Overcoming drug delivery challenges with lipid-based nanofibers for enhanced wound repair

Abstract

Wound healing is a dynamic, multi-phase process that includes haemostasis, tissue regeneration, cellular proliferation, and matrix modification. Traditional wound care procedures frequently encounter complications such as delayed healing and infection, demanding new therapeutic approaches. In this context, nanomaterial-based devices provide considerable benefits due to their capacity to improve medication delivery and tissue healing. We suggest a lipid-based nanofiber formulation for wound treatment that overcomes the restricted skin penetration of hydrophilic niacin, a strong wound healing agent. Niacin-loaded nanofibers (NLNFs) were manufactured utilizing glyceryl monostearate (GMS) by a self-assembly process, which included high-pressure homogenization and probe sonication for optimum nanostructure creation. The NLNFs were physicochemically characterized using Fourier transform infrared spectroscopy (FTIR), X-ray diffraction, scanning electron microscopy (SEM) and surface profilometry to determine their morphology and homogeneity, and a drop shape analyser was used to determine hydrophobicity. In vitro tests revealed prolonged drug release, great cytocompatibility, and strong antioxidant activity, indicating superior free radical scavenging capacity. Ex vivo tests, such as the Draize skin irritation test, skin permeation test, and drug retention assays, revealed low skin irritation, increased permeability, and efficient drug retention in skin layers. In vivo experiments showed rapid wound closure and positive histological results, which were backed by hemocompatibility tests such as hemolysis and whole blood clot analysis, validating the formulation's safety. ELISA results indicated that the NLNF-treated group had higher levels of critical wound-healing indicators than the controls. Overall, our findings suggest that NLNFs have tremendous potential as a unique and effective treatment alternative for controlling and improving wound healing processes.

---

Introduction

Wound healing is a complex process that involves a number of well-coordinated activities such as haemostasis, inflammation, proliferation, and remodelling. Despite the body's natural capacity to heal wounds, problems such as infections, persistent inflammation, and insufficient angiogenesis can slow the healing process, resulting in longer recovery periods and higher healthcare expenses. Wound healing and tissue regeneration are thus critical biological processes essential for maintaining the skin's integrity and functionality, the human body's largest organ.^1,2 Cellular activity, growth factors, and cytokines interact to restore tissue integrity after damage. The wound-healing process is categorized into three overlapping stages: inflammation, proliferation, and remodelling.³ The inflammatory phase is initiated to prevent infection and remove debris, followed by the proliferative phase, which concentrates on tissue development and revascularization. Finally, the remodelling phase reinforces the newly created tissue, which frequently results in scar formation rather than real tissue regeneration.⁴ Despite the body's extraordinary ability to heal, the process frequently produces fibrotic tissue or scars that lack the complete functioning of the original tissue. This is most visible in adult wound healing, as the repair process focuses on patching up rather than real regeneration. In contrast, embryonic wound repair can proceed without fibrosis, revealing insights into possible regenerative therapeutics.⁵

Recent advances in tissue engineering and regenerative medicine attempt to enhance wound healing results by introducing novel medicines and technologies. These include the use of scaffolds activated with growth factors, bioactive compounds, and genetically engineered cells to speed up the healing process while promoting tissue regeneration.^6,7 Furthermore, developing techniques such as in situ 3D printing, portable bio-printers, nanomaterials and electro-sprayers are being investigated in order to develop more effective skin replacements and enhance therapeutic results. Also, understanding the cellular and molecular processes that govern wound healing and tissue regeneration is critical for creating novel treatment techniques. Research continues to examine the effects of other cytokines, such as IL-22, which have a dual role in encouraging healing and perhaps contributing to carcinogenesis if not correctly controlled.⁸ While great progress has been made in understanding and improving wound healing and tissue regeneration, current research and technological developments offer the possibility of attaining more effective and functional tissue repair, eventually improving patient outcomes and quality of life.

Wound healing and tissue regeneration are crucial parts of medical treatment, especially for patients with chronic wounds, burns, and surgical injuries. Traditional wound care treatments, while successful in certain cases, sometimes fall short in addressing the complications of chronic wounds and infections. This has resulted in an increased interest in nanomedicine and the use of nanomaterials to improve wound healing and tissue regeneration.⁹ Nanomedicine uses nanomaterials’ unique qualities, such as their high surface area-to-volume ratio, adjustable physical and chemical properties, and molecular interaction, to generate novel treatment techniques.¹⁰ Engineered nanoparticles, including polymeric, lipid, metallic and ceramic-based nanomaterials, have also been investigated for their possible role in wound healing and infection prevention.¹⁰ These materials can deliver medicinal drugs and growth factors in a regulated manner, promoting tissue regeneration. Furthermore, nanomaterial-based scaffolds like nanofibers imitate the extracellular matrix, providing structural support while stimulating cell adhesion and differentiation, which are important for tissue healing.¹¹ Recent breakthroughs in nanotechnology have resulted in the creation of complex wound dressings, including nanoparticles and nanofibers.¹² These dressings not only protect the area from external pollutants, but also promote healing by retaining moisture, lowering inflammation, and administering therapeutic substances directly to the wound. For example, hydrogel/nano silver-based dressings and copper oxide-infused dressings have been shown to be effective in treating difficult-to-heal lesions including diabetic foot ulcers and burn injuries.¹³

Niacin, generally known as vitamin B3, has been shown to improve wound healing by boosting angiogenesis, lowering oxidative stress, and enhancing cellular metabolism. Research has looked into numerous niacin methods and formulations to help in the healing process. The combination of niacin and nanomaterials has shown encouraging effects in expediting wound healing and increasing overall outcomes. Niacin increases angiogenesis and possesses antioxidant properties that are necessary for tissue regeneration, whilst nanoparticles improve the drug's delivery efficiency, allowing for longer release and more targeted action at the wound site. This combination not only accelerates healing but also minimizes infection risks by leveraging nanoparticles’ antibacterial properties, resulting in a more effective treatment outcome than through conventional approaches.¹⁴ The key results include its antibacterial and antioxidant effects, which are produced by niacin-loaded nanoparticles, enclosed in microcapsules or hydrogels.^15,16 These nanoparticles assist in killing microorganisms, decreasing oxidative stress, and stimulating angiogenesis, all of which improve wound healing. For example, combining niacin with selenium nanoparticles (Se NPs) can minimize the wound area and increase VEGF and collagenase I levels, both important for tissue regeneration.¹⁷ The synergistic actions of niacin and nanoparticles not only accelerate the healing process, but also allow for regulated and targeted administration of therapeutic substances, reducing possible side effects and toxicity. Furthermore, niacin can change macrophage morphologies, moving them from pro-inflammatory (M1) to anti-inflammatory (M2) states, which lowers inflammation and increases tissue repair, both of which are favourable for wound healing. Overall, niacin has shown several benefits in wound healing, making it a viable drug for clinical wound treatment.

