New Forms of Electrospun Nanofiber Materials for Biomedical Applications

Abstract

Over the past two decades electrospinning has been emerged as an enabling nanotechnology to produce nanofiber materials for various biomedical applications. In particular, therapeutic/cell loaded nanofiber scaffolds have been widely examined in drug delivery, wound healing, and tissue repair and regeneration. However, due to the insufficient porosity, small pore size, noninjectability, and inaccurate spatial control in nanofibers of scaffolds, many efforts have been devoted to exploring the new forms of nanofiber materials including expanded nanofiber scaffolds, nanofiber aerogels, short nanofibers, and nanofiber microspheres. This short review discusses preparation and potential biomedical applications of new forms of nanofiber materials, and future directions.

Graphic Abstract

This review article discusses new forms of electrospun nanofiber materials and their biomedical applications.

1. Introduction

Tissue regeneration is the rejuvenation of damaged tissues or organs in response to injury.¹ However, some tissues or organs, such as bone, cartilage, heart, nerve, and tendon, have limited or no capacity for regeneration, particularly in large tissue defects.² In order to address the challenges of these non-self-healing tissues and organs, regenerative medicine has emerged as an interdisciplinary subject that combines scaffolds (i.e., biomaterials),³ signaling molecules,⁴ and living cells⁵ to fabricate functional tissue constructs to repair the anatomical damages.² One of the steepest challenges in regenerative medicine is the precise recapitulation of the native extracellular matrix (ECM) in scaffolds. A variety of scaffolds have been developed for tissue regeneration, including hydrogels,⁶ freeze-dried materials,⁷ membranes,⁸ particles,⁹ microspheres,¹⁰ decellularized ECM,¹¹ and nanofibers.¹² Among them, the nanofiber scaffold is the most capable of architecturally mimicking the native ECM while remaining immunologically inert, as native ECMs are mostly made of collagen nanofibers.¹³ Given their ability to match ECM morphologies, many methods have been developed to fabricate nanofibers, such as phase separataion, electrospinning, meltblowing, bicomponent spinning, forcespinning, flash-spinning,¹⁴ template synthesis¹⁵ and self-assembly.¹⁶ Compared to other methods, electrospinning shows several advantages: i) versatile fiber compositions; ii) ease of incorporation of signalling molecules; iii) fiber size ranging from several nanometers to tens of microns; iv) ease of control of nanofiber alignment by either electrical field or mechanical force; v) ease of engineering secondary structures (e.g., using modified spinnerets); vi) low cost for the laboratory experimental setup; and vii) feasible to scale up the production rate.

Owing to the biomimetic property and aforementioned advantages, electrospun nanofibers have been widely applied in wound healing, tissue engineering/modeling, regenerative medicine, and cancer research.^{2, 17–19} Cha et al. developed a mussel adhesive protein-blended polycaprolactone (PCL) nanofiber membranes for skin wound healing.²⁰ Kim and Park et al. fabricated a scaffold consisting of tri-layered composite nanofiber membranes mimicking bone ECM for bone tissue regeneration.²¹ In a separate study, Fan et al. fabricated conduits comoposed of nerve growth factor-encapsulated poly(lactic acid-caprolactone) nanofiber membranes for peripheral nerve regeneration.²² However, these traditional 2D electrospun membranes composed of densely packed nanofibers, resulting in the hindrance of both cell infiltration and growth throughout the nanofiber scaffolds.²³ Unlike the 2D nanofiber membranes, the 3D nanofiber scaffolds allow cells to proliferate and interact with their surrounding microenvironment in all three dimensions, similar to the in vivo settings where cells live.²⁴ Moreover, signaling and other cellular functions differ in 3D systems compared with 2D systems,²⁵ which indicates dimensionality mediates physiological responses regulating cell growth, migration, differentiation, survival, tissue organization and matrix remodeling.²⁶ Though traditional 2D nanofiber membranes may have some applications, they are not ideal for regenerative medicine as the complex spatial interactions found in 3D system could be missing between mats and cells. Thus, researchers are seeking to develop 3D electrospun nanofiber scaffolds to better recapitulate the architecture and morphology of ECM.

This article describes recent progress in fabricating new forms of nanofiber materials. It starts with a brief introduction on the advantages of nanofiber scaffolds, and the limitations of traditional 3D nanofiber scaffolds. Subsequently, recent advances in the fabrication and application of new forms of 3D nanofiber scaffolds are presented. In addition, this article discusses the perspective of these novel 3D nanofiber scaffolds in regenerative medicine. Finally, this article concludes with remarks on the challenges and future directions for development of electrospun nanofiber materials in biomedical applications.

2. Current electrospun nanofiber materials and their limitations in biomedical applications

Perhaps the most primitive electrospun nanofiber scaffolds are nanofiber mats which are often produced by direct deposition of nanofibers to a substrate for different periods of time during electrospinning.^24,25 Most 2D electrospun nanofiber scaffolds made of natural or synthetic polymers possess poor stability in an aqueous environment or media, as when they contact water, they swell and collapse into films, with a considerable decrease in the number of interconnected pores. Due to the limited thickness (often less than several milimeters), such nanofiber mats are generally considered to be 2D. Although nanofiber mats can have relatively low porosities, small pore diameters typically hinder cell infiltration throughout the entire mats, which limits its applications in the biomedical field.²⁷ Unlike the 2D nanofiber membranes, the 3D nanofiber scaffolds allow cells to proliferate and interact with their surrounding microenvironment in all three dimensions, similar to the in vivo settings where cells live. Moreover, signaling and other cellular functions differ in 3D systems compared with 2D systems, which indicates dimensionality mediates physiological responses regulating cell growth, migration, differentiation, survival, tissue organization and matrix remodeling.^{23, 28} The criteria used to classify 3D are nanofiber materials with certain shape (not limited to mat/membrane) and/or large pore size usually larger than the size of cells (enhance cellular infiltration).

