Effect of pore sizes of PLGA scaffolds on mechanical properties and cell behaviour for nucleus pulposus regeneration in vivo

Authors

  • Hye Yun Kim,

    1. Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, Jeonju, Korea
    Search for more papers by this author
  • Ha Neul Kim,

    1. Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, Jeonju, Korea
    Search for more papers by this author
  • So Jin Lee,

    1. Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, Jeonju, Korea
    Search for more papers by this author
  • Jeong Eun Song,

    1. Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, Jeonju, Korea
    Search for more papers by this author
  • Soon Yong Kwon,

    1. Department of Orthopaedic Surgery, Yeouido St Mary's Hospital, Catholic University of Korea, Seoul, Korea
    Search for more papers by this author
  • Jin Wha Chung,

    Corresponding author
    1. Department of Orthopaedic Surgery, Yeouido St Mary's Hospital, Catholic University of Korea, Seoul, Korea
    • Correspondence to: Gilson Khang, Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, 567 Baekje-daero, Jeonju 561–756, Korea. E-mail: gskhang@jbnu.ac.kr

      Jin Wha Chung, Department of Orthopaedic Surgery, Yeouido St Mary's Hospital, Catholic University of Korea, Seoul 150-713, Korea. E-mail: koreafoot@gmail.com

    Search for more papers by this author
  • Dongwon Lee,

    1. Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, Jeonju, Korea
    Search for more papers by this author
  • Gilson Khang

    Corresponding author
    1. Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, Jeonju, Korea
    • Correspondence to: Gilson Khang, Department of BIN Fusion Technology, Polymer Fusion Research Centre, Chonbuk National University, 567 Baekje-daero, Jeonju 561–756, Korea. E-mail: gskhang@jbnu.ac.kr

      Jin Wha Chung, Department of Orthopaedic Surgery, Yeouido St Mary's Hospital, Catholic University of Korea, Seoul 150-713, Korea. E-mail: koreafoot@gmail.com

    Search for more papers by this author

Abstract

This study investigated the influence of pore sizes of poly(lactic-co-glycolic acid) (PLGA) scaffolds on the compressive strength of tissue-engineered biodiscs and selection of the best suitable pore size for cells to grow in vivo. PLGA scaffolds were fabricated by solvent casting/salt-leaching with pore sizes of 90–180, 180–250, 250–355 and 355–425 µm. Nucleus pulposus (NP) cells were seeded on PLGA scaffolds with various pore sizes. Each sample was harvested at each time point, after retrieval of PLGA scaffolds seeded with NP cells, which were implanted into subcutaneous spaces in nude mice at 4 and 6 weeks. MTT assay, glycosaminoglycan (GAG) assay, haematoxylin and eosin (H&E) staining, safranin O staining and immunohistochemistry (for collagen type II) were performed at each time point. As the pores became smaller, the value of the compressive strength of the scaffold was increased. The group of scaffolds with pore sizes of 90–250 µm showed better cell proliferation and ECM production. These results demonstrated that the compressive strength of the scaffold was improved while the scaffold had pore sizes in the range 90–250 µm and good cell interconnectivity. Suitable space in the scaffold for cell viability is a key factor for cell metabolism. Copyright © 2014 John Wiley & Sons, Ltd.

1 Introduction

The intervertebral disc (IVD) is composed of a central glycosaminoglycan (GAG)-rich nucleus pulposus (NP) and an outer collagen-rich annulus fibrosus (AF) (Anderson and Tannoury, 2005). The outer AF is primarily responsible for withstanding tensile strength (circumferential, longitudinal and torsion), functioning as an intervertebral ligament resisting vertebral motion. The NP has a more amorphous consistency, with randomly orientated collagen fibrils and a large content of proteoglycan, providing it with the ability to resist compressive stress (Cheng et al., 2010). The hydroxyproline content appears to contribute to the compressive modulus to a much smaller degree. These trends are consistent with studies of structure–property relationships in engineered articular cartilage (Anderson and Tannoury, 2005; Cheng et al., 2010). The high content of GAG in the NP provides the mechanical properties needed for the tissue to counter the effects of compression. Loss of GAGs from the NP is the first sign of disc degeneration, resulting in a loss of compression pressure and absorption ability.

For NP tissue regeneration, a scaffold must provide both biological and physical properties. The biophysical properties of a scaffold can significantly influence cell migration, adhesion, viability and new extracellular matrix (ECM) production in a three-dimensional (3D) environment (Jones et al., 2009). Pore size, porosity and density in a scaffold play important roles in tissue engineering (Hu et al., 2002). When the scaffold has a highly porous microstructure, the microscale pores offer interconnections, 3D architecture, suitable mechanical properties and sufficient space (Murphy et al., 2000; Freed et al., 1994). The pore sizes of the scaffolds need to meet general requirements, which are sufficient space, compressive strength for greater stability and bioactivity for cell seeding, as well as the delivery of nutrients and oxygen to the seeded cells within the scaffold to form new NP tissue. Therefore, the scaffold must be designed to help cells proliferate and secrete ECM. Mechanical properties are important parameters in 3D scaffolds, because the implant could undergo diverse compressive forces. The mechanical properties of a scaffold depend on the inherent structure and bulk mechanical properties of the materials. The mechanical properties of the structured scaffolds are highly affected by pore structure and the bulk mechanical properties of the materials (Jang et al., 2004).

