Sage Journals: Discover world-class research

Abstract

The study focuses on a comparison of the electrospinning of various polylactide and polycaprolactone (PLCL) copolymers and poly-L-lactide (PLLA) and polycaprolactone (PCL) homopolymers. The chemical characterisation, electrospinnability, fibrous morphology, degradation rate and interactions with fibroblasts were assessed with respect to copolymers and homopolymers with both lower (around 50,000) and higher (around 95,000) molecular weights. The research investigated commercially available as well as synthesised copolymers. The results revealed that the electrospinnability of polymeric solutions depends on both the molecular weight and the PLA/PCL ratio in the final copolymer. It was determined that PLCL copolymers with a higher content of PCL (≥20%) were not spinnable via the electrospinning process. With the exception of PCL, the resulting fibrous materials were found to be homogeneous and with fibre diameters of slightly more than 1 µm with respect to both the tested molecular weights. The degradation rate was tested under simulation conditions via the utilisation of the lipase and Proteinase K enzymes. The degree of degradation was found to depend on the molecular weight, the crystallinity of the polymer and the specificity of the enzyme applied. While lipase was responsible for the degradation of the PCL polymer, it exerted a minor impact on the PLLA and the copolymers. Proteinase K degraded all the tested polymers with a higher specificity towards PLLA and the PLCL copolymers. All the tested polymers were affected by the surface erosion degradation process via fibrous morphology changes and mass loss with no accompanying shift in the molar mass. The electrospun PLLA materials supported both fibroblast adhesion and proliferation. All the tested materials were determined to be cytocompatible with 3T3 mouse fibroblasts.

Keywords

Polyesters electrospinning copolymers biodegradability cytocompatibility

Introduction

The electrostatic spinning (electrospinning) method is employed for the production of sub-micron fibres which have a high application potential in the fields of tissue engineering and regenerative medicine [1 –3]. Electrospun fibrous mats are capable of mimicking the extracellular matrix due to their similarity to native tissue [4 –6]. The main principle behind tissue engineering consists of the design of scaffolds with the appropriate material properties that enable the regeneration of the target tissue. The scaffold material must be able to support the cells that will penetrate into the scaffold and eventually replace it with their own extracellular matrix before the scaffold itself completely degrades [7,8]. Such applications are particularly demanding in terms of both cell adhesion and material properties. The scaffold must exhibit the ideal level of wettability (moderately hydrophilic) [9] and the appropriate mechanical properties, must be able to degrade over the desired specific time period and be inert with respect to interactions with antibodies [5]. The use of natural polymers is limited by the batch-to-batch variation in molecular mass and molecular mass distribution, i.e. two of the principle parameters influencing the electrospinning process [10]. The molecular mass of the polymer influences the diameter of the fibre which, in turn, affects the mechanical properties, degradation profile and cell adhesion of the fabricated layer. This study focused, therefore, on the research of synthetic biocompatible, biodegradable polymers, namely a subset of aliphatic polyesters [11].

Synthetic aliphatic polyesters are widely used in a range of medical applications [12,13], with the most commonly used polyesters consisting of poly(lactic acid) (PLA), poly(glycolic acid) (PGA) and a copolymer thereof poly(lactic acid-co-glycolic acid) (PLGA), and polycaprolactone (PCL). Notably, both PGA and the L form of PLA (PLLA) degrade too rapidly in the body, decreasing the local physiological pH due to a rapid increase in the local concentration of degradation products, which may lead to inflammation [14]. Conversely, the degradation of PCL depends on the enzymatic activity of the point of emplacement in the body and may last up to two years [15]. Aliphatic polyesters constitute an important group of polymers with respect to use in tissue engineering applications due to their cytocompatibility and degradability under physiological conditions. The direct use of such homopolymers, while straightforward, suffers from the inherent drawback that the approach does not lend itself to the precise control of degradability, mechanical properties, wettability, morphology. On the other hand, many of these characteristics can be optimised by means of the judicious use of copolymer constructs; for example, the rate of physiological degradation can be predetermined and controlled according to the molar ratio (and sequence) of the monomeric building blocks thereof. Significant interest has, therefore, been devoted over recent years to the use of L-lactide and ɛ-caprolactone copolymers: poly(lactic acid-co-caprolactone) (PLCL) [16,17]. However, no systematic study comparing different PLCL copolymers has yet been conducted, in contrast to that of copolymers of the afore-mentioned PLGA which have been subjected to extensive research [18 –20]. The previous research of the use of PLCL copolymers with respect to various tissue engineering applications considered selected ratios of PCL and PLA. Indeed, our previous study compared the impact of ethylene oxide sterilisation on an electrospun PLCL copolymer with a composition of 70/30 LA/CL to that of electrospun polycaprolactone with two different molecular weights with concern to morphological, mechanical and degradation changes and interactions with fibroblasts [21]. In addition, the construction of vascular grafts fabricated from PLCL has been researched [22,23], scaffolds for use in skin tissue engineering utilising a PLCL copolymer have been developed [24], materials for use in nerve tissue engineering based on electrospun PLCL have been characterised [25] and the transplantation of mesenchymal stem cells using a construct made of silk fibroin, PLCL and polyethylene oxide enhanced with growth factors has been researched [26].