The application of nanoparticles in wound healing goes beyond niacin. Nanomaterials, such as nanoparticles, nanocomposites, and nanofibers, have been studied for their antibacterial, anti-inflammatory, and regeneration capabilities.^18,19 These materials have the potential to change wound treatment by creating an ideal wound microenvironment, promoting cell migration and proliferation, and preventing microbial infections. The addition of niacin to nanomaterial-based treatments marks a substantial leap in wound healing. The unique features of nanomaterials, along with the therapeutic effects of niacin, represent a viable approach for addressing the issues associated with chronic and acute wounds. Despite nanoparticles’ great promise in wound healing, numerous hurdles remain, including the need for standardized protocols, better knowledge of the long-term consequences of nanomaterial exposure, and dealing with possible toxicity issues. Thus, in this work, we investigated the therapeutic potential of lipid-based, niacin-loaded nanofibers to modify and improve wound healing. Using Wistar rats as an in vivo model, we used a multimodal method to assess the effectiveness of these nanofibers. A thorough physicochemical analysis was performed to analyse their morphology, structural integrity, and drug release kinetics. Furthermore, we conducted in vitro experiments to investigate cellular interactions such as biocompatibility, cytotoxicity, and proliferation. Ex vivo tests confirmed the nanofibers’ potential to stimulate tissue regeneration, while in vivo investigations demonstrated their influence on wound closure, inflammation, and tissue repair. This comprehensive method demonstrates nanofibers’ promise as a viable biomaterial for wound healing applications.

Materials and methods

Materials

Niacin, glyceryl monostearate, phosphate buffered saline (PBS), Tween-80, dimethyl sulphoxide (DMSO), ethanol, methanol, ascorbic acid, isopropanol, 2,2-diphenyl-1-picrylhydrazyl (DPPH), anhydrous calcium chloride (CaCl₂), hematoxylin, eosin, sodium chloride (NaCl), paraformaldehyde, chloroform, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), ketamine and xylocaine were procured from Merck and Sigma Aldrich. Enzyme-linked immunosorbent assay (ELISA) kits were bought from Elabscience®.

Animals

We purchased 300–350 g Wistar rats from the Indian Veterinary Research Institute (IVRI) in Bareilly. Wistar rats were housed at the Teerthankar Mahaveer College of Pharmacy & Research Centre animal house facility. The rats were given a five-day window to become used to the new environment in the animal house. With four rats per cage, they were kept in group cages with unconstrained access to drinking water and food (bulgur wheat and rat pellets from Hindustan Animal Feeds, India). The environmental conditions were carefully regulated, maintaining humidity at 50% relative humidity and a temperature of 25 ± 2 °C, with a 12 hour light/dark cycle. Throughout the study, the rats were provided with a standard diet and maintained in a fed state. All procedures followed the protocols approved by the Institutional Animal Ethics Committee at Teerthankar Mahaveer College of Pharmacy & Research Centre (CPCSEA/1205/2022/17) and adhered to the guidelines set forth by the Council for the Purpose of Control and Supervision of Experiments on Animals, Ministry of Environment and Forests, Government of India.

Synthesis of nanofibers

A novel synthesis procedure for nanofibers was developed and optimized using high-pressure homogenization and self-assembly approaches.^20–22 To prepare a lipid–surfactant mixture, we weighed 1 g of glyceryl monostearate (GMS) and 0.1 g of Tween-80 and placed them in a sterile beaker. Briefly, GMS was allowed to melt in a water bath, and after the temperature in the beaker reached 50–60 °C, 20 mL of ethanol was progressively added and stirred for 10 minutes. Tween-80 was then added dropwise to the lipid melt. To get a homogeneous mixture, the solution was stirred for another 10–15 minutes at 1200 rpm, while at room temperature. To ensure uniform dispersion and promote the formation of nanofibers during the self-assembly process, 50 mL of deionized water was added dropwise to the mixture while continually stirring. A probe sonicator was utilized for 10–15 minutes at 40% amplitude in pulse mode (30 seconds on, 30 seconds off) to further increase self-assembly. After the sonication procedure, the mixture was cooled to room temperature, allowing the hydrophobic and van der Waals interactions between GMS and Tween-80 to stabilize the nanofibers. Thus, plain nanofibers (NFs) were synthesised. Drug loading was achieved at the first stage before the addition of Tween-80 to produce niacin loaded nanofibers (NLNFs). To purify the nanofibers, we re-suspended the pellet in deionized water after centrifuging it for 15 minutes at 10 000 rpm and discarding the supernatant. To remove large aggregates, we filtered the nanofiber solution via a 0.2 μm syringe filter. The nanofiber suspension was then lyophilized for long-term storage or additional analysis.

Drug loading and interaction studies

Prior to lyophilization, the formulations were centrifuged at 10 000 rpm for 10 minutes at 4 °C, and the supernatant was collected. The encapsulation and drug loading efficiencies were determined using the equations listed below (eqn (1) and (2)). The pellet was then dispersed in deionized water and lyophilized.where DL is the drug loading capacity and EE is the encapsulation efficiency. D_t denotes total drug in the formulation, D_s denotes the free drug present in the supernatant whereas D denotes total components of the formulation.

