Polymer microspheres with the size of 1–500 μm have been used widely to encapsulate drugs. SF microspheres of different sizes have changeable penetrability, which can deliver drugs and active factors to different lesion sites in the human body. Compared with synthetic polymeric microspheres, biodegradable SF microspheres have several unique advantages in the delivery of various drugs, including proteins, nucleic acids, polypeptides, and other bio-macromolecules. Wang et al. evaluated the delivery of growth factors from polylactic co-glycolic acid (PLGA) and SF microspheres on human bone marrow-derived mesenchymal stem cells (hMSCs)-derived osteochondral differentiation. The result showed better osteogenic differentiation of hMSCs in the SF microsphere system because SF microspheres showed quicker sustained release of growth factors than PLGA microspheres. However, the fabrication process of SF microspheres is complicated (Wang et al., 2009). Gong et al. developed a polyethylene glycol (PEG) assisted emulsification method to encapsulate octreotide acetate with SF microspheres and octreotide could release slowly from SF microspheres for more than 28 days in rats after intramuscular injection (Gong et al., 2019). Shen et al. prepared BMP-2 encapsulated with SF microspheres in SF/nHAp porous scaffolds, which showed a sustained release of BMP-2 for 3 weeks (Shen et al., 2016). In another study, the composite alginate hydrogels combined with SF microspheres could provide a more sustained release of IGF-1 than microspheres without SF (Feng et al., 2020). The SF microspheres have a promising future as long-acting controlled release delivery. Hino et al. reported that the secondary structure of SF microspheres obtained by spray-drying changed from a β-sheet to a random structure. Due to water absorption and subsequent recrystallization, its conformation changes into β-sheet again. The β-folded structure is the most stable structure in the secondary structure of SF, which may affect its sustained release performance. The deficiency of the spray drying method is that the conformations conversion to the β-folded structure in SF microspheres may be obstructed if the particles are dried too fast. Lammel et al. produced SF particles by salting SF solution with potassium phosphate. Compared with emulsification, salt leaching voids the use of organic solvents and is friendly and safe. However, the whole experiment condition is in the environment of PH = 8, and the particles will gather into non-dispersible aggregates with poor pelletizing if PH = 5, suggesting that the change of PH value will affect the aggregation of SF microspheres (Lammel et al., 2010). Therefore, the specific production process for SF microspheres limited its practical application in biomedicine to a certain extent.

With unique biocompatibility and biodegradability, SF hydrogels are one of the effective candidate materials for biomedical engineering, with many properties such as capacity to hold water, stability of shape and modifications to the functionality possibility. Consequently, a three-dimensional microenvironment that mimics extracellular matrix can be created, allowing cellular behavior and tissue function to be regulated. In addition to physical and chemical hydrogels, the SF hydrogels make promising biopolymeric matrices that are capable of controlling drug release (Tomeh et al., 2019). Wu et al. incubated the diluted silk solution at 60°C for 24 h to make it self-assemble to form a physical hydrogel and then dissolved doxorubicin (DOX) directly into the hydrogel to form hydrogel-loaded DOX. DOX can be released in and out of the body for more than 4 weeks and tumors treated with DOX-loaded filamentous hydrogels showed a larger necrotic area than those treated with free DOX (Wu et al., 2016). Lovett et al. mixed SF with bevacizumab solution to induce gelation by ultrasound and the SF hydrogels achieve drug-release for more than 3 months (Lovett et al., 2015). Attention should be paid that the influences, such as an increase in local temperatures, an engineering shear, and gas-liquid interface increase, affected the gelation of SF during the ultrasound process.