Biodegradable polymer scaffolds play major roles in the field of tissue engineering as a 3D template to guide tissue regeneration for different applications. Poly(lactic-co-glycolic acid) (PLGA) scaffold is a well-known biodegradable material of excellent mechanical strength (Lee et al., 2007; Bolland et al., 2008; Kim et al., 2003), good biocompatibility and biodegradability in vivo, which can be readily modulated by variation of the copolymer ratio lactic:glycolic acid (Oh et al., 2006). Also, PLGA has been approved for human clinical application by the Food and Drug Administration (FDA) and has been used in tissue engineering as surgical sutures and some implantable devices (Chen et al., 2000). PLGA-based composite scaffolds, originally developed for NP regeneration, and PLGA scaffolds have demonstrated great potential for NP tissue engineering due to their ability to promote mechanical properties and tissue development.

In this study we evaluated the influence of pore size on the structural and mechanical properties and ECM secretion of scaffolds with controlled pore sizes. PLGA scaffolds were prepared using the solvent casting/salt-leaching method, whereby various sodium chloride (NaCl) particle sizes are used to produce scaffolds with a homogeneous pore structure. Porous PLGA scaffolds were constructed and long-term in vivo experiements were designed to evaluate NP regeneration performance. The effects of controlled pore size on NP reconstruction by seeded NP cells were examined in an in vivo subcutaneous implantation model.

2 Materials and methods

2.1 Materials and reagents

PLGA (molecular weight 90 000 g/mole, 75:25 mole ratio of lactide:glycolide; Resomer® RG 756) was purchased from Boehringer Ingelheim (Germany). Sodium chloride (NaCl; Orient Chemical Co., Korea) and methylene chloride (MC; Tedia Co Inc., Phillipsburg, USA) as solvents were of HPLC grade.

2.2 Preparation of the porous PLGA scaffold

In our study, PLGA scaffolds with different pore sizes were prepared by the solvent-casting/salt-leaching method. A prefabricated porogen matrix of NaCl was used to generate an open pore structure. The porogen NaCl was sieved to obtain granules of 90–180, 180–250, 250–355 and 355–425 µm diameter, respectively. After that, 9 g NaCl was added to the mixture of PLGA (1 g) and MC (5 ml). The mixture was poured into a silicone mould (5 mm diameter × 5 mm thick) and then pressurized using a Lab Press machine (MH-50Y, CAP 50 tons; Masada, Tokyo, Japan) at 60 kg/cm2 for 24 h at room temperature. The mixture was incubated in distilled water to remove the NaCl for 48 h, resulting in porous PLGA scaffolds. PLGA scaffolds with four different ranges of pore diameters were lyophilized for 24 h (Jang et al., 2004).

2.3 Scaffold characterization

The pore architecture of non-degraded and degraded PLGA scaffolds with different pore sizes (90–180, 180–250, 250–355 and 355–425 µm) was observed using scanning electron microscopy (SEM; Bio-LV, Model SN-3000, Hitachi Co., Tokyo, Japan). Before microscopy, the dried scaffolds were plasma sputter-coated with gold (Emscope SC 500K, London, UK) under argon gas and the morphology was examined using Bio-LV SEM (Munirah et al., 2008).

Water uptake and degradation studies were conducted on PLGA scaffolds with different pore sizes. The dried scaffolds were weighed (Wd) and each was placed in a 15 ml conical tube, 10 ml phosphate-buffered saline (PBS; Gibco, Grand Island, NY, USA) was added to each tube, and the tubes were incubated at 37°C and 60–70 rpm. The wet weight (Wt) of each sample was measured after different periods of time (1, 3, 7, 14, 21 and 30 days in vitro). The scaffolds were taken out of the solutions and lyophilization was performed, using a freeze-drying method, to obtain the dry weight (Wi). Percentage degradation was calculated using the formula:

display math

In addition, the water uptake was also calculated using the formula:

display math

The pH of the degradation buffer was measured using a pH meter (Thermo Scientific, USA) at scheduled times (Ko et al., 2007; Lee et al., 2006).

