Design, fabrication and comprehensive testing of biodegradable 3D printable hybrid polymer airway splints

Abstract

The trachea plays a critical role in respiration and airway protection but is susceptible to damage from pathological conditions such as stenosis, fistula, obstruction, and malacia. While existing treatment options are useful, they often have limitations, driving the need for innovative alternatives. This study introduces a novel approach using 3D printing technology to create hybrid degradable tracheal splints made of pectin-g-polycaprolactone (pec-g-PCL). We synthesized and characterized various compositions of pec-g-PCL to assess their physicochemical properties and tested their suitability for 3D printing. The resulting materials demonstrated the potential for use as tracheal splints. Using CAD software, we created two distinct designs, which were then fabricated according to those specifications. Micro-computed tomography (micro-CT) imaging revealed splint porosities ranging from 80% to 90%, highlighting their intricate internal microarchitecture. Design verification was conducted through numerical simulations, based on finite element modeling (FEM), to evaluate mechanical properties and computational fluid dynamics (CFD) for assessing the airflow dynamics of the fabricated tracheal splints. Degradation studies indicated that the 3D-printed scaffolds exhibited approximately 30% degradation over a period of 35 days. In vitro, biocompatibility assessments confirmed the scaffold's compatibility with biological systems. These findings demonstrate the potential of pec-g-PCL-based tracheal splints as a promising solution to overcome limitations in current treatments. This research paves the way for advanced biomaterials that could revolutionize patient care by offering more effective solutions for managing tracheal disorders.

---

1. Introduction

In the field of biomedical engineering, the combination of 3D printing technology with biodegradable materials has brought about a new wave of innovation, particularly evident in the creation of intricate designs.^1,2 The trachea's architecture, with its C-shaped cartilage rings and flexible tissues, is vital for maintaining airway integrity and flexibility, making it challenging to replicate.³ 3D printing provides a solution by enabling the creation of patient-specific tracheal splints with precise geometries and complex shapes that are difficult to produce by traditional methods.⁴ It allows for controlled porosity essential for tissue integration and cellular infiltration, resulting in biocompatible, bioabsorbable splints that degrade over time, reducing the need for further surgeries. The rapid prototyping capabilities also allow for quick design iterations based on clinical feedback, making 3D printing a powerful tool in tracheal splint development and potentially lowering medical costs. Various 3D printing techniques have been explored for biomedical applications, each offering unique advantages and limitations.⁵ Stereolithography (SLA) and digital light processing (DLP) provide high-resolution fabrication but require photocurable resins, which may raise concerns regarding biocompatibility. Selective laser sintering (SLS) enables the construction of complex geometries without the need for support structures; however, it involves high processing temperatures and is limited to powder-based polymers.^5,6 Fused Deposition Modeling (FDM) and extrusion-based printing are among the most widely used techniques for fabricating thermoplastic biomedical devices due to their operational simplicity, cost-effectiveness, and compatibility with biodegradable polymers.⁷

Tracheal disorders encompass a wide range of conditions affecting the structure and function of the trachea.^8,9 One common example is tracheomalacia, a condition characterized by weak cartilage in the trachea that leads to the collapse or narrowing of the airway during breathing. Another condition, tracheal stenosis, involves the narrowing of the tracheal lumen due to scarring or inflammation, leading to difficulty breathing.^10,11 Surgical techniques for tracheal disorders can be effective but have inherent disadvantages. Tracheal reconstruction and resection may lead to complications such as infection and anastomotic issues. Tracheostomy can result in infections and scarring.^12,13 Treatment options, such as stenting the trachea with non-biodegradable materials like silicone or metal, are effective in providing immediate support to the airway.¹⁴ However, these materials may lead to long-term complications such as inflammation, tissue damage, or the need for additional surgeries to remove the stent.

The emergence of biodegradable tracheal splints offers a promising alternative for the management of tracheal disorders.^15–17 These splints are designed specifically for each patient using their own imaging data, ensuring a perfect fit to their unique tracheal anatomy. This customized support helps to strengthen the weakened trachea, maintain an open airway, and enhance respiratory function. Additionally, the biodegradable materials used in these splints have several advantages over traditional non-biodegradable options.^18,19 As the splint slowly degrades within the body over time, there is no need for additional surgeries to remove it once it has served its purpose. This reduces the risk of long-term complications and streamlines the overall treatment process for patients with tracheal disorders.

Various polymers are explored in tracheal tissue engineering to develop innovative solutions for tracheal repair and regeneration. Among the polymers studied, polylactic acid (PLA)²⁰ and its copolymer poly(lactic-co-glycolic acid) (PLGA)^21,22 have received significant attention due to their biocompatibility, biodegradability, and adjustable mechanical properties. These polymers offer flexibility in fabrication methods, such as electrospinning and 3D printing, which enable the creation of scaffolds with customized structures and degradation rates.²³ Additionally, polycaprolactone (PCL) has been explored for its favorable mechanical properties and slower degradation, making it suitable for long-term support in tracheal tissue engineering applications.^20,24,25 Other polymers, like polyethylene glycol (PEG) and polyurethane (PU), have also been studied for their distinct features, such as hydrophilicity, flexibility, and biocompatibility.^18,26 By utilizing these polymers and their intrinsic properties, biomimetic scaffolds are developed to facilitate tracheal tissue regeneration and ultimately enhance outcomes for patients with tracheal disorders. Proteins and other natural polymers, such as polysaccharides, offer an advantage over synthetic polymers because they are more similar to living tissue.^27,28 Tracheal tissue engineering extensively explores natural polymers such as proteins and polysaccharides, including chitosan, gelatin, silk fibroin, collagen, hyaluronic acid, and fibrin.^29–35 These natural polymers are studied in various forms such as hydrogels, foams, coatings, and electrospun fibers. However, while these natural polymers offer biocompatibility and support cellular growth, they alone often lack the mechanical strength required to maintain the patency of the airway.^36,37

As per our knowledge among natural polymers, pectin remains unexplored in the field of tracheal tissue engineering. Pectin is a preferred material for the building of scaffolds for tissue engineering applications because it shares structural similarities with the glucosaminoglycans (GAG) in the extracellular matrix of human cells, which promote inter- and intracellular connections.³⁸ Currently, a wide variety of pectin-based scaffolds are used for this purpose, including sponge, electrospun fiber, hydrogel, 3D printed structures, and others.³⁹ Due to a lack of bioactive functional groups and their inherent hydrophobicity, PCL's ability to promote cell adhesion and proliferation is severely constrained. The limitations caused by the surface features of PCL-based scaffolds have led to the development of numerous modifications, such as coating or blending of PCL with other biocompatible materials to generate biocompatible scaffolds with improved mechanical strength and inherent bioactivity for tissue engineering.⁴⁰

However, challenges include the potential compromise of mechanical properties, variations in degradation rates leading to uneven tissue breakdown, and processing complexities to achieve uniform distribution.^41,42 Inherent polymer chain modification is a more efficient technique compared to surface modification. In a mass modification, cell signaling peptides are integrated directly into the biomaterials, ensuring that recognition sites are present throughout the material, both within its mass and on its surface. Efforts have been dedicated to achieving the necessary mechanical properties and tissue regeneration in tracheal tissue engineering by exploring diverse combinations of natural and synthetic polymers. However, the inherent modification of PCL with pectin within the realm of tracheal tissue engineering remains unexplored. Addressing these gaps will be instrumental in developing innovative tracheal substitutes.

An attempt is made here to include pectin into PCL to improve the properties of PCL. To obtain pectin-graft-PCL (pec-g-PCL) polymers, ε-caprolactone (CL) was polymerized by ring-opening polymerization with pectin. Pectin incorporation into PCL resulted in novel copolymers that improved the environment for tissue engineering. Moreover, after optimization of polymeric conditions and evaluating its diverse physico-chemical properties, we tested its printing ability to assess its potential possibility as tracheal splints. Two distinct tracheal splint designs are conceptualized and fabricated, with a focus on studying their mechanical behavior through numerical simulations. The biocompatibility of the polymers utilized in the splint designs was thoroughly examined. These combined efforts demonstrate the potential of these splints in facilitating tracheal tissue regeneration.

2. Materials & methods

2.1 Materials

High methoxyl pectin (galacturonic acid ≥74.0% on a dried basis, degree of esterification ≅60%, molecular weight 30 000–100 000 g mol⁻¹), ε-caprolactone (>99%, dried over calcium hydride for 24 hours and distilled under reduced pressure), stannous octoate (catalyst) were all procured from Sigma-Aldrich, Australia. All other reagents were used as received unless specified otherwise.

The L929 fibroblast (mouse) cell line was procured from Sigma-Aldrich, Australia. Dulbecco's Modified Eagle's Medium GlutaMax (DMEM), trypan blue, fetal bovine serum (FBS), and penicillin–streptomycin (PS) were sourced from Life Technologies, USA. Paraformaldehyde and sodium cacodylate buffer were obtained from ProSciTech, Australia. Phosphate-buffered saline (PBS), ActinGreen™ 488 ReadyProbes™ Reagent (Invitrogen, USA) and NucBlue™ Live ReadyProbes™ Reagent (Invitrogen, USA), Bone marrow-derived hMSCs (Lonza, Switzerland) were cultured and maintained at 37 °C and a humidified atmosphere with 5% CO₂ in alpha Minimum Essential Media (αMEM, Gibco, USA), 1% Penicillin–Streptomycin (Gibco, USA), and 10% FBS (Gibco, USA) until 80% confluency.

2.1.1 Synthesis of pec-g-PCL. Bulk ring-opening polymerization was performed to synthesize pec-g-PCL, in accordance with the procedure outlined in Scheme S1 (ESI†). Pre-determined amounts of pectin (Table S1, ESI†) were weighed and transferred to a 50 mL round-bottom flask under an inert nitrogen atmosphere. Stannous octoate (1 mol%) was then added as a catalyst, followed by the incremental addition of the CL monomer. The polymerization process was executed at a temperature of 100 °C for a duration of 16 hours, employing various weight ratios of pectin to CL, ranging from 1 : 5 to 1 : 100 (w/w). Upon completion of the designated reaction period, the reaction mixture was dissolved in acetone and precipitated in cold water. The resulting pec-g-PCL polymers were centrifuged at 10 000 rpm for 10 minutes, then dried under vacuum for 4 hours and left to dry overnight in the vacuum oven to ensure complete removal of residual solvents. The yield percentage and grafting rate (GR%) were calculated using eqn (1) and eqn (S2) (ESI†).