Hence, many efforts have been devoted to increasing the pore size of electrospun nanofiber scaffolds. Bellis et al. used acupuncture needles (160 μm in diameter) to mechanically stab 150 micropores in a 15 mm diameter circular nanofiber mat. The average pore size ranged from 150–175 μm, and the average spacing between pores was 375 μm. However, cells still struggled to infiltrate within the areas without micropores (Figure 1A).²⁹ In another study, Xie et al. prepared a 3D basket-weaved nanofiber scaffold composed of many fine strips/bundles consisting of either random or aligned nanofibers by making use of a ‘noobing’, or 3D weaving technique. The obtained 3D nanofiber scaffold can be tailored with desired architectures, alignment, and pore size (Figure 1B).³⁰ Based on the literatures, the appropriate pore size for scaffolds depends on the specific applications. For example, for hard tissue regeneration, the appropriate pore size should be larger than 400 μm.³¹ For soft tissue regernation, the appropriate pore size should be smaller than 300 μm. Normally smaller pores result in more fibrotic tissue formation. Larger pore size is beneficial for the growth of blood vessels, and the diameter of a pore smaller than 400 μm limits the growth of blood vessels, and results in a smaller blood vessel diameter.³¹

Figure 1.

The representative traditional 3D electrospun nanofiber scaffolds. (A) (i) Electrospun PCL nanofiber mat with 160 μm pores created by acupuncture needles, (ii) High magnification image of created 160 μm pore.²⁸ (B) (i) 3D nanofiber scaffold composed of 15 layers of fiber stripes with basket-weaved structure and regular pores, (ii) 3D nanofiber scaffold composed of 2 layers of fiber stripes.²⁹ (C) (i) Fluorescently-labeled PCL (red) and PEO (green) fibers (sacrificial fiber) showed pronounced alignment and interspersion. (ii) The residual PCL nanofibers after remove PEO nanofibers in an aqueous solution.³⁰ (D) (i) Schematic illustration of the cold-plate electrospinning technique used to produce highly porous nanofibers. (ii) Comparison of electrospun scaffold made by traditional electrospinning, salt leaching electrospinning and cold-plate electrospinning techniques.³¹ (E) (i) Scheme for creating a cotton ball-like electrospun scaffold using spherical dish and metal array. (ii) A cotton ball-like 3D PCL nanofiber scaffold.³² (F) (i) Photograph of 3D hybrid scaffolds, (ii) it composes of microfibers produced by melt electrospinning and nanofibers generated by electrospinning.³³

In the macrostructure, these woven scaffolds had tailored pore sizes, which was able to significantly improved the cell infiltration. In the microstructure, however, the scaffolds were made of fine nanofiber strips or bundles, and cells were incapable of penetrating into the strip materials. Using a different strategy, Mauck et al. develop a sacrificial templating method to prepare a 3D electrospun scaffold composed of PCL (a water-insoluble polymer) and PEO (a water-soluble polymer) nanofibers. The higher porosity of nanofiber mats were obtained by removing the sacrificial PEO nanofibers, and the porosity increased with increasing the concentration of PEO nanofibers from 5% to 80%. The infiltration and distribution of seeded mesenchymal stem cells were dramatically improved compared to PCL mats (Figure 1C).³²

Besides the above-mentioned 3D scaffolds made of nanofiber mats, several studies attempted to fabricate porous 3D nanofiber scaffolds. Park et al. employed a cold plate to collect the crystallized silk fibroin nanofibers during electrospinning which subsequently form a 3D fibrous scaffold after removing the ice through lyophilization (Figure 1D).³³ Jun et al. fabricated a cotton ball-like 3D porous nanofiber scaffold using a specially designed collector - a spherical dish embedded with an array of metal probes. Cells infiltrated into the cotton ball-like scaffold after 7 days of culture, while no cell penetration was observed in traditional electrospun mats (Figure 1E).³⁴ Hutmacher et al. developed a hybrid 3D scaffold consisting of microfibers and nanofibers by combining melt electrospinning writing and electrospinning. The melted polymer was used to build a deck of microfibers, and then nanofibers were deposited on the surface before the next layer of microfibers was laid on (Figure 1F).³⁵ Although the pore size and porosity of 3D nanofiber scaffolds made by aforementioned technologies are obviously increased compared to the traditional nanofiber mats, these porous nanofiber scaffolds are mainly composed of random nanofibers. Topographic cues rendered by aligned nanofibers play a siginificantt role in regulation of cell behavior (e.g., neurite outgrowth). For examples, the comparative study of neurite outgrowth in the random and aligned nanofibers indicated that the random nanofibers could interfere with the aligned nanofibers to influence the pattern of neurite outgrowth and impede axonal regeneration.³⁶ The influence of the random nanofibers could be diminished by preseeding the inner surface of a bilayer NGC with Schwann cells because both the morphology and cytoskeleton structure of Schwann cells were only affected by the topmost layer of fibers.³⁶ In another study, the direction of neurite growth on uniaxially aligned, electrospun nanofibers was demonstrated to be either parallel or perpendicular to the direction of fiber alignment. The direction is determined by a set of parameters including fiber density, surface chemistry of the fibers, and surface property of the supporting substrate.³⁷ At this point the porous 3D scaffolds consisting of aligned nanofibers would be an appealing choice to allow the cells to interact with their surrounding microenvironment in all three dimensions, similar to the in vivo settings where cells live, in particular for neurons, muscle cells, and tenocytes.