2.4 Isolation and culture of NP cells

Intervertebral disc (IVD) tissue for cell isolation was obtained from female, 4 week-old New Zealand White rabbits, as described previously (Xu and Jr, 2005). Briefly, the extracted discs were washed in PBS, pH 7.4, containing 2% penicillin (Invitrogen), to remove blood or other contaminants. The NP tissue was separated from surrounding AF using a scalpel and was washed with PBS. After being washed several times with PBS containing 2% penicillin, NP tissue was digested using enzymatic-based methods (collagenase type A). The NP tissue was incubated at 37ºC in a humidified atmosphere of 5% CO2 for 12 h in 10 ml Dulbecco's modified Eagle's medium (DMEM: Gibco):nutrient mixture F12 (Invitrogen,USA) at a 1:1 ratio, containing 0.6% v/v collagenase type A (Roche Applied Science, Germany) and supplemented with 10% v/v fetal bovine serum (FBS; Biochrom, Germany), 100 g/ml penicillin (Invitrogen) and 100 g/ml streptomycin (Invitrogen). The cell–collagenase solution was filtered through a 100 µm nylon mesh and centrifuged at 1200 rpm for 5 min. The supernatant was decanted and the cell pellet was washed three times in culture medium to remove the remaining enzyme. The cells were cultured in a mixture of equal a volumes of F12 and DMEM supplemented with 10 volumes of F12 and DMEM supplemented with 10% FBS, 100 g/ml penicillin and 100 g/ml streptomycin. The cultured cells were incubated at 37°C and 5% CO2 while they underwent two passages.

Scaffolds with four different ranges of pore diameter were immersed in 70% ethanol for 30 min, then washed three times with PBS. The scaffolds were immersed in culture medium for 1 h and then placed in a 24-well plate and seeded with NP suspension. After incubation for 30 min, 1 ml culture medium was added to each well and culture took place in a 5% CO2 incubator at 37°C.

2.5 Cell penetrability and proliferation assay

Scaffolds immersed in culture medium for 1 h were placed on a 24-well plate and seeded with 20 µl NP suspension at a density of 5 × 105 cells/scaffold. After incubation for 30 min, 1 ml culture medium was added to each well and culture took place in a 5% CO2 incubator at 37°C. The penetrability and proliferation of the cells on each specimen were determined after carrying out a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazoliumbromide (MTT; Sigma Chemical Co., St. Louis, MO, USA) assay after culture periods of 1, 3, 7 and 14 days in vitro. All the samples were transferred to new plates and incubated in 1 ml new culture medium. 100 µl of MTT solution (5 mg/ml stock in PBS) was added to each well and the samples were incubated for 4 h at 37°C to allow the crystals to solubilize. Before testing for cell viability, each sample was observed to determine cell migration within the porous structure of the scaffold; then each sample of 100 µl was pipetted into a 96-well plate and the absorbance at 570 nm, using an ELISA microplate reader (E-Max, Molecular Devices, Sunnyvale, CA, USA), was yielded as a function of viable cell number. The rate of tetrazolium (MTT) reduction is proportional to the rate of cell viability (Jeong et al., 2011).

2.6 Morphology of cells adhering to PLGA scaffolds

To observe the attachment and morphology of NP cells on PLGA scaffolds with different pore sizes, NP cell suspensions were seeded at a concentration of 5 × 105 cells/scaffold and cultured for 1 and 3 weeks in vitro. The samples were then washed three times with PBS and immobilized with 2.5% v/v glutaraldehyde (Sigma) for 24 h at 4°C. Subsequently the scaffolds were subjected to sequential dehydration for 30 min each with a series of ethanols (50%, 60%, 70%, 80%, 90% and 100%). The morphology of the cells on the scaffolds was examined by Bio-LV SEM (SN-3000 Hitachi, Japan) at 10 kV (Lee et al., 2010).

2.7 In vivo animal evaluation

NP cells contained in PLGA scaffolds were implanted into subcutaneous spaces of male 4 week-old athymic nude mice (Orient Animal, Korea). The mice were sacrificed at 4 and 6 weeks. Any fibrous capsules surrounding the implants were completely removed prior to analysis. After harvesting the samples, observation of gross morphology, histological evaluation, compressive strength testing and biochemical analysis were performed.

2.8 Histological evaluation

To evaluate histological analyses, 1 × 106 NP cells were seeded on the scaffolds and the scaffolds were cultured for 4 and 6 weeks in vivo. Each scaffold specimen was fixed in 4% formalin solution and embedded in paraffin. Sections of 4 µm thickness were prepared, then deparaffinized, rehydrated and stained with haematoxylin and eosin (H&E) to assess cell viability, and with safranin O with fast green counterstaining to identify the presence of proteoglycan-rich matrix (Bae et al., 2011).