Our study aims to provide both new insight into the use of poly(lactic acid)/polycaprolactone-based copolymers with various molecular masses and monomer sequences and to investigate the effect of molecular composition and compare them with other widely-used polyesters. The study investigated the following parameters: spinnability using the needleless electrostatic spinning procedure, the biodegradation of fibrous layers, crystallinity, morphology and in vitro biological behaviour.

Materials and methods

The study considered four polyester copolymers (three with a low molecular mass, approx. 50 kg/mol, and one with a higher molecular mass, approx. 95 kg/mol) and their corresponding homopolymers. An overview of the various polymers, their abbreviations as used in the study and their suppliers is provided in Table 1. PLCL(49) with Mn 48,900 g/mol was synthetized via ring opening bulk polymerization using stannous octoate as catalyst at 120℃ for 72 h.

Table 1.

Polymers used in the study used for electrospinning (non-spinnable polymers).

Polymer type	Abbreviation used in the study	Supplier	Indication
Polycaprolactone	PCL(36)_16 PCL(36)_22	Merck KGaA, Germany	704105
Poly-L-lactide	PLLA(46)	Merck KGaA, Germany	94823
Poly(L-lactide-co-caprolactone)	PLCL(47)	Polyscitech	AP047
Poly(L-lactide-co-caprolactone)	PLCL(48)	Polyscitech	AP074
Poly(L-lactide-co-caprolactone)	PLCL(49)	CMU Bioplasorb	Lab synthesis
Polycaprolactone	PCL(95)	Merck KGaA, Germany	440744
Poly-L-lactide	PLLA(80)	Polyscitech	AP123
Poly(L-lactide-co-caprolactone)	PLCL(95)	Corbion	Purac PLC 7015

Note: The number in brackets following the polymer abbreviations indicates the approximate relative number-average molecular mass (M_n) in g/mol as measured by means of GPC. In the case of PCL(36), the number following the lower dash represents the concentration of the polymeric solution used for electrospinning purposes.

Characterisation of selected polymers

Prior to the commencement of the electrospinning process, the homopolymers (PLLA and PCL) and the PLCL copolymers listed in Table 1 were chemically characterised with respect to molecular weight and distribution, the ratio of lactide (LA) and caprolactone (CL) units and crystallinity.

Gel permeation chromatography

Gel permeation chromatography (GPC), which was applied for the measurement of approximate relative molecular mass, was performed using an Agilent 1100 Series HPLC Value System with a refractive index (RI) detector. The GPC apparatus was composed of two polyester (PL) gel mixed-B columns of 10 µm 300 x 7.5 mm and one mixed-C column of 5 µm 300 x 7.5 mm operated at 40℃. The Styragel columns were calibrated with Easi Vial Polystyrene PS-M Standards (Agilent). Tetrahydrofuran (>99.5%, Fisher Chemical) containing 2% (v/v) triethylamine (≥99%, Sigma Aldrich) was used as the eluent at a flow rate of 1.0 ml/min.