3. New forms electrospun nanofiber materials and their biomedical applications

3.1. Expanded nanofiber scaffolds

In order to address the aforementioned problems that have plagued the electrospinning field for decades, several novel methods based on 2D mat expansion were developed.^{23, 38, 39} These new approaches not only solved the problem of compact structure, but also achieved control over fiber alignment within scaffolds, which may greatly expand their applications in the biomedical field [Table 1].

Table-1:

New forms of nanofiber materials and their biomedical applications.

Nanofiber Composition	Postprocessing methods	Advantages	Ref.
Collagen/PCL	Micropore generation	Rapid cellular infiltration	28
PCL	Noobing	Uniform cell expansion	29
PCL/PEO	Sacrificial template	Rapid cellular infiltration	30
Silk Fibroin	Salt leaching	Rapid cellular infiltration	31
PCL	NaBH4-Hydrogen gas	Increased pore size-Rapid cellular infiltration	35,36,37, and 38
PCL	CO2	Increased pore size-Rapid cellular infiltration	34
PLGA/Collagen/gelatin	Mechanical cutting-Short nanofiber	Injectability	81
PLGA/Collagen/gelatin/Bio-glass	Short nanofiber-Aerogel	Increased pore size	88
SiO2-Chitosan	Short nanofiber-Aerogel	Increased pore size	91
PCL	Short nanofiber-TISA-Aerogel	Increased pore size	86, and 87
PCL/gelatin	Short nanofiber-Nanofiber microsphere	Injectability	95, and 96

3.1.1. Gas foaming in an aqueous solution

The first developed expansion method was based on an innovative gas foaming technology. Briefly, the hydrolysis reaction of NaBH₄ generates hydrogen gas bubbles capable of expanding both aligned and random electrospun nanofiber mats in the z direction, resulting in the formation of porous architectures and simultaneous maintenance of imparted anisotropic cues (Figure 2A(i–v)).^{23, 40} Based on the same principle, different gas-foaming approaches were used to generate gas bubbles for expansion of nanofiber mats. He et al. and Zhao et al. utilized the alcoholysis reaction of NaBH₄ to produce hydrogen bubbles to expand nanofiber scaffolds.^{41, 42} In other studies, Liu et al. and Mi et al. reported a CO₂ gas foaming technology based on different solubilities of CO₂ in ethanol and water.^{43, 44} Briefly, electrospun nanofibers were deposited into an ethanol/dry ice bath to form CO₂ saturated nanofiber scaffolds. Then, the scaffolds were immersed into a water bath and the escaped CO₂ gas bubbles caused expansion of scaffolds.

Figure 2.

New forms of 3D electrospun nanofiber scaffolds with high porosity and good alignment. (A) The expanded 3D nanofiber scaffold generated by modified gas foaming technology. (i) The schematic shows the 3D structure of expanded PCL nanofiber peanuts. The red arrow indicates the direction of fiber alignment. (ii) Photographs showing the morphology of PCL nanofiber mats before and after expansion. (iii) Photograph showing a pile of PCL nanofiber peanuts. (iv) The thickness distribution of PCL nanofiber peanuts. (v) The cross-section structures (Y-Z, X-Z, X-Y planes) of PCL nanofiber mats before and after expansion.³⁶ (vi) The H&E staining of unexpanded, 3-mm-thick and 10-mm-thick PCL expanded nanofiber scaffolds after 8 weeks subcutaneous implantation.⁴¹ (B) The expanded 3D nanofiber scaffold generated by CO2 depressurization technology. This method is suitable to expand both hydrophobic (i) and hydrophilic (ii) polymers. (iii) The CO₂ depressurization method can maximum retain the loaded coumarin 6 compared to the modified gas foaming technology. (vi) (a) H & E staining and Masson’s trichrome staining of 3D expanded nanofiber scaffolds with arrayed holes after 1 week, 2 weeks and 4 weeks subcutaneous implantation. Green dots indicate the boundary of cell filtrated area. Green arrows indicate collagen deposition.³⁴

Subcutaneous implantation to rats revealed that expanded nanofiber scaffolds had outstanding histocompatibility(Figure 2A(vi)).²⁸ The 85% and 100% of the regions in 3-mm and 8-mm thick expanded scaffolds exhibited cellular infiltration after 8-week implantation, respectively. While cell penetration only occurred on the surface of traditional nanofiber mats, infiltration occurred from the sides of expanded scaffolds. Within the cell infiltrated area, a number of new blood vessels and a large amount of collagen were detected, suggesting the formation of new dense tissue. Moreover, no obvious fibrotic tissues were observed in the edge between host tissues and implanted scaffolds (Figure 2A(vi)).²⁸ In a separate study, Xie et al. reported superelastic and injectable PCL nanofiber “peanuts” fabricated by gas-foaming for potential applications in management of noncompressible torso hemorrhage and junctional hemorrhage. These PCL nanofiber “peanuts” not only had excellent shape-recoverability in air, water, and blood, but also exhibited greater blood absorption capacity compared to hemostatic materials (e.g., gauze, gelfoam) used in clinics today. Furthermore, the nanofiber “peanuts” demonstrated the efficacy of hemostasis in a swine liver injury model.⁴⁰ In another study, Xie et al. attempted to use expanded PCL/gelatin (50:50) nanofiber scaffolds with arrayed holes for engineering skin tissue. The cells migrated from seeded GFP-labelled human fibroblast spheroids to the surface and surrounding nanofiber walls of punched holes, and proliferated well in the scaffolds.⁴⁵ Similarly, Wen et al. developed layered Chitosan/PVA (3:7) scaffolds by gas foaming for skin wound healing.⁴⁶ The nanofiber scaffolds with layered structures promoted regeneration of the dermis, and reduced scar formation in the regenerated skin.⁴⁶ Other than the applications in hemostasis and skin wound healing, the expanded (PLA)/silk fibroin nanofiber sponge was filled into a hollow nerve guidance conduit to form an engineered nerve graft.⁴⁷ The graft was not only able to promote the proliferation of Schwann cells in vitro, but also improve the functional recovery of regenerated peripheral nerve in vivo.⁴⁷