Immunohistochemical analysis was performed according to the manufacturer's protocol (UltraTek HRP Kit, Immunotech, Marseille, France). The samples were deparaffinized, rehydrated and pretreated with 0.4% v/v pepsin in 0.01 n hydrochloric acid (Sigma) at room temperature for 30 min, then the slides were washed with PBS and incubated with hydrogen peroxidase for 10 min to reduce the non-specific background due to the endogenous peroxidase. All sections were treated with peroxidase block (Super Block, UltraTek HRP Kit; ScyTek Laboratories, Logan, UT, USA) for 10 min prior to antibody incubation. Monoclonal mouse anti-rabbit collagen type II (Calbiochem®, EMD Biosciences Inc., La Jolla, CA, USA) was diluted (1:1000) with antibody diluent (Dako Cytomation, Carpinteria, CA, USA) and then applied to the sections for 60 min. After washing with PBS, they were treated with UltraTek anti-polyvalent biotinylated antibody (UltraTek HRP Kit, Immunotech) for 10 min. After washing with PBS, the signal was finally visualized as a brownish precipitate, using freshly prepared chromogen substrate 3-amino-9-ethylcarbazole (AEC; UltraTek AEC Kit, Immunotech). The sections were counterstained with Mayer's haematoxylin and mounted using permanent aqueous mounting medium (Scy Tek Laboratories).

2.9 Mechanical properties

The compressive strengths of 5 mm PLGA scaffolds of pore sizes 90–180, 180–250, 250–355 and 355–425 µm were measured at a loading rate of 1.0 mm/min, using a Texture Analyser (FTC, Sterting, Virginia, USA) at room temperature. A stress–strain curve was used to define the compressive strength of each scaffold; four specimens/sample were tested. As a control, the same procedure was performed on PLGA scaffolds with the same pore size ranges soaked in culture medium at 37°C.

2.10 Determination of sGAG and collagen content

The scaffolds were harvested at 4 and 6 weeks in vivo. After mechanical analysis, the samples were frozen at −80°C and lyophilized for 24 h, then digested with 300 µl papain solution (125 µg/ml papain, 5 mm l-cysteine hydrochloride, 100 mm disodium hydrogen phosphate, 5 mm sodium EDTA, pH 6.8; Sigma) for 24 h at 60°C. The amount of sulphated GAGs was determined by quantifying with 200 µl 1,9-dimethylmethylene blue (DMMB; Sigma) using an ELISA microplate reader (E-max) at 490 nm. The standard curve for the analysis was generated using chondroitin sulphate (Sigma) in the range 0–50 µg/ml. To evaluate the collagen content of the digested sample, papain solution was added to 2 n sodium hydroxide (Sigma) and 50 µl of this mixture was autoclaved at 120°C for 20 min, then 450 µl chloramine T and p-dimethylaminobenzaldehyde were added to the hydrolyzed samples. The mixtures were mixed slowly for 25 min at room temperature. The standard curve for the analysis was generated using hydroxyproline (Sigma) and absorbances were detected by an ELISA microplate reader at 570 nm.

2.11 Statistics

All data were analysed using Student's t-test; p < 0.05 was considered significant.

3 Results

3.1 Scaffold characterization

3.1.1 Morphology of porous PLGA scaffolds

Cross-sections of porous PLGA scaffolds with different pore sizes were examined by Bio-LV SEM. Figure 1A shows the morphology of PLGA scaffolds prepared using NaCl porogen with four different particle sizes (90–180, 180–250, 250–355 and 355–425 µm). The PLGA scaffolds with small pore sizes (90–180 and 180–250 µm) showed a higher density of pores than those with larger pore size (250–355 and 355–425 µm). The pore sizes of the PLGA scaffolds were estimated by SEM (Figure 1B). The exact pore sizes of these PLGA scaffolds were precisely determined and noted to be 169 ± 7.8, 200 ± 3.3, 328 ± 17.7and 410 ± 19.3 µm, respectively. Pores on the surface of scaffolds fabricated with porogen particle sizes of 250–355 and 355–425 µm were larger than those of scaffolds fabricated in a range of 318–420 µm, while much smaller micropores (169–200 µm) were observed between the walls of the micropores. The lack of organized porous structures is likely due to the good interconnectivity of the salt-leached samples, which reduced the presence of well-organized, largely closed-off pores.

Figure 1.

(A) Photographs of porous PLGA scaffolds prepared using NaCl porogen with four different particle sizes. (B) Cross-sectional morphology of porous PLGA scaffolds with pore sizes of (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm; magnification = (A) × 100, (B) × 200

3.1.2 Morphology of degraded PLGA scaffolds

The cross-sectional surface morphology of degraded porous PLGA scaffolds was imaged using SEM after 0 and 21 days to evaluate how the structural morphology of these scaffolds changes over time in PBS (Figure 2). We were able to observe that some pores diminish, and the remaining pores were smaller in all four groups. After 21 days of degradation, PLGA scaffolds with smaller pore sizes (90–180 and 180–250 µm) were largely shrunken and with closed-off pores. Also, the number and size of the pores decreased and some fibre-like connections appeared among the pores (Figure 2B). On the other hand, scaffolds with larger pore sizes (250–355 and 355–425 µm) were slightly shrunk during the dehydration process and maintained some porous structure, but finally became soft and with thinner pore walls.

