Sciatic nerves, and lumbar dorsal root ganglia (DRGs) were harvested from female and male 20–25 weeks old Sprague-Dawley rats for SC and sensory neuron isolation. The sacrifice of the animals was conducted by deep isoflurane narcosis before guillotine decapitation. The breeding, housing, transport, and sacrifice of animals was conducted in compliance with the Austria’s Animal Testing Law (TVG 2012, §2, 1.c) and Article 3 of the Directive 2010/63/EU of the European Parliament and of the Council on the Protection of Animals Used for Scientific Purposes (European Parliament, 2010).

The culture of rat SCs was performed as described before (Haertinger et al., 2020; Weiss et al., 2020). Briefly, sciatic nerves were excised and washed in 1x Dulbecco’s Phosphate Buffered Saline (1x PBS, GIBCO, Waltham, MA, United States) + 1% antibiotic-antimycotic (Thermo Fisher Scientific, Waltham, MA, United States). The fascicles were pulled out of the epineurium and were digested overnight in MEM∝ (Minimum Essential Medium alpha, GIBCO, Waltham, MA, United States) supplemented with 10% fetal calf serum (FCS, LINARIS, Wertheim Bettingen, Germany), 1% Penicillin-Streptomycin (P/S, GIBCO, Waltham, MA, United States), 1% Sodium Pyruvate Solution (NaP, GIBCO, Waltham, MA, United States), 2.5% 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid buffer solution (HEPES, Sigma-Aldrich, Burlington, MA, United States), 0.125% (w/v) collagenase type IV (GIBCO, Waltham, MA, Unites States), 1.25 U/ml Dispase II (Sigma-Aldrich, Burlington, MA, United States) and 3 mM calcium chloride (Merck, Kenilworth, NJ, United States) at 37°C and 5% CO₂. The next day, cells were washed and seeded on 0.01% poly-L-lysine hydrobromide (PLL, Sigma-Aldrich, Burlington, MA, United States) and 5 μg/ml laminin (Sigma-Aldrich, Burlington, MA, United States) coated dishes in SC culture medium (SCM) consisting of MEM∝ supplemented with 1% P/S, 1% NaP, 2.5% HEPES, 0.5% N-2 Supplement (GIBCO, Waltham, MA, Unites States), 2 μM forskolin (Sigma-Aldrich, Burlington, MA, United States), 10 ng/ml recombinant Heregulin ß-1 (PeproTech, Rocky Hill, NJ, United States), 10 ng/ml recombinant FGF-basic (fibroblast growth factor basic, PeproTech, Rocky Hill, NJ, United States), 5 ng/ml PDGF-AA (platelet-derived growth factor, PeproTech, Rocky Hill, NJ, United States) and 5% FCS. Separation of SCs from fibroblasts was achieved during passaging using a two-step enrichment procedure (Weiss et al., 2018), which usually resulted in a SC culture purity of over 95% (Weiss et al., 2020). SCs were passaged upon reaching 80–90% confluence and cultures from passage 2 (p2), but not higher than p6 were used for these experiments. Half of the medium was changed three times a week. SCs (1 × 10⁴/cm²) were seeded on PLL/laminin coated wells of 8-well μ-slide (ibidi, Gräfelfing, Germany) or T25 flask (GIBCO, Waltham, MA, United States) in SCM and allowed to attach 5 h before NMRT or control treatment (see Figure 1). SCs grown in 8-well μ-slides were immunostained for purity and proliferation analysis, SCs grown in T25 flasks were used for supernatant harvest and gene expression analysis.