3.1.2. Depressurization of subcritical CO₂ fluid

The expanded scaffolds fabricated by gas foaming in an aqueous solution overcome some of the shortcomings of traditional nanofiber mats. However, there are still some unresolved issues with the expansion procedures: (i) This method applies only to nanofibers made of hydrophobic polymers; (ii) The NaBH₄ may react with polymers or loaded substances; (iii) The encapsulated bioactive agents may lose functionality during the process of expansion in the aqueous solution, as NaBH₄ is a strong reducing agent and may impair the biactivities of loaded components. In order to avoid these problems, the depressurization of subcritical CO₂ fluid was developed. In brief, the nanofiber mat was immersed in subcritical CO₂ fluid followed by rapid depressurization. During the depressurization, the volume of CO₂ rapidly expands, leading to expansion of nanofiber scaffolds.³⁸

As shown in Figure 2B, the CO₂ depressurization method was suitable for expanding nanofiber mats made of both hydrophobic polymers (PCL nanofibers) (Figure 2B(i)) and hydrophilic polymers (PVP nanofibers) (Figure 2B(ii)). The porosity of the nanofiber scaffolds increased from 78.5% for the raw nanofiber mats to 92.1% and 99.0% after the first and second expansion by depressurization of subcritical CO₂ fluid, which are comparable to the scaffolds fabricated by gas foaming in the aqueous NaBH₄ solution. In addition, this method was able to retain the integrity of encapsulated chemicals (coumarin 6) and bioactive molecules (antibacterial protein LL37) to a greater extent compared to the NaBH₄ gas foaming method (Figure 2B(iii)). Moreover, the antibacterial activity of loaded LL37 was maintained before and after expansion by depressurization of subcritical CO₂ fluid. Tissue engineering and regenerative medicine often necessitate the topical delivery of signalling molecules.^{48, 49} For instance, Jansen et al. developed a stromal cell-derived factor-1α (SDF-1α) encapsulated PCL/gelatin electrospun membrane to promote bone marrow stromal cells (BMSCs) migration for accelerating bone regeneration.⁵⁰ Sethuraman et al. reported a VEGF and bFGF co-loaded poly (l-lactide-co-caprolactone) (PLCL)/poly (2-ethyl-2-oxazoline) (PEOz) nanofiber matrix for cardiac regeneration, which significantly enhanced angiogenesis and aided the functional recovery of ischemic heart.⁵¹ Nguyen et al. fabricated a VEGF and PDGF-BB loaded chitosan/poly(ethylene oxide) (PEO) nanofiber scaffold for normal and diabetic wound healing, faster granulation tissue formation and collagen deposition were observed.⁵² Interestingly, significant cellular infiltration, neotissue formation, blood vessel formation and ECM deposition were observed in the expanded 3D nanofiber scaffolds without growth factor loading after subcutaneous implantation. (Figure 2B(iv)). On this premise, we can speculate that the expanded nanofiber mats loaded with growth factors may achieve better efficacy in bone, cardiac and skin regeneration.^50–51 Besides growth factors, herbal extracts,^{53, 54} antibiotics,⁵⁵ enzymes,⁵⁶ micro-RNAs,⁵⁷ DNA plasmids,⁵⁸ peptides,⁵⁹ and nanoparticles⁶⁰ can also be incorporated into 3D expanded nanofiber scaffolds to achieve anti-bacterial, anti-inflammation, anti-oxidant, pro-adhesion, pro-proliferation, pro-angiogenesis and other functions that play complementary roles in regenerative medicine.

