Effect of Pore Characteristics and Alkali Treatment on the Physicochemical and Biological Properties of a 3D-Printed Polycaprolactone Bone Scaffold

Abstract

Polycaprolactone scaffolds were designed and 3D-printed with different pore shapes (cube and triangle) and sizes (500 and 700 μm) and modified with alkaline hydrolysis of different ratios (1, 3, and 5 M). In total, 16 designs were evaluated for their physical, mechanical, and biological properties. The present study mainly focused on the pore size, porosity, pore shapes, surface modification, biomineralization, mechanical properties, and biological characteristics that might influence bone ingrowth in 3D-printed biodegradable scaffolds. The results showed that the surface roughness in treated scaffolds increased compared to untreated polycaprolactone scaffolds (R_a = 2.3–10.5 nm and R_q = 17– 76 nm), but the structural integrity declined with an increase in the NaOH concentration especially in the scaffolds with small pores and a triangle shape. Overall, the treated polycaprolactone scaffolds particularly with the triangle shape and smaller pore size provided superior performance in mechanical strength similar to that of cancellous bone. Additionally, the in vitro study showed that cell viability increased in the polycaprolactone scaffolds with cubic pore shapes and small pore sizes, whereas mineralization was enhanced in the designs with larger pore sizes. Based on the results obtained, this study demonstrated that the 3D-printed modified polycaprolactone scaffolds exhibit a favorable mechanical property, biomineralization, and better biological properties; therefore, they can be applied in bone tissue engineering.

1. Introduction

In recent years, the increasing prevalence of bone fractures and segmental bone defects as a result of trauma, congenital disorders, or cancer-related bone resections has become a significant concern in healthcare systems.¹ To address this problem, bone tissue engineering (BTE) applies biomaterial-based substitutes and scaffolds, which is a promising approach.² Bone scaffolds as temporary matrices should be biocompatible, biodegradable, osteoinductive, and osteoconductive as well as porous with interconnected pores and with sufficient mechanical properties for improved bone remodeling.³ In this context, both materials and fabrication techniques can be adjusted to provide desired properties. Substantial research is focused on the potential utility of electrospun and freeze-dried scaffolds for BTE. Electrospun scaffolds provide a three-dimensional fibrous matrix that mimics the native extracellular matrix (ECM). Nevertheless, electrospun scaffolds are relatively poor mechanically compared to bone tissue, which is one major disadvantage.^4,5 On the other side, freeze-dried scaffolds have porous structures, but their pore sizes are not controllable.^6,7 Furthermore, 3D printing has been explored for scaffold fabrication, which allows for the adjustment of the pore size, pore shapes, pore interconnectivity, and mechanical properties.^8,9 It has been shown that the pore size, porosity, and pore shapes not only influence bone regeneration by providing sufficient space for cell and tissue invasions and nutrient transport but also affect mechanical strength.^10,11 The SLM-produced Ti6Al4V scaffolds with six distinct designs such as pore shapes and pore sizes were evaluated. The results demonstrated that the mechanical and biological properties depended on the pore shape, pore size, and permeability.¹² In BTE, these characteristics of scaffolds have been shown to significantly influence the physicomechanical properties and the rate of bone tissue regeneration. Additionally, these parameters lead to a more altered mechanical strength of non-metallic scaffolds; therefore, the mechanical strength and biological properties are further dependent on other properties such as geometrical features and materials.¹³ Apart from the scaffold characterization, material selection is also essential. Three-dimensionally printed synthetic polymer scaffolds are most widely employed due to their controllable physical, chemical, and mechanical properties. Among the synthetic polymers, FDA-approved polycaprolactone (PCL) has potential due to its low melting point, processability, biocompatibility, and bioresorbability.¹⁴ The 3D-printed PCL scaffolds have demonstrated customized anatomical shapes, controllable porous architectures, processability, and desirable mechanical properties used for bone tissue engineering grafts in preclinical and clinical investigations. Recently, the commercialized products of 3D-printed PCL including Osteoplug and Osteomesh (Osteopore) implants have been employed as bone void fillers.^15,16 However, the PCL application was limited by hydrophobicity, poor cell affinity, and cell surface recognition sites. Therefore, the surface treatment by changing the surface topography and surface chemistry should be applied to circumvent these limitations. Various methods have been used for physical or chemical modification of PCL scaffolds such as those of grafting polymerization, surface coatings, laser treatment, plasma treatment, aminolysis, and hydrolysis.^17,18 Among these techniques, alkali treatment by NaOH is a simple and effective method for improving the wettability, surface roughness, cell interactions, and coating with other materials. The NaOH-treated PCL scaffold surfaces can be immobilized with bioactive molecules, such as proteins, peptides, polysaccharides, and other materials to create bioactive scaffolds or an environment for cell proliferation.¹⁸⁻²⁰

Various studies evaluated the effects of hydrolysis of scaffolds on the surface roughness, surface hydrophilicity, material immobilization, and in vitro performance. A study proposed the NaOH etching to modify the PCL surface for an apatite layer deposit. The reported apatite formation indicated that NaOH hydrolysis induced a dense and uniform apatite layer formation due to the introduced carboxylate groups on the PCL surfaces.²¹ Another report has shown that the concentration of NaOH, the reaction temperature, and the treatment time influence the porosity and the roughness of 3D-printed PCL surfaces. Hence, the proliferation of cells increased with increasing the surface roughness and porosity, whereas the scaffold’s mechanical strength decreased.²² Surface modifications of PCL composite scaffolds can also improve the contact between cells and other blended materials. In another study, the O₂ plasma and NaOH hydrolysis were used on HA/PCL scaffolds to etch the PCL surfaces. This demonstrated that the NaOH treatment with modulating the surface characteristics more effectively exposed hydroxyapatite particles and promoted hydrophilicity, cell proliferation, and differentiation of human dental pulp stem cells (hDPSCs) than the O₂ plasma treatment.²³ However, another study showed that collagen and apatite can be located on hydrolyzed 3D-printed PCL scaffolds owing to the wettability and rough surface leading to better interactions with other materials, which is suitable for BTE.²⁴

