Magnetic properties are generated by electrons, spinning around the nuclei of atoms and their axis (16). The magnetic field intensity of a magnetic material is related to its atomic structure and temperature. In some atoms, the magnetic dipoles, arising from the spinning of electrons, do not cancel each other out, thereby creating a permanent dipole. The permanent dipole aligns with the external magnetic field. The magnetism of superparamagnetic materials disappears with the disappearance of the external magnetic field. This provides a basis for strengthening the magnetic field intensity and reducing the adverse effect of a magnetic field (17).

Static magnetic fields (SMFs) or pulsed magnetic fields (PMFs) are usually used in the studies of nerve cells. PMFs are safe and efficient to promote nerve regeneration (18–20). SMFs have constant intensity and direction and a frequency of 0 Hz. Recently, studies have been conducted on the effects of SMFs on nerves at the cellular level. Numerous studies showed that the effects of exposure to SMFs on cellular proliferation varied depending on the cell type (21, 22). Moreover, most studies, investigating the effects of SMFs on cell cycle distribution, showed that there was no statistical difference between the exposed and control group (23). There are two necessary pathways for the regeneration of peripheral nerves: the growth of neurites and myelin sheaths. Mann et al. provided new ideas for the correlations between these two enhanced pathways (24). They reported that a slight nuclear magnetic resonance therapy (NMRT) could induce an increase in SCs proliferation; however, its effect on repair-associated genes was not statistically significant. Furthermore, NMRT could not only enhance the neuronal maturation of individual neurons but also encouraged neurite outgrowth. Interestingly, they also showed that NMRT could induce the secretion of neurotrophic and neuritogenic factors in SCs, leading to the survival of dorsal root ganglia (DRG) neurons and neurite outgrowth.

These studies suggested that the differences in growth direction in response to magnetic field direction depended on the cell type. Macias made two coils of individual copper sheets folded into a square coil to establish a PMF (25). DRG showed a growth tendency parallel to the electric field (perpendicular to the magnetic field). Interestingly, Eguchi showed that SCs and collagens exhibited different growth directions under strong SMFs (8-T). The directions of the SCs and collagen groups were parallel and vertical to that of the magnetic field, respectively. Moreover, SCs oriented in the direction vertical to the magnetic field in a mixture of Schwann cells and collagen under 2 h of magnetic field exposure. DRG showed the same growth tendency as that of collagens, thereby providing theoretical support for the use of a magnetic field to guide peripheral nerve regeneration. Other studies demonstrated that fibrin (26, 27), PC12 cells (12), erythrocytes (28), and osteoblasts (9, 29) aligned parallel to the magnetic field. This property of the magnetic field might be significant in selecting different ways to guide peripheral nerve regeneration.

A strong magnetic field has a destructive effect on cells and can cause genetic mutation and DNA damage (30, 31). Liu et al. reported that the high-intensity magnetic field might be unfavorable for the growth of Schwann cells. When PMF was set to a frequency of 50 Hz, different from 5.0 or 10.0 mT (T = tesla), 0.5, 1.0, or 2.0 mT each was safe for the growth of Schwann cells (32). Interestingly, Eguchi reported that SCs and collagen exhibited normal growth under strong SMFs (8-T) (11). The reasons for these dramatic differences have not been explored yet, and it was speculated that the effects of magnetic field intensity on cellular growth were related to the type of magnetic field. SCs could better withstand SMF as compared to PMF.

A major challenge in peripheral nerve regeneration is the local and sustained delivery of bioactive factors. Studies on MNPs showed that the intervention of an external magnetic field was a type of conventional means. Due to good biocompatibility, MNPs can also be used alone as drug-loading tools to promote peripheral nerve regeneration (33). Magnetic liposomes could enhance neurite outgrowth with a nerve growth factor (NGF) through phosphorylated extracellular regulated protein kinases (ERK) 1/2, which simultaneously upregulated the β-tubulin and integrin β1 (34). Hu et al. showed an increased extension of spiral ganglion neurons (SGNs) in MNPs and MNPs + magnetic field groups as compared to those in the control group (2). Mitrea et al. used oral chitosan-functionalized MNPs to promote neural regeneration in the PNI experimental model and obtained satisfactory results (35). The mechanism for this phenomenon might be the release of Fe ions from iron oxide NPs inside the cells, promoting neurite outgrowth. Numerous approaches used MNPs to assist in drug or gene delivery, including liposome or micelle encapsulation, polymer coatings, and direct surface functionalization (16). Other applications of MNPs include the directional migration of cells induced by external magnetic fields (2, 36–39). In a review, Gilbert divided MNP-mediated cellular manipulations into two categories: magnetic cell guidance and magnetic cell transplantation (16). In essence, both these methods are based on the combination of MNPs and target cells induced by external magnetic fields.