The yield (%) of pec-g-PCL polymers was calculated using eqn (1).

2.1.2 Design and 3D printing of external airway splints. The unique designs of the tracheal splints were created using FreeCAD software. The design incorporates a bellow-shaped cylinder with strategically placed rectangular convolutions, mimicking the natural architecture of the trachea and enabling the external placement of the splint in cases of airway collapse. The designs incorporated a fixed opening angle of 90° to ensure proper positioning of the device over the airway.

For the first model (M1), each convolution was designed with a longitudinal thickness of 1.5 mm and a separation distance of 2 mm. Key specifications included an inner diameter of 10 mm, a length of 10 mm, a wall thickness of 3 mm, and suture holes spaced at intervals of 2 mm.

In designing the second model (M2), we took inspiration from the composite structure of the trachea, which contains both rigid and flexible elements. To mimic this, the inner surface of the lumen of the tracheal splint was patterned with equilateral hexagons, each with an interior angle of 120 degrees. These hexagons were marked with 1 mm diameter circles, which helped create porous splints with optimized airflow and tissue integration. Additionally, the lumen surface was separated by double convoluted rectangles with thicknesses of 2 mm and 1 mm. The incorporation of double convoluted rectangles between these hexagonal patterns adds an interesting structural element, likely providing both flexibility and stability to the overall design. This combination of geometric shapes was found to be well-suited for mimicking the natural properties of the trachea while ensuring functionality and compatibility with surrounding tissues.

A 3D Bioprinter, the Allevi 3 model from Allevi Inc. in Philadelphia, PA, USA, equipped with hot melt extrusion capabilities ranging from 20 °C to 160 °C, was utilized for printing 3D splints with two distinct geometries using pec-g-PCL. The predetermined designs of the tracheal splints were uploaded as STL files to the Bioplotter. Each scaffold was fabricated employing various 3D printing parameters, including printing temperature, pressure, printing speed, layer height, and nozzle gauge size (Table S4, ESI†). To initiate the printing process, polymers were loaded into the nozzle and allowed to heat up to 70 °C for 10 minutes, reaching a molten state for extrusion. The fabrication of each splint type took approximately 30 minutes to complete. The printing-induced shear rate at the nozzle wall was calculated using eqn (2),⁴³

where r is the nozzle radius and S is the printing speed. The strand widths of printed scaffolds were estimated using the ImageJ software to calculate mean printing accuracy (P) (from four measurements) using eqn (3)⁴⁴where W_i is the printed width and W is the experimental width.

2.2 Characterization

The chemical structures of the pec-g-PCL samples were analyzed using a 400 MHz ¹H-NMR Bruker Spectrometer, employing deuterated chloroform and acetone [CDCl₃ and (CD₃)₂CO] as the solvents. Laboratory-based FTIR spectroscopic analysis was conducted using a PerkinElmer FTIR spectrometer at room temperature. The ATR-FTIR spectra were obtained using 16 co-added scans within the 4000–400 cm⁻¹ range. The synchrotron-based macro ATR-FTIR microspectroscopy was subsequently employed for high-resolution chemical imaging of these polymer samples at 2 μm pixel resolution, using an in-house developed macro ATR-FTIR device equipped with a 250-mm-diameter facet germanium (Ge) ATR crystal (n_Ge = 4.0), and a 20× IR objective (NA = 0.60).⁴⁵ In brief, the synchrotron-FTIR experiments were performed at Infrared Microspectroscopy (IRM) beamline, Australian Synchrotron (Victoria, Australia), using a Bruker Vertex 80v spectrometer coupled with a Hyperion 3000 FTIR microscope and a liquid nitrogen-cooled narrow-band mercury cadmium telluride (MCT) detector (Bruker Optik GmbH, Ettlingen, Germany). All the synchrotron-FTIR spectra were recorded using a projected aperture of 3.1 mm, 16 co-added scans and 4-cm⁻¹ spectral resolution, within a spectral range of 3800–700 cm⁻¹. Blackman–Harris 3-Term apodization, Mertz phase correction, and zero-filling factor of 2 were set as default acquisition parameters using OPUS 8 software suite (Bruker).⁴⁶

To evaluate the crystallinity of polymers, XRD analysis was performed using an XRD machine (Rigaku in Tokyo, Japan). X-ray intensity was measured between 2θ angles of 10 to 30°, using CuKα radiation with a wavelength of 1.54 Å and a scanning rate of 0.05 s⁻¹. The degree of crystallinity (X_C) was calculated using eqn (4)

where A_c represents the crystalline area and A_a represents the amorphous area. Additionally, contact angle measurements were obtained using a goniometer system equipped with a digital camera, employing the sessile drop method. A drop of distilled water (30 μL) was carefully placed on the polymer film, allowing it to rest for 1 minute to ensure wetting equilibrium. The static contact angle was then captured and computed within 2 to 5 seconds after this resting period.

2.2.1 Thermal analysis. A Thermogravimetric Analyzer (TGA, PerkinElmer, Pyris Diamond System) is used to assess the thermal stability of pec-g-PCL polymers. The samples, weighing around 10 mg, were subjected to a thermal ramp from 50 °C to 900 °C at a heating rate of 10 °C min⁻¹, under a constant nitrogen flow of 50 mL min⁻¹. Differential scanning calorimetry (DSC) using a TA instruments discovery DSC250 was employed to investigate the thermal behavior of various compositions of pec-g-PCL polymers. All experiments were conducted under a nitrogen atmosphere with a flow rate of 50 mL min⁻¹ to prevent oxidative degradation. For DSC analysis, each sample weighing between 9 and 10 mg was heated from −65 °C to 100 °C at a rate of 10 °C min⁻¹, held for 5 minutes to erase the thermal history, cooled to −65 °C and subsequently reheated to 100 °C. The data from the second heating ramp was used for analysis. The degree of crystallization (X_c) of the polymers was determined using eqn (5).where ΔH_m represents the enthalpy of fusion of PCL samples (in J g⁻¹), and ΔH_100% represents 100% crystallinity (139.5 J g⁻¹) of PCL. The enthalpies of fusion were derived from the area under the melting endotherm curve.

2.2.2 Rheological testing of PCL. The rheological properties of the synthesized polymers were evaluated using HR-2 Discovery Hybrid Rheometer (TA Instruments, Deleware) equipped with parallel plate geometry. Polymer discs with a 25 mm diameter and 2 mm thickness were hot-pressed for the study. An amplitude sweep experiment was conducted at 70 °C using a frequency of 1 rad s⁻¹ to determine the linear viscoelastic region (LVR). Subsequently, oscillatory experiments were performed using strain percentages within the LVR. The polymeric samples were subjected to angular frequencies ranging from 0.1 to 100 rad s⁻¹ at four different temperatures: 70 °C, 80 °C, 90 °C, and 100 °C, with a constant strain value of 1%.

To quantify the shear-thinning behavior of the polymers, the power law model was utilized, η([small phi, Greek, tilde]) = K[small phi, Greek, tilde]⁽ⁿ⁻¹⁾, where K is the consistency index and n is the power law index. In oscillatory temperature sweep mode, a temperature ramp from 60 °C to 140 °C at a heating rate of 5 °C min⁻¹ and a strain amplitude of 1% was applied to evaluate the dynamic complex modulus and viscosity. Additionally, the isothermal stability of the material at 70 °C was assessed through time sweep experiments. The time-temperature superposition principle was applied to the frequency sweep data to generate horizontally shifted master curves, using 70 °C as the reference temperature. The rheological master curves revealed the plateau modulus and relaxation behavior.

Before starting each experiment, all samples were equilibrated for 2 minutes to ensure a uniform temperature throughout. The measured data were processed using the TRIOS software. In the case of pec-g-PCL copolymers, TTS was used to create master curves by measuring the elastic (G′) and viscous (G′′) moduli at various temperatures and frequencies. The theory of time-temperature superposition (TTS) states that the linear viscoelastic properties of the polymer can be aligned to a specific reference temperature, T₀, through a single temperature-dependent shift factor, aT(T). The changes in the factors, a T(T), are evaluated using the following criteria (eqn (6)):

ln(aT) = E_a/R(1/T − 1/T₀), (6)

to generate a master curve for the evolution of the G′ and G′′ versus shifted frequency using 70 °C as the reference temperature. In eqn (6)R represents the universal gas constant (8.314 J·K⁻¹·mol⁻¹), E is the activation energy of flow indicating the temperature sensitivity of the flow process, T is the experimental temperature, and T₀ in kelvin is the reference temperature. The slope of the best-fit line (with R² = 0.99) from the plot of ln(aT) versus the reciprocal of the test temperatures provides the value of E_a/R.

2.2.3 Micro-computed tomography (μ-CT) imaging of tracheal splints. The 3D printed tracheal splints were analyzed using X-ray tomographic imaging using a Bruker Skyscan 1275 Micro CT, at 30 kV with no filter, 360° and 2 frames per second scanning rate. The procedure began with the acquisition of scan data, followed by reconstruction utilizing SkyScan NRecon GPU software. Representative 2D projection and reconstructed images were generated by applying a threshold between the two local maxima in the grey-scale histogram. For optimal alignment, datasets underwent potential reorientation using SkyScan Dataviewer software. Subsequently, the data analysis was done in both 2D and 3D using SkyScan CTAn software. To visualize the structures comprehensively, volumetric 3D models were crafted utilizing SkyScan CTVox software.

2.2.4 Mechanical testing. All tracheal splints fabricated by 3D printing underwent mechanical testing using a Mecmism adhesion tester (MultiTest dv (u)). A load cell of 500 N and a strain rate of 1 mm min⁻¹ were applied during the tests. Mechanical properties were assessed under radial and lateral compression. Splints measuring 10 mm × 10 mm × 10 mm (length × width × height) were compressed at a constant strain rate of 1 mm min⁻¹ between flat steel plates. Each specimen underwent at least 6 replicates for compression tests. Additionally, to evaluate recovery capability, cyclic compression was conducted under a 10% strain using the same load cell. For a specific strain level, splints underwent 50 cycles of compression, and stress–strain curves were recorded.