As a result, PCL as a common thermoplastic polymer for 3D printing has drawn considerable attention for tissue engineering scaffolds. Therefore, it is motivating to evaluate the PCL scaffold’s outcomes based on the roles of the pore size, porosity, pore shapes, and surface modification on mechanical and biological properties. This study aims to thoroughly understand the role of scaffold architecture and surface modification to explain the mechanical strength, biomineralization, and cell functions. For this purpose, the PCL scaffolds were developed with four different designed geometries by a 3D-printing technique and modified by NaOH solution in different concentrations. Then, the morphological and physicochemical properties of scaffolds were evaluated and the in vitro degradation and bioactivity of the scaffolds were characterized. The biocompatibility and mineralization were performed to validate potential applications for bone tissue engineering. More importantly, our objective was to determine which 3D-printed structure is more effective in promoting bone tissue regeneration as a framework (i.e., roughness, bioactivity, proliferation, spreading, and mineralization). It was hypothesized that the 3D porous modified PCL scaffolds could serve as a matrix for cell seeding, osteoconductivity, and mechanical strength.

2. Materials and Methods

2.1. Fabrication of PCL Scaffolds

The PCL scaffolds were fabricated by Abtin Teb bioprinting (Abtin Teb company, Iran). In brief, 3D cube and triangle pore geometries were designed using Abaqus 2018 (Dassault Systèmes, USA) and imported in a stereolithography (STL) file format into 3D printing software (Repetier-Host) for G-code generation. The PCL pellets (Sigma-Aldrich, M_n: 80,000) were inserted into the stainless-steel nozzle, melted in the oven, and extruded through the bioprinting nozzle. The melting temperature of the PCL pellets was 80 °C, and the printing temperature was applied at 90 °C. The PCL scaffolds with 20 mm diameters and 2 mm heights were printed with a nozzle diameter of 400 μm, pore sizes of 500 and 700 μm, and orientations of 0/90° and 0/60/120° (Figure 1A). Afterward, the scaffolds were punched into discs with 8 mm diameters.

Figure 1.

Designed and 3D-printed PCL scaffolds. (A) Representative images of the 3D models with orientations of 0/90° and 0/60/120°. (B) Representative images of the printed scaffolds.

2.2. NaOH Treatment of Scaffolds

The 3D-printed scaffolds were modified with NaOH (Sigma-Aldrich) solutions in different concentrations. The scaffolds were cleaned using 70% ethanol (Sigma-Aldrich) for 15 min followed by immersion in 1, 3, and 5 M NaOH solutions for 6 h at a shaking speed of 80 rpm. Following NaOH treatment, the scaffolds were rinsed with deionized water and dried in an oven at 37 °C. Table 1 shows the whole design and printing scenarios.

Table 1. Scaffold Pore Shape and Size and NaOH Concentration.

sample name	pore shape	pore size (μm)	NaOH concentration (M)
C700	cube	700
C500	cube	500
T700	triangle	700
T500	triangle	500
C7001	cube	700	1
C5001	cube	500	1
C7003	cube	700	3
C5003	cube	500	3
C7005	cube	700	5
C5005	cube	500	5
T7001	triangle	700	1
T5001	triangle	500	1
T7003	triangle	700	3
T5003	triangle	500	3
T7005	triangle	700	5
T5005	triangle	500	5

2.3. Morphological Characterization

The surface morphology of the PCL and NaOH-treated PCL scaffolds was evaluated using a scanning electron microscope (SEM). Before and after treatment, the PCL scaffold surface morphology, pore size, and design were evaluated using SEM (Philips XL30 SEM, Philips, Netherlands). First, the scaffolds were sputter-coated with gold to make the surfaces conductive for SEM observation.

2.4. Surface Pore Size and Porosity

The pore size distribution and the thicknesses of scaffold filaments were determined from SEM images of each scaffold using ImageJ software (Wayne Rasband, National Institute of Health, USA). The scaffold porosity (%) was calculated by analyzing the densities of printed scaffolds by measuring the weights and volumes of the constructs using eq 1(24)

1

where ρ_e is the experimental density of the scaffolds and ρ_t (1.145 g/cm³) is the theoretical density of PCL. The experimental density is evaluated by measuring the weight, height, diameter, and volume of scaffolds based on the following relations. Equation 2:

2

2.5. Surface Roughness Evaluation

The surface roughness of PCL and NaOH-treated PCL was evaluated by the atomic force microscopy (AFM) technique (Bio-AFM, Ara-Research Company, Iran). The scaffolds were located on the sample holder, and three distinct areas (10 μm × 10 μm) on each sample surface were scanned in the non-contact mode at the ambient conditions. The surface roughness was determined by Imager version 1.00 software supplied by the Ara-Research Company. The arithmetical mean deviation of the assessed profile (R_a) and root mean square roughness (R_q) values were obtained.

2.6. Surface Chemistry Analysis

The PCL and NaOH-treated PCL scaffolds were analyzed using ATR-FTIR (Specac Golden Gate ATR) spectroscopy to characterize the functional groups and interactions. The scanning wavenumber range measured was 650–4000 cm^–1.

2.7. In Vitro Biodegradation

For evaluating the degradation properties of the scaffolds, the samples were immersed and incubated in simulated body fluid (SBF) for up to 60 days at 37 °C. First, the scaffolds were weighed. After soaking in SBF for 1 and 4 days as well as 1, 2, 3, 4, 5, 6, and 8 weeks, all scaffolds were taken out, rinsed with distilled water, and then dried completely, and the weight loss was calculated. The media was refreshed every week. At various intervals, the weights of the scaffolds were recorded (w_t). The degradation ratio of scaffolds was calculated as eq 3:

3

In the above equation, w₀ represents the weight of the sample before soaking.

2.8. In Vitro Bioactivity

For bioactivity evaluation assay, the scaffolds were soaked in SBF at 37 °C for up to 60 days. To calculate the volume of SBF required for the samples, the apparent area of the specimen was divided by V_SBF = . The SBF solution was exchanged each week. Afterward, the specimens were washed gently with distilled water and then thoroughly dried in a vacuum oven at 50 °C. The apatite formation on the scaffold surface was examined by energy-dispersive X-ray spectroscopy (EDX) and FE-SEM (ZEISS, Germany).