¹H NMR spectroscopy

The molar ratio of LA:CL was measured by means of ¹H nuclear magnetic resonance (NMR) spectroscopy. The spectra were recorded on a Brüker Avance spectrometer at 300 MHz using 5 mm diameter sample tubes. All the spectra were obtained at room temperature using an 8 mg/ml sample in deuterated chloroform (CDCl₃, 99.8%, Cambridge Isotope Laboratories Inc., USA) with 0.05 v/v% tetramethylsilane. The ¹H NMR spectra showed signals at δ ∼ 1.35–1.67 ppm (CH₂CH₂CH₂, PCL), 1.57 ppm (CH₃, PLA), 4.00 ppm (CH₂O, PCL) and 5.2 ppm (CH, PLA). The LA:CL ratio of the PLCL copolymers was calculated from the integrals of the peaks at 4.0 ppm (CL unit) and at 5.2 ppm (LA unit). The multiplet at approximately 2.30 ppm (CH₂COOCH₂) is related to CL-CL junctions, whereas the peak at 2.50 ppm (CH₂COOCH) indicates CL-LA junctions. These signals were employed in order to obtain the monomer sequence distribution.

Crystallinity measurement

The degree of crystallinity of the fibrous polymer layers was measured by means of X-ray diffraction (XRD) using Diffrac.Eva software. Diffraction measurements were performed in a vertical powdered θ-θ D8 Discover diffractometer equipped with a 1D detector placed behind the nickelβ filter for the absorbance of diffracted CuKβ radiation. The procedure involved the use of a Bragg–Brentan goniometer setup and the samples were placed on a non-diffracted layer fabricated from silicon single crystal. The estimation of the relative error of the mass fraction of the crystalline and amorphous phases was 10 wt% with respect to the method employed assuming that the elemental composition of the crystalline and amorphous phases was the same. It should be noted that only the spinnable polymers were analysed via XRD.

Electrospinnability of the polymers and subsequent characterisation

Further analysis was based on the study of the electrospinning process using characterised polymers and the testing of the degradation process thereof and interactions with a fibroblast cell line.

Electrospinning process and the morphology assessment of the resulting fibres

All the polymers were tested for their electrospinnability using a Nanospider NS 1WS500U machine produced by Elmarco a.s. (Czech Republic). All the polymers were dissolved in the same chloroform/ethanol/acetic acid (CHETAA, Penta s.r.o., Czech Republic) solvent mixture at a weight ratio of 8/1/1. The final polymeric concentration used for obtaining the continuous fibrous layer is shown in the final column of Table 1. Polycaprolactone with a lower molecular weight was electrospun from two different concentrations of 16 and 22 wt% leading to varying layer morphologies, while the other polymers were electrospun from one optimised concentration. A positive voltage was applied to the wire (spinning electrode) and a negative voltage to the collector; the potential difference was 40 ± 5 kV. The distance between the spinning electrode and the collector was maintained at between 160 and 180 mm. The temperature during the experiments was 22 ± 5℃ with a relative humidity of 40 ± 5%. The fibres were analysed by means of scanning electron microscopy (SEM, Tescan VEGA3 SB, Czech Republic) and the fibre diameters were measured (100 fibres per sample) using NIS elements software (LIM s.r.o., Czech Republic).

Degradation studies

The degradation profiles of all the fabricated fibrous layers were tested according to the methodology described in our previous article [21]. Briefly, 50 ± 0.5 mg samples were extracted from the fibrous layers fabricated as described above. Enzymatic degradation was performed employing individual enzymes [2 U/ml of Proteinase K (velek P, Thermo Scientific cat. no. 17916, 30 U/mg) and 5 U/ml of lipase (Sigma Aldrich cat. no. 62309, 30 U/mg)] in 5 ml phosphate-buffered saline (PBS) at a pH of 7.4. The PBS contained 0.02 v/v% sodium azide as the antibacterial reagent. The polymer samples were submerged in the prepared solutions and maintained at a constant temperature of 37℃. The buffer/enzyme solution was replaced daily so as to maintain the enzymatic activity at the desired level throughout the entire experiment. The mass loss was measured every 24 h over five days of incubation. The samples were removed from the solution, carefully rinsed in distilled water and dried at room temperature. The dry samples were weighed in order to observe the losses in mass, analysed via SEM so as to detect the effect of degradation on surface changes and further analysed by means of GPC in order to monitor the molecular mass. Samples placed in PBS containing no enzymes were used as the control reference.