To solve the problem of instability, aggregations, and cellular toxicity of MNPs, Qin et al. synthesized a novel nanomedicine composed of NGF-functionalized Au-coated MNPs. They showed that both the static and rotation magnetic groups could significantly increase the number of differentiated cells and neurites, exhibiting longer average neurite length and clearer directionality, as compared to the no magnet group (36). Liu et al. designed a type of fluorescent–magnetic bifunctional superparamagnetic iron oxide nanoparticles (SPIONs) to control the phenotypic stability of repaired SCs (37). The SMF-induced SPIONs group showed a significantly higher mRNA expression level of brain-derived neurotrophic factor (BDNF), glial cell-derived neurotrophic factor (GDNF), oligodendrocyte transcription factor 1 (Olig1), and vascular endothelial growth factor (VEGF) (qRT-PCR and ELISA). Western blot analysis showed that the expression levels of Beclin1 and lc3b in the experimental group were higher than those in the control group, while the expression level of p62 protein was lower, suggesting the activation of autophagy in Schwann cells by the magnetic stimulation SPIONs. Moreover, for their effects on elongation and branching, there was a statistically significant increase in the expression levels of immune-related cytokines and transcription factors associated with repair phenotypes in the SPION + MF groups. In vitro experiments also verified the effects of SPIONs on promoting peripheral nerve regeneration. Hu et al. reported that the combination of SGNs with poly-L-lysine-coated SPIONs could promote the extension of neurites and growth along the direction of a magnetic field (2). Huang et al. reported that the number of SCs with ChABC/PEI-SPIONs, which migrated into the astrocyte region, was 11.6- and 4.6-fold higher than that of control groups under the driven effect of a directional magnetic field (40). Raffa et al. observed that PC12 cells also exhibited more neurite extension and orientation under the guidance of MNPs (39).

Due to the advantageous features of nanofibers and MNPs, numerous researchers have combined MNPs with biodegradable nanofibers to produce magnetic nanofiber scaffolds. Nanofibers have the characteristics of a large surface-to-mass ratio, high porosity, and superior mechanical performance (41–43). Based on electrospinning (ES), there are three techniques for obtaining iron oxide-loaded composite fibers: the introduction of pre-synthesized SPIONs in a polymer solution before ES (used in biodegradable polymers mostly), mixing a precursor before a post-ES process to yield the SPIONs within the fibers, and in situ synthesis techniques (44, 45). There are two major drawbacks in the classical precipitation of NPs on the surface of biomaterials, including difficulty in controlling the morphology and aggregation of NPs and the incompatibility of SPIONs formation with a wide range of biopolymers (44). Silanization, polymer brush coating, and grafting have been used for ameliorating the dispersibility of NPs in polymeric matrices (44). Sodium citrate, polyacid, or oleic acid has also been applied for the dispersion of iron oxide NPs. Nottelet et al. combined thiolyne photoaddition with free ligand exchange to anchor SPIONs on the surface of nanofibers, which maintained the initial fiber morphology and avoided the unwanted aggregation of MNPs.

A variety of materials have been used in the preparation of magnetic nanofibers, such as chitosan/poly (vinyl alcohol) (PVA), poly (ε-caprolactone) (PCL), hydroxyapatite (HA), magnetic poly (L-lactic acid) (PLLA), poly (D, L-lactic acid) (PDLLA), and poly (glycolide-co-L-lactide) (PGLA) (46–50). Due to the tunability, porosity, hydrophilicity, capacity for the incorporation of biological factors, and the polymeric nature of hydrogels, they are increasingly used in bone or nerve regeneration, cancer therapy, and tissue engineering (51). Gilbert et al. designed a method to inject the small conduits of aligned fibers within a hydrogel to reduce fiber tangling (52). They observed that SPION could increase the elongation of neurites by 30%. In addition, they also concluded that SPION content in the electrospinning fibers could be set between 2 and 6% by weight to balance the magnetization and fiber diameter, alignment, and/or density. Gilbert et al. also synthesized aligned control PLLA and SPION-grafted PLLA electro-spun fibers to induce DRG growth. They set up three different types of magnetic fields to study the induction of nerve regeneration by magnetic nanofibers under different magnetic fields: SMF, alternating magnetic field, and linearly moving magnetic field. They observed that neurites could extend 24 and 30% farther on SPION-grafted fibers as compared to that of untethered SPIONs on PLLA fibers under the static or alternating magnetic field. Moreover, 40 and 27% longer neurites were observed in the alternating and linearly moving magnetic fields as compared to that in the SMF (53).