2.9. Mechanical Testing and Analysis

The mechanical properties of non-treated and treated scaffolds were measured using a Universal Testing Machine (Instron Ltd., High Wycombe, UK) with a 5 KN load cell. Three specimens were tested from each group having a height-to-diameter ratio of 1 (diameter: 8 mm and height: 8 mm). The scaffolds were compressed at a loading rate of 0.03 mm/min. The compressive strain was recorded up to 70% of the original height. The stress–strain (σ–ε) curves were recorded, and the compressive moduli were determined from the initial linear region slope of the stress–strain curves.

2.10. Biological Assessments

2.10.1. Scaffold Sterilization

For sterilization, the scaffolds (three scaffolds from each group) were transferred to 48-well plates and sterilized in 70% ethanol for 15 min followed by PBS washing and drying. This procedure was conducted under a laminar flow bench. Then, the scaffolds were sterilized using a UV light for 20 min on each side.

2.10.2. Cell Culture

Pre-osteoblast MC3T3-E1 cell lines (mouse C57BL/6 calvaria, ECACC, Sigma-Aldrich, Sweden) were cultured in the complete medium of Dulbecco’s Modified Eagle medium or (DMEM; Gibco Life Technologies, USA) supplemented with 10% fetal bovine serum (FBS; Gibco Life Technologies, USA), penicillin (100 IU), and streptomycin (100 μg/mL), and 2 mM l-glutamine (Gibco Life Technologies; USA). In an incubator, the cells were maintained at 37 °C in a humidified atmosphere of 5% CO₂ and 95% humidity.

2.10.3. Cell Viability Assay

Cell viability of the 3D-printed scaffolds was assessed by an MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] cell proliferation kit (Cell Growth Determination Kit, Sigma Life Science) after three and seven days. First, the scaffolds (three scaffolds from each group) were placed in a complete growth medium (DMEM + 10% FBS) for three days by incubation at 37 °C in a humidified atmosphere containing 5% CO₂ and 95% humidity to achieve medium extracts.

For cell viability, 5 × 10³ MC3T3-E1 cells/well were seeded in 96-well plates and allowed to make a monolayer within 24 h. Afterward, the complete medium was replaced with the medium extracts. After the incubation (for three and seven days), 100 μL of MTT solution was added to each well and incubated at 37 °C for 4 h to form the formazan. Finally, 100 μL of isopropanol/hydrochloric acid (0.04 N) was added to dissolve the formazan, and the absorbance was quantified at 570 nm using a microplate ELISA reader (Synergy H1 Hybrid Multi-Mode Microplate Reader, BioTek, USA).

2.10.4. Cell Adhesion Study

Cell adhesion was assessed after three days of incubation. For cell adhesion, MC3T3-E1 (2 × 10³ cells/well) were seeded directly onto the scaffolds and kept in an incubator at 37 °C in a humidified atmosphere containing 5% CO₂ and 95% humidity. After incubation, the scaffolds containing cells were fixed with 4% paraformaldehyde for 20 min and then rinsed with PBS three times. Afterward, the samples were dried for 2 h, and the gold-coated samples were observed under SEM.

2.10.5. Alizarin Red Assay

Alizarin red staining (indicative of calcified matrix deposition) was assessed after incubation for seven days. MC3T3-E1 cells (5 × 10⁵ cells/well) were cultured in 48-well tissue culture plates and kept overnight at 37 °C in CO₂ and 95% humidity to form a cell monolayer. Afterward, the medium was exchanged with medium extracts containing an osteogenic cell culture medium (ascorbic acid, 50 μg/mL, and β-glycerophosphate, 10 mM (Sigma-Aldrich, USA)). After seven days, the cells were fixed in paraformaldehyde for 10 min in dark conditions, and the fixed samples were then washed with PBS and stained with 1% (w/v) alizarin red for 30 min at room temperature. Finally, the samples were washed with PBS to remove unreacted alizarin red and observed under a loop microscope.

2.11. Design of Experiment and Statistical Analysis

A customized design of experiment (DOE) was conducted to evaluate the effect of factors (pore size, pore shape, and NaOH concentration) on different responses (R_a, R_q, compressive strength, compressive modulus, cell viability, and mineralization). Two levels were considered for the pore size (500 and 700 μm) and pore shape (cube and triangle), and three levels were taken into account for the NaOH concentration (1, 3, and 5 M). In this study, the presented data was expressed in triplicate. The mean values and standard deviations were considered for the pore size, porosity, R_a, R_q, biodegradation, and mechanical strength. The statistical analyses were obtained through a two-way analysis of variance (ANOVA) followed by a Tukey pairwise comparison test using Minitab V17 software. GraphPad Prism version 9.00 was used to draw the graphs. P values of <0.05 were considered to be statistically significant differences.

3. Results

3.1. Scaffold Characterization

In this study, two designed pore shapes and different pore sizes were produced by 3D printing to investigate the physical, mechanical, and biological properties of the scaffolds. The images revealed that the 3D-printed PCL scaffolds had a filament orientation of 0/90° and 0/60/120° with dimensions of 20 mm × 2 mm (Figure 1A). Scaffolds with two types of pore shapes were obtained by changing the filament offset. The filament orientation affected the printing configuration, and during extrusion, the printed layers formed cube and triangle pores in accordance with the software design (Figure 1B). This indicated that the geometrical features could provide adequate space for cell function properly. In this context, pore shapes and the surface curvature could influence cell spreading and bone regeneration processes. Moreover, the 3D-printed scaffold pore size strongly depended on the design. By decreasing the scaffold pore size from 700 to 500 μm, the distance between filaments decreased and the number of filaments increased. The pore size and interconnectivity also play an essential role in bone regeneration, which can provide enough space for bone ingrowth and mass transport.

The SEM images of PCL scaffolds showed that the cube and triangle pore shapes with pore sizes of 700 and 500 μm were in good accordance with their software design in Figure 2. These 3D-printed scaffolds displayed well-interconnected architectures, and the macropore wall surface was approximately rough. Additionally, the cross-sectional area of the PCL scaffolds confirmed uniform pores and interconnectivity.

Figure 2.

SEM images of the PCL scaffolds before surface modification (top view) and a cross-section of scaffolds (side view, bottom).