Finite element analysis was further conducted on the tracheal splints to understand their stress concentration under compression using the ANSYS 2022 R2 workbench. The SOLID185 element type was used with a meshing size of 0.2 mm. The boundary conditions are shown in Fig. S1 and Table S2 (ESI†). In radial compression tests, the symmetry plane of the splint was constrained, which can only move vertically. A displacement of 10 mm was applied on the top rigid plate, and the bottom rigid plate was fully fixed. In the lateral compression test, the symmetry plane of the splint was fully fixed, and a horizontal displacement of 10 mm was applied to the rigid plate on the right side. The total recorded displacement was considered as the platen's displacement value. Structural analysis was performed with a step size of 0.01 over 100 time steps. Reaction forces at the platens, along with displacement data, were used to generate force–displacement curves. These curves were then compared to experimental force–displacement data to assess the accuracy of the determined effective moduli in predicting mechanical behaviour.

To study the airflow through the trachea after placement of the tracheal splint on tracheal wall, the STL mesh is imported into ANSYS Fluent 14.5 (ANSYS Inc., Pennsylvania, USA), where tetrahedron elements and a patch-independent algorithm are utilized for mesh generation. The base mesh size is set to 0.2 mm, but the final size is determined through a mesh-independent evaluation, ensuring a tolerance of less than 0.1%. Depending on the model size, the mesh typically comprises 400 000 to 1 400 000 elements. Inlet velocity of 5 m s⁻¹, the calculations are conducted using the laminar model. The inlet and outlet are maintained at a constant pressure of 1.0 atm. The airway wall is assumed to exhibit a no-slip boundary condition. Air is treated as a Newtonian fluid with a constant density of 1.225 kg m⁻³ and viscosity of 1.8 × 10⁻⁵ kg m⁻¹ s⁻¹, which are reasonable assumptions considering the low pressure within the airway.

2.2.5 Scanning electron microscopy. The surface morphology of degraded samples was investigated using scanning electron microscopy (SEM) (FEI Nova NanoSEM). Samples were cross-sectioned, mounted on SEM stubs, and sputter-coated with Iridium for analysis. For cell culture samples, a series of steps were followed: after being washed thrice with sterile PBS and once with sodium cacodylate buffer, they were fixed with 100 μL of 1% osmium tetroxide for 30 minutes in the dark. Subsequently, samples were washed with Milli-Q water and then sequentially dehydrated with 50%, 70%, 90%, 95%, and 100% ethanol (10 minutes for each step) before being air-dried overnight.

2.2.6 Cell culture studies. Cytotoxicity testing was conducted in vitro following the guidelines outlined in ISO-10993-5 to evaluate the overall biocompatibility and safety of the splints, using the direct contact method. The mouse fibroblast cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) with high glucose, GlutaMAX™ supplement, and pyruvate. The medium was supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin (5000 U mL⁻¹).

The cells were grown in T25 or T75 culture flasks (Corning Inc., Corning, NY, USA) at 37 °C in a humidified atmosphere with 5% CO₂ and 95% air until they reached approximately 80% confluency. To subculture, the media was aspirated, and TrypLE™ was added to cover the cells. The flask was then incubated at 37 °C in a humidified atmosphere with 5% CO₂ for 5–10 minutes. Fresh DMEM was added, and the cell suspension was centrifuged at 200 × g for 5 minutes. After removing the supernatant, the cell pellet was resuspended in fresh DMEM and transferred to a new culture flask.

Bone marrow-derived hMSCs were cultured and maintained at 37 °C in a humidified atmosphere with 5% CO₂. The cells were grown in alpha Minimum Essential Media (αMEM) supplemented with 1% penicillin-and 10% FBS until they reached 80–90% confluency. Upon reaching confluency the cells were passaged using TrypLE™. For all the experiments described below, the highest passage number was passage 6 with the culture medium being carefully changed every 3 days. The polymeric samples were first sterilized by immersion in ethanol, followed by rinsing with sterile distilled water. The samples were then dried under laminar airflow before being exposed to UV light on each side for 10 minutes.

1. Preliminary cytotoxicity testing. To assess the viability of L929 cells on different compositions of the pec-g-PCL polymers and 3D printed polymeric splints. Control wells containing only cells and cell-free controls with only media were also included. After treatment, the cells were washed three times with phosphate-buffered saline (PBS), and fresh cell culture media containing 15 μg mL⁻¹ resazurin sodium salt was added to each well, including the cell-free controls. The plates were incubated at 37 °C for 3 hours before measuring fluorescence using a ClarioStar plate reader (BMG Labtech, Ortenberg, Germany) with 560 nm excitation and 590 nm emission. Each sample had three replicates, and the experiment was independently repeated two times. The proliferation of cells on polymeric splints was also evaluated on days 1, 3, and 7 using a similar resazurin-based assay described above.

The percentage cell viability at each time point was calculated using the following formula:

For SEM imaging, the cells were cultured in a 24-well plate containing polymeric scaffolds at a density of 10 000 cells per well, and the morphology was evaluated on day 7 after fixing them using fixative (details of fixatives are shown in SEM sample preparation).

2. Efficacy of the scaffold in supporting hMSC growth and proliferation. The hMSC cells were seeded onto polymeric scaffolds at a density of 10 000 cells per well in 96-well plates and incubated in a 5% CO₂ incubator at 37 °C for 24 hours for cell viability studies. For cell proliferation studies, the splints seeded with cells were incubated for 1, 3, 7, and 14 days. At each time point, the cell viability was assessed using resazurin sodium salt as previously mentioned. A minimum of three replicates were used for each assessment.

Cellular function and interaction with the scaffolds were assessed by staining actin filaments and counterstaining cell nuclei. Briefly, hMSC cells at a density of 5000 cells per well were seeded onto the splints and incubated for 7 days under standard conditions. Cellular adhesion and morphological observations were conducted using SEM analysis. Their morphology was examined after fixation on day 14, respectively.

3. Fluorescence imaging and alignment analysis. For imaging, cells were fixed in 100 μL of 1% osmium tetroxide for 30 minutes in 0.1 M sodium cacodylate buffer (pH 7.4) at 37 °C, followed by three rinses with sodium cacodylate buffer, each for 5 minutes. NucBlue™ Live ReadyProbes™ Reagent and ActinGreen™ 488 ReadyProbes™ Reagent were added to the cell dishes with 2 drops per mL to stain the nuclei and cytoskeleton for 20 minutes.

For actin staining, cells were permeabilized with 0.1% (v/v) Triton X-100 in sodium cacodylate buffer for 90 seconds, before actin staining. The cells were washed three times with sodium cacodylate buffer, and imaging was performed in the same buffer. LSM800 Airyscan (Zeiss, Germany) using a 60× water immersion objective and Axio Observer Z1Carl Zeiss microscope. Laser excitation and fluorescence emission settings were as follows: DAPI (Ex: 405 nm/Em: 415–495 nm) and Alexa Fluor 488 (Ex: 488 nm/Em: 500–554 nm). The images were processed using Fiji ImageJ software. To assess cellular alignment, F-actin images were analyzed using the OrientationJ and Directionality plugin (version 2.3.0) in Fiji, which applies Fourier spectrum analysis across an angular range of 0° to 180°. Image brightness and contrast were optimized, and background subtraction was performed in Fiji to minimize background interference in the analysis.

2.2.7 Degradation studies. A comprehensive experimental degradation protocol was employed to investigate the degradation pattern of pec-g-PCL cylindrical scaffolds, using six replicas of each 5 mm × 5 mm × 6 mm scaffold (width × diameter × height). For accelerated degradation studies, the replicas were then immersed in 10 mL of 5 M NaOH solution at 37 °C for 35 days for initial degradation studies. At specific time intervals, the samples were removed, dried overnight at 30 °C to remove any remaining liquid, and then weighed to determine weight loss.

The morphological characteristics of the degraded scaffolds were then examined using SEM. In addition, degradation was assessed using thermal and mechanical evaluation methods, following the procedures outlined in Sections 3.1 and 3.4.

2.3 Statistical analysis

Numerical values are presented as the mean ± standard deviation. Statistical analysis was performed using either one-way or two-way analyses of variance (ANOVA), followed by the Tukey post hoc tests. Data were expressed in the form of mean value ± standard deviation (SD). Statistical significance was considered at p < 0.05.

3. Results and discussion

Biodegradable natural and synthetic polymers are highly valued for creating tissue engineering scaffolds for tracheal regeneration, owing to their compatibility with biological systems.^47–49 Synthetic polymers, such as, PCL provide significant flexibility in terms of fabrication and modification; however, their hydrophobic nature may obstruct effective cell adhesion.⁴¹ On the other hand, natural polymers such as pectin provide excellent biocompatibility but lack mechanical strength and processing convenience.³⁹ The integration of natural and synthetic polymers in the development of tracheal regeneration splints creates a unique synergy that enhances both biocompatibility and mechanical properties, which are crucial for effective tracheal support and healing. To overcome the limitations of homopolymers, a diverse range of pec-g-PCL polymers are synthesized by grafting pectin using ROP of CL at 100 °C, with stannous octoate as a catalyst (Fig. 1 and Fig. S2, ESI†). It is observed that the yield of the polymer increases with varying compositions of pectin and CL, and a maximum yield of ∼97% (Table S1, ESI†) which is proportional to the decrease in monomer-to-initiator ratio.^50,51 Further, the structure of pec-g-PCL is confirmed using ¹H-NMR spectroscopy, which validated the successful ROP and attachment of CL onto the pectin polysaccharide. Fig. 2(a) and Fig. S2 (ESI†) show the ¹H-NMR spectra of pec-g-PCL with varying pectin to CL ratios, ranging from 1 : 5 to 1 : 100. The PCL displays CH₂ peaks from the opened ring of CL monomer at δ = 1.38 ppm, 1.55 ppm, 1.65 ppm, 2.3 ppm, and 3.63 ppm. The peak at δ = 4.05 ppm corresponds to the CH₂ group bonded to an oxygen atom in the PCL backbone. The presence of pectin is confirmed by its characteristic CH peaks at δ = 3.46 ppm, 3.81 ppm, 3.76 ppm, 4.21 ppm, 4.28 ppm, 4.53 ppm, 4.72 ppm, and 5.28 ppm. It is important to note that the peaks corresponding to pectin and PCL displayed minor shifts due to chemical interactions in the pec-g-PCL sample. The grafting percentage of pectin varied between 16% and 79% (see Table S1, ESI†) calculated from NMR spectra. The FTIR spectra of the synthesized pec-g-PCL polymers are shown in Fig. 2(b). The shift in OH stretching from 3375 cm⁻¹ (pectin) to 3450 cm⁻¹ (pec-g-PCL) suggests hydrogen bonding between the grafted PCL and the pectin chain.⁵² The pec-g-PCL copolymer spectra show an increased intensity of the ester C=O bond at 1724 cm⁻¹, along with a slight decrease in the broad OH signal at 3432 cm⁻¹ after copolymerization, indicating successful grafting through the hydroxyl groups on the surface of pectin. Notably, as the concentration of CL increased, the intensity of the absorption band associated with the OH group decreased significantly.