Chen et al. mixed SF hydrogels with carboxymethyl chitosan (CMCS) into SF/CMCS hydrogels, which can be easily controlled in terms of thickness, quality, and shape by electrodeposition on graphite (Chen et al., 2021). Guziewicz et al. developed a lyophilized SF hydrogel, which makes SF hydrogels suitable for long-term sustained release applications in antibody release (Guziewicz et al., 2011). In terms of physical strength and stability, hydrogels made by chemicals are highly effective compared with physical hydrogels. The common preparation methods for chemical hydrogels include photopolymerization, chemical crosslinking agent, and irradiation. He et al. found that tannic acid-reinforced methacrylated chitosan/methacrylated SF hydrogels (TA-CSMA/SFMA) had a significant effect on wound healing in a full-thickness skin defect model (He et al., 2020). It should be noted that the addition of a photoinducer may cause adverse reactions, such as surface oxidation. Kim et al. optimized bone tissue engineering using SF composite hydrogels with HAP nanoparticles (NPs) by irradiating SF composite hydrogels with HAP nanoparticles (NPs) and these hydrogels enhanced the osteogenic differentiation of hMSCs in vitro (Kim et al., 2017). Zhou et al. prepared porous polyvinyl alcohol/SF/nano-hydroxyapatite (PVA/SF/N-HA) composite hydrogels by genipin (GP) crosslinking and the composite hydrogels could be a novel approach to corneal tissue engineering (Zhou et al., 2019). Therefore, both physical and chemical cross-linking can induce gelation of SF solution, and the corresponding SF gels loaded with drugs can be prepared with the conditions optimized using improved methodology according to the performance of loaded drugs and experimental purposes. In biomedical applications, SF hydrogels are limited by their poor mechanical properties and swelling behavior (Yang et al., 2022). The biological safety of SF hydrogels and their composite hydrogel materials, hydrogels’ drug loading efficiency, and the excellent injection performance of hydrogels is important challenges in the field of regenerative medicine (Liu et al., 2022a).

SF coating can be obtained by dipping coating and layer-by-layer assembly. The application of SF coating can improve the “ initial burst release” of drugs and prolong the release time of drugs. Meanwhile, the behavior of drug-release can be precisely controlled by adjusting the thickness and crystallinity of SF coating. Pritchard et al. coated adenosine tablets with SF solution and put them into phosphate buffer solution (PBS) after methanol treatment for in vitro release. They found that the increase of the crystal content or the thickness of SF coating could deliver adenosine for several weeks and then degrade in the body, requiring no surgical removal (Pritchard et al., 2010). Wang et al. prepared SF coating by a step deposition process and used it as a carrier to incorporate drugs and the release duration of these adulterated molecules can be prolonged by controlling the coating structure and suppressing the initial burst (Wang et al., 2007b).

SF membrane can be obtained by solution pouring and freeze-drying. Lee et al. investigated antithrombogenicity and surface properties of SF and S-carboxymethyl keratin (SCMK) polymers by solvent cast, which show better antithrombotic properties than SF or SCMK alone (Lee et al., 1998). Seib et al. poured the SF solution onto the substrate and dried it to generate a SF membrane with different crystallinity. After different steam temperatures, dried SF membranes were immersed in doxorubicin solution and then were implanted in the corresponding tumor sites in breast tumor mice. The result showed that the different crystallinity of SF membrane produced at different temperatures provided a controlled release of doxorubicin and limited adverse reactions in the treatment of human breast cancer (Seib and Kaplan, 2012). The result suggests these systems as a good candidate material for SF coating or film to control drug release, when considering the all-aqueous process involved, coatings with conformal properties, and the degradability and biocompatibility of SF coating or film. A number of drawbacks, however, include expensive coatings, which together reduce affordability and commercial viability, and prevention of direct contact between loaded drugs and cells by coatings to the observation that insufficient manifestation of their effects may occur (Kwon et al., 2021).