The hydrophobicity and surface smoothness of 3D-printed PCL scaffolds restrict cell spreading. Surface treatments of hydrophobic PCL were employed to improve the wettability and cell interaction. PCL scaffolds were treated with NaOH solution to create hydrophilicity and COOH/OH groups on the surface. For hydrolysis, the surface topography changes when hydroxide anions from NaOH solution hydrolyze the ester bonds of PCL, exposing the carboxylic acid (COOH) and hydroxyl (OH) groups associated with the polymer chains. NaOH treatment of the cube geometry is shown in Figure 3. In the cube geometry, smooth surfaces with few holes were observed in different NaOH concentrations in comparison with the triangle geometry. Smooth structures and lesser wall roughness are created due to rapid flow rates because the higher flow velocity will not affect the wall roughness as much. Furthermore, a large pore size is expected to influence the diffusion process and flow rate; therefore, the scaffolds with a pore size of 700 μm were smoother than that of 500 μm. Under higher magnification, the treated PCL scaffolds in a cube pore geometry showed increased surface roughness compared to the untreated PCL.

Figure 3.

SEM image of 3D-printed scaffolds with the cube pore geometry after surface modification (top view).

In the NaOH-treated PCL with a triangle pore geometry, rough surfaces with holes were observed on the scaffold walls (Figure 4). When the pore size decreased from 700 to 500 μm, the surface roughness increased and quite rough surfaces with deep holes were seen on the PCL scaffolds. This was generated during penetration from the scaffold surface to the inside. As the triangle geometry showed higher surface roughness, this confirmed the intensity of the velocity flow rate and exposure of the walls. In addition, compared with the PCL scaffolds without NaOH treatment, the surfaces of macroporous walls of the treated scaffolds with a triangle pore geometry became completely rough. In conclusion, it is worth noting that the surface alteration intensity was dependent on the pore shape and pore size of the PCL scaffolds.

Figure 4.

SEM image of the 3D-printed scaffolds with the triangle pore geometry after surface modification (top view).

Following the observation of treated samples with different concentrations of NaOH, the PCL scaffolds treated with 1 M NaOH had the desirable surface roughness and holes with structural integrity (no change in the filament shape). However, hydrolysis in a 3 M concentration did not obviously change the wall roughness, whereas treating in a 5 M concentration created a significant change in the structure of scaffolds due to a higher diffusion and interaction of NaOH into the walls.

The pore size and filament thickness distribution of all fabricated scaffolds are shown in Table 2. It was found that there was no noticeable difference in the pore size and filament thickness for untreated scaffolds compared to the software design. Moreover, the NaOH concentration did not significantly influence the pore size or filament thickness in the cube geometry. By contrast, NaOH treatment had an obvious effect on the pore size and filament thickness of the scaffolds with the triangle shape and 500 μm pore size. Hence, increased penetration of NaOH into the wall surfaces of scaffolds with the triangle geometry resulted in an increased deformation and altered filament thickness and pore size compared to the untreated PCL and cube geometry. Consequently, the decrease in the filament distance and slow flow velocity together with a higher exposure of NaOH had an increasing variation in the smallest pore sizes.

Table 2. Physical Properties of the 3D-Printed PCL Scaffolds.

sample no.	pore size (μm)	filament thickness (μm)	measurement density (g/cm3)	porosity (%)	theoretical porosity (%)	Ra (nm)	Rq (nm)
C700	688 ± 6	383 ± 3	0.57 ± 0.03	49.7 ± 3	64.53	2.3 ± 1	17 ± 3
C500	504 ± 5	397 ± 4	0.52 ± 0.03	54.1 ± 2	64.05	3.2 ± 1	23 ± 2
T700	693 ± 6	407 ± 2	0.59 ± 0.09	47.9 ± 4	64.54	3.1 ± 2	22 ± 3
T500	514 ± 5	397 ± 2	0.57 ± 0.02	50.1 ± 2	56.86	3.8 ± 6	27 ± 4
C7001	689 ± 3	398 ± 3	0.46 ± 0.04	59.2 ± 4	64.53	3 ± 1	21 ± 6
C5001	479 ± 2	421 ± 2	0.47 ± 0.06	58.3 ± 5	64.05	4.4 ± 1	31 ± 5
C7003	681 ± 3	404 ± 3	0.48 ± 0.1	58.1 ± 2	64.53	2.5 ± 1	18 ± 2
C5003	487 ± 6	396 ± 6	0.54 ± 0.01	52.6 ± 1	64.05	3.5 ± 3	25 ± 3
C7005	692 ± 4	390 ± 3	0.51 ± 0.1	55.3 ± 4	64.53	2.7 ± 0.1	20 ± 1
C5005	466 ± 2	414 ± 2	0.52 ± 0.05	54.2 ± 4	64.05	4.1 ± 1	29 ± 3
T7001	661 ± 5	400 ± 4	0.52 ± 0.05	54.3 ± 4	64.54	8.8 ± 6	64 ± 4
T5001	407 ± 1	425 ± 3	0.52 ± 0.06	54.1 ± 5	56.86	10.5 ± 4	76 ± 3
T7003	666 ± 4	395 ± 4	0.41 ± 0.04	63.6 ± 3	64.54	3.2 ± 0.05	23 ± 0.3
T5003	308 ± 3	477 ± 7	0.47 ± 0.04	58.7 ± 4	56.86	6.8 ± 1	49 ± 9
T7005	665 ± 4	389 ± 2	0.4 ± 0.09	61.5 ± 7	64.54	6.1 ± 0.3	44 ± 2
T5005	360 ± 3	477 ± 2	0.52 ± 0.02	54.2 ± 1	56.86	8.1 ± 2	58 ± 8

In Table 2, the scaffold porosity is listed as well. All the untreated PCL scaffolds had similar porosities (∼50%). The porosity analysis demonstrated that the porosity of scaffolds after treatment with NaOH increased compared to the untreated PCL, whereas no apparent, significant difference in porosity for scaffolds with different treatments was found.