The lyophilized formulation was subsequently tested for inter-excipient interactions, and its crystallinity and amorphous nature were assessed using Fourier transform infrared spectroscopy [(FTIR); Bruker Vertex 70v spectrophotometer] and X-ray diffraction investigations (Bruker, D8 Advance X-ray diffractometer), respectively.^23,24

Morphological analysis

1 mg mL⁻¹ samples were collected, and a substantial amount was mounted on silicon wafers. The morphology and shape of the prepared nanofibers were evaluated using a scanning electron microscope (JEOL-JSM-840A, Japan) and surface profiler (Bruker DektakXT Vision 64®). Surface charge and zeta potential were examined using a Zetasizer (Malvern Instruments, Malvern, UK) to determine the probability of agglomeration or aggregation.²³

Drug release kinetics

The kinetics and pattern of drug release were investigated in vitro using a Franz-cell diffusion system.²⁵ Briefly, 1 mg of formulation was dissolved in 1 mL of deionized water, and the resulting component was placed between the apparatus's cells. A membrane made of cellulose acetate with a molecular weight limit of 12 000–14 000 Da was used. The recipient and donor compartments were clamped together, the open side of the donor compartment facing outside, and the suspension was filled according to the orientation of the donor component. 20 mL of buffer (pH 7.2, phosphate buffered saline) was introduced into the recipient chamber. The device was maintained at 37 ± 2 °C while being constantly stirred at 200 rpm using a magnetic stirrer. For a maximum of 60 hours, samples (1 mL) were taken at prearranged intervals, and an equal quantity of fresh media was added to maintain the sink conditions.²⁶ A UV spectrophotometer with a wavelength of 262 nm was used to measure the quantity of medication in the samples.²⁷ The cumulative percentage of drug release was computed and shown over time using a standard calibration curve (Fig. S3†).

Ex vivo permeation studies

Following the administration of a ketamine–xylazine injection to induce anesthesia, the animals were euthanized by spinal dislocation. The ex vivo skin permeation experiment was conducted using excised Wistar rat skin and a Franz diffusion cell system.²⁸ We utilized male rats measuring 300–350 g, with full-thickness abdominal skin. Before removing the skin, the hair on the abdomen was softly removed using an electric trimmer. The dermal layer of the skin was carefully cleaned with water to remove any residual connective tissue or blood vessels. The skin was prepared for the experiment by immersing it in phosphate buffer at pH 7.2 for 1 hour, after which it was placed on a magnetic stirrer with a magnetized probe to ensure uniform dispersion of the diffusant. The detached rat skin was placed between the sections of the diffusion cell and maintained at a constant temperature of 37 ± 2 °C, with the stratum corneum oriented towards the donor compartment and the dermis facing the receptor compartment.²⁹ To maintain sink conditions, 1 mL samples were regularly withdrawn from the receptor compartment and replaced with an equal volume of fresh medium. The materials were passed through a Whatman filter before being evaluated using a UV spectrophotometer.³⁰

To understand how well substances can pass through the skin, we need to calculate many different factors. These factors help us compare how much of a substance can penetrate the skin and how quickly it does so. We look at things like the steady flow of the substance, how easily it can pass through the skin, and the total amount that ends up inside the body.³¹ The permeation flow was computed using eqn (3).³² The amount of a substance that passed through the skin was divided by the total amount of that substance in the starting area. This calculation helped us determine how easily the substance could penetrate the skin.

where J is the permeation flux (μg min⁻¹), δM is the quantity of medication that penetrated, C is the film's cross-sectional area, and T is the duration.

Ex vivo drug retention studies

Drug retention in the skin was measured after the ex vivo permeation tests.³⁰ The skin was taken out of the testing device and washed with salt water. The outer layer of the skin was separated from the inner layer and dried with hot air. The substance that had penetrated the skin was removed from both layers using a special liquid called “methanol”.^33,34 The amount of drug was measured using a spectrophotometer at a wavelength of 262 nm.

Free radical scavenging assay

The antioxidant activity of the formulated NLNFs was assessed using the DPPH free radical scavenging assay.^35,36 Briefly, a stock solution of DPPH was prepared in methanol, and different concentrations of NFs, NLNFs, niacin, and ascorbic acid were prepared. A total of 100 μL of DPPH solution was added to 100 μL of each standard, including the NF solution, NLNF solution, and niacin solution, in a 96-well plate. After incubating for 30 minutes, the absorbance values were measured at 517 nm using a microplate UV spectrophotometer. The inhibition of the free radical scavenging process was then computed utilizing eqn (4).³⁷where A_d is the optical density of DPPH radicals in methanol and A_ds is the optical density of DPPH radicals with formulations.

Determination of ROS

To identify the generation of reactive oxygen species, the carboxy-H2DCFDA test was employed. A new carboxy-H2DCFDA solution was prepared in sterile DMSO before the experiment. To get rid of any remnants of the original medium, seeded cells were rinsed with PBS. Following a 30 minute dark incubation period under standard conditions (5% CO₂ at 37 ± 2 °C), cells were treated with a dye at a concentration of 1 μM. The cells were then treated with varying doses of nanoformulations (LCNPs and MD-LCNPs) and plain medication (MD) after being twice rinsed with PBS to get rid of the dye containing carboxy-H2DCFDA. Carboxy-H2DCFDA-treated and untreated cells were used as controls. The ROS were examined using fluorescence microscopy after an hour of incubation.

Cellular toxicity assessment

The cytocompatibility of NLNFs with the L929 fibroblast cell line was determined using the MTT test.²³ Cells were seeded at a density of 5 × 10⁴ cells per well in a 96-well plate and incubated for 24 hours under normal culture conditions (5% CO₂, 37 ± 2 °C). Cells were treated with NFs, NLNFs, and niacin at different doses (31.25–1000 μg mL⁻¹), whereas untreated cells served as controls. After 24 hour of exposure, 20 μL of MTT reagent (5 mg mL⁻¹) was added to each well and cells were incubated for an additional 4 hours. Formazan crystals were dissolved in 100 μL of DMSO, and absorbance was measured at 570 nm using an Infinite 200 PRO microplate reader. This test offers useful information on the effects of various doses of niacin, NFs, and NLNFs on cell viability, as evidenced by absorbance measurements at the appropriate wavelength.

Hemolysis assay

The human body is prone to immunological reactions, and any foreign entity can cause significant deformation of blood cells; thus a hemocompatibility experiment was performed to determine whether our formulation has any negative effects on blood cells. A previously established hemolysis assay was applied with slight modifications.³⁸ Blood was collected from healthy donors with prior consent and all the experiments were performed in accordance with the ethical guidelines. Blood (5 mL) was collected in EDTA coated vials to avoid possible coagulation. The blood was then diluted with PBS (1 : 2). Following dilution, centrifugation was used to settle down the erythrocytes and the supernatant was collected in a separate vial. After washing the pellet with PBS, red blood cells were re-dispersed (1 : 10) in the same. Simultaneously a dispersion of nanofibers (1 mg ml⁻¹) in PBS was prepared and 800 μL of it was added to micro-centrifuge tubes. 40 μL of erythrocyte dispersion was added gradually to the tubes containing the sample dispersion. These micro-centrifuge tubes were incubated for 1 hour at 37 ± 1° C. The supernatant was collected, and the absorbance was measured at 570 nm with a spectrophotometer. The calculation was performed using eqn (5).where A_nf is the absorbance of the sample, A_n is the absorbance of the negative control and A_p denotes the absorbance of the positive control.