In addition to inducing nerve regeneration under a paramagnetic action, the magnetic field can also be used to form aligned columnar structures in biological scaffolds, especially systematically tuning the mechanical properties of hydrogel scaffolds (54–56). Schmidt et al. used polyimide, magnetic alginate microparticles, and glycidyl methacrylate hyaluronic acid to invent a magnetically aligned regenerative tissue-engineered electronic nerve interface (MARTEENI) (57). This new technology combined TEENI technology with magnetically templated scaffolds to overcome the challenges of prosthetic-limb neural-interfacing technology, such as stiffness mismatch, low axonal population sampling, and long-term signal decay. After 6–8 weeks of exposure, MARTEENI groups showed similar results in the expression levels of extracellular matrix (ECM) proteins and axonal density as compared to the autograft groups. Under the induction of an external magnetic field, magnetic nanofibers can accelerate drug release. Ming-Wei Chang et al. reported that magnetic hollow fibers could induce greater drug release as compared to the non-treated samples under AMF (~40 kHz) (58). This might be due to the magnetocaloric effect caused by superparamagnetism. This kind of nanofiber might have a good prospect in drug delivery with precise positioning.

Since the 1970's, increasing evidence has confirmed the neuron-promoting effects of SMF and alternating magnetic fields. The mechanism of the magnetic field, including both external or micromagnetic fields based on superparamagnetism, which affects peripheral nerves, is unclear. A comprehensive understanding of the mechanism at the cellular level might help in developing strategies to use magnet-mediated therapy for PNI. This part of the review highlights the mechanism of stimulating the effects of a magnetic field on cells (Figure 1).

One of the mechanisms of the magnetic field on cells is its effect on the molecular structure of excitable membranes, modifying the function of embedded ion-specific channels (59). Strong magnetic fields can alter the preferred orientation of different diamagnetic anisotropic organic molecules. Low magnetic field intensity relies on the molecular structure of excitable membranes. Researchers have focused on the changes in ion channels under a magnetic field. The voltage-gated channels, including potassium, sodium, and calcium ion channels, are affected by magnetic field exposure, thereby making neurons highly sensitive to magnetic field exposure (60–62). Numerous studies have investigated the internal flow of calcium ions (3, 63). In previous studies, Aldinucci simultaneously exposed human lymphocytes to 4.75 T SMFs and 0.7 mT PMF at 500 MHz for 1 h. The combined SMF and PMF groups could increase the Ca²⁺ influx (64). Balassa et al. reported that the calcium ion channels might play significant roles in synaptic functions under magnetic field stimulation (65). Finberg et al. reported that SMFs could mediate the Ca²⁺ influx through L-type voltage-gated calcium channels (VGCCs), providing a neuroprotective activity related to their anti-apoptotic capacity. They reported that the cellular apoptosis of primary cortical neurons, which were exposed to SMF (50 G), decreased, showing a lower expression level of pro-apoptotic markers, including cleaved caspase-3, cleaved poly ADP ribose polymerase-1, the phospho-histone H2A variant (Ser139), and active caspase-9. Moreover, the expression levels of Cav1.2 and Cav1.3 channels were significantly enhanced (63). Currently, these protective effects on these cells are only reflected in the SMF, while the strong PMF accelerates cellular apoptosis. Prasad et al. reported that the moderate-intensity SMF (0.3 T) could induce the oligodendrocytes precursor cells to increase the intracellular Ca²⁺ influx by increasing the expression levels of L-type channel subunits—CaV1.2 and CaV1.3 (3). Numerous studies showed that the changes in ion channels were an important part of this process.