Fig. 1 Schematic illustration of the design, synthesis, and characterization of pec-g-PCL tracheal splints, biocompatibility, and degradation studies.

Fig. 2 (a) Representative ¹H NMR spectra of P3 and pectin; (b) reference ATR-FTIR spectra of the synthesized copolymers and pectin; (c) DSC curves of P3 and PCL; (d) XRD analysis of the synthesized copolymers; (e) contact angle measurements of synthesized copolymers; synchrotron macro ATR-FTIR results of PCL (top-row: (f)–(h)) and P3 (bottom row: (i)–(k)) including microscopic visible images, 2D and 3D high-resolution chemical maps of ν(C=O) band of PCL based on integrated area under the peak in 1765–1670 cm⁻¹ (left to right).

Advanced synchrotron macro ATR-FTIR microspectroscopy enables molecular characterization of heterogeneous materials with high spatial resolution, using step intervals as small as 250 nm.⁴⁵ In this study, the technique was employed to investigate the spatial chemical distribution of PCL and pectin segments within the pec-g-PCL (P3) copolymers at the grafting sites. These insights provide a deeper understanding of grafting efficiency, phase compatibility, and structural heterogeneity in the copolymer system.

Fig. 2 illustrates the differences in surface morphology and chemical composition between PCL and the P3 copolymer. Both 2D and 3D synchrotron macro ATR-FTIR chemical maps were generated based on the integrated area under the ν(C=O) stretching absorption band of carbonyl groups (∼1750 cm⁻¹), which is the key characteristic of PCL.

For PCL (top row: Fig. 2(f)–(h)), the microscopic image reveals a relatively smooth surface morphology, consistent with a homogenous polymer structure. Their corresponding 2D and 3D chemical images further confirm its chemical homogeneity, as evidenced by the even and intense distribution of ν(C=O) absorption across the surface. In contrast, the incorporation of PCL onto pectin copolymer (bottom row: Fig. 2(i)–(k)) introduces significant changes to both morphology and chemical distribution. The microscopic image highlights a textured and irregular surface morphology, indicative of structural disruptions caused by the presence of pectin. The 2D chemical map reveals substantial heterogeneity in the distribution of ν(C=O) absorption, with reduced intensity across the majority of the surface. This heterogeneity is further emphasized in the 3D chemical map, which illustrates an uneven chemical landscape with localized regions of reduced carbonyl group concentration.

These findings strongly suggest successful grafting of PCL onto pectin, as evidenced by the altered morphology and distinct changes in the ν(C=O) distribution. The reduced intensity and heterogeneous pattern of ν(C=O) absorption in the P3 copolymer are consistent with the presence of pectin chains disrupting the uniformity of the PCL matrix. This structural and chemical evidence highlights the impact of pectin incorporation on the overall matrix, confirming its role in modifying both the morphology and chemical composition of the copolymer.

The thermal behavior of pec-g-PCL polymers was analyzed using TGA under a nitrogen atmosphere. Fig. S3a and b (ESI†) show the characteristic TGA and derivative TGA curves for PCL and pectin-g-PCL polymers (P1 to P5). The pec-g-PCL polymers, with varying feed ratios, exhibit two stages of degradation. The first stage, occurring at T_max1 = 270–305 °C, corresponds to the degradation of the pectin backbone due to saccharide ring dehydration. The second stage, at T_max2 = 348–412 °C, corresponds to the decomposition of PCL. Pectin, with its weak glycosidic bonds, exhibited a low initial degradation temperature of around 197 °C, but the incorporation of PCL chains significantly improved the thermal stability of the resulting polymers.^38,52 The melting and crystallization properties of pec-g-PCL polymers were analyzed through DSC, which provides valuable insights into the polymers’ chain mobility. Fig. 2(c) and Fig. S3(c), (d) (ESI†) illustrate the DSC curves of pec-g-PCL with various feed ratios (P1 to P5). The melting temperatures varied from 46 °C to 56 °C, while the crystallization temperatures ranged from 19 °C to 29 °C. It's evident that when the amount of CL incorporation increased, the melting temperature approached the T_m of pure PCL. Moreover, the DSC data shows that increasing the PCL arm length raises the enthalpy of fusion, significantly influencing the degree of crystallinity (Table S3, ESI†). Pectin's amorphous nature disrupts the arrangement of PCL chains, restricting chain mobility and hindering crystallization. Reducing pectin content allows greater PCL chain movement, promoting crystallization and increasing the proportion of crystallizable PCL, which enhances the overall crystallinity of the polymer.^51,53

Interestingly, the double melting peak observed in pec-g-PCL polymers (Fig. 2(c)) is likely due to lamellar thickening, which occurs during heating in thermal analysis. This observation suggests the coexistence of two distinct lamellar structures or thicknesses in the copolymer, possibly formed through annealing during the heating process.^51,54 The diffractogram (Fig. 2(d)) of the pure pectin shows two broad halos at 12.9° and 22° of 2θ, which is in line with what has been reported in the literature for pure pectin films.¹⁷ In the XRD spectrum of pec-g-PCL (Fig. 2(d) and Fig. S4, ESI†), the typical PCL peaks at 2θ = 21.5° and 24.5°, corresponding to the semi-crystalline diffraction of the (110) and (200) lattice planes of PCL's orthorhombic crystal lattice, indicate lower crystallinity compared to pure PCL samples. This decrease in crystallinity in pec-g-PCL compositions is promising, as it is in line with earlier reports suggesting that lower crystalline structures create favorable conditions for cell adhesion and tissue growth.⁵⁵

Contact angle measurements are used to assess the wettability of the polymers. Fig. 2(e) displays the contact angle measurements for pec-g-PCL polymers with varying compositions. As the concentration of CL increases, the contact angle rises, indicating a gradual increase in hydrophobicity.⁵⁶ The decrease in contact angle for polymers with higher pectin content is due to pectin's hydrophilic functional groups, such as carboxyl and hydroxyl groups, which attract water molecules and enhance the polymer's interaction with water. Additionally, pectin's amorphous structure also contributes to higher water absorption, enhancing surface wettability.^56,57 These observations provide clear evidence of PCL grafting onto pectin, significantly altering the surface properties and confirming the modified chemical landscape essential for biomedical applications. Thus, the successful grafting of PCL onto pectin not only confirms chemical modification but also provides a pathway to optimize the 3D printability of these biopolymer systems for advanced biomedical applications.

3.1 Rheological and 3D printability test of pectin-g-PCL polymer

The 3D printability of grafted polymers is primarily influenced by their flow properties. Ideal inks for 3D printing applications must exhibit viscoelastic behavior. During extrusion, they should behave as viscoelastic fluids, displaying shear-thinning behavior under applied stress. Once extruded, these inks should transition into self-supporting viscoelastic solids with sufficient viscosity to maintain their structural.⁵⁸ The rheological analysis further provides insights into the macromolecular structure of the pec-g-PCL, enhancing our understanding of its printability. The elastic and viscous nature of various pec-g-PCL compositions (P1 to P5) is analyzed by evaluating the storage modulus (G′) and the loss modulus (G′′) at 70 °C to understand the extrudability of the printing materials. Before conducting oscillatory measurements, the LVR region is identified to isolate regimes where the material structure remains intact. The results of the amplitude sweep, depicted in Fig. 3(a), showed that the storage modulus (G′) and loss modulus (G′′) remained almost constant, suggesting that the polymer structure was unaffected at low strain amplitude. However, both G′ and G′′ decreased at high strain amplitude. The rheological behavior of the pec-g-PCL copolymers exhibits notable variations depending on the percentage of pectin incorporated. When a higher proportion of pectin is present (for example, P1 and P2), the pec-g-PCL tends to display a transition from predominantly elastic to viscous behavior as strain increases. Initially, the storage modulus (G′) surpasses the loss modulus (G′′), indicating a more elastic response. However, with increasing strain amplitudes, the viscous nature of pec-g-PCL becomes increasingly prominent, causing a crossover where the loss modulus exceeds the storage modulus. This shift implies the material's transition from predominantly elastic deformation to more viscous flow. In the pec-g-PCL compositions from P3 to P5, both the storage modulus (G′) and loss modulus (G′′) remain relatively constant. This behavior suggests that at higher PCL concentrations, the material mainly demonstrates viscous properties. As the proportion of PCL increases in the pectin-g-PCL polymers, the loss modulus (G′′) exceeds the storage modulus (G′), indicating that the material becomes more viscous than elastic under the applied strain conditions. This change reflects the predominant influence of PCL, which overrides the elastic characteristics imparted by pectin. Overall, the viscoelastic behavior of the polymers shifts towards a predominantly viscous response. As a result, a 1% strain (within the LVR) was used for the remaining experiments. A similar trend is evident in the frequency sweep experiments, as illustrated in Fig. 3(b). When the pectin content in the pec-g-PCL composition increases, the intermolecular distance between pectin molecules decreases. This allows for the formation of hydrogen bonds and electrostatic interactions, creating a cohesive, structured network that gives the material gel-like and elastic properties. Conversely, when the pectin content decreases and the PCL content increases, the intermolecular distance increases, leading to a reduction in these interactions and a weakening of the network structure.^59,60

Fig. 3 Rheological testing of pec-g-PCL polymers (a) strain sweep (0.1–100%) (b) angular frequency at 1% strain amplitude (c) temperature sweep at 1% strain and 1 rad s⁻¹ frequency (d) TTS curve of P3 at 70 °C (e) apparent viscosity of pec-g-PCL polymers (f) activation energy and shear yield stress of pec-g-PCL polymers.