DRG neurons were cultured as described before (Millesi et al., 2021). Briefly, 3–4 lumbar DRG pairs (L3–L6) were harvested from the spinal roots and washed in 1x PBS + 1% antibiotic-antimycotic. DRGs were cut into smaller pieces and digested overnight using the same digestion solution as described for the nervous tissue (see above). The next day, the pooled cell suspension was washed, resuspended in 1 ml Neurobasal™-A medium (GIBCO, Waltham, MA, United States) and centrifuged through a 20% Percoll solution (Merck, Kenilworth, NJ, United States). The pellet was resuspended in 500 μl DRG medium (DRGM) consisting of Neurobasal™-A medium supplemented with 10 ng/ml recombinant NGF (Invitrogen, Waltham, MA, United States), 1x B27 supplement (Invitrogen, Waltham, MA, United States), 2 mM L-glutamine (Invitrogen, Waltham, MA, United States) and 1% P/S. 20 μl of the DRG cell suspension was plated per 8 well of a 0.01% PDL and 5 μg/ml laminin coated 8-well μ-slide. The DRG cell suspension was allowed to adhere for a period of 5 h before 10 h of NMRT or control treatment (see Figure 1). DRG cultures were then immunostained for neurite feature analysis.

Phase contrast images of SC and DRG cultures were regularly taken with a benchtop microscope (NIKON Eclipse Ts2R, Tokyo, Japan).

Cells were treated with a therapeutic low intensity NMR device adapted for cell cultures (MBST, MedTec Medizintechnik GmbH, Wetzlar, Germany, mean dynamic magnetic field strength of the sweep). The device parameters were chosen based on a previous in vitro study demonstrating favorable effects on cells to a field of 0.4 mT and a radiofrequency of 16 kHz (Oliva et al., 2019). One treatment cycle consisted of 1 h NMRT at ambient conditions followed by a 1.5 h recovery phase in the incubator at 37°C and 5% CO₂. The 5 h treatment involved five treatment cycles: 2 cycles on day 0 and 3 cycles on day 1. The 10 h treatment involved 10 treatment cycles: 2 cycles on day 0, 4 cycles on day 1, and 4 cycles on day 2. The 15 h treatment involved 15 treatment cycles: 2 cycles on day 0, 4 cycles on day 1, 4 cycles on day 2, and 5 cycles on day 3. The control cultures were also kept at ambient conditions for the respective NMRT treatment time. After the last treatment cycle of the respective condition, cells were incubated at 37°C and 5% CO₂ overnight before further analysis. A schematic overview of the NMRT treatment regimen can be found in Figure 1.

The proliferative effect of NMRT on SCs was assessed by visualizing DNA synthesizing cells using the Click–iT Plus EdU Alexa Fluor 555 Imaging Kit (Invitrogen, Waltham, MA, United States). SCs were cultured in SCM and treated according to the 5, 10, and 15 h NMRT treatment regimen. The day after the last treatment cycle, 10 μM EdU was added to the SC cultures and incubated for 2 h. EdU detection was performed according to the manufacturer’s protocol.

The staining procedure was carried out at room temperature unless otherwise noted. A washing step involved a sequential incubation with 1x PBS for 5 min each. All antibodies and details are listed in Supplementary Table 1. Before the immunofluorescence staining, NMRT treated and untreated (control) SCs and DRG neurons were washed with 1x PBS and fixed with 4.5% formaldehyde solution (SAV Liquid Production GmbH, Flintsbach am Inn, Germany) for 15 min. Blocking and permeabilization was performed in one step with 1x PBS containing 1% bovine serum albumin (BSA, Sigma-Aldrich, Burlington, MA, United States), 0.3% Triton X-100 (Sigma-Aldrich, Burlington, MA, United States), and 5% goat serum (DAKO, Glostrup, Denmark) for 15 min. Subsequently, the cells were incubated with primary antibodies diluted in staining solution (1x PBS containing 1% BSA, 0.1% Triton X-100, and 1% goat serum) overnight at 4°C. The next day, cells were washed and incubated with respective secondary antibodies diluted in the staining solution for 1 h. For nuclear staining, 50 μg/ml 4,6-Diamidino-2-Phenylindole solution (DAPI, Thermo Fisher Scientific, Waltham, MA, United States) was added for 10 min. After washing, the cells were embedded in Fluoromount-G mounting medium (Invitrogen, Waltham, MA, United States). Immunofluorescence images were taken using an Eclipse Ti (Nikon, Tokyo, Japan) microscope. Images from one donor in different conditions were acquired at the same day with the same settings to enable comparative image analysis. Images are depicted as maximum intensity projections of total z-stacks, contrast and brightness have been adapted in a homogenous manner, and pseudo coloring was applied to make multicolor figures comprehensible for color-blind readers.