3.1.3. Solids of revolution inspired expansion

The gas foaming in an aqueous solution and depressurization of subcritical CO₂ fluid technologies are able to transform traditional nanofiber scaffolds into expanded, layered-structured nanofiber scaffolds with and without containing bioactive molecules available for tissue engineering and regenerative medicine. However, the nanofiber scaffolds generated by these technologies are still associated with several limitations: i) limited to the uniaxial fiber alignment; ii) limited to cuboid shape; iii) limited to the cell infiltration from the sides of each layer.³⁹ Inspired by solids of revolution, rotating expansion was further developed to manufacture complex nanofiber scaffolds with controlled alignment, porosity, and various shapes.³⁹ For example, a nanofiber mat was cut into a rectangular shape, then thermal treatment was used to fix one side of the rectangle mat. When the fixed side was perpendicular to the direction of nanofibers (Figure 3A(i)), a cylindrical-shaped nanofiber scaffold was obtained after expansion in NaBH₄ solution (Figure 3A(ii) and Figure 3A(iii)). The X−Y plane of the cylinder consisted of radially aligned nanofibers, while the X−Z and Y−Z planes showed a highly porous structure (Figure 3A(iv)). Similarly, when the fixed side was parallel to the direction of nanofiber alignment (Figure 3C(i)), a cylindrical-shaped scaffold was also obtained following expansion in NaBH₄ solution (Figure 3C(ii) and Figure 3C(iii)). In contrast, the X−Y plane of the cylinder showed a highly porous structure, while the X−Z and Y−Z planes showed a channel structure (Figure 3A(iv)). Subcutaneous implantation of these two types of nanofiber scaffolds in rats showed significant cell infiltration and migration throughout the scaffolds, even after 1 week. Especially in the transection, the morphologies of formed neo-tissues showed radial and porous structures in the radially and vertically aligned scaffolds, respectively (Figure 3B and Figure 3D). Xia et al. developed radially aligned PCL nanofiber membranes as dural substitute for wound closure and tissue regeneration by utilizing a specially designed collector composed of a central point electrode and a ring electrode.⁶¹ The thickness of membranes fabricated by this method is often less than 1 mm and pore sizes are small due to densenly packed nanofibers, thus limiting their potential applications. By comparison, there is no limitation for the cylindrical-shaped scaffolds consisting of radially aligned nanofiebrs in terms of thickness, and the pore size and porosity can be tuned by simply changing the initial membrane thickness. With few limitations, such scaffolds hold great potential for wound healing by accelerating the migration of keratinocytes and fibroblasts to promote re-epithelialization and granulation tissue formation. In addition, nerve fiber layers in the rat retina exhibits radially-aligned axons extending towards the optic nerve head. Similarly, rat retinal ganglion cells cultured on radially aligned nanofiber membranes could recapitulate the axonal orientation of the nerve fiber layer of the native retina.^{62, 63} The regeneration of corneal stroma was also investigated in other studies. In order to orient and integrate the retinal ganglion cells and damaged corneal stroma, a corneal stroma was designed with a radially aligned structure.⁶⁴ Ahearne et al. prepared a decellularized corneal ECM modified radially aligned PCL nanofiber membrane for corneal stroma regeneration, which was able to significantly enhance the migration of human corneal stromal cells.⁶⁵ Park et al. fabricated a transparent hemispherical 3D PCL/Collagen nanofiber scafolds with radially aligned pattern, which could mimic the 3D morphology of the eye. It was not only capable of promoting the migration of rabbit corneal cells, but also enhancing the expression of zona ocludin-1 (ZO-1) that plays an important role in cellular organization or corneal epithelium.⁶⁶ It is expected that the cylindrical-shaped scaffolds consisting of radially aligned nanofibers made with a biocompatible polymer (e.g., type I collagen and hyaluronic acid) could be promising for corneal regeneration.

Figure 3.

The emerging 3D electrospun nanofiber scaffolds with controlled alignment. (A) The expanded 3D radially aligned nanofiber scaffold generated by thermo fixation and rotating expansion. (i) Schematic of one side fixed rectangular PCL nanofiber mat, the fixed side is perpendicular to the direction of nanofibers. (ii) Schematic of expanded 3D radially aligned nanofiber scaffold. (iii) Photograph of the resultant cylinders. (iv) SEM image showing the X-Y plane made of radially aligned nanofibers and the porous structure of X-Z and Y-Z plane. Arrows indicate the direction of fiber alignment. (B) H&E staining showing cell infiltration in the expanded, radially aligned PCL nanofiber scaffolds. (C) The expanded 3D vertically aligned nanofiber scaffold generated by thermo fixation and rotating expansion. (i) Schematic of one side fixed rectangular PCL mat, the fixed side is parallel to the direction of nanofibers. (ii) Schematic of expanded 3D vertically aligned nanofiber scaffold. (iii) Photograph of the resultant cylinders. (iv) SEM image showing the porous structure of the X-Y plane and channel structure of the X-Z and Y-Z plane. Arrows indicate the direction of fiber alignment. (D) H&E staining showing cell infiltration in the expanded, vertically aligned PCL nanofiber scaffold.³⁵

Both gas foaming in an aqueous solution and depressurization of subcritical CO₂ fluid technologies can be applied to solids of revolution-inspired expansion. The depressurization of subcritical CO₂ fluid method is more suitable for expanding drug loaded nanofiber membranes for biomedical use. Based on this approach, Xie et al. prepared a 25-hydroxyvitamin D₃ (25(OH)D₃) eluted, radially-aligned PCL nanofiber scaffolds and evaluated their host response after subcutaneous implantation of the scaffolds to human immune system-engrafted mice.⁶⁷ The local delivery of 25(OH)D₃ from the radially-aligned PCL nanofiber scaffolds was able to regulate the immune response to a pro-regenerative status, indicated by higher pecentrage of infiltrated macrophages in M2 phase. Meanwhile, the topically released 25(OH)D₃ from scaffolds increased the production of human cathelicidin LL-37 at or near the implantation site. This strategy could potentially help improve the failure rate of current medical devices by reducing inflammatory responses and enable implementation of drug-eluting 3D scaffolds/tissue constructs for modulating immune response during tissue regeneration. In a different study, Bellamkonda et al. engineered a “tumor guide” device composed of aligned PCL nanofiber membrane and collagen hydrogel to guide the invasive tumor cell migration away from the primary tumor site to an extracortical location, then induce apoptosis.^{17, 68} The cylindrical-shaped scaffolds consisting of longitudinally aligned PCL nanofibers could attract more cancer cells compared to the 2D nanofiber membrane. Moreover, a variety of molecules can be immobilized to the surface of scaffolds in a graded fashion to further improve the cancer cell attraction. In addition, different molecules could be incorporated to 3D nanofiber scaffolds for local chemotherapy, immunotherapy, and/or radiation therapy.