In 3D printing, it is important to maintain continuous ink flow, which depends on the shear-thinning behavior of the printing inks once the ink flow begins.^58,61Fig. 3(c) shows the complex viscosity curve of pec-g-PCL (P1 to P5) over a temperature range of 60–140 °C. For polymer compositions of P1 and P2, the complex viscosity remained relatively constant at elevated temperatures, which makes extrusion challenging due to the lack of shear-thinning behavior. For P3 to P5, the viscosity of the polymer melts initially decreased significantly with rising temperatures, followed by a more gradual decline. Consequently, a printing temperature of 70 °C was selected for extrusion, as it optimizes the reduction in viscosity and promotes the onset of shear-thinning behavior. In pec-g-PCL compositions, the ratio of pectin to PCL had a significant impact on printability and structural integrity, as shown in Fig. 3(d). P1 and P2 resulted in gel-like, elastic properties that improved shape retention and stability. However, this elasticity led to nozzle blockage and complicated extrusion. Intermediate ratios, such as P3 and P4, showed a balanced blend of elasticity and viscosity, supporting effective extrusion and shape retention. Conversely, higher PCL content ratios, like P5, led to dominant viscous behavior, improving extrusion while causing poor shape retention and spreading due to a higher G′′ than G′. For optimal 3D printing, especially for applications like tracheal splints, an intermediate ratio like P3 strikes a suitable balance, combining elasticity with viscosity. Further, the thermal stability of the copolymers was tested for assessing their suitability in the 3D printer nozzle through dynamic time sweep experiments at a constant frequency of 10 rad s⁻¹ and a temperature of 70 °C (Fig. S6b, ESI†). The results indicated consistent measurements for both the storage modulus (G′) and the loss modulus (G′′), demonstrating the material's stability over time, allowing enough time for 3D printing experiments to be completed. Typically, a splint of dimensions 12 mm × 10 mm × 10 mm can be completed within 30 minutes.

Additionally, the time-temperature superposition (TTS) principle has been applied to understand the viscoelastic behavior of copolymers across different time scales and temperatures. It helps to predict long-term material behavior based on short-term, high-temperature tests. The shifted data collected at various temperatures for pec-g-PCL (P3) (Fig. 3(d)) was observed to be superimposable on each other by a simple horizontal shift, suggesting that the principle of time-temperature superposition is applicable. The temperature dependence of the shift factor (aT) for pec-g-PCL (P3) also follows the Arrhenius-type equation (Fig. 3(f)). The E_a value for pec-g-PCL (P1 to P5) varied between 40 and 31 KJ mol⁻¹, which is consistent with values reported in previous literature.⁶² The shift factor, determined using the Arrhenius equation, indicated that higher G′ values at lower frequencies suggest that pectin enhances the elastic properties of the polymer over longer time scales. This increased elasticity is due to stronger intermolecular interactions resulting from pectin's functional groups. Moreover, the activation energy of flow increases with higher pectin content, making the material more difficult to process as these interactions resist flow. In Fig. 3(f), the shear yield stress of pec-g-PCL polymers is shown at different pectin concentrations. The data demonstrates that the yield stress increases with higher pectin content. For example, polymer P5 have low yield stress and low viscosity, leading to issues such as lateral spreading during deposition. On the other hand, polymers with higher pectin content, such as ratios P1 and P2, have higher viscosity, resulting in clogging of the printing nozzles (as indicated in Fig. 3(e)) Therefore, the polymer composition of P3 strikes a balance, offering optimal rheological properties suitable for 3D printing. This balance is crucial as it prevents both excessive spreading and nozzle clogging. The printing accuracy of pec-g-PCL compositions ranged from ∼66 to ∼80% (Table S4, ESI†). By carefully controlling the ink formulation and fine-tuning the printing parameters, it is possible to create 3D printable inks that maintain uniform properties in the printed structures. Based on these findings, the 3D printability of different compositions of pec-g-PCL copolymers is examined. A geometric design in the form of an STL file was imported into a bioprinter, where the copolymer was fed into the nozzle and extruded through a cylindrical metal tip using pneumatic pressure. The crosshead speed and temperature parameters used are detailed in Table S4 (ESI†). We studied pec-g-PCL polymers with different ratios (P1 to P5). Rheological investigations indicated that compositions with elevated pectin content (P1 and P2) resulted in nozzle blockages, thereby hindering their printability where as P5 polymer composition was deemed unsuitable for 3D printing due to its excessively low viscosity. In contrast, compositions P3 and P4 exhibited successful 3D printing capabilities, demonstrating commendable shape fidelity and extrusion quality. Consequently, P3 composition of the pec-g-PCL polymer has been selected for further investigation due to its superior hydrophilicity, rheological properties, and printability, making it our focus for further study.

3.2 Design and fabrication of tracheal splint

Two distinct splint models (Fig. 4(a)) were created using CAD modeling to investigate the influence of geometrical design and porosity on mechanical properties and cellular behavior. Both models, measuring 10 mm × 10 mm × 10 mm, were fabricated from the P3 composition using an Allevi 3 bioprinter (Fig. 4(b)). The printing parameters included a temperature of 70 °C, a pressure of 10 psi, a 30-gauge nozzle, a layer height of 0.2 mm, a print speed of 1 mm s⁻¹, and a print bed temperature of 25 °C. Micro-CT imaging provided detailed 3D visualizations of the 3D printed P3 splints, enabling a comprehensive comparison with commercial PCL splints to assess manufacturing precision (Table S5, ESI†).

Fig. 4 Design and fabrication and micro-CT analysis of tracheal splint (Model 1-M1 and Model 2-M2). (a) CAD models of tracheal splint [front view and top view]; (b) 3D printed pec-g-PCL tracheal splints; (c) micro-CT imaging of tracheal splints, porosity and printing resolution of 3D printed tracheal splints calculated using micro-CT data; (d) porosity levels and printing resolutions; (e) Euler number and volume discrepancy of 3D printed tracheal splints obtained from micro-CT analysis.

The images showed good layer-to-layer adhesion, confirming the successful integration of individual layers as shown in Fig. 4(c). Fewer voids were visible in the micro-CT images both on the interior and surface of the splints (Fig. S8, ESI†), with quantitative analysis revealing porosity levels between ∼65% to ∼80% respectively (Fig. 4(d)). Splint M2 exhibited higher porosity than M1, likely due to differences in their design specifications. Both pec-g-PCL splints (P3-M1 and P3-M2) displayed greater porosity compared to the commercial PCL splints (PCL-M1 and PCL-M2). This increased porosity in pec-g-PCL may result from the grafting density, which disrupts the uniform polymer network. Additionally, altered crystallization behavior in pec-g-PCL could further contribute to porosity during solidification. The higher printing resolution of 99% for commercial PCL, compared to 80% for pec-g-PCL (Fig. 4d), indicates that commercial PCL achieves finer details and sharper edges. The lower resolution for pec-g-PCL may be due to differences in material properties, such as altered viscosity or crystallization, affecting the precision of layer deposition during 3D printing, as seen in Fig. 4(d). Micro-CT was further utilized to evaluate structural parameters such as the connectivity factor (Euler characteristic) and pore distribution.^63,64 For Model 1, the high porosity was attributed to hollow structures, with a high positive Euler number suggesting that these hollow areas did not form large interconnected voids, thereby maintaining topological simplicity. In Model 2, the presence of thin walls and lattice-like features led to high porosity and a high Euler number, as the structure comprised many small solid regions separated by porous sections, adding complexity (Fig. 4(e)). Model 1 tracheal splints exhibited less volume discrepancy compared to Model 2 splints, with discrepancies ranging between 12–18% (Fig. 4(e)). Some of these discrepancies can be attributed to the intricate design features, where the material deposition may not be uniform or may fail, leading to variations between the intended and actual printed volumes. Hence, micro-CT imaging effectively characterized the structural integrity and porosity of 3D-printed pec-g-PCL splints. The observed micropores and porosity levels provide valuable insights into the splint architecture, which is critical for optimizing performance in tissue engineering, particularly for tracheal regeneration.

3.3 Mechanical testing of tracheal splints

The mechanical characterization and Finite elemental analysis (FEA) conducted in this study provided valuable insights into the performance of C-shaped tracheal splints under radial and lateral compression. The trachea exhibits distinct mechanical behaviors in these two directions, which are primarily influenced by its elliptical lumen shape and the membranous portion.^32,65 To analyze the impact of different geometrical shapes on the mechanical properties of the splints, the stress distribution in M1 and M2 splints under radial and lateral compression, both subjected to identical displacement, was evaluated in Ansys Workbench.⁶⁶ The PCL material was assumed to be isotropic, with a Young's modulus of 200 MPa, determined experimentally through compression testing on a cylindrical scaffold, and a Poisson's ratio of 0.3 based on values reported in previous literature.⁶⁴ The boundary conditions are detailed in Fig. S1 (ESI†).

Fig. 5(a)–(d) and Fig. S9 (ESI†) show the stress distribution of the porous splint for M1 and M2 under radial and lateral compression respectively. Stress analysis under radial compression deformation revealed that both types of porous splints experienced stress concentration primarily on the top and lateral sides. However, the maximum stress was localized at the corners of the convoluted grooves in M1 splints and the corners of the hexagonal array in M2 splints. The higher magnitude and wider distribution of maximum stress in M1 splints likely contributed to their earlier yielding under radial compression, as shown in Fig. 5(e). A similar trend was observed for M1 and M2 splints under lateral compression. The mechanical properties and structural integrity of the splints were assessed experimentally by applying controlled loading conditions. The tracheal substitute underwent mechanical tests, including radial and lateral compression. In the radial compression tests, the load–deformation curves of P3 (M1 and M2) were slightly lower than those of PCL in both models (Fig. 5(e)–(h)), indicating reduced stiffness for P3 compared to PCL. The reduced stiffness of P3 compared to PCL can be attributed to several key factors related to its composition and structure. The introduction of pectin into the polymer matrix disrupts the crystalline structure of PCL, leading to a less ordered molecular arrangement, which in turn lowers stiffness. However, no significant differences were observed in the load–deformation curves or stiffness during lateral compression tests. Splints that achieved at least 20% of the maximum displacement, corresponding to a 10% separation between platens under a 10 N load, satisfied the requirement to accommodate growth over eight months in a porcine model.⁶⁷ This threshold of 10% displacement under a 10 N load is a conservative estimate to allow for pediatric tracheal growth.⁶⁷ The 3D-printed splints displayed slightly greater stiffness than that reported for native tracheas in the literature, suggesting that these 3D-printed pec-g-PCL splints could effectively preserve tracheal shape and endure physiological pressures in vivo.⁶⁶

Fig. 5 Mechanical study of 3D printed splints (a) Von Mises elastic stress distribution in M1 under radial compression (b) total deformation of M1 under radial compression (c) Von Mises elastic stress distribution in M2 under radial compression (d) total deformation of M2 under radial compression (e) and (f) experimental radial compression test (g) and (h) experimental lateral compression test (i) and (j) CFD analysis of airflow through the splinted tracheal lumen (k) and (l) cyclic radial compression of M1 and M2.