In vitro testing

Prior to cell seeding, the fibrous layers were cut into round patches (6 mm in diameter) and sterilised by means of immersion in 70% ethanol for 30 min followed by double rinsing in PBS (pH 7.4). Mouse 3T3 fibroblasts (ATCC, USA) were cultivated in Dulbecco’s Modified Eagle Medium (DMEM, Lonza Biotec s.r.o., Czech Republic) supplemented with 10% foetal bovine serum (Lonza Biotec s.r.o., Czech Republic), 1% glutamine (Biosera, Czech Republic) and 1% penicillin/streptomycin/amphotericin B (Lonza Biotec s.r.o., Czech Republic). The cells were placed in a humidified incubator under an atmosphere of 5% CO₂ at 37℃. Once the cells had become confluent, they were suspended using trypsin-EDTA (Lonza Biotec s.r.o., Czech Republic). The fibroblasts (passage 10) were seeded on the scaffolds which were placed in a 96-well plate at a density of 5 × 10³ per well. The medium was changed twice per week during the experiment (total of 11 days). Materials incubated in the complete medium without the presence of cells served as negative controls for the MTT test (n = 2 per each testing day).

The viability of the cells seeded on the scaffolds was analysed via the MTT test after 3, 7 and 11 days of the culture period. MTT solution (50 µl) was added to 150 µl of complete DMEM and the samples were incubated at 37℃ for 4 h. The violet crystals of formazan thus formed were solubilised in acidic isopropanol and the resulting solution was measured spectrophotometrically (λ_sample 570 nm, λ_reference 650 nm). Eight samples per tested material were incubated on each testing day with the MTT solution and the final absorbance was calculated as the difference between the absorbance of the sample (570 nm) and the reference (650 nm). The absorbance of the negative controls was subtracted from the measured absorbance values of the tested samples in order to compare the viability of the cells cultured on the fibrous materials. The data were expressed as the mean value of measured absorbance ± standard deviation and plotted in the form of a graph.

The fibroblasts were fixed onto the surface of the tested materials after 3, 7 and 11 days in 2.5% glutaraldehyde for 30 min followed by double rinsing in PBS and staining with propidium iodide (Merck KGaA, Germany, dilution 1:1000). Following double rinsing in PBS, images of the materials were captured using a Nikon ECLIPSE Ti-E/B inverted fluorescence microscope. Ten images per each material were subsequently quantified by means of MATLAB software as described in [21], Supplementary material II. The results thus enabled the comparison of the cell density of the tested materials.

After 11 days of cultivation, the samples were rinsed twice in PBS and fixed in 2.5% glutaraldehyde for 30 min and, subsequently, 0.1% Triton X-100 and 0.1% bovine serum albumin solution in PBS was employed as a blocking buffer for 10 min. The samples were stained using phalloidin-FITC (Merck KGaA, Germany, dilution 1:1000, 1 h, room temperature) which binds to the actin filaments of the cells. DAPI (Merck KGaA, Germany, dilution 1:1000, 5 min, room temperature) was then used for the counterstaining of the cell nuclei. The stained cells were observed by means of inverted fluorescence microscopy using a Nikon ECLIPSE Ti-E/B.

Results and discussion

The paper provides a description of the extensive study of a range of PLLA and PCL homopolymers and copolymers aimed at the exploration of their potential for use in biomedical applications. More specifically, the study focuses on the chemical characterisation of polymers, their electrospinnability, degradation profiles and biological in vitro behaviour.

Characterisation of the input polymers

A study of molecular mass was conducted in order to enable the detailed comparison of the polymers considered. While the suppliers provided information on the polymers including molecular weight (number averaged M_n or weight averaged M_w) and viscosity, the conditions under which the information is obtained tend to vary considerably. Our method, however, applied the same conditions for all the polymers considered in the study thus allowing for their direct comparison. Both the molecular mass (M_n and M_w) and the polydispersity index (Ð) were assessed and the results are shown in Table 2. The polymers were divided into two subgroups according to their molecular weights, i.e. lower molecular weights of around 50,000 g/mol and higher molecular weights of approximately 95,000 g/mol.

Table 2.

Molecular mass data for all the polyesters studied.