3.2. Origami inspired transformation

Other than expansion technology, 2D electrospun nanofiber membranes can be converted to 3D scaffolds with certain shapes based on the origami and kirigami concept. Origami and kirigami mainly involve rolling, bending, folding, stacking, wrinkling, buckling, or cutting of sheets. Based on this concept, various nanofiber scaffolds were prepared including tubular-shaped scaffolds³⁶ and layer-by-layer stacked scaffolds⁶⁹. For example, Wang et al. constructed a 3D multilayered tissue using on-site layer-by-layer cell assembly while electrospinning.⁶⁹ Briefly, a layer of collagen/PCL (1 : 3) was first electrospun onto the wire loop, and then 1 mL of fibroblast suspension (1×10⁵ cells/mL) was seeded onto the nanofibers. By repeating the above steps, nanofibers/fibroblasts alternated multilayered artificial dermal tissue were formed. Similarly, the tissue engineering skin consisting of nanofibers-fibroblasts mimicking dermal / nanofibers-keratinocytes mimicking epidermal was also fabricated. With the assistance of a mold, Wang et al. fabricated a 3D tissue construct by stacking human fetal osteoblast-seeded PCL/nHA composite nanofiber membranes for construction of 3D tissues.⁷⁰ The cells seeded on both sides of nanofiber membranes facilitated the bonding of the adjacent membranes due to the secreted ECM. Some potential limitations are associated with this approach. Mechanically, some nanofiber membranes may not be rigid enough to bear the transformed 3D structures. The prepared 3D nanofiber scaffolds are still composed of 2D nanofiber membranes, which can seed cells only on the surface, severely limiting cell infiltration within membranes in vivo. In addition, mass production of such 3D scaffolds could be tedious, difficult, and require a very large amount of cells.

3.3. Short nanofibers

Over the past two decades, research into minimally invasive therapy has intensified due to enabling technological advances in tissue engineering and the ease at which clinical implementation, namely injection, can occur.^{71, 72} Compared to injectable hydrogels, electrospun nanofiber scaffolds are less flexible and unable to be injected and have since garnered much less interest as minimally invasive therapeutics.^73–75 To circumvent these limitations, researchers attempt to make new forms of electrospun nanofiber materials through various post-processesing procedures. In this scenario, the post-processing methods are always required to produce the short fibers from the as-spun continuous nanofibers. The initial attempt was reported by Stoiljkovic et al.⁷⁶ In their study, short electrospun fibers were prepared from a nanofiber membrane using a UV-cutting method. Photo cross-linkable polymer nanofibers resulted after irradiating their crosslinkable moieties using UV light in the presence of mask with defined slits can produce short nanofibers. However, using UV cutting is limited to polymers with UV crosslinking moiety. Therefore, Yoshikawa et al. utilized mechanical cutting to create smaller pieces of nanofiber membranes followed by homogenization in water, giving rise to individual fibers.⁷⁷ Both abovementioned techniques always need a second or third step to make a well-dispersed short nanofiber slurry for further applications. In order to reduce the additional processing steps, the direct fabrication of short fibers using the electrospinning method has been reported by Luo et al.⁷⁸ This method directly produced short micro-fibers with an aspect ratio of 10–200 μm by altering the molecular weight of polymethylsisesquioxane and using a volatile solvent. In another study, Fathona et al. reported an interesting way to make short nanofibers with electric spark-induced cutting method, this study demonstrates that the length of the short nanofiber can be tuned by changing the needle size.⁷⁹ Neither methods of direct short fiber synthesis wield precise control of fiber length, necessitating future research related to individual short fiber preparation.

Nanofiber scaffolds as implant have been widely examined in bone regeneration.^80–82 However, few studies have applied electrospun nanofiber materials for regenerating bone in a minimally invasive approach, likely due to the laborious processing and lack of functional group for cargo conjugation.^73–75 Recently, Xie et al. prepared hydroxylapatite-coated short nanofibers using mechanical cutting followed by mineralization in SBF.⁸³ The mineralized short nanofibers, after chelating calcium-binding osteoinductive peptides, allowed for sustained peptide release for 4 weeks. Injection of short functionalized nanofibers as bone fillers demonstrated increased healing of critical-sized alveolar bone defects in rats. The functionalized short nanofiber fragments may have a great potential as minimally invasive therapies to treat bone and tissue defects as they are injectable and capable of filling irregular-shaped defects without necessitating surgical implantation.

3.4. Nanofiber aerogel

Directly applying short nanofibers as injectable therapies may offer some valuable non-surgical tools for regnerative medicine, but in other cases, rigid structures may be desirable. In this case, short nanofibers can be used as building blocks to fabricate porous, 3D nanofiber scaffolds called nanofiber aerogels. Two main approaches have been developed. One method involves the following steps: i) nanofiber production by electrospinning, ii) preparation of short nanofibers, iii) short nanofiber dispersing, templating or molding, and iv) freezing, freeze-drying, and cross-linking.^84–86 The other method differs in the short nanofiber dispersing and templating step by forming short nanofiber aggregates through thermally induced self-assmebly.^{39, 87–89} Generating electrospun nanofiber mats from the desired materials is the initial step, then the fiber mats would be fragmented into short nanofibers either mechanically or by another method, such as with electric sparks, as described in the previous section (Short nanofiber). If the nanofiber mat is brittle, simple homogenization under ultrasonication is enough to make short nanofibers with lengths ranging from 20 μm to 50 μm.⁹⁰ The pore size of nanofiber aerogels can be readily tuned by changing freezing temperatures. In addition, the anisotropic/directional freezing on the well-homogenized short nanofiber slurry at different temperatures can be used to generate oriented porous structures with different pore sizes. However, after freeze-drying, the long-term stability of the aerogel may be poor due to the lack of weak bonds between the fibers. Opting to chemically or thermally crsslink or pyrolyzed nanofiber-based aerogels will significantly enhance their mechanical stability for long periods of time.⁸⁴

