilin-2 (Pfn2). The Tubulin polymerization-promoting protein family member 3 (Tppp3), which regulates tubulin dynamics, was also downregulated. We believe that this is because of in vivo priming favoring major cytoskeletal reorganization, a key prerequisite for neurite elongation. We found that both actin polymerizing (Arpc2) and depolymerizing (Dstn) proteins were downregulated indicating a balanced scenario. We also noted that the neurite outgrowth promoters, such as, caveolin-1 (Cav1) and elongation factor 2 (Eef2) were also downregulated after in vivo priming (Iketani et al., 2013;Wang et al., 2019). How their alterations influence the overall outgrowth of the in vivo primed neurons is unknown to us at present and may need further investigation.

Unique molecular changes in the in vitro primed adult DRGs
Expression of 81 proteins was uniquely altered in the in vitro primed DRGs with 51 and 30 proteins upregulated and downregulated, respectively, compared with control. The list of upregulated and downregulated proteins and their associated individual molecular functions identified from the UniProt database are given in Extended Data Figure 1-1. STRING database-based functional enrichment analysis clustered most of these proteins at the biological processes, such as, cellular and metabolic processes, response to stimulus, cellular component organization, response to chemical, response to stress, and establishment of localization, etc. (Fig. 3). Further, our specific focus on nerve growth-related biological processes, such as, anatomic structure development, cell communication, cell differentiation, cytoskeleton organization, nervous system development, neurogenesis,  neuron projection development, positive regulation of cell growth, response to stress and signaling revealed 53 potential proteins for future investigation (Fig. 6). The upregulated proteins in this model were mainly associated with actin binding, intracellular transport, and transcriptional regulation (Extended Data Fig. 1-1). For example, the actin binding proteins, such as, coactosin-like protein (Cotl1) and coronin-1B (Coro1b) were upregulated. Similarly, the transport proteins sodium/potassium-transporting ATPase subunit b À1 (Atp1b1), solute carrier family 2 facilitated glucose transporter member 1 (Slc2a1), and importin subunit b À1 (Kpnb1) were upregulated. Among the transcriptional regulators, prohibitin-2 (Phb2) and activated RNA polymerase II transcriptional coactivator p15 (Sub1) were upregulated. Transcriptional activation and actin-dependent cytoskeletal modifications are critical events for peripheral  nerve regeneration. Similarly, molecular transport between Golgi and endoplasmic reticulum, and nuclear and cytoplasmic compartments is also critical for efficient nerve regrowth. We also found that the axon growth promoter integrin b À1 (Itgb1) and the dendritic spine maintenance protein cortactin-binding protein 2 (Cttnbp2) were upregulated in the DRGs after in vitro priming (Y.K. Chen and Hsueh, 2012;Nieuwenhuis et al., 2018). The upregulation of these proteins might contribute to the enhanced outgrowth of the in vitro primed neurons and substantiates the validity of this model for future studies addressing axon regeneration.
We found that the growth modulators, such as, microtubule-associated protein 6 (Map6), protein RUFY3 (Rufy3) and neural cell adhesion molecule L1 (L1cam) were downregulated in the in vitro primed DRGs (Extended Data Fig.  1-1). However, their exact roles in adult peripheral nerve regeneration are not clearly understood and need additional investigation.

Validation of the expression and growth promoting role of MANF in adult PNS neurons
We next validated the expression and growth promoting roles of the commonly upregulated protein MANF in adult PNS neurons. We were particularly interested in MANF because of its known neuroprotective and growth promoting roles in CNS neurons (Tseng et al., 2017;Eesmaa et al., 2021). However, its role in PNS neurons was not established. We examined its basal expression in adult primary sensory neurons from rats and found that it is expressed at moderate to high levels in NF200 low small caliber neurons and lacked expression in most of the NF200 high large caliber neurons (Fig. 7A, upper panel). Three-day injury to sciatic nerve increased the expression of MANF in NF200 low small caliber neurons, but strikingly, most of the NF200 high neurons still had no or lower MANF expression (Fig. 7A, lower panel). Given the known neurotrophic and neuroprotective actions of MANF in CNS neurons, we speculated that supplementation of MANF, especially to NF200 high large caliber neurons, may improve neurite outgrowth. To examine this, we supplemented MANF (100 ng/ml) in adult primary sensory neuron cultures and found that it promotes neurite outgrowth, especially in NF200 high neurons, significantly (Fig.  7C,D). MANF also improved the number of branches and branch length in adult neurons substantiating its growth promoting roles in the PNS (Fig. 7E,F).
We found very low to almost no detectable expression of MANF in satellite glial cells, especially after sciatic nerve injury (Fig. 7B). Overall, the validation of the expression of MANF in injured DRGs, and demonstration of its novel growth promoting role in the PNS, indicate the promise of the in vitro priming model as an ideal tool for exploring novel growth promoters in the PNS.
We also asked whether MANF is secreted by adult neurons in our culture conditions. To address this, we performed LC-MS/MS-based detection of rat MANF in the supernatants of adult rat primary sensory neuron cultures grown for 14 h. We found no detectable levels of MANF in the supernatant (data not shown), indicating that no significant release of MANF occurs in the culture conditions we employed.