For the quantification of proliferating SCs, the CellCounter plugin of ImageJ 1.47¹ was used. At least 300 DAPI⁺ nuclei were counted per experimental condition by excluding burst nuclei and nuclei cut by the image boarder. Subsequently, SOX10⁺/DAPI⁺ nuclei (SCs) and SOX10⁺/DAPI⁺/EdU⁺ nuclei (proliferating SCs) were counted.

The total number of DRG neurons was evaluated by manual counting of neuronal cell bodies that grew within one well of an 8-well chamber slide without and with NMRT treatment (n = 4). For the quantification of neurite features (primary neurites, branching points, neurite length), DRG cultures were immunostained for tubulin β-3 and 15–20 DRG neurons were analyzed in control and NMRT treated DRG cultures, respectively, of 4 individual donors (n = 4). The number of primary neurites per DRG neuron and mean number of branching points per DRG neurons were determined by manual counting. The mean length of all neurites per DRG neuron visible per field of view was evaluated using the “free line tool” of ImageJ. The mean neurofilament fluorescence staining intensity of NMRT treated and control DRG cultures (n = 4) was calculated using the “measure” function of ImageJ.

RNA isolation of NMRT treated and untreated SCs was performed with the RNeasy kit (Qiagen, Hilden, Germany) and DNase-I (Qiagen, Hilden, Germany) according to the manufacturer’s manual. Subsequently, RNA concentration was measured with a Nanophotometer (Implen, Munich, Germany) and 50 ng RNA was reverse transcribed (iScript™ cDNA Synthesis Kit, Biorad, Hercules, CA, United States). The Reverse Transcription Polymerase Chain Reaction (RT-PCR) was performed using SYBR Green Supermix (BioRad, Hercules, CA, United States) and the 7500 Fast Real-Time PCR System (Applied Biosystems, Waltham, MA, United States). The respective primer sequences are listed in Supplementary Table 2. Reactions were performed in duplicates. The quantities of target gene expression were calculated relative to the geometric mean of the internal control genes, GAPDH and β-actin, and presented as gene expression fold change between NMRT treated and control samples [2^–(ΔΔCT)].

Conditioned medium of SC cultures after a 10 h NMRT treatment (condSCM-NMRT) and of untreated control SC cultures (condSCM-CTRL) was harvested and mixed 1:1 with DRG culture medium (DRGM) without NGF, respectively. SC medium (SCM) mixed 1:1 with DRGM without NGF served as additional control. Freshly dissociated DRGs were seeded in the condSCEM-NMRT/DRGM, the condSCEM-CTRL/DRGM and the SCM/DRGM media, respectively, and cultured for 24 h before the analysis of neurite features. A schematic overview of the medium experiment set-up can be found in Figure 2.

The normality of data distribution was assessed by quantile-quantile plots and data were logarithmized. The significance of differences of the measured parameters between experimental conditions were evaluated with Wolfram Mathematica using a two-way ANOVA approach. Two-way ANOVA allows the estimation of the impact of two different categorical independent variables on one continuous dependent variable (Norton and Strube, 1986), here, the impact of the experimental condition and the cell donors on the measured parameter values. An additional statistical analysis was performed with GraphPad Prism 8 using a grouped analysis two-way ANOVA. The donor dependent visualization of measured parameter values for statistical analysis is shown in Supplementary Figure 1. The results are depicted as single values for each donor ± standard deviation (SD); *p < 0.05, ^**p < 0.01, ^***p < 0.001.

In order to analyze and compare the morphology between control and NMRT treated SC cultures, phase contrast images from all conditions were taken before treatment and after the 5, 10, and 15 h treatment regimen. In all conditions, SCs possessed a spindle-shaped morphology with di- and tri-polar extensions and formed the typical swirled parallel alignment with increased culture density (Figures 3a–h), no changes were noted in the NMRT treated group in comparison to the control group. Thus, NMRT did not affect the morphology of SCs or their capability to align to each other.