Sample name^a	M_n according to producer (g/mol)	Refractive index detector
Sample name^a	M_n according to producer (g/mol)	M_w (g/mol)	M_n (g/mol)	Ð
PLLA(46)	45,000–55,000	100,000	46,100	2.19
PCL(36)	45,000	52,500	35,700	1.46
PLCL(47)	45,000–65,000	77,000	47,400	1.62
PLCL(48)	45,000–50,000	90,800	47,900	1.90
PLCL(49)	Lab synthesis	93,500	48,900	1.91
PLLA(80)	75,000–85,000	^b	^b	^b
PCL(95)	80,000	149,000	94,800	1.57
PLCL(95)	1.2–1.8 dl/g^c	193,000	94,900	2.03

Note: M_n is the number averaged molar mass, M_w is the weight averaged molar mass and Ð is the dispersity (M_w/M_n); the homopolymer entries are shaded in grey for the sake of clarity. PLLA: poly(L-lactic acid); PCL: polycaprolactone; PLCL: poly(lactic acid-co-caprolactone); GPC: gel permeation chromatography.

^aThe number in brackets represents the approximate relative molecular mass (M_n) measured via GPC.

^bPLLA(80) was not soluble in tetrahydrofuran; therefore, no GPC analysis was conducted.

^cInherent viscosity (CHCl₃, 25℃, 0.1 g/dl).

In addition to molecular weight, the molar ration of lactide and the caprolactone unit as well as the crystallinity of the polymers make up further important parameters with concern to the electrospinning process and the further use of these materials for the fabrication of tissue engineering scaffolds; these parameters are shown in Table 3. Those polymers with 100% of CL exhibit the highest degree of crystallinity. It is possible to predict a decrease in crystallinity in accordance with an increase in the LA content in the copolymer. While this relationship is valid with respect to those polymers with lower molecular weights, the PLLA(80) and PLCL(95) polymers exhibited an inverse tendency; however, the tendency was within the range of relative error, i.e. 10%. While selected detailed studies of the correlation between crystallinity and the electrospinning process can be found in [27,28], the effect of the LA/CL ratio on crystallinity and that of the monomer content on the electrospinning process have not been studied to date. Moreover, statistical copolymers only were used in our experiments. The effect of molecular weight on fibre morphology is already well known and can be found in the literature [29,30].

Table 3.

Monomer ratio and monomer sequence distribution, measured via ¹H NMR spectroscopy.

Sample name	Molar ratio of LA:LC according to supplier (%)	Molar ratio of LA:LC according to ¹H NMR (%)	Monomer sequence distribution (%)	Crystallinity (wt%)	Electrospinning concentration (%)
Sample name	LA/CL	LA/CL	CLLA/CLCL	Crystallinity (wt%)	Electrospinning concentration (%)
PCL(36)	0/100	0/100	–	52	16;22
PLLA(46)	100/0	100/0	–	10	10
PLCL(47)	30/70	60/40	40/60	–	–
PLCL(48)	50/50	39/61	25/75	–	–
PLCL(49)	70/30	83/17	41/59	34	12
PCL(95)	0/100	0/100	–	49	10
PLLA(80)	100/0	100/0	–	28	6
PLCL(95)	70/30	82/18	67–73/33–27	8	10

Note: The degree of crystallinity of the fibrous layers was measured via XRD. PLLA: poly(L-lactic acid); PCL: polycaprolactone; PLCL: poly(lactic acid-co-caprolactone); LA: lactide; CL: caprolactone.

Electrospinning process

Each polymer was dissolved in a solvent mixture of chloroform/ethanol/acetic acid (8/1/1 by weight) and optimised in terms of concentration for the needleless electrospinning process (see Table 3). All polymeric concentrations were optimized to obtain uniform fibres. Below these concentrations, some beads were obtained and higher concentrations increase fibrous diameters [31]. Notably, it was not possible to electrospin all the polymers in the selected series from this solvent system. Our findings revealed that the PLCL copolymers with a relatively low CL content (<20%) were spinnable (with respect to both lower and higher molecular masses), whereas the PLCL copolymers with higher CL contents (≥40%) could not be spun under the test conditions, i.e. relative humidity of 10%–60% and temperature of 20℃–30℃. The PLCL(47) and PLCL(48) polymer solutions disintegrated into droplets via the Rayleigh instability effect [32] after covering the wire of the Nanospider without the occurrence of the electrospinning process, which can probably be attributed to the low interaction parameter between the polymeric segment and the selected solvent system segment according to polymer physics theory [33]. CL content appears to constitute the dominant factor with concern to the electrospinning process. One of the explanations for why copolymers with amounts of CL higher than 20% cannot be spun consists of the greater interaction energy between the LA-CL monomers than that between the polymer and the solvent. The polymers therefore create individual globules which resist the formation of fibres via the electrospinning process. Only statistical copolymers had been used in all experiments and it is hard to predict any behaviour and compare it with block copolymers. Some studies for block copolymers can be found for example in [34]. However, we did not find any studies for electrospinning of statistical copolymers.