3.2. Chemical Analysis

The ATR-FTIR spectra in Figure 5 show the untreated and treated PCL characteristic peaks at different NaOH concentrations. The characteristic peaks of the untreated PCL were observed at 2944 and 2865 cm^–1 due to the CH₂ stretching, while ester bonds (C=O) were confirmed at 1720 cm^–1. During NaOH treatment, the characteristic peaks of the PCL scaffolds were preserved.^25,26 After NaOH treatment, the characteristic peaks of PCL scaffolds were well-retained; however, no broadband is clearly observable in the 3200–3600 cm^–1 range, suggesting a lower content of OH groups.²⁷

Figure 5.

ATR FTIR spectra of PCL scaffolds before and after surface modification.

3.3. Surface Roughness

Figure 6 shows the roughness and changes in the scaffold surface topography before and after NaOH treatment. It shows that the scaffold surfaces became rougher after immersion. The R_a and R_q of the untreated and treated PCL scaffolds are presented in Table 2. The R_a and R_q ranges were found to be from 2.3 to 10.5 nm and from 17 to 76 nm by varying the NaOH concentration and pore shape and size, respectively. AFM measurements showed that the R_a range of the 3D-printed PCL scaffolds was from 2.3 to 3.8 nm, which was generated during the printing process. The AFM profile indicated that the PCL scaffolds have relatively smooth surfaces. Moreover, the AFM profile of the cube geometry has a roughness of 2.5–4.4 nm, while the triangle geometry has a highly roughened surface of 3.2–10.5 nm. The surface roughness was increased at small pore sizes, but large pore sizes presented relatively smooth surfaces. The untreated PCL scaffolds had a smooth surface morphology, whereas the treated scaffolds were rough. According to the SEM images, the roughness of scaffolds with the triangle pore geometry increased compared to the cubic pore geometry. The results showed that decreasing the scaffolds’ pore size increased the surface roughness. The AFM results also revealed that the pore shape and pore size and the concentration of NaOH solution had a significant effect on the surface roughness. As a result of incorporating functional groups and etching of the surface, a more nanoscale surface roughness was generated. The PCL scaffold surface became rougher because the NaOH solution could penetrate inside the PCL filament.

Figure 6.

AFM images of the surface roughness of PCL scaffolds before and after surface modification.

3.4. In Vitro Biodegradation

After 60 days of degradation in SBF at 37 °C, the untreated PCL scaffolds showed no visual mass loss, so the lack of biodegradability is a challenge to re-create new bone. On the contrary, all treated PCL scaffolds showed gradual biodegradability from 1 to 3% after 60 days of immersion in SBF. Considering the slow biodegradability of the scaffolds in the current study, the fabricated scaffolds can act as frameworks for the new tissues and provide enduring support. At the same time, as they degrade slowly over time, the new bone can achieve sufficient mechanical strength.²⁸

3.5. In Vitro Bioactivity

The FE-SEM images show the surfaces of untreated PCL, cube, and triangle scaffolds after immersion in SBF for 60 days in Figures 7, 8, and 9, respectively. Due to the lack of bioactivity of PCL scaffolds, a prolonged time (60 days) was selected to increase the quantity of apatite deposition. The amorphous apatite particles were observed on both the untreated and treated PCL scaffold surfaces, and the precipitation on the scaffolds was flower-like apatite. After 60 days of immersion, bone-like apatite precipitates were observed, which covered the entire filament surface evenly. Generally, the scaffold surfaces were covered by thick flower-like apatite layers, whereas no precipitates or compositional changes were found on C7003 and T5003 scaffolds. It was observed that, at a 1 M concentration, apatite precipitations completely covered the scaffolds. By using EDX, the elemental composition of the biomineralization layer was determined as apatite. According to EDX analysis, the precipitates were primarily composed of Ca and P. A carbon peak could be seen in the spectra of the scaffolds that can be related to either the carbon content of PCL or the formation of hydroxycarbonate apatite. Results of the bioactivity test showed that the Ca/P atomic ratio in mineralized scaffolds was from 1.15 to 1.61, which was close to the theoretical value of hydroxyapatite (1.67) (Figure 11A). In the case of the T5003, the Ca and P deposition was too low, and the Ca/P atomic ratio was not detected in comparison with the other scaffolds. In addition, the PCL scaffolds with large pore sizes showed a higher Ca/P atomic ratio than those with a small pore size. Furthermore, the Ca/P ratio in the cube pore geometry was less than that of the triangle pore geometry. According to these results, the bioactive properties of the scaffolds make them suitable for use as frameworks.

Figure 7.

FE-SEM images (top view) and EDX analysis of the PCL scaffolds before surface modification.

Figure 8.

FE-SEM images (top view) and EDX analysis of the cube pore-geometry scaffolds after surface modification and immersion in SBF for 60 days.

Figure 9.

FE-SEM images (top view) and EDX analysis of the triangle pore-geometry scaffolds after surface modification and immersion in SBF for 60 days.

Figure 11.

(A) Variations of the Ca/P ratio of scaffolds. (B) Cell viability of scaffolds (there was no statistically significant difference between the groups).

3.6. Mechanical Strength

The compressive behavior of the PCL scaffolds before and after treatment by NaOH is presented in Figure 10. The compressive strengths and compressive Young’s moduli of the treated-PCL scaffolds were slightly lower than the untreated-PCL scaffolds. For the former, mechanical strengths ranging from 9.33 to 18.95 MPa were found, while Young’s moduli ranging from 9.1 to 20.22 MPa were obtained. The results showed that small pore size, triangle pore shape, and low hydrolysis concentration improved slightly the compressive properties of the scaffolds. Actually, there was no statistically significant difference between the groups at the significance level of p < 0.05. It should be noted that the higher porosity and larger pore size led to a decrease in the mechanical properties of scaffolds. Therefore, it is important to provide enough mechanical strength with porous scaffolds. The results showed that both cube and triangle scaffolds mimic the stiffness range of native spongy bone (2–12 MPa)²⁹ and can contribute to the mechanical protection of new bone formation. In this way, the treated scaffolds can facilitate bone regeneration at low load-bearing sites.¹⁹

Figure 10.

Mechanical properties of the scaffolds. (A) Stress–strain curves of the cube scaffolds. (B) Stress–strain curves of the triangle scaffolds. (C) Compressive strength of the scaffolds. (D) Compressive modulus of the scaffolds.