Whole blood clotting assay

Whole blood clotting test was performed to see how our nanoformulations influenced the blood clotting mechanism.³⁹ Blood samples were drawn from healthy donors and placed in EDTA-coated anticoagulant vials. In a 24-well plate, coverslips were placed in each well. The nanoformulation was dispersed in PBS at a concentration of 1 mg mL⁻¹, with 50 μL applied to each coverslip within the wells. Triton X-100 was used as a positive control and PBS as a negative control. Next, 50 μL of blood and 50 μL of 0.1 M calcium chloride were added to each well. The plate was incubated at 37 °C for intervals of 5, 10, 15, 20, 25, and 30 minutes.

At the end of each time point, 2 mL of deionized water was added into each well, followed by another 10 minutes of incubation. Each well underwent spectrophotometric analysis, which involved evaluating the absorbance at a wavelength of 570 nm in order to quantify the amount of liberated haemoglobin.

Ex vivo skin irritation test

The Draize skin irritation test was slightly modified to assess the dermal irritability of the compounded NLNFs.^40,41 Rats weighing between 250 and 300 grams were used in this experiment after receiving approval from the committee that oversees animal research ethics (CPCSEA). The rats were housed under regulated conditions: 50% humidity and a constant temperature of 25 °C. Prior to the experiment, the animals were randomly assigned to four groups: a positive control group (n = 5) was exposed to 0.8% formaldehyde solution (a conventional irritant), an untreated control group (n = 5) that got a placebo, a group receiving treatment (n = 5) that was administered niacin, and an additional treatment group (n = 5) that received NLNFs. Stomach hair was removed with an electric clipper, and the exposed region was then cleansed with dry cotton. On day one, the irritants (formaldehyde, niacin, NLNFs, and a placebo) were administered to the prepared skin. After 24 hours of contact, the films were removed, the skin was cleansed with deionized water, and erythema and oedema were assessed using the Draize scoring system. The ratings went from 0 (no response) to 4 (severe erythema or oedema).⁴² Evaluations were also conducted at 48 and 72 hour intervals, and the collected data were statistically evaluated.

Wound initiation and closure rate analysis

Following ethical permission, healthy animals were chosen and acclimatized to laboratory settings. An electric clipper was utilized to shave the dorsal area, which was subsequently cleaned with 70% ethanol before wound formation. Under anaesthesia, a consistent excision incision (1–2 cm in diameter) was created, and the initial wound area was measured using transparent graph paper. The animals were divided into four groups, with six animals per group: the control group, which received a placebo, the NF treated group, Niacin treated group, and the NLNF-treated group. The wounds were treated with formulations (NF, NLNF, and niacin paste) and then covered with sterile dressings. Every day, the wound area was evaluated by sketching it on graph paper and taking photographs.^43,44 Wound area decrease was monitored until healing was complete, and the percentage of wound closure was estimated using the following formula in eqn (6).where W₀ is the wound area on day 0 and W₁₄ is the wound area on day 14, while W_CR is the wound closure rate.

Histopathological analysis

To prepare skin tissue samples for histological investigation, specimens were first washed with phosphate-buffered saline (PBS) to eliminate contaminants, and then fixed in 4% paraformaldehyde to preserve tissue structure. Following fixation, the tissues were immersed in paraffin wax and sliced into three-micron slices with a microtome.⁴⁵ These slices were stained with hematoxylin and eosin (H&E) to distinguish cellular components, with nuclei blue/purple and cytoplasm pink/red. After that, representative micrographs were taken for analysis of the dye-labelled tissues under a fluorescence microscope in order to enhance visual details and enable the study of tissue architecture and cellular characteristics.

ELISA estimation of pathological markers

Serum was extracted for ELISA testing after Wistar rat femoral vein blood was collected. Using a commercially available enzyme-linked immunosorbent assay (ELISA) kit, the quantities of ICAM, VCAM, VEGF, and PGD2/E2 were measured in accordance with the manufacturer's (Elabscience®) instructions.^46,47

Statistical analysis

The data are presented as mean ± SD. All tests were carried out concurrently and reproduced to ensure accuracy, with data analyzed using Origin 2023 and GraphPad Prism 5 (GraphPad Software, Inc., San Diego, CA, USA). Statistical significance among groups was determined using two-way ANOVA and Student's t-tests. A p-value less than 0.05 was considered statistically significant, indicating a low probability that the observed differences were due to chance.

Results

Particle synthesis and drug interaction studies

Following the successful fabrication of nanofibers via self-assembly, the nanofibers were examined for their interaction with the encapsulated medication using FTIR. The FTIR spectra revealed that our formulation underwent little reaction with niacin, but the medication was effectively loaded onto the nanofibers (Fig. 1A). The FTIR spectra clearly show that there was no major change in peaks across the spectrum. However, a shift in the carbonyl peak (–C=O) at 1600 cm⁻¹ indicated that the glyceryl monostearate and niacin groups interacted positively. Also, a very distinct peak about 3300 cm⁻¹ was seen in the case of NLNFs, which was not clearly evident in the spectra of niacin, indicating drug loading onto glyceryl monostearate plain nanofibers. Other peaks were visible at 1186 cm⁻¹, 1301 cm⁻¹, and 1463 cm⁻¹, and a cluster of peaks were visible below 1000 cm⁻¹. The peaks at 1186 cm⁻¹ correspond to –C–O stretching vibrations of the glyceryl backbone of glyceryl monostearate, whereas the peaks at 1301 cm⁻¹ belong to CH₂ twisting vibrations of methylene groups (–CH₂), which may be present as side chains. The peak at 1463 cm⁻¹ represents the bending (scissoring) vibrations of the methylene (–CH₂) group. These vibrations, along with others (such as C–H stretching at 2850–2920 cm⁻¹ and carbonyl stretching about 1700 cm⁻¹), provide specifics about the formation of NLNFs.