Figure 2.

SEM image shows cross-section of non-cell sample for degradation morphology after 0 (A) and 21 days (B). Pore sizes of PLGA scaffolds: (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm; magnification = ×100

3.2 Water uptake and degradation of scaffold

The swelling ratios of the various scaffolds are shown in Figure 3A. The water uptake behaviours in each group of scaffolds increased gradually, and the water uptake ratio increased significantly with the increase of pore size of the scaffolds between 1 and 4 weeks. The swelling properties of the scaffolds with larger pore size (355–425 µm) were doubled compared to scaffolds with smaller pore sizes (90–180 µm) at 4 weeks.

Figure 3.

(A) Water uptake of porous PLGA scaffolds and (B) weight loss of fabricated scaffolds was investigated by measuring the degradation rate of the porous PLGA scaffolds. Significant differences between groups: **p < 0.05; *p < 0.1

The in vitro degradation behaviour of the fabricated scaffolds was investigated by measuring the rate of weight loss of the scaffolds as a function of degradation time. The percentage degradations of PLGA scaffolds with larger pore sizes (250–355 and 355–425 µm) were significantly higher than those of scaffolds with smaller pore sizes (90–180 and 180–250 µm) (Figure 3B).

3.3 MTT assay

To evaluate migratory capacity and proliferation in PLGA scaffolds with various pore sizes, NP cells were seeded on scaffolds and measured at 1, 3, 7 and 14 days using the MTT assay (Figure 4). As shown in Figure 5A, cells were incubated with PLGA scaffolds to measure scaffold permeability, which increased as a function of culture time. No significant difference was observed in the cell migration in any of the four groups at day 1. Cell distribution in the porous PLGA scaffolds increased gradually from 5 day to day 14. The majority of NP cells from all four groups were found to be viable throughout the matrix.

Figure 4.

The morphology of monolayer-cultured NP cells after the second passage was identified by microscopic observation; magnifications = (a) × 40, (b) × 100

Figure 5.

(A) Photographs to measure the permeability of cells incubated in porous PLGA scaffolds with pore sizes of (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm. (B) Measurements of viability of seeded NP cells on the surface of porous PLGA scaffolds on days 1, 5, 7 and 14 in in vitro. Significant differences between groups: ***p < 0.001; **p < 0.01; *p < 0.05

Cell proliferation and attachment were measured at 1, 3, 7 and 14 days using the MTT assay (Figure 5B), although cell proliferation on each group of scaffolds was lower than that of day 2, due to a low degree of initial attachment onto the scaffold matrix. The cell number seeded on the scaffold significantly increased gradually at all culture periods. During days 5–14 the trend of average cell number/construct was as follows: NP cells proliferated rapidly in groups with both smaller and larger pores; when the proliferation of cells on the scaffolds was measured, it was determined that the cell growth rate on the largest pore size scaffolds (355–425 µm) increased less than on other scaffolds (90–180, 180–250 and 250–355 µm) over the 7 days.

3.4 Adhered cell morphology on scaffolds

As shown in Figure 6, the morphology of NP cells grown on the four different PLGA scaffolds showed some differences at 1 and 3 weeks in vitro. After 1 week, NP cells showed flatter and irregular morphology on scaffolds with pore sizes 90–180, 180–250 and 250–355 µm, but some of the attached cells showed better adhesion and activity on the scaffold with pore size 355–425 µm. After 3 weeks, most of the cells seemed to be adhered to the scaffolds with larger pore sizes (250–355 and 355–425 µm) than on those with smaller pore sizes (90–180 and 180–250 µm). In the scaffolds with larger pore sizes (250–355 µm and 355–425 µm), the NP cells changed their growth direction and extended toward the pores. On the other hand, NP cell morphology changed obviously in scaffolds with pore size 250–355 µm and the cells were clustered and abundant, as shown in Figure 5. It seemed that the pore size of scaffold had significant effect on the activity and proliferation within inner pores of scaffolds.

Figure 6.

SEM images of NP cell morphology on PLGA scaffold surfaces with pore sizes of (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm at cultivation times of (A) 1 week and (B) 3 weeks: magnifications = (A) × 1000, (B) × 2000

3.5 Histological evaluation

To evaluate in vivo application of scaffolds, the specimens were implanted into subcutaneous spaces in nude mice. The implants were allowed to develop for up to 6 weeks and harvested at defined time points. After implantation, only PLGA constructs with pore size 180–250 µm maintained their original rounded cylindrical shape compared with in vitro specimens. However, the strength of the interface could not be quantitatively measured between 4 and 6 weeks (Figure 7). In contrast, the sizes of all of the PLGA scaffolds (90–180, 180–250, 250–355 and 355–425 µm) when excised were gradually decreased compared with the original volumes, and the scaffolds became more strained and flat in appearance. However, the tissue layers surrounding all the constructs were completely attached to the scaffolds. The areas of integration were much more favourable for the scaffolds with larger pore sizes (250–355 and 355–425 µm).