3.7. Cell Behavior

3.7.1. Cell Viability

A prerequisite for the use of PCL scaffolds in bone tissue engineering is to increase the scaffold’s cytocompatibility. MC3T3-E1 cell viability on the scaffolds was assessed by MTT assay after three and seven days (Figure 11B). Two-way ANOVA results revealed that the scaffold (P values of 0.000) was a significant factor in cell viability, whereas time (P value = 0.917) was not an important factor in cell viability. Hence, the statistical analysis showed significant differences between scaffolds but not between time intervals. Furthermore, it can be concluded that hydrolysis of the PCL scaffolds had no adverse effect on the cytocompatibility. Considering the pore size results, cell viability increased at 500 μm in comparison to 700 μm to a limited extent. Furthermore, it was demonstrated that cell viability was higher in the cube pore geometry than in the triangle pore geometry. In the triangle pore geometry, cell viability in a higher NaOH concentration and 500 μm pore size decreased due to the loss of structural integrity compared with the cube pore geometry. Furthermore, cell proliferation decreased compared to the cube pore geometry. Moreover, the results showed that the hydrolysis of the PCL scaffolds not only supports the bone cells but also improves cell viability.

3.7.2. Cell Attachment

The cell adhesion on the surfaces of the scaffolds after three days is shown in Figure 12. In the case of the untreated PCL, the cells showed a spheroid morphology due to the flat and relatively smooth surface of the scaffolds. The flat surface of PCL scaffolds led to poor filopodium formation and cell spreading. After scaffold treatment, the cells grew on porous and rough surfaces. It appears that cells attached to the spaces between the pores on the treated scaffolds or on the open pores on their surfaces (3 and 5 M). In some groups, cells spread on rough surfaces and attached to the etched scaffold surface (1 M). Additionally, higher cell numbers were found on the rough surfaces than on the untreated PCL scaffolds, confirming that the surface topography of the treated scaffolds favor cell adhesion. Moreover, the scaffold characteristics and rough surfaces of scaffolds affect the cell anchoring and improve cell adhesion and response. Large pore sizes resulted in higher flow velocity, allowing the cells less time to attach to the surface. A larger pore size results in higher medium diffusivity in the center of a scaffold, resulting in less pore occlusion over time. However, cell bridging increased in the cube pore geometry due to the more concave surfaces. It has been shown in vitro that cells can be anchored and can grow better in smaller porosity owing to cell bridging over the pores. To conclude, the treated scaffolds offer small pore sizes for the initial attachment of the cells and larger pore sizes for preventing pore occlusion. Further, the treated-PCL scaffolds in 1 M demonstrated improved cell spreading due to their optimal modifications, roughness, and structural integrity.

Figure 12.

SEM images of cell-seeded scaffolds after three days of culture (top view).

3.7.3. Alizarin Red Assay

Figure 13 shows the mineralization and calcium deposition of MC3T3-E1 cells in the osteogenic cell culture medium after seven days. The results of alizarin red staining indicated that the mineralized nodule formation and the mineralization caused by the scaffolds in the 1 M group and large pore size scaffolds were significantly higher than with the other scaffolds (Figure 14). The C7001 and T7001 groups showed statistically significant differences compared to the other groups. The results of the alizarin red staining are in complete agreement with the trends observed in terms of the pore size, pore shape, roughness, and cell proliferation.

Figure 13.

Qualitative analysis of mineralization by MC3T3-E1 cells caused by different scaffolds.

Figure 14.

Percentage of the mineralization area of alizarin red staining (p < 0.05).

3.8. Influential Parameters

In this study, three factors (pore size, pore shape, and NaOH concentration) were selected, and the DOE was also used to evaluate the effect of these factors on different responses. Herein, we introduced diverse 3D-printed PCL scaffolds with different pore shapes and pore sizes. The effects of pore shapes and pore sizes along with the surface modification of PCL scaffolds on the physical, chemical, and biological performance were studied. Please note that 3D printing has the potential to produce scaffolds with good control over the structure by changing the filament orientation angle and filament distance, thus having porous PCL scaffolds with various shapes and pore sizes. An optimization of the pore size, pore shape, and surface modification was introduced in our study by affecting the performance properties such as the roughness and mechanical and biological behavior. Results indicated that the properties could be affected by changing factors. As a result, porous structures with the appropriate size, shape, and porosity led to better performance due to altering the permeability and supplying enough space for cell function, triggering biological processes, cell proliferation, and bone ingrowth. In the case of R_a, only the pore size and pore shape were significant since the P values related to their main effects were 0.023 and 0.008, respectively. Regarding the main effect (Figure 15A), it was found that an increase in the pore size caused a decrease in R_a. In addition, when the pore shape was triangle, the R_a increased. The P values for the pore size and pore shape for R_q were estimated to be 0.021 and 0.007, respectively, which indicated a high R_q in small pore sizes and the triangle geometry (Figure 15B), similar to R_a. For compressive strength, the main effect of the pore shape with a P value = 0.011 was significant; however, it was found that the significant interaction was between the pore size and NaOH concentration (P value = 0.040). Figure 15C shows that, in the triangle geometry, the compressive strength increased. Moreover, high NaOH concentrations and large pore sizes led to a reduction in the compressive strength, and large pore sizes had a lower compressive strength in higher NaOH concentrations (Figure 15D). The main effects of the factors of the pore size and NaOH concentration on the compressive modulus were significant (pore size P value = 0.000 and NaOH concentration P value = 0.043). Figure 15E demonstrates the reduced compressive modulus with the increase in the pore size. Furthermore, in low NaOH concentrations, the compressive modulus was high and then decreased up to the center point and became constant up to the high level of this factor. The cell viability of three days showed that the terms for the NaOH concentration and pore shape were significant (pore shape P value = 0.047 and NaOH concentration P value = 0.036), and in contrast, none of the parameters in the cell viability of seven days were significant. Therefore, the triangle geometry and increase in the NaOH concentration were considered to decrease the cell viability at three days (Figure 15F). In the case of mineralization, NaOH concentrations with P value = 0.013 in the main effects indicated that an increase in this factor decreased the mineralization (Figure 15G).

Figure 15.

