Preparation and characterisation of zein/polyphenol nanofibres for nerve tissue regeneration

Abstract

Bridging strategies are required to repair peripheral nerve injuries that result in gaps >5–8 mm. Limitations such as donor‐site morbidity and size mismatches with receptor sites for autografts, together with immunological problems associated with allografts and xenografts, have created an increased interest in the field of manufactured nerve guide conduits. In this study, zein, a plant protein‐based polymer, was electrospun to prepare nanofibrous mats. An important challenge with zein mats is the rapid change from fibre to film under aqueous conditions. Tannic acid (TA), which is a polyphenol, was selected to prepare a blend of zein/TA with different weight ratios to investigate its effect on the wetting resistance of nanofibres. The electrospun mats were characterised and evaluated by Fourier transform infrared spectroscopy and scanning electron microscopy (SEM). Also, degradation and mechanical properties of the mats were studied. Results showed that TA had a significant effect on the resistance to film formation in nanofibres. Moreover, the degradation and elongation at break of mats were increased with increase in TA concentration. For the investigation of the peripheral nerve regeneration potential, Schwann cells were selected for cytotoxicity evaluation by the 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5 diphenyltetrazolium bromide assay and cell morphology by SEM. Schwann cells had good biocompatibility with zein/TA blends (%) of 90/10 and 80/20.

1 Introduction

Peripheral nerve damage, especially in neurotmesis injuries, results in the loss of sensory and motor functions and may lead to permanent disabilities [1, 2]. For lesions >∼5–8 mm, end‐to‐end suture treatment of nerve stumps (neurorrhaphy) remains a clinical challenge because of generated tension across the suture lines that are known to inhibit regeneration [3, 4].

The gold standard in longer gaps includes nerve autograft transfers; however, they are limited by the supply of donor nerves, the need for a second surgery, donor site morbidity, and a mismatch between the donor nerve and the recipient site [5, 6]. Also, immunogenic mismatching and possible transmission of disease are risk factors associated with using allografts and xenografts [6]. Manufactured nerve guide conduits (NGCs) are an alternative treatment for peripheral nerve regeneration [7].

Natural and synthetic polymers such as collagen, chitosan, poly(caprolactone), and poly(l ‐lactic acid) have been investigated to prepare NGCs by researchers [7, 8, 9]. Natural polymers have excellent properties such as biocompatibility, biodegradability, encouraging cell attachment, capability to attach to other molecules, and simple and cheap manufacturing that have promoted their application for tissue engineering [7, 8]. Silk protein is a natural polymer that is investigated for nerve regeneration [7] or hair keratin hydrogel (as a filler) showed results similar to autografts for nerve regeneration in a 2 cm and 4 mm tibial nerve defect in rabbit and mice, respectively [10]. Most Food and Drug Administration (FDA)‐approved nerve guides are manufactured from cross‐linked bovine collagen type I known as NeuroGen^TM, NeuroMatrix^TM, NeuroFlex^TM (conduit based), NeuraWrap^TM, and NeuroMend^TM (wrapped based) [2, 3]. However, these commercial conduits have limitations; for example, NeuroMatrix^TM is nonflexible with low kink resistance [1, 2], and NeuroGen^TM and NeuraWrap^TM are expected to fully resorb for a period of 3–4 years, which can raise the risk of long‐term complications like nerve compression and fibrosis in some instances [3].

Of natural polymers, zein is a natural corn protein with good cell compatibility, low cost, and easy fabrication [11, 12]. Also, it is not derived from human or animal sources, and thus it presents less risk for disease transmission. Studies have reported zein applications in drug delivery systems [13]; moreover, zein porous scaffolds have been investigated for use as a bone tissue substitute [14]. Recently, Wang et al. reported a biodegradable porous zein conduit for rat sciatic nerve defect repair [15]. These porous scaffolds were prepared by casting/leaching‐based techniques. Electrospinning is another method to prepare NGCs [16]. Random or aligned forms of fibres and hollow tubular structures with various dimensions can be obtained by electrospinning [16, 17, 18]. Studies have reported prepared zein mats by the electrospinning method [19, 20]. For example, in 2005, Miyoshi et al. prepared ribbon‐like zein fibres ∼1 μm from the zein solution with a concentration of 21wt% [21]. However, water instability of fibrous electrospun zein structures is a challenge [20, 22]. It has been shown that cross‐linking improves morphological properties of zein fibres in a wet environment, but most cross‐linkers are toxic [22]. A few studies have demonstrated blending zein with polymers that can promote the stability of zein mats in a wet environment.