All the other polymers were successfully electrospun. As mentioned above, two different concentrations (16 and 22 wt%) were used to spin the PCL(36) so as to obtain nanofibrous and microfibrous structures, respectively. It has been reported previously that a higher concentration of the polymer in the solution leads to a larger fibre diameter; more detailed studies of viscosity, surface tension, voltage, etc. can be found for example in [35] or [31].

To summarise, seven fibrous mats were fabricated for the purpose of the study; the fibre diameter distributions (measured from at least 100 fibres per sample) are plotted in Figure 1 and SEM images of the produced layers are provided in Figures 2 and 3. Whereas PCL creates fibres with diameters in the submicron range, PLLA and PLCL were found to create primarily microfibres. The morphology of PLLA(80) and PLCL(49) is similar and comparable with microfibres produced from PCL(36) (22 wt%). Surprisingly, it appears that the molecular weight of the initial polymer does not influence the diameters of the fibres; thus, it is possible to predict a certain correlation between the crystallinity of the fibrous layer and its morphology, which is probably due to the better alignment of the macromolecules in the fibres [36].

Figure 1.

Fibre diameter of the electrospun materials. (a) Box plots showing the minimal and maximal values, upper and lower quartile and median of the measured data. (b) Histograms of fibre diameters with its normal distributions. PLCL: polylactide and polycaprolactone; PLLA: poly-L-lactide; PCL: polycaprolactone; SD: standard deviation.

Figure 2.

SEM analysis, where the negative control consists of a layer produced following the electrospinning process immersed in PBS and the lipase and Proteinase K images show the various materials following five days of degradation induced by the selected enzyme; magnification 5000×. PLCL: polylactide and polycaprolactone; PLLA: poly-L-lactide; PCL: polycaprolactone.

Figure 3.

Detailed view of the degraded layers with degraded lumps and their SEM images prior to degradation, where (a) represents NC of PCL(36)_16 and (b) represents PCL(36)_16 following two days of lipase treatment respectively, and (c) and (d) represent NC of PLCL(49) and its proteinase catalysed degradation after three days. SEM: scanning electron microscopy.

Figure 4.

Weight loss caused by the degradation of the fibrous layers as catalysed by means of lipase (above) and Proteinase (below). The PLCL95 experiment was terminated after four days due to the total degradation of the material prior to the fifth day. PLCL: polylactide and polycaprolactone; PLLA: poly-L-lactide; PCL: polycaprolactone.

The resulting fibres were found to have smooth surfaces (see Figures 2 and 3). The fibre diameter was evaluated with respect to each of the electrospun mats as shown in Figure 1. Most of the materials produced exhibited a mean fibre diameter of slightly above 1 µm. The only polymer creating fibres which can be characterised as nanofibers is polycaprolactone. The diameter of fibres depends on its molecular weight and concentration (and other electrospinning conditions, such as the solvent from which the polymer was spun). This is mostly visible for PCL(36)_16. We attribute this to its high crystallinity which is approximately 50% (see Table 3). This polymer creates both nano and micro fibres depending on the arrangement of the macromolecules affected by electrospinning.