SF porous scaffolds have well-interconnected pores and can well simulate the three-dimensional network environment in organisms (Moisenovich et al., 2019), to facilitate cell adhesion, proliferation, and growth (Abdel-Naby et al., 2017). The preparation methods of SF porous scaffold include freeze-drying, salt leaching, and porogen leaching. A frozen-dried SF/hyaluronan scaffold was prepared, and 3 weeks of mesenchymal stem cell culture were performed on the scaffolds by Garcia-Fuentes et al. In comparison to plain SF scaffolds, stem cell cultures on SF/hyaluronan scaffolds produced more effective tissue formation (Garcia-Fuentes et al., 2009). A salt leaching method was used to prepare demineralized bone matrix-SF (DBM/SF) scaffolds. The scaffolds can promote cell attachment and growth, and it may be a suitable candidate material for bone tissue engineering (Ding et al., 2015). However, traditional porous scaffolds generated by porogen can hardly control the distribution of scaffolds’ pores. Ze et al. combined supercritical carbon dioxide (SC-CO2) technology with porogen leaching using ammonium bicarbonate (AB) particles as a pore-forming agent, which could overcome the disadvantages of traditional porogen leaching and prepared scaffold was loaded with Schwann cells to demonstrate its good cytocompatibility (Li et al., 2019). Zhang et al. prepared bioactive glass (MBG)/SF scaffolds with aspirin to repair calvarial defects in the mice model. The MBG/SF scaffolds have been shown to enhance drug loading in vitro, increase drug release rate, and stimulate bone regeneration in vivo (Zhang et al., 2014).

SF conduits are more widely used in the preparation of nerve catheters (NC). The use of artificial nerve conduits (NCs) can be useful in the treatment of damaged peripheral nerves in certain cases, however, their therapeutic effect is often not enough effective (Madduri and Gander, 2012). Therefore, it is necessary to manufacture NCs to achieve a better therapeutic effect. Studies have shown that SF NCs promote axon regeneration in the short nerve gap thanks to the slow degradation of SFs (Wang et al., 2008; Cao and Wang, 2009). Schwann cells, in particular, can survive, adhere, and migrate along silk fibers due to their affinity for them (Yang et al., 2007). SF conduits are produced in three major ways: mold casting, staining, and electrospinning. Yang et al. used stainless steel casting molds by lyophilized demoulding to provide SF-NC and SF vessels that are implanted into the back of adult rabbits could degrade rapidly and meet the requirements of peripheral nerve regeneration compared with SF fibers (Yang et al., 2009). Ghaznavi et al. used catheterization immersed in SF/polyethylene oxide to examine the cellular inflammatory response and functional recovery in models of sciatic nerve defects and the catheter had physical and biological properties of promoting nerve repair from proximal to distal nerve stump (Ghaznavi et al., 2011). Wang et al. combined SF with PLGA to prepare nerve guidance conduits (NGCs), which showed the strength of good biocompatibility, stiffness, and high porosity (Wang et al., 2015). In addition to the traditional methods mentioned above, Carvalho et al. developed an SF-based cross-linking catheter for the delivery of neurotrophic factors. Studies have shown that this design has significantly improved retrograde transport, neuronal protection, and motor nerve reinnervation, and to some extent solved the defect of incomplete nerve regeneration caused by a lack of neurotrophic factors in the proximal region (Carvalho et al., 2021). You et al. fabricated a multi-channel bioactive SF nanofiber catheter, which can promote neuron-like development and directional neurite extension in vitro and in vivo spinal cord injury models (You et al., 2020). Despite its advantages, the scaffold can also be disadvantageous. It is difficult to distribute cells evenly and remain underpopulated in the central or core regions (Luetchford et al., 2020). The scaffold design relies on a bottom-up approach that may spread cells evenly and shape them into the final 3D structure by layering different cells or drugs (Chen et al., 2011). Likewise, simple hollow SF NGCs cannot enhance functional and sensory recovery after peripheral nerve injuries (PNIs). It has been shown that including neurotrophic factors (NTFs) in NGCs can significantly improve nerve regeneration (Carvalho et al., 2021). It is important, therefore, to find optimal delivery methods for drugs such as neurotrophic factors (NTFs) that are extracted from SF conduits.