Influential parameters. (A) Main effects of parameters for R_a. (B) Main effects of parameters for R_q. (C) Main effects of parameters for compressive strength. (D) Interaction effects of parameters for compressive strength. (E) Main effects of parameters for compressive modulus. (F) Main effects of parameters for cell viability (3D). (G) Main effects of parameters for mineralization.

4. Discussion

This study aimed to develop 3D-printed PCL scaffolds with different pore sizes, shapes, and NaOH modifications to achieve proper mechanical properties and biological functions concurrently. In BTE, porous scaffolds with biocompatible and biodegradable materials and high mechanical properties along with an optimal pore geometry (size and shape), porosity, and interconnectivity are necessary for cell proliferation, mass transport, and bone tissue regeneration.³⁰ In the last decade, additive manufacturing has been proven to be promising in the building of custom-tailored architectures with favorable mechanical properties.³¹ The previous studies confirmed that the 3D-printed PCL scaffolds as bone grafts could develop biodegradable and porous structures with desired mechanical properties.³²⁻³⁴ Generally, PCL scaffolds are unfavorable for cell adhesion and bone tissue regeneration due to the hydrophobicity and lack of molecular motifs for biological performance.³⁵ Therefore, PCL scaffold modification can increase the surface roughness and hydrophilicity to enhance cell proliferation and adhesion.¹⁸

The present study involved developing a porous PCL scaffold with interconnected pores and proper mechanical strength using 3D printing. The PCL scaffolds with pore sizes of 500 and 700 μm and pore shapes of a cube and triangle were designed and successfully printed. The porous structures of the PCL scaffolds were well-controlled and interconnected. The results showed that the untreated PCL scaffolds maintained their pore size, pore shape, and interconnectivity in accordance with the software design; moreover, the mean porosity percentage of these scaffolds was approximately 50%. Previous studies have shown that scaffolds with a wide range of pore size, that is, between 100 and 1500 μm, can be used suitably for bone tissue engineering. Alternatively, the in vitro and in vivo studies indicated that pore sizes of greater than 300 μm, microporosity of <20 μm, and interconnected open pores facilitate enhancing new bone formation and the generation of capillaries.^36,37 In addition, an interconnective porosity of 60–75% is suggested for optimal cell growth.¹¹ A small pore favors hypoxic conditions and induces chondrogenesis, whereas a large pore benefits in vivo due to better nutrient supply, delayed pore occlusion, angiogenesis, and stimulation of new bone formation.^38,39 It has also been suggested that larger pores (>500 μm) promote bone healing and ingrowth. For this reason, large-pore-size 3D-printed scaffolds may be the most favorable bone substitute by providing a desirable environment for the ingrowth of natural bone.³⁷ In addition, it is known that the pore geometry and curvature could control the cell response and bone regeneration process, and the increased pore corner numbers lead to higher cell growth and pore occlusion.¹²

To improve cell seeding efficiency and osteoconductivity, alkaline hydrolysis was used to 3D-print PCL scaffolds to overcome the hydrophobicity and lack of specific cell recognition sites.⁴⁰ When PCL scaffolds were immersed in different concentrations of NaOH solutions (1, 3, and 5 M), the surface roughness increased and the scaffold surfaces improved the cell attachment and mineralization. This study demonstrated the combination of SEM and AFM observations on the pore distribution and surface properties. The SEM observation shows the pores’ morphology and surface of scaffolds; however, SEM cannot fully reflect information on the surface roughness of the scaffolds. Therefore, AFM results are effective in nanometer spatial resolution to measure the surface topography, surface characteristics, and nanopore morphology. Based on SEM results, the cube geometry showed smooth surfaces with few holes in different NaOH concentrations; moreover, the scaffolds with a pore size of 700 μm were smoother than scaffolds with a pore size of 500 μm. In contrast, rough surfaces with holes on the scaffold walls were observed on the triangle pore geometry and small pore sizes. Then, information on the surface morphology variation of the scaffolds was obtained. In this context, compared with the untreated PCL, the surface roughness of the treated PCL scaffolds in the triangle pore geometry and of a smaller pore size increased due to higher diffusion of NaOH into the polymer matrix. In this study, the results suggested that the higher fluid velocities and rapid diffusivity in larger pore sizes and the cube pore geometry led to low roughness, while higher permeation and lower rates occurred in the triangle pore geometry and smaller pore size. The higher NaOH concentrations altered porous architectures and dimensions especially in the triangle pore shape and 500 μm pore size. Nevertheless, in the 1 M concentration, the surface roughness was maintained without a loss of the structural integrity and deformation. Moreover, the plots of the main effects of R_a and R_q responses illustrated that the small pore sizes and triangle geometry improved surface roughness. It has been indicated that NaOH treatment by alkaline degradation of the ester bonds provides polar functional groups to the surface roughness of PCL scaffolds.²⁰ Consecutively, the roughness affects the physicochemical properties of the scaffold, the cellular response, and the anchoring of the cells to the scaffold surface.²²

The ideal scaffolds for bone tissue engineering should degrade at the same rate as new bone is formed when implanted in vivo. Therefore, for bone regeneration, a rate of degradation that promotes gradual development of the ECM is crucial.⁴¹ PCL is a biomaterial with potential uses in biomedical applications. The biodegradation rate reported for PCL, however, differs based on the degree of hydrophobicity and crystallinity. It has been shown that a slower degradation ratio of approximately 2–5 years correlated to the hydrophobic nature and higher crystallinity.⁴² In the current study, the alkaline hydrolysis-treated PCL scaffolds are expected to provide additional mechanical support for neo-tissues. In addition to the slow biodegradation of the 3D-printed PCL scaffolds, they tend to provide enduring support to the patient and promote ECM formation. In vitro bioactivity evaluation of the 3D-printed PCL scaffolds is another essential factor in understanding cell–material interactions and for orthopedic applications. A crystalline phase of hydroxyapatite is expected to form upon an interaction with SBF, which could enhance bone-bonding abilities. The untreated and treated PCL scaffolds were coated with hydroxyapatite, resulting in a chemically and structurally equivalent mineral phase to that of bone, improving the interfacial bond between the scaffolds and bone tissues. The aim of the bioactivity test was to examine the bone-like apatite formation on the surface of the treated scaffolds upon immersion in the SBF, which was achieved in almost all designs. Before soaking, there were no ions such as calcium or phosphate groups on the scaffolds as they were made of pure PCL. The ions on the surface of scaffolds are all related to the SBF solution after soaking. The chemical elements on the scaffold surfaces identified by EDX analysis showed the carbon content that can be related to PCL or the hydroxycarbonate apatite formation. However, ATR FTIR analysis could be useful to identify the carbon groups and their association with the deposited calcium phosphate.