Studies on MNPs in various magnetic fields have shown that they have great potential for nerve regeneration. After the binding of MNPs to target cells, the magnetic force acting on the target cells becomes the most intuitive mechanism. Schwann cells, which are the commonly used target cells, are essential for the rapid saltatory propagation of action potential. Studies have focused on the effects of mechanical stimulation on Schwann cells (66). Recent studies showed that Schwann cells could create “stimuli” for themselves even in the absence of external mechanical stimulation (67, 68). They secrete basal lamina, an essential component of the SC's ECM, to enhance mechanical resistance. Mechanical stimulation might have a positive impact on SCs based on mechano-sensors and mechano-transducers (66). The stimulation can activate ECM, cell adhesion molecules, and some mechano-transduction pathways (69) and can enhance the expression levels of β1 integrins (70, 71). Moreover, low-level mechanical stimulation can promote the proliferation of SCs accompanied by demyelination and apoptosis (72–74). Xia et al. reported that miR-23b-3p from the extracellular vesicle of SCs might be a key factor of mechanical stimulation, promoting peripheral nerve regeneration (75). Meanwhile, force generation is downstream of many signaling cascades activated by neurotrophic factors, such as netrin-1 and NGF (76, 77). Raffa et al. quantified the correlations between the magnitude of mechanical force and nerve elongation (78, 79). This provided a theoretical basis for the quantitative design of MNPs to promote axonal regeneration.

Another possible mechanism is the magnetocaloric effects of MNPs in an external magnetic field, promoting nerve regeneration. This effect has been applied in bioengineering for a long time. Lin et al. completed and applied the localized heating of magnetic nanofibers to inactivate tumor cells (80). Although there is a lack of studies on the correlation of the magnetocaloric effect with nerve regeneration under a magnetic field, the correlation between temperature and neurite growth has been confirmed. He et al. reported that suitable temperature increased the serum levels of TGF-β and IL-10 and decreased the serum levels of TNF-α and IL-1β (81). This result is the same as the cytokine changes of nerve cells under the magnetic field. Numerous studies proved that the increase in temperature could positively affect the growth of neurites, and the temperature of 37–42°C might be the most suitable temperature for nerve growth (82).

The effects of the magnetic field on peripheral nerves are also reflected in growth factors and inflammatory factors. Zhang et al. reported that the low-intensity PMF could enhance the expression levels of myelin basic protein (MBP), myelin oligodendrocyte glycoprotein expressions (MOG), and transforming growth factor (TGF) (TGF-β1, TGF-βR1, and TGF-βR2) in the central nervous system (83). The TGF-β inhibitor could reduce the expression levels of NGF in a mouse model, which suggested that the stimulation of a magnetic field might promote the release of NGF (4). Mert et al. reported that the low-frequency PMF (1, 3, 5, 7Hz) not only increased compound action potential (CAP) amplitude and sciatic nerve conduction velocity (SNCV) but also reduced the expression levels of chemokines, such as CXCL1 and CCL3, which prevented the infiltration of immune cells and migration of neural progenitors (84). Moreover, PMF treatment could increase the expression levels of bFGF and decrease that of VEGF in the sciatic nerve. Interestingly, VEGF and bFGF indicated different results in diabetic mouse models, showing decreased expression levels (85). The mRNA expression levels of BDNF, GDNF, and VEGF were higher in the magnetic field-induced Schwann cells than those in the control group, while those of NT-3 were similar (86). There was no reasonable explanation for this phenomenon. It was speculated that this might be related to the differences in the growth directions of Schwann cells and neurites in the magnetic field.

During peripheral nerve regeneration, macrophages gradually convert from the pro-inflammatory (M1) phenotype to the anti-inflammatory (M2) phenotype (87–90). They are also associated with cellular proliferation and differentiation as well as the release of growth factors (91). Dai et al. reported that the combination of MNPs and alternating magnetic fields could promote macrophage to M2 polarization (92). In addition, they also suggested that this regulation might be directly related to the internalization of MNPs involved in activating the expression of interleukin 10 (IL-10), which might be amplified by the external magnetic field.

Brief exposure to SMF at 100 mT for 15 min led to a marked but transient potentiation of binding of a radiolabeled probe for activator protein-1 (AP1) in immature cultured rat hippocampal neurons. It caused a high expression level of growth-associated protein-43 and increased AP1 DNA binding through the expression of Fra-2, c-jun, and jun-D proteins. A study by Hirai provided a basis for finding the signaling pathway of the magnetic stimulatory effects on neurons (93). Liu et al. studied the mechanism of remyelination using in vitro and in vivo experiments on MNPs bound to SCs under SMFs. They suggested that Raf-MEK-ERK1/2, Rac1-MKK7-JNK-c-Jun, and TORC1-c-Jun pathways might be related to peripheral nerve regeneration under superparamagnetism. Notably, although there was a statistically significant increase for regeneration-related protein in the SPION + MF group as compared to the pure magnetic field group, it was not possible to determine whether this effect was due to superparamagnetism or mechanical stimulation (37).

Numerous high-quality studies are still required to establish a complete system for the mechanistic effects of magnetism on peripheral nerves, which should include all the aforementioned experimental phenomena.