Initially, nanofiber aerogels were developed in catalysis and oil-water separation.^{86, 91, 92} Due to the biomimetic property and highly porous structure, resarchers explored the applications of nanofiber aerogels in tissue regeneration. Xie et al. recently prepared a hybrid aerogel composed of electrospun PLGA, collagen, gelatin, and Sr–Cu co-doped bioactive glass nanofibers with immobilized BMP-2 peptides attached to aerogels through calcium-binding E7 domain. The bone regeneration of the hybrid aerogels were evaluated critical-sized cranial bone rat defect model.⁹⁰ Figure 4A shows the fabrication process and SEM images of the primary pores and secondary pores in the aerogel. In this study, the porosity, mechanical property, and composition of aerogels were optimized by varying the freezing temperatures, the ratios between polymer and glass nanofibers, cross-linking temperatures, and duration of crosslinking. Calvarial bone healing and defect closure significantly improved following implantation of E7-BMP-2 peptide-incorporated hybrid aerogels after 8 weeks, compared with unfilled defects, and aerogels alone. More recently, Li et al reported a superelastic, ceramic, nanofiber aerogel consisting of intrinsically rigid, structurally flexible electrospun SiO₂ nanofibers using chitosan as bonding sites (SiO₂ NF-CS) via a lyophilization technique.⁹³ The superelastic SiO₂ NF-CS aerogels were able to self-fit to mandibular defects in rabbits and promote calvarial bone formation in an osteoporotic rat model Figure 4B. In their study, BMSC-loaded SiO₂ NF-CS (Figure 4B, (i–iv)) scaffolds showed enhanced calvarial bone regeneration after 10 weeks.

Figure 4.

(A) Schematic representation of the four main processing steps for the preparation of hybrid aerogel and its structure at different length scales.⁸⁸ (B) (i) Schematic representation of hMSC loaded aerogel for cranial bone repair. (ii) Photographs of shape recovered SiO2 NF-CS scaffold upon implantation in round shaped mandibular defects in rabbits, indicating an implantation via minimal invasion. (iii) 2D sectional CT image of bone defects, showing close contact of scaffolds with host bones. (iv) 3D reconstructed CT images of rat cranial bone defects at 5 and, 10 weeks post-surgery. Red circles labeled surgery sites. (v) Bone volume fraction and (vi) bone density at 5 and, 10 weeks post-surgery.⁹¹

In different studies, Fong et al. prepared PCL nanofiber aggregates through thermally-induced self-assmebly (TISA) and examined their bone regeneration capacity in an ectopic bone model.⁸⁷ The TISA-derived nanofiber scaffolds were both porous and mimicked the fibrous architecture of the ECM to guide the cellular infiltration and form new bone tssue. Using a similar method, the group also tested PCL/PLA-3D nanofiber aggregates for repairing a 5-mm calvarial bone defect in rats. The PLA composed PCL/PLA-3D shows better mechanical property as compared with PCL-3D group.^{88, 89} Additionally, the product of PLA degradation, lactate, induced angiogenesis and ALP activity. The in vivo results revealed that PCL/PLA-3D scaffolds facilitated better bone formation in a cranial bone defect mouse model as compared to PCL-3D scaffolds, though the mechnism by which regeneration was improved is unknown. In addition to bone regeneration, Mo et al. reported the use of nanofiber aerogels for cartilage regeneration.⁹⁴ They first prepared a 3D scaffold (3DS-1) built from electrospun gelatin/PLA nanofibers and tune the mechanics of 3DS-1 by heat and water treatments. The 3DS-1 possessed porous and nanofibrous structure, which could mimic the structure of natural ECM; in addition, it presented superabsorbent properties and could improve the growth of chondrocytes in vitro. To further improve the regenerative efficacy, a modified scaffold (3DS-2) cross-linked with hyaluronic acid (HA) was also fabricated. The 3DS-2 with and without seeding chondrocytes were tested in a rabbit articular cartilage injury model. However, the 3DS-2 group showed thicker new cartilage and significant deposition of type collagen II and aggrecan, which indicated 3DS-2 would promote the regeneration and remodeling of the cartilage defect. HA is a native component of ECM in cartilage, which may explain how the HA-based scaffold further enhanced cartilage repair compared to scaffolds lacking HA.

3.5. Nanofiber microspheres

Deriving a simple and versatile scaffold for multiple applications is a major focus of biomaterials science.¹ Particularly, simple ECM-mimicking scaffolds with nanofibrous web-like architectures were widely studied due to the rapid reintegration with host tissue after implantation.⁹⁵ Moreover, the injectability and biomimetic nature of nanofiber microsphere (NMs) received increased attention compared with hydrogels and other scaffolds as their small size and ability to be injected using small needles rendered them an important tool in minimally invasive therapy. Ma group was the pioneer in the development of NMs from PLLA polymers using self-assembly.⁹⁴ Such NMs have various biomedical applications including cell and growth factor delivery. However, NMs fabricated with self-assembly is specifically limited to polymers containing certain functional groups (-OH, -NH₂, and -COOH).⁹⁶ To circumvent these drawbacks, Xie et al. reported a new method for fabricating NMs by combining electrospinning and electrospraying and explored their potential applications in biomedical therapies.⁹⁷ NMs were formed by electrospraying aqueous dispersions of length-defined electrospun nanofiber segments (obtained by cryocutting or homogenization) into liquid nitrogen followed by freeze-drying and thermal treatment (Figure 5A). The NMs size can be controlled by varying the applied electrical field and solution flow rate during the electrospraying process. NMs with different structures - porous and hollow - were achieved. Furthermore, a broad range of polymer and inorganic bioactive glass nanofiber-based NMs were fabricated by electrospraying short nanofiber dispersions, indicating tremendous design flexibility. Very recently, Xie et al. enhanced the bioactivity of NMs by tethering BMP-2 and vascular endothelial growth factor (VEGF)-mimicking peptides onto poly(ε-caprolactone) (PCL):gelatin:(gelatin-methacryloyl)(GelMA)(1:0.5:0.5) NMs by photocrosslinking the methacrylic group in GelMA and octenyl alanine (OCTAL) in the modified peptides.⁹⁸ Figure 5B shows the SEM images of PCL:gelatin:GelMA nanofiber-composed functional NMs. The conjugated BMP-2-OCTAL on NMs significantly promoted osteogenic differentiation of BMSCs Figure 5C show the OPN expression (pink) of the differentiated BMSCs. Moreover, human umbilical vein endothelial cells (HUVECs) seeded on VEGF-mimicking peptide QK-OCTAL-tethered NMs significantly up-regulated vascular-specific marker CD31, indicating enhanced micro vascularization (Figure 5D).