The structural stability of scaffolds is also necessary for bone replacement to bear loads and support cell adhesion. Bone scaffolds are typically made of porous degradable materials that act as mechanical supports during tissue formation and repair. To provide the necessary mechanical support for a load-bearing bone defect, the implanted scaffold should have sufficient mechanical strength.^43,44 The low mechanical strength of porous scaffolds is a major challenge in BTE.⁴⁵ In addition, the crystallinity degree can affect the mechanical properties. Specifically, PCL is an aliphatic semi-crystalline polyester with a melting temperature of 55–60 °C for extrusion-based 3D printing. According to the previous studies, PCL pellets, PCL filament feedstock, and printed scaffolds do not exhibit significant differences in crystallinity. Furthermore, several studies revealed that printing processes such as the print speed, liquefier temperatures, and filament feed rate affected the tensile strength and modulus. Also, geometrical features such as the pore size and filament diameter play important roles in mechanical properties.^46,47 The compression strength and compression modulus of spongy bone are reported to be approximately 2–12 MPa and 0.05–0.5 GPa, respectively. Multimodal non-uniform gradient porosity patterns are created by spongy bones because of their range of pore sizes and porosity. Ideally, scaffolds should be able to support the growth of cells and tissues in different regions by varying pore sizes and porosity parameters. In this concept, different pore sizes and porosities would exhibit different potentials while different pore parameters could be examined using a single scaffold to facilitate the ingrowth of cells and tissues.⁴⁸ In this study, no significant statistical difference (p < 0.05) in the compression strength was found between the triangle and cube pore geometries, and designed scaffolds mimic the compression strength of spongy bone properties. However, the main and interaction effects for compressive strength showed that the pore shape and the interaction between the pore size and NaOH concentration were significant. It was found that the triangle geometry improved the compressive strength, while high NaOH concentrations and large pore sizes reduced the compressive strength. On the other hand, large pore sizes caused the compressive modulus to decrease. Moreover, low NaOH concentrations increased the compressive modulus. Consequently, to facilitate a match between the natural bone tissue and the bone graft substitute, its mechanical properties should be carefully tailored.⁴⁹

The scaffolds should provide a temporary physical environment for the expansion of cells while facilitating the formation of new tissues.⁵⁰ The in vitro cell viability indicated that the cytocompatibility was scaffold-dependent. Cells are sensitive to biochemical and biomechanical signals and the geometry of their environment.¹³ It has been reported that cell viability is significantly improved by the surface roughness, which results in more nanoscale structures for cell action.⁵⁰ Our findings showed that the cell viability was approximately increased in lower NaOH concentrations as well as a pore size of 500 μm and cube pore shape. In addition, the main effects’ plots indicated that the cell viability decreased in the presence of a triangle geometry and higher NaOH concentrations. However, it was still above 70% for the triangle geometry. The number of concave surfaces in the cube geometry is greater than the triangle geometry, which leads to a faster pore occlusion due to a better cell bridge and growth. Moreover, the treated-PCL scaffolds displayed higher cell attachment, indicating that the structural integrity and surface roughness might affect cell growth. On the one hand, smaller pores increase proliferation since they provide a greater surface for cellular adhesion and growth, while on the other hand, smaller pores limit the scaffold permeability, which affects nutrient and waste exchanges as well as cell migration and capillary formation. Despite this, high flow rates can also result in cell damage or detachment.¹² This study also showed that scaffolds with a cube pore geometry and the smallest pores (500 μm) enabled the greatest ingrowth of cells over time.

Cell adhesion assay revealed that limited cells adhered to the PCL strands due to the hydrophobicity of PCL, which does not support cell adhesion or retention.⁵¹ Meanwhile, cells on the treated scaffolds (particular in 1 M) not only are distributed on the pores but also are spreading on the surfaces of the PCL scaffolds. However, the number of cells attached to the surface-treated PCL scaffolds was notably higher than in the untreated PCL scaffolds. As demonstrated by alizarin red analysis, the treated PCL scaffolds improved cellular properties and highly enhanced mineralization. The biomineralization of cells in 1 M groups and larger pore sizes was considerably higher than that in the PCL group on day 7. Alizarin red staining showed that great dispersion of calcium deposits was only found in 1 M groups and larger pore sizes, and limited calcium depositions were found on the other scaffolds. DOE analysis showed the main effect of the NaOH concentration on mineralization was significant, and increasing the NaOH concentration could decrease the mineralization. Furthermore, it has been indicated that an even cell distribution, better cell retention, and a rigid support of the treated scaffolds enhanced the formation of the bone matrix and calcification.³⁹ A sufficient medium flow can be assumed through the larger-pore-size scaffolds due to the high medium permeation in the scaffold center and adequate oxygen provision.⁵² Apart from these findings, it is suggested that future studies should focus on assessing the effectiveness of the treated scaffolds in vivo.

5. Conclusions

This study developed PCL scaffolds by 3D printing and treated them with NaOH solutions in different concentrations. The 3D-printed scaffolds were designed in cube and triangle pore shapes with 500 and 700 μm pore sizes. The results showed that the scaffolds possessed well-interconnected macropores and controllable porous structures. The surface modification in the high concentration greatly affected the triangle geometry and small pore size due to slow velocity and higher exposure. The treated scaffolds showed favorable characteristics with the surface roughness, porosity, interconnectivity, mechanical properties like those of cancellous bone, bioactivity close to hydroxyapatite, cell viability, and biomineralization, which are outstanding in promoting bone regeneration in orthopedic applications. Out of three NaOH concentrations, the scaffolds in the 1 M concentration exhibited optimal bioactivity, biomineralization, and cytocompatibility. Overall, the attained results showed that the treated scaffolds with customized structures and desirable modification demonstrated great potential as an effective bone defect treatment.

The authors declare no competing financial interest.