Next, we used a well-established multicolor immunofluorescence staining analysis (Weiss et al., 2020, 2021) to qualitatively and quantitatively assess whether the NMRT treatment influences the purity or proliferation of SC cultures over time. The SC cultures were characterized by the expression of the S100 protein and the glial cell associated transcription factor SOX10 demonstrating a highly enriched SC population (Figure 4A). The few fibroblasts present in SC cultures are negative for S100 and SOX10, while both SCs and fibroblasts express the intermediate filament vimentin (VIME) (Figure 4A). As the nuclear staining of SOX10 is more convenient for image analysis, the following proliferation assay used SOX10 and not S100 for the determination of SC identity. DNA synthesizing cells were visualized by EdU detection. The VIME staining was used to show all cells. Coimmunostainings and EdU detection of SC cultures were performed after the 5 h (Supplementary Figure 2A), 10 h (Figure 4B) and 15 h (Supplementary Figure 2B) NMRT treatments alongside with respective controls. The immunofluorescence images illustrated SOX10 positive SCs, EdU positive nuclear signals in DNA synthesizing cells, and VIME positive cytoskeletal constituents. The number of SCs (SOX10 positive cells) and proliferating SCs (SOX10 and EdU positive cells) were quantified for each condition. The results showed a purity of about 95% in both control and NMRT treated SC cultures after the 5 h (CTRL: 94.50 ± 2.83, NMRT: 5 h 95.88 ± 2.85), 10 h (CTRL: 94.13 ± 3.40, NMRT: 93.88 ± 2.75) and the 15 h (CTRL: 95.00 ± 2.10, NMRT: 94.33 ± 2.88) treatments (Figure 4C). The amount of proliferating SCs was similar in the 5 h control and NMRT groups (CTRL: 31.04 ± 5.74, NMRT: 35.80 ± 7.34). Notably, proliferation was slightly but significantly increased in the 10 h NMRT group (CTRL: 28.83 ± 5.36, NMRT: 32.21 ± 4.58) and 15 h NMRT group (CTRL: 23.58 ± 8.17, NMRT: 30.04 ± 7.55) (Figure 4D). We further tested whether NMRT affects the mRNA expression levels of repair SC associated tumor necrosis factor receptor superfamily member 16 (Ngfr), receptor tyrosine-protein kinase erbB-3 (Erbb3) and transcription factor AP-1 (Jun). No difference between control and NMRT treated SC cultures were found for the expression of Ngfr (CTRL: 1, NMRT 5 h: 0.95 ± 0.08; NMRT 10 h: 0.93 ± 0.03; NMRT 15 h: 0.94 ± 0.04), Erbb3 (CTRL: 1; NMRT 5 h: 0.97 ± 0.11; NMRT 10 h: 0.97 ± 0.04; NMRT 15 h: 0.99 ± 0.11) and Jun (CTRL: 1, NMRT 5 h: 0.87 ± 0.05; NMRT 10 h: 0.98 ± 0.08; NMRT 15 h: 1.00 ± 0.06) (Figure 4E). These findings show that the purity of SCs and the expression of repair SC associated genes was similar in NMRT treated and untreated SC cultures. However, NMRT slightly enhanced the proliferation rate of cultured SCs.

We further investigated if NMRT exerts an effect on primary DRG neurons. To determine early differences in neuronal behavior, we chose the 10 h NMRT treatment regimen for further investigations. Five hours after the seeding of digested DRGs, the cultures were either exposed to NMRT or left untreated. The subsequent phase contrast images suggested that the treated cultures contained more neurons with a denser network of neurites (Figure 5A). Counting of all attached neuronal cell bodies per well confirmed a significantly increased number of DRG neurons after NMRT (CTRL: 62.75 ± 20.59, NMRT: 178.75 ± 48.72) (Figure 5B). For qualitative assessment of the neurite outgrowth, we stained the DRG cultures for the axonal protein tubulin β-3 (TBB3), typically expressed in developing axons. The comparison between control and NMRT treated DRG cultures visualized an advanced neurite network in the latter (Figure 5C). Quantification of neurite features further demonstrated that NMRT treated neurons showed significantly increased numbers of primary neurites (CTRL: 2.18 ± 0.68, NMRT: 4.86 ± 0.51) (Figure 5D) and neurite branching points (CTRL: 0.92 ± 0.75, NMRT: 2.66 ± 0.45) (Figure 5E) alongside with longer neurites (CTRL: 247.49 ± 20.59 μm, NMRT: 505.01 ± 71.26 μm) (Figure 5F). These results indicate that the NMRT treatment had beneficial effects on the survival and neurite network formation of DRG neurons.