Computational fluid dynamics (CFD) was employed to analyze airflow in the trachea under physiological conditions. The velocity distribution within the tracheal lumen is depicted in Fig. 5(i) and (j), showing that airflow remained uninterrupted after the placement of the splint based on velocity fields, while noting that a detailed wall shear stress analysis would further strengthen this observation and is recommended for future work. The highest velocity was observed at the central region, gradually decreasing towards the periphery due to wall shear stress exerted by the tracheal walls. Comparing both splints, M2 exhibited higher turbulence kinetic energy than M1, attributed to differences in their geometrical features. This study confirmed that the splints do not obstruct airflow but instead support patients suffering from tracheal collapse.

Resilience after deformation is another crucial feature for external airway splints in medical applications. To evaluate this, both radial and lateral compression tests involving 50 cycles were conducted on the tracheal substitutes. The load–displacement curves for M1 and M2 under radial compression are shown in Fig. 5(k) and (l). After the initial compression cycle, there was a rapid reduction in load, with M1 dissipating more energy than M2 under both radial and lateral compression. Notably, the mechanical response stabilized even when subjected to 50 compression cycles at a radial deformation of 10%.

3.4 Degradation and biocompatibility studies

Assessing the biodegradability of scaffolds is pivotal for gauging their compatibility with the temporal demands of cell differentiation and tissue regeneration. Following this, the impact of accelerated degradation on the characteristics of P3-based scaffolds was evaluated by analyzing weight loss, surface morphology, and compressive modulus of both radial and lateral compression tests, in comparison to those of commercial PCL scaffolds used as controls. Fig. 6(a) illustrates the 3D cylindrical model used for degradation studies and Fig. 6(b) shows the photographic images of degraded samples. In Fig. 6(c), there is a comparison of weight loss between P3 scaffolds and commercial PCL scaffolds. Initially, the degradation process showed non-uniform mass loss, mainly due to cleavage in the non-uniform regions of the polymer. However, over time, the P3 based scaffolds degraded more quickly, with a mass loss of 35 ± 0.4% by day 28, compared to the commercial PCL scaffolds, which had a mass loss of 21 ± 0.9%. By day 35, the P3 scaffolds completely broke down, resulting in a loss of mechanical strength. The faster degradation of pec-g-PCL scaffolds under hydrolytic conditions is caused by several factors. Firstly, the presence of hydrophilic pectin in the PCL scaffold leads to increased water absorption, which accelerates the breakdown of polymer chains.^38,57,68 This increased breakdown is due to the easily broken ester linkages in the graft copolymer's pectin and PCL segments. As a result, increased water absorption and ester linkages lead to faster degradation of pectin-g-PCL scaffolds, exceeding the breakdown rate of conventional PCL scaffolds.

Fig. 6 (a) A 3D cylindrical model used for degradation studies; (b) photographic images of degraded samples; (c) residual weight % of degraded samples under accelerated degradation condition; (d) Young's modulus of degraded samples under radial compression; (e) cell proliferation studies using L929 cells for 7 days; (f) confocal imaging of L929 cells on 3D printed splints after 7 days of incubation; (g) SEM imaging of L929 cells attached on 3D printed splint after 7 days of incubation; (h) cell proliferation of hMSC cells on 3D printed splints for 14 days; (i) confocal imaging of hMSC cells on 3D printed splints after 7 days of incubation; (j) hue, saturation and brightness (HSB) maps generated with the OrientationJ plugin (the color code is indicated in the bottom left); (k) alignment profile of hMSCs cultured on 3D printed splints; (l) SEM imaging of hMSC cells attached on 3D printed splints after 14 days of incubation.

Fig. 6(d) shows how the compressive modulus of degraded scaffolds changes over time, confirming the degradation process observed in weight loss studies. Initially, the mechanical properties remained stable for the first three days. However, after 28 days, there was a noticeable reduction in the elastic modulus for all samples. The P3 scaffolds experienced a 50% decrease, while PCL scaffolds experienced a 30% decrease. Unfortunately, it was not possible to analyze the mechanical properties on day 35 due to scaffold disintegration. This decrease in modulus is likely due to the significant mass loss caused by the accelerated degradation mechanism. SEM findings confirm that surface erosion is the main way these structures break down under alkaline conditions (Fig. S10(a), ESI†). Initially, the surface of the scaffold was smooth, but over time it became rougher and developed pores and cracks. This change was more noticeable in P3 scaffolds compared to commercial PCL scaffolds, suggesting that the lower crystallinity and higher water accessibility in P3 scaffolds are the main reasons for their breakdown. These degradation studies clearly show that the breakdown of the fabricated scaffolds is influenced by their chemical composition.

It is also important to note that the composition of the polymer framework can significantly impact cell viability. Preliminary tests for cytotoxicity of the pec-g-PCL polymers using the L929 fibroblast cell line, showed positive results. The L929 cells were grown on various compositions of pec-g-PCL polymers for 24 h, and their viability was evaluated. The results showed cell viability > 90%, confirming that the developed polymers are non-cytotoxic (Fig. S11a, ESI†). Among them, pec-g-PCL (P3) was selected for further studies. Similarly, L929 cells cultured on 3D printed splints demonstrated high cell viability (Fig. S11b, ESI†). Among the different compositions, pec-g-PCL (P3) was used to evaluate its potential for cell proliferation and was compared to commercial PCL scaffolds. Fig. 6(e) reveals a promising progression of viable cells on 3D-printed scaffolds from day 1 to day 7. However, the proliferation of cells on PCL-based splints was significantly hindered by the 7th day, suggesting that the hydrophobic nature of the scaffold could impede further proliferation once cell density reaches a certain threshold (Fig. 6(e)). The influence of geometry on cell proliferation was evaluated by seeding cells onto 3D-printed splint models made from pec-g-PCL and comparing them to commercial PCL.

Confocal imaging revealed that among the scaffolds tested, cell proliferation was significantly enhanced on the P3 scaffolds, with the highest proliferation observed on the P3-M2 splint (Fig. 6(f)). Cell adhesion activity of L929 cells on 3D-printed splints was assessed at 2 and 24 hours (Fig. S11c, ESI†). After 24 hours of cell seeding, the number of cells adhering to P3-based splints were higher than on commercial PCL splints. However, no significant difference was observed between the two groups of P3-based splints at 24 hours, highlighting the excellent early cell adhesion activity of P3 splints (Fig. S11b, ESI†). The SEM analysis (Fig. 6(g)) of the scaffolds provides additional support for cellular adhesion onto the 3D-printed scaffolds. These findings suggest that the pec-g-PCL scaffolds, with a more favorable 3D microenvironment, promoted better attachment and proliferation. The combined results suggest that the developed pec-g-PCL polymeric scaffold has favorable cytocompatibility properties.

In tracheal tissue engineering, the interaction between cells and their surroundings plays a crucial role in determining the effectiveness of the engineered tissue.^9,69 This interaction is facilitated by the extracellular matrix (ECM), which enables specific interactions between integrins and ligands, impacting cell proliferation and other functions.⁷⁰ Therefore, a 3D scaffold must have high porosity, appropriate pore size, an interconnected pore network, and mechanical properties to support optimal cell growth, migration, attachment, and proliferation.^71–73

The study evaluated the effectiveness of two different tracheal splint designs in supporting the growth and proliferation of human mesenchymal stem cells (hMSCs). hMSCs were chosen due to their extensive use in the development of 3D-printed tracheal constructs, attributed to their exceptional regenerative and immunomodulatory properties. These multipotent cells can differentiate into critical cell types, such as chondrocytes and fibroblasts, which are vital for the structural and functional integrity of the trachea's cartilage and connective tissues. Over a 14-day culture period, the polymer-based splints demonstrated comparable levels of cellular metabolic activity, with no significant differences observed, as shown in Fig. 6(h). The P3 splints showed a higher proliferation rate at day 14 than commercial PCL. Finally, high coverage of green viable cells on the surface of P3 splints was observed when stained with actin green and nucblue (Fig. 6(i)). This further confirmed the biocompatibility of the scaffolds, with sustained cell growth and proliferation maintained on modified PCL splints. The P3 splints exhibited a greater number of adhering hMSCs after 24 hours of incubation compared to the commercial PCL splints (Fig. S11e, ESI†). The observed growth and proliferation of hMSCs on the 3D printed splints can be attributed to the substrate's topographical features and chemical composition, creating a supportive microenvironment that promotes cellular proliferation. hMSCs possess the ability to self-renew and differentiate into multiple cell lineages in vitro. This process is influenced by factors such as growth factors, the surrounding microenvironment, and the availability of substrates with varying topographies and stiffness. Therefore, achieving proper cell orientation is crucial for efficient differentiation while minimizing potential side effects from external growth factors and inducers present in the culture medium.

In Fig. 6(j) and (k), the directionality analysis indicates that the surface topography of the 3D-printed splints significantly influences cellular behavior, promoting elongation and alignment along the microgrooves formed during the 3D printing process. The hMSCs cells on commercial PCL splints align with orientation angles ranging from 73° to 19°, while on P3 splints, the orientation angles are higher, at 105° and 126° (Fig. 6(k)). This observation suggests a difference in how the cells interact with each type of material surface. The lower orientation angles on commercial PCL imply that the cells may be aligning more uniformly or in a restricted manner, which could be due to a smoother or less complex surface topography. In contrast, the higher orientation angles on P3 splints suggest that the modified surface properties of pectin-g-PCL, such as increased roughness or chemical interactions, allow for a broader range of cell alignment, potentially enhancing cell adhesion and spreading. Further, SEM analysis also confirmed the elongated cell morphology and extensive cell–cell networking, supporting these observations. Additionally, higher hMSCs adhesion was evident on P3 splints, as shown in Fig. 6(l). These findings support the potential of pectin-g-PCL scaffolds to provide a supportive and biocompatible environment for tracheal tissue engineering applications.