It has been reported that nanoparticles are effective carriers for drugs that promote efficacy. It shows unique characteristics due to the “size effect”. The common methods for preparing SF nanoparticles include desolvation, salt leaching, and supercritical fluid technologies. Kundu et al. prepared SF nanoparticles loaded with VEGF using a desolvation technique using dimethyl sulphoxide (DMSO) as the desolvation solvent. In vitro release experiments showed that there was no sudden release of VEGF in the drug delivery system, which could be used as an efficient release system of VEGF (Kundu et al., 2010). Lammel et al. produced SF particles by salting the SF solution with potassium phosphate. Compared with emulsification, salt leaching avoids the use of organic solvents and is friendly and safe. The whole experimental condition is in the environment of PH = 8. Increasing SF concentration will lead to larger particles. When PH = 5, particles will gather into non-dispersible aggregates with poor spheronization (Lammel et al., 2010). The preparation of SF nanoparticles was carried out by solution-enhanced dispersion by supercritical CO₂ (SEDS). The SEDS process was used to encapsulate the non-steroidal anti-inflammatory drug Indomethacin (IDMC) in SF nanoparticles. The experimental results showed that SF nanoparticles prepared by the SEDS process were suitable for drug delivery as a biocompatible carrier and feasible for drug delivery control (Zhao et al., 2012). There is a great potential for SF microparticles/nanoparticles to be used as delivery systems for a wide variety of therapeutic agents, such as small molecules, protein drugs, genes, and vaccines, which enhances the biological stability and improves the drug delivery (Pham and Tiyaboonchai, 2020). The unspecific targeting of SF microparticles and nanoparticles may result in low therapeutic efficiency and systemic toxicity, due to their inefficiency as drug delivery systems.

In order to deliver localized treatment, microneedles (MNs) were used to penetrate the layers of the cuticle, which can achieve subcutaneous injections and continuously releases the drug directly into interstitial fluid or blood vessels by adjusting the swelling properties of MNs. The production method of SF microneedles is relatively simple and usually cast by mold. Qi et al. poured proline/melatonin/SF mixture into a polydimethylsiloxane microneedle mold to obtain drug-loaded SF microneedles and microneedles incorporated the drug into the body through the skin and maintained a high concentration of melatonin in the blood of rat insomnia model (Qi et al., 2021). Yavuz et al. reported that levonorgestrel was loaded into microparticles and then injected into silk-based microneedles, which could achieve continuous release for up to a year for unwanted pregnancy (Yavuz et al., 2020). Microneedle devices can also provide highly reproducible results and therapeutic benefits by penetrating through the cuticle into the interstitial fluid. However, there are some limitations to microneedles in biomedical applications, such as skin irritation or allergy to sensitive skin and broken microneedle tips left in the skin (Waghule et al., 2019).

In addition to the above-mentioned forms of SF carriers for drug delivery, Hu et al. used the electrostatic spinning method of preparation of the regenerated SF nanofibers and cultivated Schwann cells on SF nanofibers. They found that the regeneration of electrostatic spinning nanofibers simulates the structure and function of extracellular matrix form and the adhesion, growth, and proliferation of Schwann cells, showing excellent biocompatibility (Hu et al., 2012). In addition, studies have explored that straining flow spinning (SFS) can be used to regenerate high-performance silk fibers as scaffolds for axonal growth induction and guidance and spontaneous cellular connection of dissociated neurons (Mercado et al., 2020). Zhu et al. prepared a porous scaffold called SF sponge to grow primary cortical cells (with or without microglia removal). The results showed that neuronal maturation and synaptic development were reduced in fibroin sponges depleted by microglia (Zhu et al., 2020). Overall, a variety of SF forms, as biomaterial carriers, can be generated in a variety of applications using the method described. We summarized the advantages and disadvantages of different forms of SF carriers for drug delivery (Table 1).

SF hydrogels bring drugs into the site of neuronal damage in the brain (targeted administration, and provides the microenvironment for stroke recovery to realize treatment. Stroke, including ischemic and hemorrhagic strokes, is one of the causes of neuron damage in the human body, and the regeneration capacity of neurons is very limited. Current strategies aimed at enhancing brain tissue’s limited capacity for spontaneous regeneration, include using biologics, small molecules, and cells to treat diseases and SF has been used as an effective delivery vector. Noh et al. reported that SF loaded with Brain Factor-7 peptide reduces brain damage and improves neurological function after stroke by reducing the generation of ROS (Noh et al., 2020a). Gorenkova et al. reported that self-assembled SF hydrogels filled the stroke cavity conducive to stroke recovery and there was no significant microglia/macrophage reaction (Gorenkova et al., 2019). Sun et al. reported that a 3D scaffold made of IKVAV-modified SF hydrogels may be useful for the differentiation and maturity of neuronal stem cells (Sun et al., 2017). Fernández-García et al. found that SF hydrogel-loaded with bone marrow mesenchymal stem cells (MSCs-SF) was better to implant into an infarcted hemisphere than SF hydrogels without MSCs, suggesting that SF is an effective delivery vector for stem cells in stroke (Fernández-García et al., 2018). Our group has recently prepared injectable SF hydrogels loaded with hydrogen sulfide (H₂S@SF) for the treatment of severe intracerebral hemorrhage. The result showed that the neuroprotective effect of H₂S@SF hydrogels on intracerebral hemorrhage was superior to that of SF hydrogels alone because of sustained H₂S release and localized targeted delivery from H₂S@SF hydrogels (Zhang et al., 2022). In general, it is possible to use SF-based drugs for the development of brain tissue engineering in stroke patients.