Figure 7.

Gross morphology of PLGA scaffolds with pore sizes of (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm after implantation and post-harvest after (A) 4 weeks and (B) 6 weeks

The morphology of cells and distribution of ECM were evaluated by means of histological staining; H&E and safranin O staining and collagen type II immunolocalization. Figure 8 shows the results of H&E staining of scaffolds after cell culture in vivo for 4 and 6 weeks; the cells have occupied most of the spaces in the scaffolds with larger pores, while the scaffolds with smaller pores were still empty throughout.

Figure 8.

H&E staining in porous PLGA scaffolds for evaluation of in vivo application of scaffolds with pore sizes of (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm after (A) 4 and (B) 6 weeks of implantation in vivo. All images are × 100 magnification (scale bars = 100 µm)

Safranin O staining was used to detect the anionic GAG chains of the proteoglycan. Figure 9A shows the results of safranin O staining of the scaffolds after 4 and 6 weeks of cell culture. After 4 weeks, GAGs (red or orange) in scaffolds with smaller pore sizes only covered a few pores, gradually increasing with time, whereas GAGs in scaffolds with larger pore sizes covered most of the pores and secreted more of the GAG contents of the ECM. As the culture time extended, the GAGs of scaffolds with larger pore sizes began to be distributed all over the constructs and formed a more uniform matrix distribution compared with PLGA scaffolds with smaller pore sizes.

Figure 9.

(A) Safranin O staining and (B) collagen type II immunohistochemistry staining in porous PLGA scaffolds with pore sizes of (a) 90–180, (b) 180–250, (c) 250–355 and (d) 355–425 µm after 4 and 6 weeks of implantation in vivo, stained red for collagen type II. All images are × 200 magnification (scale bars = 100 µm)

As shown in Figure 9B, collagen secretion was found to be similar to sulphated GAG secretion. Collagen type II exhibited strong immunopositivity at the specific region of in vivo PLGA scaffolds with larger pore size, mainly localized in the pericellular and interterritorial matrix. Moreover, the collagen distribution in scaffolds with larger pore sizes was more uniform than that in smaller pore size scaffolds, especially after 6 weeks in vivo. Also, scaffolds with larger pore sizes showed more a homogeneous collagen type II distribution than in scaffolds with smaller pore sizes.

3.6 Determination of sGAG and collagen content

At each time point of harvesting, the samples were stored in cryotubes at −70°C for GAGs content testing and thawed immediately prior to testing. The GAGs content of each sample increased gradually with time (Figure 10A). By 6 weeks, in scaffolds with larger pores (250–355 and 355–425 µm) the amount of GAGs was high. There were significant differences between each group, except 90–180 and 180–250 µm at 4 and 6 weeks. The amount of GAGs for the 355–425 µm pore size scaffolds had significantly increased from 4 and 6 weeks, while that in the 90–180 and 180–250 µm pore size scaffolds had slightly increased. The amount of collagen gradually increased significantly with time in PLGA scaffolds with larger pore sizes. However, in the two groups with smaller pore sizes (90–180 and 180–250 µm) the collagen content did not show any significant difference between 4 and 6 weeks (Figure 10B).

Figure 10.

The secretion of (A) GAG and (B) collagen in porous PLGA constructs with four different pore sizes after 4 and 6 weeks. The values have been normalized to dried scaffold weight in the constructs. Significant differences between groups: ***p < 0.001; **p < 0.01; *p < 0.05

3.7 Mechanical properties

After 4 and 6 weeks in vivo implantation, the normalized compressive strength was determined for wet PLGA scaffolds with different pore sizes. However, as a control, PLGA scaffolds were measured in the dry state after fabrication. Evaluation of mechanical properties indicated that compressive strength increased with time in all implantations, as shown in Figure 10. After 4 weeks of in vivo culture, with the increase of porogen particle size from 90 to 425 µm within the controls, hardness with all four kinds of particle size of salt decreased significantly. The strength of the scaffolds with smaller pores (90–180 and 180–250 µm) increased 4.5-fold in the course of 6 weeks in vivo, while that of those with larger pores (250–355 and 355–425 µm) increased < 3.5-fold. It was clearly seen that the scaffolds with smaller pore had higher compressive strength.

Uniaxial compression tests were performed to observe the effect of pore size on the mechanical properties of the structured PLGA scaffolds. As shown in Figure 11A, the compressive strength decreased as pore size increased, i.e. compressive strength was closely related to porosity.

Figure 11.