In the current study, the nanofibres of zein/tannic acid (zein/TA) blends were fabricated and characterised as completely plant‐based natural scaffolds for peripheral nerve regeneration. TA is a polyphenol that belongs to a group of hydrolysable tannins and contains digalloyl ester groups connected to a glucose core (Fig. 1) [23, 24, 25].

Fig. 1.

Schematic of tannic acid structure

It is documented that in collagen/chitosan mixtures, the cross‐linking process by TA leads to the modification of swelling and mechanical properties [26]. It is reported that TA can form hydrogen bonds with carbonyl groups of proteins and hydrophobic interactions with the pyrrolidine ring of proline in protein (Fig. 2) [27, 28]. Additionally, the antioxidant and antibacterial properties of TA have attracted the attention of researchers [29, 30].

Fig. 2.

Schematic of interactions between tannic acid and protein

In this study, nanofibres of zein/TA blends were characterised by Fourier transform infrared spectroscopy (FTIR) and SEM. Biodegradation of electrospun mats was evaluated in phosphate buffered saline at 37°C during 30 days; moreover, tensile testing rendered their tensile strength and fracture strain. Also, biocompatibility for the purpose of peripheral nerve regeneration was studied with Schwann cells (SCs) via an MTT assay and cell morphology by SEM.

2 Materials and methods

2.1 Materials

Zein (from maize (Z3625)) and TA (1701.206 g/mol) were supplied from Sigma–Aldrich company (Germany). Acetic acid was obtained from the Merck company (Germany). The 3‐(4, 5‐dimethylthiazol‐2‐yl)‐2, 5 diphenyltetrazolium bromide (MTT) solution, Dulbecco's Modified Eagle's Medium F‐12 (DMEM/F12), fetal bovine serum (FBS), phosphate buffered saline (PBS), trypsin–ethylenediaminetetraacetic acid (EDTA), penicillin streptomycin, were purchased from Sigma‐Aldrich company (Germany).

2.2 Preparation of electrospinning solutions

Pure zein solutions were prepared at 20, 30, and 40% (w/v) in an 85% acetic acid (AcOH) solvent. Also, the blends of zein/TA were designed for the further studies (Table 1). The formulations were under constant stirring at ambient temperature for 6 h to ensure the complete dissolution of the constituents. Also, Table 1 shows the solution viscosity before electrospinning. The viscosities were measured using a Brookfield viscometer using spindle LV‐3 rotated at 50 rpm.

Table 1.

Compositions and viscosity values of prepared solutions before electrospinning

Sample	Total concentration (%, w/v)	Weight ratio (zein/TA)	Viscosity (mPa s)
PZ20	20	1/0	558
PZ30	30	1/0	656
PZ40	40	1/0	810
Z9/T	40	0.9/0.1	917
Z8/T	40	0.8/0.2	1748
Z7/T	40	0.7/0.3	2100

2.3 Electrospinning process

The polymer solutions were spun into nanofibres by electrospinning through a syringe pump (Fanavaran Nano‐Meghyas, Iran). In brief, the prepared solutions were electrospun at a 0.2 mL/h feed rate. A 20‐gauge blunt end stainless steel needle was used for solution delivery and was connected to the syringe pump. A voltage value of ∼15 kV was applied between the needle and the collector (copper plate) with 10 cm separation. All electrospinning processes were performed at ambient temperature.

3 Characterisation

3.1 Fourier transform infrared spectroscopy (FTIR)

Zein and TA powders as well as electrospun Z8/T mat were characterised using an FTIR spectrometer (Frontier model, Perkin Elmer, USA) at a frequency range of 4000–500 cm⁻¹ and a resolution of 4 cm⁻¹ to confirm the zein and TA molecular interactions.