At present, the application of SF in AD is the detection and diagnosis of the disease. SF can also be used for the treatment of Parkinson’s disease, and using SF to deliver drugs is simpler and more effective than the traditional treatment of Parkinson’s disease. Alzheimer’s disease (AD) is pathologically defined by the presence of neurofibrillary tangles and the accumulation of amyloid-beta (Aβ) in the brain (Hardy and Higgins, 1992). Goncalves et al. developed a new prototype of an electrochemical immunosensor with aminoacid-long amyloid-β peptides fixed in layer-by-layer (LbL) films containing SF to make a preliminary diagnosis of AD (Gonçalves et al., 2016). Liu et al. developed an immune electrochemical interface of conductive SF-based immunoparticles (CSIPs) by electrophoretic/electropolymerization strategy that can be easily integrated into point-of-care testing (PoCT) devices. The interface comprises electropolymerized Poly 3,4-ethylene dioxythiophene (PEDOT) bridged conductive SF-based immune particles (CSIPs) to achieve the Aβ detection in the blood of patients with AD (Liu et al., 2018). There is no valid data to support the application of SF-loaded drugs in the treatment of AD, and as a result, SF-based materials can be used in treating AD in a much broader range of applications.

Parkinson’s disease (PD) is characterized by the progressive loss of dopaminergic neurons of the substantia nigra pars compacta that project to the striatum, which translates into reduced dopamine levels. Levodopa, a precursor of dopamine, is the most common method for increasing dopamine concentration in PD patients (Hayes et al., 2019). Acharya et al. devised a new method for converting tyrosine to L-DOPA that is cheaper and more efficient. By fixing tyrosinase to SF, they found the optimal conditions for L-DOPA production and also showed greater stability, allowing mass production (Acharya et al., 2008). However, it is urgent to fill the gaps in the application of SF-loaded drugs in the treatment of PD.

After SF hydrogels and porous scaffolds are injected or implanted into the site of traumatic brain injury, it can play its anti-inflammatory effect and the role of related drugs to promote the regeneration and differentiation of damaged neurons to achieve the purpose of treatment. Sultan et al. showed that human MSCs with brain-derived neurotrophic factor (BDNF) can be encapsulated with SF-based hydrogels and BDNF-hMSC was transplanted into rat brain to treat traumatic brain injury (TBI), which represents a critical setup towards brain injury treatment for clinical application. Encapsulated BDNF-HMSC induces neurogenesis by secreting BDNF in vitro cell culture and contributes to the growth and differentiation of neurons at the injured site by eliminating oxidative stress in vivo TBI model (Sultan et al., 2021). Jiang et al. fabricated porous scaffolds using cross-linked collagen and SF and co-cultured them with human umbilical cord MSCs (hUCMSCs) to form implantable complexes. The SF implant complex is injected directly into the site of injury in a TBI canine model. The result showed that the porous SF scaffolds loaded with hUCMSCs could suppress inflammation, inhibit glial proliferation, promote vascular regeneration, induce neuronal cytoskeleton and axon regeneration, and enhance synapses and myelination (Jiang et al., 2021). Our group has recently prepared injectable SF hydrogels loaded with hydrogen sulfide (H₂S@SF) to treat TBI and surfaces filled with H₂S@SF hydrogels can achieve local administration and sustained-release H₂S, avoiding volatile and toxic side effects. Moreover, surfaces filled with H₂S@SF hydrogels are capable of penetrating the blood-brain barrier, enabling non-invasive local delivery of H₂S after brain injury (Chen et al., 2022b). SF can be formulated into various material formats and different forms of SF carriers, as promising drug delivery systems, should be further studied in the future.