Compressive strength of dried porous PLGA scaffolds seeded with NP cells for 4 and 6 weeks. Significant differences between groups: ***p < 0.001; **p < 0.01; *p < 0.05

4 Discussion

In our study, the salt-leaching process used in conventional methods was selected to investigate preparing porous scaffolds in tissue engineering. The salt porogen dispersed randomly in the PLGA solution and modified solvent crystallization by reducing and limiting crystal growth, making the crystals of solvent smaller and asymmetrical. After sublimation of the solvent, the polymer-rich phase becomes the skeleton of the scaffold and the spaces left by the dissolved crystals develop into pores. The sodium chloride can affect the polymer chains within the polymer matrices during hydration. As the polymeric matrix hydrates, the NaCl reacts with PLGA chains, finally resulting in the crosslinking of PLGA molecules (Rossella et al., 2010).

The fabricated PLGA scaffolds showed that their porosities increased with decreasing porogen particle size. The porogen particle size, determined as the pore size in the scaffolds, had effects on the mechanical properties (Krebs et al., 2009). Therefore, scaffolds with different pore size ranges had different mechanical properties and increased with increasing relative density. It was necessary to study the effect of pore size and relative density on the mechanical properties of scaffolds. We found that higher porosity corresponded to a lower Young's modulus and lower compressive strength, which was consistent with the inverse tendency between porosity and mechanical properties that has been described by other studies (Wu and Ding, 2005; Goins et al., 2005).

Although few studies have shown the effects of porosity on bone, higher porosity is usually associated with greater bone formation (Kujala et al., 2003). In a non-fully interconnected scaffold, bone ingrowth should generally be faster when larger macropores are present. Although increased porosity and pore size facilitate bone ingrowth, the result is a reduction in mechanical properties, since the structural integrity of the scaffold is compromised.

For tissue regeneration, all scaffolds must be tested for their ability to take up and retain water, because this capability makes them suitable scaffolds for the exchange of nutrient and metabolic in the medium flow. Both water uptake and degradation are essential to studies on degradation behaviour. They are aimed to be used in biomedical applications, particularly following tissue engineering. Scaffolds should be designed be to completely replaced by the regenerated ECM on incorporation with the surrounding tissue and to need no further surgery to remove them. The scaffolds should be adjusted for degradation rate and degradation production. The effect of the degradation rate is associated with cellular interaction, including cell proliferation and migration and tissue regeneration. The degradation rate of 3D porous scaffolds reflect that a fast degradation rate negatively affects cell proliferation and viability in the scaffold in vitro and in vivo. On the other hand, when degradation does not occur, tissue regeneration is not properly achieved. Therefore the materials must process an appropriate degradation rate to assure that scaffolds maintain a necessary structure and strength required in the initial period of implantation until the new ingrowth's tissue substitutes the support (Gong et al., 2006). The mechanical properties are a useful prediction of the degradation process when the polymer is degraded. The smaller pore size scaffolds have the higher compressive strength (Jiang et al., 2007). Since a scaffold with macro-porosity has a high wetting rate, the water molecules become more closed to the polyester bond and this reaction starts to be susceptible to hydrolysis as the rate-limiting step. As the water uptake of the scaffold changes the polymer chains. It affects porosity and 3D structural feature of scaffolds. A higher water uptake enhanced the hydrolysis reactions. It is usually useful to evaluate the water uptake of biopolymers, because it represents the affinity for water of the polymer. Therefore, our PLGA scaffolds with four different pore size ranges, made of biocompatible synthetic polymer, may interact with the medium flow in vivo. Consequently, highly interconnective pores are useful in NP tissue engineering to achieve new tissue regeneration within the scaffold system.

NP cells reside as single, paired or multiple cells in a contiguous pericellular matrix (PCM), whose structure and properties may significantly influence cell and extracellular matrix mechanics (Goins et al., 2005). In this study, the effect of porous PLGA scaffolds was investigated using cell invasion and proliferation for NP regeneration.

The larger interconnective pores improve the cellularity and matrix content within the scaffold. The matrices containing pores 13–85 µm in diameter formed cartilaginous tissue at the surface of matrix, with little penetration into the microporous structures. Macroporous scaffolds have been found to improve cell attachment, proliferation and biosynthetic activity, secondary to enhanced diffusion of cells and nutrients into the centres of the scaffolds (Murphy et al., 2010; Karande et al., 2004; Anita et al., 2005). Lack of cell migration and tissue ingrowth within 3D scaffolds remains a major constraint in the clinical application of these structures. High cell density at the surface of the construct may deplete nutrient supply before these nutrients reach the inner cells. Therefore, suitable pore sizes have been studied in search of scaffold designs and culture techniques improving cell distribution throughout the matrices, as a basis for uniform tissue regeneration. However, with increasing culture time, the degree of cell migration and the rate of cell proliferation on the structured scaffolds surpassed those of the PLGA scaffolds.