Fig. 1 (A) FTIR spectra of niacin, NFs and NLNFs showing minimal but positive interaction between the drug and other excipients, (B) XRD spectra of niacin, NFs and NLNFs showing amorphous nature of the formulation.

The formulations were subsequently evaluated for their crystallinity, as crystalline structures may pose challenges for skin-related applications. X-ray diffraction (XRD) analysis revealed a significant reduction in the crystalline peaks of niacin upon encapsulation within the nanofibers, indicating a shift towards an amorphous state (Fig. 1B). Additionally, the overlapping peaks of both niacin and glyceryl monostearate in the niacin-loaded nanofibers confirmed successful drug encapsulation.

The formulations were than assessed for encapsulation and drug loading efficiencies. Among the various formulations, NF2 demonstrated the highest encapsulation efficiency and drug loading capacity, with values of 92.8 ± 2.7% and 13.6 ± 1.3%, respectively (Fig. 2A and Table 1). This formulation was selected for further investigation and designated as NLNF. Stability of the nanofibers was assessed via zeta potential measurements, where NLNF exhibited a negative zeta potential of −29.5 ± 7.1 mV (Fig. 2C) indicating that the formulation is resistant to self-agglomeration and aggregation.

Fig. 2 (A) Drug loading and encapsulation efficiencies of various formulations, (B) scanning electron micrographs of nanofibers (NFs and NLNFs), (C) zeta potential analysis of nanofibers.

Table 1 Formulation optimization and drug content analysis

Formulation names	Formulation components			Encapsulation efficiency (%)	Drug loading (%)
	GMS (w/w)	Niacin (w/w)	Tween-80 (w/w)
NF1	1	0.5	0.1	74.4	4.2
NF2	1	0.25	0.1	92.8	13.6
NF3	1	0.125	0.1	89.5	11.2
NF4	1	0.1	0.1	51.3	9.4
NF5	1	1	0.1	56.4	10.2

---

Morphological analysis

Morphological analysis was carried out using a scanning electron microscope and it was found that the nanofibers ranged from 200 nm to 1000 nm in diameter and were well formed with a distinct appearance. Also drug loading was observed on the surface of nanofibers (Fig. 2B). Furthermore, we evaluated the nanofibers for their applicability on wounds and for that the nanofiber paste must be smooth so that it cannot cause irritation to the skin tissues and thus we used a surface profiler to assess the same and it was found that our formulations were smooth in texture with an average roughness of 575 nm (Fig. S1†).

Contact angle measurement

The stratum corneum is the skin's outermost layer and is predominantly made up of lipids, making it more permeable to lipophilic (hydrophobic) compounds. Hydrophilic formulations seldom enter the skin, whereas hydrophobic formulations are more likely to do so. As a result, we examined our formulations for water sorption propensity using a drop shape analyser and discovered that even when niacin, a water soluble vitamin, was loaded onto nanofibers, they exhibited a nearly hydrophobic nature with a contact angle more than 70° (Fig. 3). In contrast, NFs showed a higher contact angle of 99° while pure niacin showed a contact angle with respect to water around 54°. As a result, our formulations proved to be reliable when employed as topical pastes in wound healing dressings.

Fig. 3 Representative photographs of contact angle measurement using a drop shape analyser.

Drug release kinetics

Drug release analysis was performed using a Franz diffusion cell apparatus. From the release kinetics, it was found that the drug was released from the nanofibers in a very slow and sustained manner with an initial burst release up to 12 hours. It was observed that more than 45% of the loaded niacin was released in the first 24 hours which was followed by a slow release up to 48 hours (Fig. 4A). Different plots were drawn to assess the release model and it was found that the drug release followed a Higuchi model of drug release (Fig. S2†).

Fig. 4 (A) Drug release profile of NLNFs, (B) cumulative drug permeation through the skin, (C) DPPH antioxidant assay results showing the % antioxidant potential of NFs and NLNFs, (D) percentage cellular viability assessment using L929 fibroblast cell lines. Three independent experimental trials were performed, and the representative data are presented as mean ± standard deviation (SD). Statistical significance is indicated by *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001, corresponding to progressively higher levels of significance, respectively.

Skin permeation studies

Skin permeation studies were conducted using excised rat skin and a diffusion cell apparatus to determine how niacin is released through the skin. Within 24 hours, over 52.16 ± 1.8% of the drug penetrated the skin and reached the buffer solution whereas raw niacin reached its peak even before 24 hours (Fig. 4B). These results suggest that niacin passively diffuses through the skin, indicating its potential for transdermal delivery.^48,49

Drug retention studies

Following the permeability test, a drug retention test was performed to measure the proportion of drug entrapped inside the skin layers. After 24 hours, the skin retained 23.17 ± 6.5% of the drug. It was determined that the bulk of the medication got past the stratum corneum barrier and into the bloodstream.

Antioxidant assay

Antioxidants are essential for maintaining proper repair of tissues without excessive scar formation, regulating the oxidative environment in wounds, and easing the transition through the healing stages. Thus we assessed the free radical scavenging potential of our formulations through DPPH assay. We observed that although compared to standard oxidants our formulation showed a lower level of free radical scavenging activity, it was found to be significantly high enough for use. Compared to the plain formulation (37.1 ± 0.7%), NLNFs showed significantly higher level of free radical scavenging activity (69.0 ± 2.9%) which was comparable with that of niacin itself (Fig. 4C). Thus, our formulation can be used for wound healing regimens without causing any oxidative damage.

ROS determination through H2DCF-DA assay

Because ROS function as signalling molecules, moderate ROS levels aid in wound healing; nevertheless, excessive ROS production must be managed. Both endogenous (like superoxide dismutase and catalase) and therapeutic (topical) antioxidants aid in balancing ROS levels, which facilitates rapid and effective wound healing. Therefore, the H2DCF-DA test was carried out to examine ROS production in live cells that results from oxidative stress caused by niacin and the resultant formulations. The association between fluorescence intensity and ROS generation measurement was clarified. The results show that, in comparison with niacin given alone, NLNFs increased ROS levels but not to a significant level. Both NLNF and niacin treated groups showed very little ROS generation, but it was marginally higher than in the control group. However, the group treated with NFs showed significantly higher levels of ROS (Fig. S4†). These results support our nanofiber's potential as an effective niacin transporter and it presents a viable treatment option for wound healing by reducing problems brought on by oxidative stress.