The MANF receptor neuroplastin is expressed in the PNS
The knowledge about the specific receptor for MANF was unknown for a long time. However, a recent study by Yagi et al., demonstrated that MANF binds to neuroplastin receptor for mediating its anti-inflammatory actions (Yagi  A, Immunostaining of normal DRG shows the expression of neuroplastin in NF200 low (purple arrows) and NF200 high (yellow arrows) sensory neurons. B, Immunostaining of 3-d axotomized DRG shows the expression of neuroplastin in NF200 low (purple arrow; mild expression) and NF200 high (yellow arrows; intense expression) sensory neurons and satellite glial cells (positions of SGCs are indicated using white arrows). et al., 2020). The expression of neuroplastin in PNS tissues has not been examined previously. Therefore, we examined the expression of neuroplastin in normal and injured (axotomized or in vivo primed) DRGs and found that it is expressed in both NF200 low and NF200 high sensory neurons (Fig. 8A). Importantly, satellite glial cells gained its expression after axotomy (Fig. 8B). We earlier observed that NF200 high neurons and satellite glial cells have low intrinsic expression of MANF, while NF200 low neurons express it at higher levels even after injury (Fig.  7A). Collectively, our observation suggests that MANF derived from NF200 low neurons may bind to neuroplastin receptors on NF200 high neurons and glial cells and activate them for nerve regeneration. These possibilities warrant additional investigations of MANF-neuroplastin axis in the PNS.

Discussion
Despite active research in the past several decades, no drug therapy is available for the effective repair of injured peripheral nerves. While nerve reconstruction surgeries may be helpful for short distance regeneration, a combined approach of surgery and drug therapy is highly desired for long-distance regeneration and better functional recovery (López-Cebral et al., 2017). Here, by performing a comparative proteomics analysis of in vitro and in vivo primed DRGs, we were able to shortlist molecular candidates that are likely modulators of adult axon growth. We believe that this approach is more robust in identifying potential candidates for adult nerve regeneration since the candidates were derived from two independent growthpriming models.
The priming of DRGs preconditions adult neurons for regeneration, and hence the commonly upregulated proteins in our two models are likely promoters of axon growth. However, the opposite may also be true, such that the upregulated proteins may be growth inhibitory. For example, preconditioning of injured neurons also induces expression of growth inhibitors (Christie et al., 2014;Krishnan et al., 2016). Hence, follow-up studies of the commonly upregulated critical proteins are warranted. We were keen in initially studying MANF because of its previously reported neuroprotective, axon polarizing, and neurite outgrowth roles in CNS neurons (Tseng et al., 2017;Eesmaa et al., 2021). MANF was also shown to contribute to cell metabolism (Danilova et al., 2019), and inclined to this, our proteomics data revealed several molecular changes associated with cell metabolism in the primed DRGs. Additionally, we validated the expression of MANF predominantly in NF200 low sensory neurons, with very weak to no detectable endogenous levels in basal and injury activated glia. We also demonstrated the presence its receptor Neuroplastin, especially in NF200 high neurons and injury activated glia, suggesting that exogenous MANF may modify cellular signaling in these cells by binding to neuroplastin. Substantiating this argument, we demonstrated that exogenous MANF promotes the outgrowth parameters of NF200 high sensory neurons in culture. Efficient regeneration of large diameter NF200 high axons is critical for peripheral nerve repair because of them being a major population that is heavily myelinated. Therefore, the growth promoting effect of MANF in this population is promising to consider it as a potential nerve repair agent. MANF has been previously shown to regulate glial cell signaling (Walkowicz et al., 2017). Studies in the past has also showed that MANF is a secretory protein (Li-Na et al., 2017). Although we did not find its remarkable release from cultured neurons, its secretion from injured neurons in vivo cannot be ruled out, especially since its receptor is expressed in injury activated glia. Therefore, further validation of MANF in more rigorous peripheral nerve injury models in vivo is warranted. Overall, our proteomics exploration of primed DRGs identified previously unrecognized critical roles of MANF in the PNS.
We found common downregulation of several transport proteins in the DRGs after in vitro and in vivo priming. While their downregulation in the in vitro primed DRGs was not surprising, may be because of the lack of longer nerve connections to neuronal soma in this model requiring no major retrograde transport, such a downregulation in the in vivo DRGs was not expected as retrograde transport is critical for axotomy-driven nerve preconditioning (Ducommun Priest et al., 2019). However, the unique upregulation of the transport proteins dynamin-3, dynein 1 light intermediate chain 2, and KIF5A after in vivo priming indicates that these proteins may be critically required for the long-distance retrograde transport in vivo. We also found the upregulation of the transport protein importin subunit aÀ5 in the in vivo primed DRGs although our focused analysis considering selected biological processes did not capture this protein. The common downregulation of the neurite outgrowth facilitators such as contactin-1, CDK5, NCAM-1, and peroxisome modulators in the primed DRGs were unexpected, but indicate the presence of upstream growth decelerating signals in these DRGs. Supporting this argument, several previous studies demonstrated that growth inhibitory signals are active past neuronal injury and during the early regenerative reprogramming of adult neurons (Christie et al., 2014;Krishnan and Zochodne, 2015;Krishnan et al., 2016).
Overall, we identified 263 molecular candidates based on their differential expression in two growth priming models and tabulated their individual functions from the UniProt knowledgebase for further investigation. We also shortlisted most promising candidates among them based on their functional enrichments at the nerve growth-related biological processes. In addition, we identified a novel growth promoting role for MANF in the PNS along with the detection of its receptor neuroplastin in PNS tissues. This is a promising start and follow-up systematic screening of the shortlisted candidates in nerve regeneration models is warranted to explore the full potential of these candidates in improving adult nerve regeneration.