The reason for this is that large porous structures reduce the surface area of the scaffold and subsequently cell adhesion, migration and cell–cell interaction will be disturbed. This result confirmed that seeded cells on PLGA scaffolds with medium pore sizes (180–250 and 250–355 µm) are able to permeate into pores of the scaffold. It was demonstrated that larger pore size of PLGA scaffold has a good porosity. Open pores, interconnectivity and adequate pore size for cell growth and ECM secretion are required for the application of regenerative medicine strategies. All four different PLGA scaffolds made by the solvent/salt-leaching method had apparently uniform pore morphology with pores square in shape and well-controlled 3D interconnected architecture, so these PLGA scaffolds with high porosity provide a high surface area to facilitate cell adhesion and cell–polymer interactions and offer sufficient space for cell proliferation and tissue growth during in vitro culture (Murphy et al., 2000; Freed et al., 1994; Liao et al., 2005; Lu et al., 2000; Kim et al., 2007).

We also performed histological studies to examine the effects of pore size in vivo. Consequently, we found that the group of PLGA scaffolds with larger pore sizes was clearly distinguished from the group with smaller pore sizes in terms of overall NP tissue formation, cell organization and ECM distribution in the specimens. PLGA scaffolds with larger pore sizes exhibited good quality NP tissue when compared to the control group. This may cause an insufficient amount of ECM to be secreted in the smaller pores, so that the cells cannot stabilize and maintain their phenotype. Therefore, it is postulated that pores may be too small in size to be able to provide enough space for the cells to function normally and secrete enough ECM (Agrawal and Ray, 2001; Babensee et al., 1998; Susmita et al., 2003).

In order to further study the effects of porous PLGA on ECM formation by NP cells, we examined the level of newly synthesized GAGs and collagen after 4 and 6 weeks of implantation (Lee et al., 2012). As reported in Figure 9, the size of the pores in the scaffolds had a great influence on the proliferation of cells. The 250–355 and 355–425 µm pore sizes were shown to be large enough for the cells to perform their normal functions, so the cells proliferated and secreted more ECM. The larger pore sizes (380–405 µm) allow efficient transport of the nutrients or oxygen associated with cell growth (Murphy et al., 2000). In the scaffolds with smaller pores, the cells continue to increase in number and become too crowded during the middle and late stages of differentiation, therefore the ECM secretion decreases.

In fact, pore diameter is not the only important parameter in an experimental setting; other important parameters include the animal species, block size, implant chemistry, implant topography, resorption and interconnectivity. It was hypothesized that, in absorbable materials, pore density and interconnection density are more important than pore size, contrary to unabsorbable materials, in which the sizes and densities are equally important. The mechanical function of the IVD is related to ECM composition, hence the proteoglycan and collagen content of scaffolds were evaluated after mechanical evaluation.

In our study, it was noticeable that the modulus values at 4 weeks are lower than at 6 weeks. It is clear that these differences in properties are related to the ECM contents measured in the current approach, and that PLGA scaffolds have the capacity to assemble mechanically competent ECM. PLGA scaffolds with different pore sizes demonstrated accumulation of proteoglycan and collagen in both small and large pore sized implants. The compressive moduli of PLGA scaffolds with different pore sizes correlated most highly with GAGs contents, particularly in the small pore sized groups (90–180 and 180–250 µm). The collagen content appeared to contribute to the compressive modulus to a much smaller extent. These results suggest that proteoglycans contribute to the integrity of cartilaginous NP tissue in compression, while collagen acts primarily to resist tensile or shear deformation (Bolland et al., 2008). In conclusion, we have demonstrated that as the pore sizes of these scaffolds increase, cell proliferation and ECM production will accelerate, whereas the smaller pore sized scaffolds have stronger mechanical properties.

5 Conclusions

PLGA scaffolds were fabricated by solvent casting/salt-leaching with pore size ranges of 90–180, 180–250, 250–355 and 355–425 µm. The mean pore size of the scaffolds was ranged from 169 to 410 µm. SEM images showed that all four porous PLGA scaffolds had apparently uniform pore morphology with the pores square in shape and a well-controlled 3D interconnected network. As the pore size increased, the rate of cell growth and the amount ECM secretion increased for GAGs and collagen. The scaffolds with larger pore sizes showed a dedifferentiated form, but the scaffolds with smaller pore sizes had stronger mechanical properties than those with larger pore sizes. NP cells prefer the group of scaffolds with the pore sizes of up to 250 µm for better proliferation and ECM production. This research suggests that a PLGA scaffold with pores of 180–250 µm may serve as a suitable carrier for NP cells in the treatment of NP degeneration, such as herniated lumbar disc.

In this study, we found that the porous biopolymer scaffold had the mechanical properties to control pore size of scaffold in an in vivo environment.

Conflict of interest

The authors declare no conflicts of interest.

Acknowledgements

This research was supported by the Biotechnology and Medical Technology Development Programme of the National Research Foundation (NRF), funded by the Korean Government (MEST; Grant No. 2012M3A9C6050204) and by the Musculoskeletal Bioorgan Centre of the Korean Ministry of Health and Welfare (Grant No. A040003).

Ancillary