3.2 Morphology of electrospun nanofibres

The surface morphologies of electrospun matrices were observed by scanning electron microscopy (SEM; TESCAN VEGA II, Czech Republic) before and after exposure to PBS at 37°C for 45 min, 1 and 3 days. The diameter and density of the electrospun fibres were characterised using Image J software.

3.3 Degradation of electrospun mats

In vitro degradation of the mats was evaluated by the weight loss ratio as a function of incubation time in PBS at 37°C. Briefly, three samples per group were cut and placed into a 6‐well culture plate. Then 6 mL PBS (replaced every 4 days) was added into each well and incubated at 37°C. Then, the samples were taken out of the solution after predetermined periods of time (5, 10, 16, 23, and 30 days), washed with deionised (DI) water, dried completely, and finally weighed. The weight loss (W _loss) of specimens was calculated using (1):

$$W_{\mathrm{loss}}=\left[\left(W_{\mathrm{init}}-W_{\mathrm{deg}}\right)/W_{\mathrm{init}}\right]\times 100$$ (1)

where W _init is the initial weight of specimens before degradation and W _deg is the weight of specimens after degradation.

3.4 Mechanical test

Tensile properties of the nanofibrous mats were determined at ambient temperature using an Instron 5566 Microtester with a load capacity of 10 N. The specimens were cut at 5 mm × 30 mm with a thickness of ∼100 μm, and then the ends of the prepared specimens were mounted vertically on mechanical gripping units for tension tests.

3.5 Cell behaviour

SCs were used for evaluating in vitro cytocompatibility of Z9/T, Z8/T, and Z7/T nanofibrous mats for the purpose of nerve regeneration. Primary rat SCs were isolated from sciatic nerves of adult male Wistar rats (weighing 200–250 g) (School of Pharmacy of Tehran University of Medical Sciences, Tehran, Iran) according to the Terraf et al. method [31]. SCs were cultured in the DMEM/F12 medium supplemented with 10% FBS and 1.0% penicillin streptomycin.

Cells were maintained in a humidified 5% CO₂ incubator at 37°C and were fed with a fresh medium every 2 days until they reached a confluency of 80%. The MTT assay, with an indirect exposure method, was used to determine the viability and proliferation of cultured cells by measuring the metabolic reduction of MTT to a coloured formazan by viable cells. Briefly, the prepared mats were sterilised with ultraviolet light and extract solutions of mats were prepared according to ISO 10993:12 (four samples of each group were used for the MTT assay, and one sample of each group was used for the cell morphology observation). SCs with the 1 × 10⁴ cells/0.15 ml culture medium were seeded into a 96‐well microplate and incubated at 37°C and 5% CO₂. After 2 days, the SC medium was removed from each well and replaced with 0.15 ml extraction media of samples (n  = 3) collected from 1, 3, and 7 days, and cell viability was monitored by the MTT assay after 1 day; moreover, culture plate containing SCs (without the mat) was considered as Ctrl + for the comparison of MTT results. Also, after 1 day of incubation, the morphology of SCs was evaluated by SEM. Briefly, the cells were fixed with 4% paraformaldehyde for ∼120 min at 4°C, washed twice with DI water, and then dried at room temperature.

3.6 Statistical analysis

All data are expressed as the means ± standard deviation (SD) of 3–5 replicates. Statistically, analysis was performed using the p value, a one‐way analysis of variance to compare any significant difference. A value of p  < 0.05 was considered significant.

4 Results and discussion

The morphologies of electrospun fibres are shown in Fig. 3. It was observed that micron‐sized irregular spherical beaded structures were obtained in PZ20. The number of spherical beads was decreased and nanofibres were more distinguished in the PZ30 sample; however, they were not uniform. As the concentration of zein reached 40%, the uniform zein nanofibres without beads and the most diameter distribution of ∼160 nm (Fig. 3 d) were observed. When ethanol is used as a solvent, zein mats have a ribbon‐like form resulted from the collapse of tubular‐like skin from the fast evaporation of ethanol in high viscosity conditions [20]. However, it is observed from Fig. 3 that acetic acid solvent leads to the formation of round‐like nanofibres probably because of less evaporation or the different self‐assemblies of the zein molecules in the nanofibres.