Cellular toxicity studies

Any formulation that is given to the body must minimize cellular toxicity, especially for wounds where toxic substances can worsen tissue deterioration and reduce the effectiveness of wound healing. Since cell proliferation is a crucial component of the healing process, the MTT test is a well-established method for determining if a formulation is safe and whether it encourages cell proliferation in the event of wound healing. To ascertain cellular viability after being treated with our formulations, we employed the MTT test. L929 fibroblast cell lines were utilized to determine the cellular viability. It was found that our formulation (NLNFs) exhibited cellular viability of more than 80% at the maximum concentration (1 mg ml⁻¹). It was additionally found that our plain formulation (NFs) also exhibited very little toxicity towards the cells, indicating that our formulations are biocompatible and suitable for use in the healing of wounds (Fig. 4D).

Hemolysis and whole blood clot assay

When developing formulations for wound healing, particularly when incorporating nanomaterials and scaffolds, it is critical to evaluate the hemocompatibility of the formulations. Thus, we used the hemolysis and whole blood clot assays to ensure that our formulation did not cause any adverse reactions to blood cells or intrinsic coagulation pathways. Ideally, formulations generating less than 5% hemolysis are considered safe. In this study, we discovered that our formulations exhibited considerably reduced hemolysis than the positive control, Triton-X100, and all the formulations were comparable to the negative control (PBS) which was less than 5% (Fig. 5A). Conversely, a formulation is considered to exhibit a positive clotting effect if the absorbance value is close to 0.1. A higher absorbance indicates greater resistance to clot formation. Results from the whole blood clot assay demonstrated that in comparison with Triton-X100, our formulations showed significantly higher clotting percentages. While NLNFs exhibited a favourable response in promoting clot formation, the plain formulation displayed a delayed reaction, suggesting that niacin loading improved clotting efficiency (Fig. 5B). These findings indicate that our formulations are safe for topical application on wounds and have the potential to enhance the healing process.

Fig. 5 (A) Hemolysis assay for the assessment of hematocompatible nature of the formulations; TX100 (positive control), PBS (negative control), NFs, niacin and NLNFs, (B) whole blood clot assay for TX100 (positive control), PBS (negative control), NFs, niacin and NLNFs.

Ex vivo skin irritation studies

The Draize irritation test is often used to examine the possible irritancy of formulations intended for topical applications, such as wound healing therapies. It helps in determining if a substance may, when applied, result in adverse effects such as redness, swelling, or harm to the skin and mucous membranes. A good result with low irritation scores assures that the formulation may be used topically without posing any risks, facilitating a more efficient and painless healing process for wounds. Thus we assessed our formulation using the same and we found that our formulation showed no adverse reaction like swelling, redness, heat, rash or cutaneous vasodilation when applied to the shaved skin. In contrast, the positive control (0.8% formalin) showed little redness and inflammation (Fig. 6).

Fig. 6 Representative images of the Draize irritation test for skin compatibility of the formulations: score 0 depicts no inflammation or irritation; score 1 depicts mild irritation and little inflammation; and score 2 indicates mild irritation and higher inflammation.

Wound closure and healing

Over a period of 14 days, the animals were well taken care of and monitored time to time for any toxic or adverse reactions. Over the period, the wound closure rate was determined using transparent graph paper and the respective areas were noted down daily and after the period of study the wound closure rate was determined. It was found that compared to the control group, the group treated with NLNFs showed a higher wound closure and scar removal rate after the 14 days. With a wound closure rate of 87.5 ± 6.4% on Day 7 and 99.6 ± 0.2% on Day 14, our nanoformulation showed significantly high wound healing efficiency and thus can be used for chronic wounds (Fig. 7 & Fig. S3†).

Fig. 7 Representative photographs of wounds in different groups taken at different times (Day 0, Day 7 and Day 14) respectively.

Histopathology

After 14 days, the animals were humanely euthanized via spinal dislocation, and skin tissue was excised for detailed morphological and histological analysis to identify any structural abnormalities (Fig. 8). The study focused on key stages of wound healing, including inflammation, tissue formation, and remodelling, providing valuable insights into the efficacy of the prepared formulation. Histopathological examination revealed substantial infiltration of granulation tissue in the dermis and underlying subcutaneous layers, particularly pronounced in the NLNF-treated group compared to other treatment groups. H & E staining showed the formation of a full-thickness epidermis over the 14 day period, with notable infiltration of inflammatory cells, fibroblast migration, and regeneration of dermal appendages and follicles. Compared to the control group, the NLNF-treated group showed significantly increased infiltration of inflammatory cells as well as more migration of fibroblasts, whereas the NF group showed less. These findings suggest that the formulation significantly enhanced wound healing without any observed histopathological abnormalities.

Fig. 8 Representative micrographs of H & E stained histological sections, (A) control, (B) NF-treated, (C) niacin-treated, and (D) NLNF-treated.

ELISA estimation of various markers

Quantification of various pathological markers was carried out via ELISA using the protocol given by the manufacturer (Elabscience®). Excised tissues were homogenized and centrifuged. Pellets were discarded to remove any cell debris and the supernatant was used for quantification of markers. As cytokines, mediators and growth factors play critical roles in the wound healing process, we analyzed the lysate for VEGF, PGD2, IL-1β, IL-6, TNF-α and IL-22 as they have pronounced roles in tissue healing and regeneration.^{50–52,52,53}

VEGF is very essential in wound healing as it promotes angiogenesis, endothelial cell proliferation and migration, thus enhancing the wound healing process through constant supply of oxygen and nutrients.⁵⁰ We observed that compared to the control, VEGF levels in the NLNF-treated group were significantly high (Fig. 9A). On the other hand, NFs showed significantly low levels of VEGF. PGD2, a lipid mediator, regulates inflammation in wound healing by inhibiting pro-inflammatory cytokine production and promoting inflammation resolution, crucial for wound repair and chronic inflammation prevention. From the ELISA tests, it was observed that the levels of PGD2 were comparable in the control, niacin-treated and NF-treated groups but were significantly high in the case of the NLNF-treated group (Fig. 9B).

Fig. 9 ELISA estimation of various markers, (A) VEGF, (B) PGD2, (C) IL-1β, (D) IL-6, (E) TNF-α, and (F) IL-22. Three independent experimental trials were performed, and the representative data are presented as mean ± standard deviation (SD). Statistical significance is indicated by *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001, corresponding to progressively higher levels of significance, respectively.