As an increased neurite outgrowth could suggest progressing neuronal maturation, we stained the DRG cultures for neurofilament heavy polypeptide (NFH), a marker for maturing axons. The immunofluorescence images illustrated that NMRT treated cultures contained neurons expressing NFH in the cell body and the neurite network (Figure 6A). Quantification confirmed a significantly higher NFH mean fluorescence intensity (MFI) per DRG neuron in the NMRT treated cultures (CTRL: 855.19 ± 509.99, NMRT: 1333.24 ± 605.99) (Figure 6B). Coimmunostaining with TBB3 and NFH further revealed that not all TBB3 positive neurons express NFH (Figure 6C). The results demonstrated that exposure to NMRT increased the attachment and neurite outgrowth of all neurons and enhanced neuronal maturation of individual neurons.

In addition to the elevated number of DRG neurons in NMRT treated cultures, an increase in non-neuronal cells could be observed. Based on the morphology of these cells, and not caring out a SC depletion step in the DRG culture protocol, we assumed that most of the non-neuronal cells represent SCs. Coimmunostaining of NMRT treated and untreated DRG cultures for the SC marker S100 and the neuronal marker TBB3 confirmed that the majority of non-neuronal cells in the DRG cultures indeed are SCs (Figure 6D). Therefore, we next investigated whether the beneficial effect of NMRT on DRG neurons could have been indirectly mediated by the increased number of SCs present in these cultures. SCs are known to produce neuritogenic and neurotrophic factors supporting neuron survival and neurite outgrowth. Hence, we analyzed if the conditioned medium derived from NMRT treated (condSCM-NMRT) and control (condSCM-CTRL) SC cultures would mimic the effect of direct NMRT treatment. As the SC medium contains growth factors that might affect DRG neuron behavior, SC medium (SCM-CTRL) was used as an additional control. To this end, freshly dissociated DRGs were seeded in normal DRG medium (DRGM) without NGF mixed 1:1 with the different SC media (condSCM-NMRT, condSCM-CTRL, and SCM-CTRL) and cultured for 24 h. A schematic overview of the medium experiment set-up can be found in Figure 2.

Phase contrast images showed that DRG neurons grew best when cultured with condSCM-NMRT as illustrated by the increased attachment and survival of cells, especially large diameter neurons (Figure 7A). Coimmunostainings for S100, TBB3 and NFH visualized DRG neurons with extended neurites and associated SCs in the condSCM-NMRT, while neurite outgrowth was poor in condSCM-CTRL and SCM-CTRL (Figure 7B). In line with these observations, quantification of neurite features identified that the numbers of primary neurites (SCM-CTRL: 1.07 ± 0.14, condSCM-CTRL: 0.82 ± 0.41, condSCM-NMRT: 2.39 ± 0.49) (Figure 7C), neurite branching points (SCM-CTRL: 0.31 ± 0.15, condSCM-CTRL: 0.256 ± 0.19, condSCM-NMRT: 1.616 ± 0.43) (Figure 7D), and neurite length (SCM-CTRL: 66.16 ± 3.97 μm, condSCM-CTRL: 61.08 ± 33.35 μm, condSCM-NMRT: 117.43 ± 36.48 μm) (Figure 7E) was significantly increased in the cultures exposed to condSCM-NMRT. However, the expression of NFH was similar in all conditions (SCM-CTRL: 0.28 ± 0.06, condSCM-CTRL: 0.22 ± 0.11, condSCM-NMRT: 0.27 ± 0.09) (Figure 7F). These results indicate that the enhanced survival and neurite outgrowth observed in NMRT treated DRG cultures is to a significant extent mediated by secreted molecules produced by SCs in response to NMRT.