Fig. 3.

SEM images of electrospun mats of

(a) PZ20, (b) PZ30, (c) , (d) Different magnifications of PZ40, (e) Histogram of size distribution of PZ40 nanofibres

Table 1 shows that the viscosity is increased when the weight ratio of the zein solutions is increased from 20 to 40%; moreover, in the 40% solution, the increase in TA leads to an increase in viscosity. Studies have shown that higher polymer concentration causes higher solution viscosity due to the presence of more polymer chain entanglements and therefore the beaded structures are eliminated [19, 32]. It is reported that an important challenge of electrospun pure zein in tissue engineering, wound healing, and drug‐delivery applications is its very low tenacity in wet environments [20, 33]. Wetting leads to a plasticisation effect on zein causing a fibrous electrospun matrix to become film‐like [34, 35]. This was confirmed in Fig. 4 where after 40 min, nanofibres of PZ40 mat were becoming film‐like (Fig. 4 a) and after 1 day, the electrospun mat showed a fully film‐like structure. To overcome this challenge, cross‐linking has been demonstrated to improve the water stability of zein fibres [22, 33]. Zein has been successfully cross‐linked using hexamethylene diisocyanate, citric acid, succinic anhydride, eugenol, and glutaraldehyde [19, 20, 22]. However, usually cross‐linkers have toxic effects; moreover, the possibility of chemical or physical alteration of the drug molecules makes cross‐linkers inappropriate for drug encapsulation [22].

Fig. 4.

SEM images of electrospun mat of PZ40 after

(a) 45 min and, (b) 1 day immersion in PBS

A few studies have reported blending pure zein with other polymers and their effect on the stability of zein fibre in wet environments. It was demonstrated that blending zein with poly(caprolactone) stabilised the protein as intended for 7 days at 37°C in PBS [36]. Yao et al. studied electrospun fibres composed of zein/gelatin blends for hemostatic application [37]. They showed that as the weight ratio of zein/gelatin increased, the fibre diameter was increased, though fibrous structure was maintained at a zein/gelatin ratio of 2/1 in PBS for 2 h; however, a tendency towards film formation was observed over time.

Since TA has cross‐linking properties along with anti‐oxidant and anti‐bacterial properties [26, 29, 30], blends of zein/TA with zein mass fractions of 70, 80, and 90 wt% in the total concentration of 40% (w/v) in 85% AcOH were used for investigating the effect of blending on physical and chemical properties of electrospun mats. Fig. 5 shows the intermolecular interactions of zein and TA by FTIR analysis. The peak at 3300 cm⁻¹ in Fig. 5 a shows the N–H stretching vibration of the amide I from zein [38]. In the infrared spectra (Fig. 5 c), the significant peak in the range of 3200–3400 cm⁻¹ is related to hydrogen bonding between zein and TA [39]. The appearing peaks at 2960 and 1652 cm⁻¹ (Fig. 5 a) belong to N–H stretching of amino acids and C  = O stretching of amide I groups, respectively [40]. Also, the peak at 1538 cm⁻¹ corresponds to N–H bending and C–N stretching vibrations [35]. In Fig. 5 b, the peaks at 1030 cm⁻¹ and 1704 cm⁻¹ likely represent the C–O stretch in ester bonds and the stretching of the C = O group in TA, respectively [41, 42, 43]. The peak at 1201 cm⁻¹ shows hydrophobic interactions between TA and proline residues of zein; moreover, it represents the C–O stretch in ester bonds of TA [44]. In the zein/TA spectrum (Fig. 5 c), the peaks at 1037 cm⁻¹ and 1325 cm⁻¹, and, moreover, the peak shift from 1652 (from the zein spectrum) to 1660 cm⁻¹ show the presence of TA in Z8/T.

Fig. 5.