Figure 5.

(A) Scheme illustrating the fabrication of peptide tethered NMs and their applications. (B) SEM images showing NMs composed of PCL:gelatin:GelMA (1:0.5:0.5) nanofiber segments (i, ii) before and (iii, iv) after crosslinking (C) Confocal microscopy images showing the OPN expression of BMSCs seeded on PCL:gelatin:GelMA (1:0.5:0.5) NMs with conjugation of BMP-2 peptides after culturing in the osteogenic differentiation medium for 14 days. (D) Confocal microscopy images showing tubular network formation from HUVECs seeded on PCL:gelatin:GelMA (1:0.5:0.5) NMs with conjugation of QK peptides in the simulated medium for 7 days.⁹⁶

4. Future directions

After reviwing several new forms of nanofiber materials and their potential biomedical applications, it is clear that enhancing the 3D structure and functionality of such materials will be the key to their clinical implementation. Moreover, stepping into the new dimension of nanofiber materials with different 3D physical forms is importnat for further understanding how ECM mimicry modulates tissue repair and regeneration. New forms of nanofiber materials will allow for the development of new materials geared toward answering fundamental biological questions and obtaining more applied technological advancements in tissue engineering/modeling. This short perspective reviews the emergence of techniques used in transforming 2D nanofiber materials into 3D, ECM-mimicking tissue regeneration scaffolds. Improvements to procedures used to transform nanofiber mats to 3D structures are still necessary. For example, during nanofiber mat expansion by gas foaming in an aqueous solution, the generated gas bubbles can be stabilized by surfactant, thus speeding up expansion rate. Future studies may explore the addition of surfactants to nanofibers during electrospinning in the hopes that the surfactants may increase expansion during gas foaming. The current expansion technology can produce 3D nanofiber scaffolds with cuboid shapes and solids of revolution-inspired shapes. Irregular-shaped 3D nanofiber scaffolds could be produced by combining gas-foaming and 3D molding technologies using 3D printing or templating. Expanded nanofiber scaffolds in previous studies were mainly made of PCL. Although PCL is biocompatible and biodegradable, the degradation rate is relatively slow and may not be suitable for regeneration of certain type of tissues. Nanofibers with different compositions should be explored, in particular, polymers with faster degradation rates or in combination with natural ECM components. Cell infiltration within expanded scaffolds was mainly demonstrated in in vitro cell culture and in vivo subcutaneous studies. Further studies should examine and identify optimal 3D nanofiber scaffolds for repairing specific tissue defects. Cylindrical-shaped scaffolds consisting of radially aligned nanofibers could provide contact guidance and enhance cell migration rendered by anisotropic topographic cues, meanwhile chemical gradients can be generated in such scaffolds to further promote cell migration through chemotaxis.

Although nanofiber aerogels show promise in tissue regeneration, they face criticisms for having poor mechanical propertieds. Particularly, the mechanical property of nanofiber aerogels is relatively poor due to their highly porous structure, making them quite brittle. Utilization of nanofiber aerogels for repairing load-bearing tissues could be a problem. Hydrogels could be incorporated to nanofiber aerogels to enhance their mechanical properties. Further, the pore size of nanofiber aerogels is not optimized for cell infiltration. Nanofiber aerogels lack nanotopographic cues that can direct cell migration despite having oriented porous structures. 3D printed scaffolds could be used as templates to generate patterned channles within nanofiber aerogels, which could promote cell infiltration.

3D-printed scaffolds have attracted a lot of attention in regenerative medicine due to the precise control over pore size, porosity, and geometric shape. However, these scaffolds lack nanoscale biomimetic topographies and insufficiently incorporate signaling molecules. Recently, Xie et al. developed a method of decorating short nanofibers to 3D-printed scaffolds with surface conjugation of BMP-2 peptides, rendering scaffolds with nanofibrous surface morphology and osteoinductive property without affecting their bulk mechanical property.⁹⁹ The decorated scaffolds significantly promoted adhesion and proliferation of pre-osteoblasts and BMSCs. The 3D-printed scaffolds with immobilized BMP-2 peptides showed enhanced mRNA expressions of osteogenic markers Runx2, Alp, OCN, and BSP in BMSCs, indicating the enhancement of BMSCs osteogenic differentiation. The amount of BMP-2 immobilized to 3D printed scaffolds and its release profile should be optimized based on in vivo test in the future.

NMs and short nanofibers shows great promise in minally invasive therapy. Therefore, fabricating porous NMs from different polymers to tune degradation and mechanical properties to exactly match defective tissues would significantly increase the clinical translation. Recently, Mao et al. reported a nanofiber-hydrogel composite that addresses these issues.¹⁰⁰ By incorporating interfacial bonding between electrospun PCL short fibers and a hyaluronic acid hydrogel network, they generated a composite that mimics the microarchitecture and mechanical properties of soft tissue ECM. Therefore, the short nanofibers and NMs bearing hydrid scaffolds may have great potential to mimic ECM for tissue engineering.