IL-1β, a pro-inflammatory cytokine, aids wound healing by recruiting neutrophils and macrophages, clearing debris, and stimulating tissue remodelling through the production of growth factors and matrix metalloproteinases (MMPs).⁵² IL-6, a cytokine, plays both pro-inflammatory and anti-inflammatory roles in wound healing, promoting immune cell activation and facilitating tissue regeneration through fibroblast proliferation and collagen synthesis. IL-22 aids in wound healing by promoting epithelial cell repair, keratinocyte proliferation, and wound surface re-epithelialization, while also maintaining skin barrier integrity and protecting against infections.⁵³ Herein, we observed that the control group was comparable with the NF-treated group and the niacin-treated group, while the NLNF-treated group had significantly higher levels of these markers thus indicating the role of NLNFs in treating wounds (Fig. 9C, D and F).

TNF-α, a pro-inflammatory cytokine, aids wound healing by enhancing immune cell recruitment, promoting angiogenesis, and stimulating fibroblast activity, but excessive production can cause chronic inflammation.⁵² We observed that the levels of TNF-α were significantly high in all the treatment groups compared to the control group though the NLNF-treated group showed excessively high levels of TNF-α (Fig. 9E). In conclusion, altogether these mediators facilitate the gradual and effective regeneration of injured tissue by coordinating the several stages of wound healing, from inflammation and tissue creation to remodelling and it was observed that our formulations were efficient in wound healing and tissue regeneration.

Conclusions

In conclusion, we successfully produced niacin-loaded nanofibers (NLNFs) utilizing a self-assembly method using glyceryl monostearate. The FTIR spectra showed a limited yet favourable interaction between niacin and the nanofibers, demonstrating effective drug encapsulation. XRD examination demonstrated that the nanofibers converted niacin to an amorphous form, making them suitable for topical applications. NLNFs outperformed the other formulations in terms of drug encapsulation and loading efficiency, and zeta potential studies validated their stability. The morphological study using SEM revealed well-formed nanofibers with a smooth surface, which is critical for wound healing applications. Contact angle measurements revealed that the nanofibers were hydrophobic, allowing for skin penetration.

The drug release profile showed a gradual, sustained release, and skin penetration experiments confirmed good transdermal administration. Antioxidant tests revealed considerable free radical scavenging activity, confirming the nanofibers’ effectiveness in lowering oxidative stress in wounds. The formulations showed good biocompatibility in cell viability studies and were non-toxic, as demonstrated by the MTT and hemolysis tests. Ex vivo skin irritation investigations verified their safe use in wound healing. Histopathological study revealed that NLNFs improved wound healing and resulted in considerable tissue regeneration. ELISA analysis confirmed these findings, revealing increased levels of critical wound healing indicators such as VEGF and pro-inflammatory cytokines. Overall, NLNFs are a viable treatment option for wound care.

Author contributions

Aaqib Javaid: design, protocol, investigation, experimentation, and writing – original draft. Prakhar Varshney: experimentation. Krishana Kumar Sharma: experimentation. Anurag Verma: experimentation and resources. Shyam Lal Mudavath: conceptualization, experimentation, original draft editing, review, resources and funding acquisition, and supervision.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

The authors have no conflict of interest to declare.

Acknowledgements

This work is supported by the Government of India for funding under DST-SERB (EEQ/2020/000563) and Indian Council of Medical research (ICMR), Government of India (DDR/IIRP23/3052). Aaqib Javaid sincerely acknowledges the doctoral fellowship provided by the Institute of Nano Science and Technology, Mohali. The authors thank Teerthankar Mahaveer University for providing the amenities of animal housing. Mr Prem Kumar and Ms Kanchan Swami's assistance in surface profiling and drop shape analysis is also gratefully acknowledged.