FTIR spectra for

(a) Zein, (b) TA and, (c) Z8/T

Fig. 6 shows that TA maintains uniform and bead‐free electrospun nanofibres in all ratios. Also, the histograms show that with an increase in the zein/TA ratio, the distribution of nanofibre diameters shifts to higher values. This can be attributed to the higher viscosity of the zein/TA solution at high TA concentrations. The ability of TA to form hydrogen bonds, as well as hydrophobic and electrostatic interactions can improve the tenacity of zein in wet environments. With the increase in TA value, interaction sites present in the protein chain saturates due to the binding of TA. This can lead to the prevention of zein fibres from readily swell and losing their fibrous morphology during exposure to water [20, 45].

Fig. 6.

SEM images and nanofibre size distribution of electrospun mats of

(a) Z9/T, (b) Z8/T, (c) Z7/T

Fig. 7 presents electrospun mat morphologies of Z9/T, Z8/2, and Z7/T after 3 days immersion in PBS. It was demonstrated that after 3 days immersion in PBS, the mats retained their nanofibre forms in all compositions. The histogram shows that after immersion, the distribution of nanofibre diameter shifts to higher values in Z9/T (Fig. 7 d) in comparison with

Fig. 7.

SEM images and nanofibre size distribution of electrospun mats of

(a) Z9/T, (b) Z8/T, (c) Z7/T after 3 days immersion in PBS

Z8/T and Z7/T (Fig. 7(e, f)). This can be attributed to an increase in phenol groups causing more hydrophobic and hydrogen‐bonding interactions, ultimately leading to a higher tenacity of fibres in the solution. It is reported that hydrophobic interactions are dominant during the initial interaction of a polyphenol with a protein, while hydrogen bonding and van der Waals interactions are important to stabilise the blend [35]. It should be noted that the mats were rinsed with DI water to remove the precipitations or purities from the surfaces of the nanofibres after immersion for imaging by SEM and weight loss evaluation. However, some precipitations resulted from immersion in PBS [46, 47] can keep in touch with the surfaces of the mats, for example, in Fig. 7 c.

When an organism regenerates, biodegradation of a biomaterial is noticeable because it negates the need for secondary surgery. For a nerve conduit, delayed degradation rates may lead to chronic inflammation (caused by foreign body reaction) and pain (caused by nerve compression), and accelerated degradation rates may lead to channel collapse and obstruction of nerve regeneration [4, 48]. However, the biodegradation rate must be considered in relation to nerve gap length.

Fig. 8 shows degradation kinetics of zein/TA blends during 30 days at 37°C in PBS. It was observed that with an increase in TA, the degradation rate was increased after each time point. Incorporation of growth factors, biomolecules, and drugs with zein/TA nanofibres can be a promising approach with tailoring release rate by adapting TA concentration.

Fig. 8.

Weight loss percent of electrospun mats of Z9/T, Z8/T, and Z7/T in PBS at 37 ˚ C during 30 days

TA − protein interactions are physical interactions (hydrophobic and hydrogen bonds); for example, the interactions of the aromatic rings of TA molecules with the hydrophobic residues of zein are hydrophobic bonds [38, 44].

However, it is considered that the wet environment can affect hydrophobic interactions and involved hydrophilic TA preferentially interact with H₂ O, causing a higher tendency for TA to elute from mats when the TA/zein ratio is increased [35]. However, in comparison with pure TA, that is a hydrophilic molecule, the stabilisation of TA through complexation with protein chains can increase the bioavailability of this molecule wherever the mats are implanted in a physiological environment [45].

Fig. 9 shows stress–strain curves of zein/TA blends. It was observed that the tensile strength of the nanofibres had a higher value in lower TA concentration. This can be attributed to the fact that with decrease in TA concentration, the fibre diameters become smaller (Fig. 6), resulting in more inter‐fibre interactions [49]. However, elongation at break was increased at a higher TA ratio. Fig. 9 presents that TA can act as a plasticiser and improve nanofibres flexibility and extensibility by reducing the intermolecular forces and increasing the mobility of the zein chains [34]. However, elongation difference was not significantly between Z8/T and Z7/T. On the other hand, the increase in the diameter of the nanofibres from Z9/T to Z7/T can be a reason in the reduction of tensile strength.