4. Conclusions

In conclusion, this study successfully developed a hybrid polymer system by grafting pectin onto PCL through a solvent-free, one-pot polymerization process. The resulting pec-g-PCL polymer exhibited tunable physicochemical properties, including optimized hydrophilicity, crystallinity, rheology, and 3D printability, making it an ideal candidate for fabricating tracheal splints. Using advanced 3D printing technology, we successfully fabricated tracheal splints with two distinct designs, achieving a porosity range of 60–80% as confirmed by micro-CT analysis. The printing resolution of pec-g-PCL reached 80%, which, although slightly lower than that of commercial PCL, was sufficient for precise scaffold fabrication. The pec-g-PCL polymer exhibited excellent cytocompatibility, customizable mechanical properties, and an enhanced degradation rate, all of which are crucial for tracheal tissue regeneration. Furthermore, the significant increase in hMSC cell proliferation highlights its suitability for promoting tissue growth and repair. These findings represent a notable advancement in the development of next-generation biomaterials for regenerative medicine, offering promising solutions for tracheal disorders and paving the way for improved patient outcomes. While 3D-printed airway splints show significant promise, their clinical translation is subject to several regulatory challenges. Additionally, long-term biocompatibility, degradation behavior, and performance must be validated through rigorous preclinical and clinical studies. With proper quality control, bioresorbable 3D printed splints like those developed in this study could be translated into clinical use for treating tracheomalacia and other airway disorders, particularly in pediatric patients, where growth accommodation is crucial.

Author contributions

Sonali S. Naik: conceptualization, data curation, formal analysis, investigation, methodology, writing original draft. Jitraporn Vongsvivut: investigation. Chaitali Dekiwadia: investigation, Amanda N. Abraham: investigation. Namita R. Choudhury: conceptualization, supervision, writing, review & editing, funding acquisition, project administration, resources, validation. Naba K. Dutta: supervision, writing, review & editing. S. Kiran: conceptualization, funding acquisition, project administration, resources, supervision, validation, writing, review & editing.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

All data have been presented in the manuscript.

Acknowledgements

The Author Sonali Naik would like to thank RMIT University for PhD scholarship as a part of AcSIR-RMIT joint PhD program. This work has been financially supported by the Center of Excellence (CoE) (F. No. 25014/2/2015-PC-II (FTS: 14428)) supported by the Ministry of Chemical and Fertilizers, Department of Chemicals and Petrochemicals, Government of India and Facility Creation Project (FCP; No. 34/FCP/NCL/2019-RPPBDD) supported by Council of Scientific and Industrial Research, Project Planning & Business Development Directorate, New Delhi. The author also thanks CSIR-National Chemical Laboratory for providing Lab facilities. The authors would like to acknowledge the financial support of Australia India Strategic Research Fund (AISRF) AIRXIV000048 for the research work carried out in Australia. The authors acknowledge the facilities and the scientific and technical assistance of the RMIT Microscopy & Microanalysis Research Facility (MMRF), RMIT Micro Nano Research Facility (MNRF). The synchrotron macro-ATR-FTIR measurement was performed on the IRM Beamline at the Australian Synchrotron, part of ANSTO, through the merit-based access program (Proposal ID. 21906).