Using SF microneedles or hydrogels can achieve targeted drug delivery, control drug release, and improve drug utilization rate (compared with traditional drug delivery), to achieve the purpose of treating nervous system tumors. Ribeiro et al. reported that SF hydrogels were able to suppress the growth of human neuronal glioblastoma (U251) in response to conformational transitions from the unordered coil to the β-sheet structure, suggesting that SF is a very useful tool in specific drug delivery to target glioblastoma (Ribeiro et al., 2018). Wang et al. prepared SF microneedle (SMN) patches to inhibit neovascularization, anti-angiogenesis, and induction of tumor cell apoptosis by remoting activation and controlling the release of Bevacizumab, thrombin, and temozolomide (Wang et al., 2022). SF microneedles provide a vehicle for drug release in the desired manner. Pandey et al. constructed SF nanoparticles loaded with DOX and the DOX-SF nanoparticles induced the exogenous apoptosis pathway of glioblastoma cell lines, C-6 and LN-229 (Pandey et al., 2020). Thermal tumor ablation can be achieved through photothermal therapy using a light absorber under near-infrared (NIR) irradiation. Nevertheless, due to leakage of light absorbent, near-infrared irradiation may damage adjacent tissues. To reduce this damage, Yao et al. developed a bioinspired green hydrogel (BVSF) integrating bioproduct biliverdin to stimulate healing wounds and produce angiogenesis by controllable NIR irradiation Yao et al., 2020. An indocyanine green-SF nanoparticle platform (ICG-SFNPs) for photo-thermal treatment of glioblastoma has been developed by Xu et al. It was effective at accumulating inside tumors in mice bearing nude C6 gliomas when ICG-SFNPs were injected into their veins (Xu et al., 2018; ZhuGe et al., 2019). Coburn et al. found that treated with SF hydrogels containing a continuous release of vincristine and doxorubicin significantly reduced the growth of neuroblastoma tumors. Vincristine delivery significantly slows tumor growth and increases the availability of oncology drugs. In addition, intratumor drug concentration was significantly higher when the hydrogel was injected into the tumor site than when intravenous (Coburn et al., 2017). Therefore, these results provide a powerful tool for the investigation of different forms of SF carriers on the programmed cell death in tumors.

Adenosine-carrying SF microspheres, nanofilm coatings, microneedles, and hydrogels can not only treat epilepsy but also monitor the delivery of drugs. Continual monitoring for symptoms and immediate in situ treatment for seizures are crucial for patients with epilepsy, chronic and non-communicable brain diseases affecting people of all ages. Wilz et al. developed a novel adenosine-release system based on SF by insetting adenosine-containing microspheres into a nanomembrane-coated SF scaffolds and were prepared. In this study, it was found that kindling acquisition occurs in a dose-dependent manner when treated with adenosine-releasing brain implants with the desired target release dose, suggesting that implant-derived adenosine might prevent epileptogenesis (Wilz et al., 2008). Zhang et al. developed sensor-equipped devices using a conductive SF hydrogel (CSFH) combination with microneedles array to treat epilepsy for real-time monitoring of the drug delivery (Zhang et al., 2020). These results, based on different SF designs, offer a rapid and sustained treatment to stop or prevent epilepsy.

Neurological diseases can also be classified as chronic or acute according to the speed of onset. Chronic injury can use various forms of SF to achieve the purpose of sustained and controlled release. Acute injury can lead to structural changes and defects. In addition to the sustained release and controlled release of the drug, it also needs to fill the damage site. Therefore, it is suitable for in situ administration with SF hydrogels to fill the defect site and provide a microenvironment suitable for nerve recovery after injury. The application of different forms of SF depends on the pathophysiological mechanisms of various neurological diseases (existing research) as follows (Table 2).