References

1. Y. W. Ding, Z. Y. Wang, Z. W. Ren, X. W. Zhang and D. X. Wei, Biomater. Sci., 2022, 10, 3393–3409.
2. X. Long, Q. Yuan, R. Tian, W. Zhang, L. Liu, M. Yang, X. Yuan, Z. Deng, Q. Li, R. Sun, Y. Kang, Y. Peng, X. Kuang, L. Zeng and Z. Yuan, Biomater. Sci., 2024, 12, 1750–1760.
3. E. M. Tottoli, R. Dorati, I. Genta, E. Chiesa, S. Pisani and B. Conti, Pharmaceutics, 2020, 12, 1–30.
4. G. C. Gurtner, S. Werner, Y. Barrandon and M. T. Longaker, Nature, 2008, 453, 314–321.
5. D. Chouhan, N. Dey, N. Bhardwaj and B. B. Mandal, Biomaterials, 2019, 216, 119267.
6. P. Martin, Science, 1997, 276, 75–81.
7. J. M. Reinke and H. Sorg, Eur. Surg. Res., 2012, 49, 35–43.
8. S. A. Eming, P. Martin and M. Tomic-Canic, Sci. Transl. Med., 2014, 6, 2656–2656.
9. S. K. Nethi, S. Das, C. R. Patra and S. Mukherjee, Biomater. Sci., 2019, 7(7), 2652–2674.
10. M. Parani, G. Lokhande, A. Singh and A. K. Gaharwar, ACS Appl. Mater. Interfaces, 2016, 8(16), 10049–10069.
11. T. Ashwini, A. Prabhu, V. Baliga, S. Bhat, S. T. Thenkondar, Y. Nayak and U. Y. Nayak, Pharmaceutics, 2023, 15, 1560.
12. X. Yue, Z. Wang, H. Shi, R. Wu, Y. Feng, L. Yuan, S. Hou, X. Song and L. Liu, Biomater. Sci., 2023, 11, 5232–5239.
13. P. Shaw, P. Vanraes, N. Kumar and A. Bogaerts, Nanomaterials, 2022, 12, 3397.
14. B. Blanco-Fernandez, O. Castaño, M. Á. Mateos-Timoneda, E. Engel and S. Pérez-Amodio, Adv. Wound Care, 2021, 10, 234.
15. A. Javaid, K. A. Abutwaibe and S. L. Mudavath, Nanophototherapy, 2025, 613–631.
16. A. Javaid, K. A. Abutwaibe, K. Kumar Sharma, P. M. Sherilraj, A. Verma and S. Lal Mudavath, ACS Appl. Nano Mater., 2024, 7, 454.
17. M. Emam, A. T. Keshta, Y. M. A. Mohamed and Y. A. Attia, Curr. Chem. Biol., 2020, 14, 169–186.
18. A. Kushwaha, L. Goswami and B. S. Kim, Nanomaterials DOI:10.3390/NANO12040618.
19. A. Naskar and K. S. Kim, Pharmaceutics, 2020, 12, 499.
20. A. Eatemadi, H. Daraee, N. Zarghami, H. M. Yar and A. Akbarzadeh, Artif. Cells, Nanomed., Biotechnol., 2016, 44, 111–121.
21. S. Yadav, A. K. Sharma and P. Kumar, Front. Bioeng. Biotechnol., 2020, 8, 127.
22. D. G. Fatouros, D. A. Lamprou, A. J. Urquhart, S. N. Yannopoulos, I. S. Vizirianakis, S. Zhang and S. Koutsopoulos, ACS Appl. Mater. Interfaces, 2014, 6, 8184–8189.
23. A. Singh, G. Yadagiri, A. Javaid, K. K. Sharma, A. Verma, O. P. Singh, S. Sundar and S. L. Mudavath, Biomater. Sci., 2022, 10, 5669–5688.
24. A. Javaid, A. Singh, S. Parvez, M. Negi and S. L. Mudavath, Colloids Surf., A, 2024, 682, 132889.
25. I. Chanda, R. Bordoloi, D. D. Chakraborty, P. Chakraborty, S. Rekha and C. Das, J. Appl. Pharm. Sci., 2017, 7, 81–084.
26. P. Panwar, B. Pandey, P. C. Lakhera and K. P. Singh, Int. J. Nanomed., 2010, 5, 101–108.
27. A. Javaid, A. Singh, K. K. Sharma, K. A. Abutwaibe, K. Arora, A. Verma and S. L. Mudavath, AAPS PharmSciTech, 2024, 25, 101.
28. P. Arora and B. Mukherjee, J. Pharm. Sci., 2002, 91, 2076–2089.
29. S. Cherukuri, U. R. Batchu, K. Mandava, V. Cherukuri and K. R. Ganapuram, Int. J. Pharm. Invest., 2017, 7, 10–17.
30. A. A. Elshall, A. M. Ghoneim, H. M. Abdel-Mageed, R. Osman and D. S. Shaker, Futur. J. Pharm. Sci., 2022, 8, 28.
31. M. Rahman and C. S. Brazel, Prog. Polym. Sci., 2004, 29, 1223–1248.
32. H. Muhammad Abdullah, M. Farooq, S. Adnan, Z. Masood, M. Asad Saeed, N. Aslam and W. Ishaq, Polym. Bull., 2023, 80, 6793–6818.
33. C. Tas, Y. Ozkan, A. Okyar and A. Savaser, Drug Deliv., 2007, 14, 453–459.
34. V. K. Rapalli, V. Kaul, T. Waghule, S. Gorantla, S. Sharma, A. Roy, S. K. Dubey and G. Singhvi, Eur. J. Pharm. Sci., 2020, 152, 105438.
35. F. Abderrahim, S. M. Arribas, M. Carmen Gonzalez and L. Condezo-Hoyos, Food Chem., 2013, 141, 788–794.
36. F. Xiao, T. Xu, B. Lu and R. Liu, Food Front., 2020, 1, 60–69.
37. A. Javaid, K. A. Abutwaibe, K. Kumar Sharma, P. M. Sherilraj, A. Verma and S. Lal Mudavath, ACS Appl. Nano Mater., 2024, 7, 454.
38. A. Julius and W. Hopper, Biomed. Pharmacother., 2019, 109, 708–715.
39. B. Mahaling, D. A. Srinivasarao, G. Raghu, R. K. Kasam, G. Bhanuprakash Reddy and D. S. Katti, Nanoscale, 2018, 10, 16485–16498.
40. M. K. Robinson, R. Osborne and M. A. Perkins, Ann. N. Y. Acad. Sci., 2000, 919, 192–204.
41. M. P. Vinardell and M. Mitjans, J. Pharm. Sci., 2008, 97, 46–59.
42. D. Hasa, S. Žakelj, I. Grabnar, F. Cilurzo, S. Dall'Acqua, A. Riva, B. Perissutti and D. Voinovich, Pharmaceutics, 2020, 12, 1–11.
43. M. Li, Y. Liang, J. He, H. Zhang and B. Guo, Chem. Mater., 2020, 32, 9937–9953.
44. F. B. R. de Moura, B. A. Ferreira, S. R. Deconte, B. C. Landim, A. B. Justino, A. A. de Aro, F. S. Espindola, R. A. F. Rodrigues, D. L. Ribeiro, F. de A. Araújo and T. C. Tomiosso, J. Tradit. Complementary Med., 2021, 11, 446–456.
45. B. Yang, J. Song, Y. Jiang, M. Li, J. Wei, J. Qin, W. Peng, F. L. Lasaosa, Y. He, H. Mao, J. Yang and Z. Gu, ACS Appl. Nano Mater., 2020, 12, 57782–57797.
46. J. Yu, Y. Liu, W. Peng and Z. Xu, J. Clin. Lab. Anal., 2022, 36, e24685.
47. C. Wu, Y. Zhang, Y. Xu, L. Long, X. Hu, J. Zhang and Y. Wang, Biomaterials, 2023, 296, 122088.
48. D. Ramadon, M. T. C. Mccrudden, A. J. Courtenay and R. F. Donnelly, Drug Delivery Transl. Res., 2022, 12, 758–791.
49. B. W. Barry, Eur. J. Pharm. Sci., 2001, 14, 101–114.
50. T. A. Wilgus, Adv. Wound Care, 2019, 8, 671–678.
51. C. S. Mantsounga, Diabetes, 2020, 69, 591-P.
52. D. Zhou, T. Liu, S. Wang, W. He, W. Qian and G. Luo, Front. Physiol., 2020, 11, 545008.
53. G. Kolumam, X. Wu, W. P. Lee, J. A. Hackney, J. Zavala-Solorio, V. Gandham, D. M. Danilenko, P. Arora, X. Wang and W. Ouyang, PLoS One, 2017 DOI:10.1371/JOURNAL.PONE.0170639.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4bm01536c

---