References

1. S. C. Ligon, R. Liska, J. Stampfl, M. Gurr and R. Mülhaupt, Chem. Rev., 2017, 117(15), 10212–10290,  DOI:10.1021/acs.chemrev.7b00074.
2. P. Ahangar, M. E. Cooke, M. H. Weber and D. H. Rosenzweig, Appl. Sci., 2019, 9, 1713,  DOI:10.3390/app9081713.
3. Y. Huo, Y. Xu, X. Wu, E. Gao, A. Zhan, Y. Chen, Y. Zhang, Y. Hua, W. Swieszkowski, Y. S. Zhang and G. Zhou, Adv. Sci., 2022, 9(29), e2202181,  DOI:10.1002/advs.202202181.
4. Z. Gu, J. Fu, H. Lin and Y. He, Asian J. Pharm. Sci., 2020, 15, 529–557.
5. W. L. Ng, J. An and C. K. Chua, Engineering, 2024, 36, 146–166,  DOI:10.1016/j.eng.2024.01.028.
6. H. Quan, T. Zhang, H. Xu, S. Luo, J. Nie and X. Zhu, Bioact. Mater., 2020, 6(6), 1789–1790,  DOI:10.1016/j.bioactmat.2019.12.003.
7. N. A. S. Mohd Pu’ad, R. H. Abdul Haq, H. Mohd Noh, H. Z. Abdullah, M. I. Idris and T. C. Lee, Mater. Today: Proc., 2019, 29, 228–232.
8. R. Machino, K. Matsumoto, D. Taniguchi, T. Tsuchiya, Y. Takeoka, Y. Taura, M. Moriyama, T. Tetsuo, S. Oyama, K. Takagi, T. Miyazaki, G. Hatachi, R. Doi, K. Shimoyama, N. Matsuo, N. Yamasaki, K. Nakayama and T. Nagayasu, Adv. Healthcare Mater., 2019, 8(7), e1800983,  DOI:10.1002/adhm.201800983.
9. J. X. Law, L. L. Liau, B. S. Aminuddin and B. H. I. Ruszymah, Int. J. Pediatr. Otorhinolaryngol., 2016, 91, 55–63,  DOI:10.1016/j.ijporl.2016.10.012.
10. M. M. Elbarbary, B. Magdy, M. ElFiky, A. M. K. Wishahy, A. Hussein, M. L. Naguib and M. Elseoudi, J. Pediatr. Surg. Open, 2023, 3, 100048.
11. J. C. Fraga, R. W. Jennings and P. C. W. Kim, Semin. Pediatr. Surg., 2016, 25, 156–164.
12. Y. G. Feng, S. L. Tao, L. Y. Mei, F. Q. Dai, Q. Y. Tan, R. W. Wang, J. H. Zhou and B. Deng, J. Cardiothorac. Surg., 2023, 18(1), 293,  DOI:10.1186/s13019-023-02369-0.
13. A. Kamran and R. W. Jennings, Front. Pediatr., 2019, 7, 512,  DOI:10.3389/fped.2019.00512.
14. L. Freitag, M. Gördes, P. Zarogoulidis, K. Darwiche, D. Franzen, F. Funke, W. Hohenforst-Schmidt and H. Dutau, Respiration, 2017, 94, 442–456.
15. K. A. Brooks, A. Y. Lai, S. J. Tucker, H. Ramaraju, A. Verga, S. Shashidharan, K. O. Maher, D. M. Simon, S. J. Hollister, A. M. Landry and S. L. Goudy, Int. J. Pediatr. Otorhinolaryngol., 2023, 169, 111559,  DOI:10.1016/j.ijporl.2023.111559.
16. R. J. Morrison, S. J. Hollister, M. F. Niedner, M. G. Mahani, A. H. Park, D. K. Mehta, R. G. Ohye and G. E. Green, Sci. Transl. Med., 2015, 7(285), 285ra64,  DOI:10.1126/scitranslmed.
17. H. F. Shieh and R. W. Jennings, J. Thorac. Dis., 2017, 9(3), 414–416,  DOI:10.21037/jtd.2017.02.53.
18. L. S. Nair and C. T. Laurencin, Prog. Polym. Sci., 2007, 32(8–9), 762–798,  DOI:10.1016/j.progpolymsci.2007.05.017.
19. Y. Chen, M. J. C. De Sá, M. Dalton and D. M. Devine, Biodegradable medical implants, in Bioresorbable Polymers: Biomedical Applications, ed. D. Devine, De Gruyter, Berlin, Boston, 2019, pp. 17–46 DOI:10.1515/9783110640571-002.
20. O. A. Sindeeva, E. S. Prikhozhdenko, I. Schurov, N. Sedykh, S. Goriainov, A. Karamyan, E. A. Mordovina, O. A. Inozemtseva, V. Kudryavtseva, L. E. Shchesnyak, R. A. Abramovich, S. Mikhajlov and G. B. Sukhorukov, Pharmaceutics, 2021, 13, 1437,  DOI:10.3390/pharmaceutics13091437.
21. Y. Tatekawa, N. Kawazoe, G. Chen, Y. Shirasaki, H. Komuro and M. Kaneko, Pediatr. Surg. Int., 2010, 26, 575–580.
22. B. Yan, Z. Zhang, X. Wang, Y. Ni, Y. Liu, T. Liu, W. Wang, H. Xing, Y. Sun, J. Wang and X. F. Li, Artif. Organs, 2017, 41, 461–469.
23. A. Kurowska, A. Nikodem, A. Jabłoński, J. Janusz, P. Szczygieł, M. Ziąbka, E. Menaszek, M. Dziadek, B. Zagrajczuk, M. Kobielarz and I. Rajzer, Mater. Des., 2022, 233, 112255,  DOI:10.1016/j.matdes.2023.112255.
24. W. Guo, M. Chen, Z. Wang, Y. Tian, J. Zheng, S. Gao, Y. Li, Y. Zheng, X. Li, J. Huang, W. Niu, S. Jiang, C. Hao, Z. Yuan, Y. Zhang, M. Wang, Z. Wang, J. Peng, A. Wang, Y. Wang, X. Sui, W. Xu, L. Hao, X. Zheng, S. Liu and Q. Guo, Bioact. Mater., 2021, 6, 3620–3633.
25. T. Khalid, L. Soriano, M. Lemoine, S. A. Cryan, F. J. O’Brien and C. O’Leary, Front. Bioeng. Biotechnol., 2023, 11, 1187500,  DOI:10.3389/fbioe.2023.1187500.
26. W. H. Chen, Q. W. Chen, Q. Chen, C. Cui, S. Duan, Y. Kang, Y. Liu, Y. Liu, W. Muhammad, S. Shao, C. Tang, J. Wang, L. Wang, M. H. Xiong, L. Yin, K. Zhang, Z. Zhang, X. Zhen, J. Feng, C. Gao, Z. Gu, C. He, J. Ji, X. Jiang, W. Liu, Z. Liu, H. Peng, Y. Shen, L. Shi, X. Sun, H. Wang, J. Wang, H. Xiao, F. J. Xu, Z. Zhong, X. Z. Zhang and X. Chen, Sci. China: Chem., 2022, 65, 1010–1075.
27. M. Gomez-Florit, A. Pardo, R. M. A. Domingues, A. L. Graça, P. S. Babo, R. L. Reis and M. E. Gomes, Molecules, 2020, 25(24), 5858,  DOI:10.3390/MOLECULES25245858.
28. Y. Xu, Z. Wang, Y. Hua, X. Zhu, Y. Wang, L. Duan, L. Zhu, G. Jiang, H. Xia, Y. She and G. Zhou, Mater. Sci. Eng., C, 2021, 120, 111628,  DOI:10.1016/j.msec.2020.111628.
29. O. A. Romanova, T. H. Tenchurin, T. S. Demina, E. V. Sytina, A. D. Shepelev, S. G. Rudyak, O. I. Klein, S. V. Krasheninnikov, E. I. Safronova, R. A. Kamyshinsky, V. G. Mamagulashvili, T. A. Akopova, S. N. Chvalun and A. A. Panteleyev, Cell Proliferation, 2019, 52(3), e12598,  DOI:10.1111/cpr.12598.
30. Z. Nematollahi, M. Tafazzoli-Shadpour, A. Zamanian, A. Seyedsalehi, S. Mohammad-Behgam, F. Ghorbani and F. Mirahmadi, Iran. Biomed. J., 2017, 21, 228–239.
31. H. Kim, J. Y. Lee, H. Han, W. W. Cho, H. Han, A. Choi, H. Hong, J. Y. Kim, J. H. Park, S. H. Park, S. W. Kim, D. S. Kim and D. W. Cho, Sci. Rep., 2021, 11, 9258,  DOI:10.1038/s41598-021-88830-3.
32. Y. Dai, G. Liu, L. Ma, D. Wang and C. Gao, J. Mater. Chem. B, 2016, 4, 4410–4419.
33. R. Satake, M. Komura, H. Komura, T. Kodaka, K. Terawaki, K. Ikebukuro, H. Komuro, H. Yonekawa, K. Hoshi, T. Takato and Y. Nakayama, J. Pediatr. Surg., 2016, 51, 244–248.
34. G. Chen, T. Ushida and T. Tateishi, J. Biomed. Mater. Res., 2000, 51, 273–279,  DOI:10.1002/(SICI)1097-4636(200008)51:2<273::AID-JBM16>3.0.CO;2-O.
35. W. Gao, K. R. Kanagarajah, E. Graham, K. Soon, T. Veres, T. J. Moraes, C. E. Bear, R. A. Veldhuizen, A. P. Wong and A. Günther, Small, 2024, 20, 27,  DOI:10.1002/smll.202309270.
36. L. Liu, S. Dharmadhikari, B. M. Spector, Z. H. Tan, C. E. Van Curen, R. Agarwal, S. Nyirjesy, K. Shontz, S. A. Sperber, C. K. Breuer, K. Zhao, S. D. Reynolds, A. Manning, K. K. VanKoevering and T. Chiang, J. Tissue Eng., 2022, 13 DOI:10.1177/20417314221108791.
37. B. Pandey, S. Chatterjee, N. Parekh, P. Yadav, A. Nisal and S. Sen Gupta, ACS Appl. Bio Mater., 2018, 1, 2082–2093.
38. M. M. Fares, E. Shirzaei Sani, R. Portillo Lara, R. B. Oliveira, A. Khademhosseini and N. Annabi, Biomater. Sci., 2018, 6, 2938–2950.
39. N. Sultana, Molecules, 2023, 28(24), 7974,  DOI:10.3390/molecules28247974.
40. L. Wang, M. Abedalwafa, F. Wang and C. Li, Rev. Adv. Mater. Sci., 2013, 34, 123–140.
41. E. Malikmammadov, T. E. Tanir, A. Kiziltay, V. Hasirci and N. Hasirci, J. Biomater. Sci., Polym. Ed., 2018, 29, 863–893.
42. G. Kozehkonan Savari, M. Salehi, S. Farzamfar, H. Ghanbari, M. Adabi and A. Amani, Nanomed. J., 2019, 6(4), 311–320,  DOI:10.22038/nmj.2019.06.000009.
43. P. Dorishetty, R. Balu, A. Gelmi, J. P. Mata, A. Quigley, N. K. Dutta and N. R. Choudhury, Mater. Today Adv., 2022, 14, 100233,  DOI:10.1016/j.mtadv.2022.100233.
44. J. Zhang, B. J. Allardyce, R. Rajkhowa, Y. Zhao, R. J. Dilley, S. L. Redmond, X. Wang and X. Liu, ACS Biomater. Sci. Eng., 2018, 4, 3036–3046.
45. J. Vongsvivut, D. Pérez-Guaita, B. R. Wood, P. Heraud, K. Khambatta, D. Hartnell, M. J. Hackett and M. J. Tobin, Analyst, 2019, 144, 3226–3238.
46. R. Balu, S. Sharma, R. Roberts, J. Vongsvivut and N. R. Choudhury, Polymers, 2024, 16(18), 2631,  DOI:10.3390/polym16182631.
47. S. O. Alaswad, A. S. Mahmoud and P. Arunachalam, Polymers, 2022, 14(22), 4924,  DOI:10.3390/polym14224924.
48. D. S. Kohane and R. Langer, Pediatr. Res., 2008, 63, 487–491.
49. M. Bernard, E. Jubeli, M. D. Pungente and N. Yagoubi, Biomater. Sci., 2018, 6, 2025–2053,  10.1039/c8bm00518d.
50. L. Liu, Y. Li, H. Liu and Y. Fang, Eur. Polym. J., 2004, 40, 2739–2744.
51. J. Tian, Y. Yang and J. Song, Int. J. Biol. Macromol., 2019, 141, 919–926.
52. G. Gorrasi, V. Bugatti, G. Viscusi and V. Vittoria, Colloid Polym. Sci., 2021, 299, 429–437,  DOI:10.1007/s00396-020-04699-0.
53. R. A. Pérez-Camargo, G. Saenz, S. Laurichesse, M. T. Casas, J. Puiggalí, L. Avérous and A. J. Müller, J. Polym. Sci., Part B: Polym. Phys., 2015, 53, 1736–1750.
54. M. Li, Y. Pu, F. Chen and A. J. Ragauskas, J. Biotechnol., 2021, 60, 189–199.
55. W. Wei, L. Xing and J. Feng, Mater. Express, 2022, 11, 1966–1974.
56. A. Benahmed, K. Azzaoui, A. El Idrissi, H. Belkheir, S. O. S. Hassane, R. Touzani and L. Rhazi, Molecules, 2022, 27, 1408,  DOI:10.3390/molecules27041408.
57. P. Wang, P. Fei, C. Zhou and P. Hong, Int. J. Biol. Macromol., 2021, 186, 528–534.
58. T. Jain, Y. M. Tseng, C. Tantisuwanno, J. Menefee, A. Shahrokhian, I. Isayeva and A. Joy, ACS Appl. Polym. Mater., 2021, 3, 6618–6631.
59. S. Laurichesse and L. Avérous, Polymer, 2013, 54, 3882–3890.
60. D. Gawkowska, J. Cybulska and A. Zdunek, Polymers, 2018, 10, 762,  DOI:10.3390/polym10070762.
61. S. Tajik, C. N. Garcia, S. Gillooley and L. Tayebi, Regener. Eng. Transl. Med., 2023, 9(1), 29–41,  DOI:10.1007/s40883-022-00267-w.
62. C. A. Kelly, S. H. Murphy, G. A. Leeke, S. M. Howdle, K. M. Shakesheff and M. J. Jenkins, Eur. Polym. J., 2013, 49, 464–470.
63. Z. Wang, J. Li, Z. Qu, B. Ayurzana, G. Zhao and W. Li, Environ. Technol. Innovation, 2024, 36, 103791,  DOI:10.1016/j.eti.2024.103791.
64. L. Zagorchev, P. Oses, Z. W. Zhuang, K. Moodie, M. J. Mulligan-Kehoe, M. Simons and T. Couffinhal, Vasc. Cell, 2010, 2, 7,  DOI:10.1186/2040-2384-2-7.
65. Y. Liu, K. Zheng, Z. Meng, L. Wang, X. Liu, B. Guo, J. He, X. Tang, M. Liu, N. Ma, X. Li and J. Zhao, Biomaterials, 2023, 300, 122208,  DOI:10.1016/j.biomaterials.2023.122208.
66. W. Liu, Z. Meng, K. Zheng, L. Wang, C. Zhang, J. Ji, X. Li, J. He and J. Zhao, Mater. Des., 2021, 210, 110105,  DOI:10.1016/j.matdes.2021.110105.
67. H. Ramaraju, A. M. Landry, S. Sashidharan, A. Shetty, S. J. Crotts, K. O. Maher, S. L. Goudy and S. J. Hollister, Biomaterials, 2022, 289, 121702,  DOI:10.1016/j.biomaterials.2022.121702.
68. S. A. Hosseini, S. Javad Hoseini, V. R. Askari, R. Salarinia, A. Ebrahimzadeh-Bideskan, F. Tara, F. Kermani, S. Nazarnezhad and S. Kargozar, J. Drug Delivery Sci. Technol., 2022, 77, 103916,  DOI:10.1016/j.jddst.2022.103916.
69. D. Xia, D. Jin, Q. Wang, M. Gao, J. Zhang, H. Zhang, J. Bai, B. Feng, M. Chen, Y. Huang, Y. Zhong, N. Witman, W. Wang, Z. Xu, H. Zhang, M. Yin and W. Fu, J. Tissue Eng. Regener. Med., 2019, 13, 694–703.
70. A. Honarvar, S. Karbasi, B. Hashemibeni, M. Setayeshmehr, M. Kazemi and A. Valiani, Mater. Technol., 2020, 00, 1–9.
71. M. Shaikh, G. Kichenadasse, N. R. Choudhury, R. Butler and S. Garg, J. Controlled Release, 2013, 172(1), 105–117,  DOI:10.1016/j.jconrel.2013.08.010.
72. M. M. Nava, L. Draghi, C. Giordano and R. Pietrabissa, J. Appl. Biomater. Funct. Mater., 2016, 14, e223–e229.
73. Q. L. Loh and C. Choong, Tissue Eng., Part B, 2013, 19, 485–502.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tb00891c

---