Fig. 9.

Tensile stress–strain curve of zein/TA mats

The first artificial conduit generation was composed of nonresorbable silicone and polytetrafluoroethylene (Gore‐Tex®) [50], but often secondary removal surgeries were required due to compression syndrome or fibrotic encapsulation of the conduit [51]. Natural biodegradable polymers are used to design second‐generation nerve conduits; for example, collagen conduits are capable of repairing nerve defects up to 20 mm [52].

However, the biosafety investigation of the degradation products of an NGC is important. Hence, in this study, SCs were used for the evaluation of the samples' biocompatibility of Z9/T, Z8/T, and Z7/T by an indirect MTT assay due to their important role in peripheral nerve regeneration through the removal of tissue debris and production of cell adhesion molecules and nerve growth‐promoting factors [53, 54]. Fig. 10 shows the results of the MTT assay after 1, 3, and 7 days. It was observed that SCs have good biocompatibility with extractions of Z9/T in all days and Z8/T after 1 day. However, viability was decreased with increases in TA concentration in Z7/T. A few studies have reported on blends of natural polymers for nerve regeneration applications. Recently, Miao et al. [40] studied zein nanofibrous membranes modified with different poly (l ‐lysine) contents of 1.46, 3.57, and 6.18% as a scaffold for nerve repair. They reported that the diameters of zein nanofibres became larger as the content of poly (l ‐lysine) increased. Also, a content of 3.57% was the best for adhesion, proliferation, and differentiation of neural stem cells. These results show that the nanofibre diameter can be effective in nerve cell behaviour and demonstrate a high potential of zein for nerve regeneration.

Fig. 10.

MTT assay results for samples of Z9/T, Z8/T, and Z7/T after 1, 3, and 7 days (*p < 0.05 compared with Z7/T, culture plate seeded with SCs without the mat as Crtl + )

Fig. 11 shows SCs’ morphology on the samples after 1 day incubation. It was observed that SCs have better morphology on Z9/T and Z8/T than that of Z7/T. It is reported that nanofibre topography can influence cell behaviour and nerve regeneration between distal and proximal nerve stumps. For example, Gnavi et al. investigated the effect of the gelatin fibre diameter on SCs. They reported that increasing the gelatin fibre diameter from 300 to 1300 nm did not affect the number of adherent cells; however, microfibres induced more elongated morphology [55]. In another study, Wang et al. indicated 400 nm fibres of silk fibroin increased cell viability and neuronal differentiation of human embryonic stem cells more than that of 800 nm fibres [56]. Investigations have shown positive effects of conductive polymer nanofibres in proliferation and axon regeneration as well [57, 58]. It is observed from Fig. 6 that the diameter distribution of the nanofibres shifts to higher values with decrease in the zein/TA ratio. The cells showed a typical SC morphology with the spindle shapes and oval‐rounded cell bodies on Z9/T and Z8/T mats, but with more distribution of nanofibres diameters ∼250 nm, SC presented relatively the coagulated shape with a weak adhesion. Nanofibres with a lower diameter value have the greater surface‐area‐to‐volume ratio that provide more surface area for a stronger cell adherence and facilitated proliferation [59, 60]. Also, more TA release from the Z7/T mat (Fig. 8) can affect SC morphology behaviour because it declined the SCs viability (Fig. 10).

Fig. 11.

SEM images of Schwann cells on nanofibres of

(a) Z9/T, (b) Z8/T, (c) Z7/T

5 Conclusions

Zein/TA solutions with weight ratios of 90/10, 80/20, and 70/30 were successfully electrospun with diameter distributions of mostly ∼100–400 nm. The viscosity and distribution of the fibre diameter increased with increasing zein/TA ratio. Blending zein with TA significantly increased nanofibre resistance to film‐like morphology formation in a wet environment in comparison with pure zein nanofibres. Results showed that with an increase in TA concentration, the degradation rate of mats and elongation at break (according to tensile test) were increased. Also, Z9/T and Z8/T nanofibres showed good cytocompatibility with SCs and cell morphologies were good on the surfaces of these samples.