Degradation studies

All the tested materials (PLLA, PLCL and PCL) consisted of semi-crystalline polyesters with a high degree of hydrophobicity; thus, the penetration of water molecules into the polymer matrix (the principle of bulk erosion) was significantly suppressed. Consequently, the autocatalytic degradation of the PLLA and PLCL was expected to be negligible [37]. The degradation rate is influenced by a number of factors beyond those inherent to the chemical structure of the polymer, including the molecular mass of the polymer, the surface area/morphology, wettability, melting point and degree of crystallinity [38]. The degradation profiles and characteristics were studied by means of measuring the mass decrease (gravimetry), molecular mass changes (GPC) and morphological changes (SEM) over time. Enzymes (lipase from Pseudomonas cepacia and Proteinase K from Tritirachium album [39]) were used to accelerate the degradation rate. Figure 4 shows the degradation profiles of the fibrous mat series. The lipase degradation data shows that whereas the PCL samples degraded, the PLLA and PLCL mats remained stable for the duration of the experiment. PCL(36)_16 was seen to degrade slightly faster than PCL(36)_22 owing to the smaller diameter (and therefore larger surface area) of the former. Interestingly, while PCL(36)_16 and PCL(36)_22 completely disintegrated within five days, the PCL(95) lost only around 50% of its initial mass during this period; moreover, the PCL (95) has smaller diameter fibres than does PCL(36)_22. It seems therefore that the slower degradation rate can be attributed to the higher molecular mass of the polymer [37]. The hydrolytic degradation of the PLLA and PLCL copolymers was observed with the use of Proteinase K, under which conditions, the PCL samples lost only around 20% of their initial mass, whereas the PLA samples lost 40%–50% and the copolymer samples over 70% of their initial mass.

The morphology of the fibrous materials changed significantly during the degradation process. In general, it was possible to observe fibre decomposition, re-crystallisation and fibre breakage (depending on the degree of crystallinity). The fibrous morphology of the negative control samples (the materials subjected to non-enzymatic treatment) was found to be smooth with no craters or pits. SEM images from the final day of the degradation process confirmed the mass loss trend of all the fibrous materials. In addition, the surface of the PCL(36)_16, PCL(36)_22 and PCL(95) treated with lipase became coarser during the degradation process.

The greatest morphological changes occurred with respect to the PCL(36)_16 and PCL(36)_22 mats, concerning which the fibres disintegrated to form lumps (Figure 3). The PCL(36)_16 and PCL(36)_22 samples exhibited very similar behaviour during the degradation process: those fibres with small diameters were hydrolysed during the first day whereupon re-crystallisation and decomposition were observed followed immediately by fibre breakage. However, the PCL(95) and PLA samples [PLLA(46), PLLA(80)], with a large proportion of submicron fibres, exhibited different behaviour, i.e. re-crystallisation was observed during the first two days followed by a more significant rate of fibre decomposition.

While all the tested materials lost mass following enzyme treatment, the molar mass, which was measured each day during the running of the degradation experiments, exhibited no significant differences (Figure 5). These results confirm therefore that the tested fibrous layers were degraded via a surface erosion process, during which the size of the matrix decreased stage by stage with the bulk part remaining unchanged. The water-soluble degradation products, which are dissolvable in an aqueous environment, remained as a result of the hydrolysis of the surface molecules [38,39].

Figure 5.

GPC analysis of the materials during degradation: PCL degradation catalysed by means of lipase and PLCL and PLLA degradation catalysed by means of Proteinase K. PLLA: poly-L-lactide; PCL: polycaprolactone; PLA: poly(lactic acid); PLCL: polylactide and polycaprolactone.

Molar mass changes

While each of the tested materials lost a certain amount of mass via the given enzyme, the molar mass, which was examined every day for the duration of the degradation experiments, exhibited no significant differences. The tested fibrous layers were degraded primarily via the surface erosion process, the reason being that extracellular enzymes such as lipase and proteinase are too large to penetrate to an intense degree into the polymer macromolecules; thus, the degradation process takes place principally on the polymeric surface.

The semi-crystalline and less hydrophilic polymers exhibited slower water penetration; thus, the mass loss affected the surface of the matrix (surface erosion). While the matrix degrades stage by stage, the bulk part remains unchanged. The basics of hydrolysis consists of bond cleavage into smaller segments, which leads to a reduction in the molar mass and an increase in the polydispersity of the polymer. The production of water-soluble products that can be dissolved in an aqueous environment represents a further step in the hydrolysis degradation process. The results of the GPC analysis for selected polymers are presented in Figure 5 so as to provide proof of the preservation of the molecular weight distribution.

In vitro testing

The electrospun materials were seeded with 3T3 fibroblasts and the interactions of the cells with the fibrous materials were assessed via the measurement of metabolic activity and cell quantification after 3, 7 and 11 days of culturing. At the end of experiment, the cells on the scaffold surface were visualised by means of the fluorescence staining of the cell cytoplasm and nuclei. The aim of submitted article was characterization of wide range of polyester based electrospun material. Therefore, fibroblast cell line was selected due to its versatility and ease of handling. Afterwards, selected materials for specific applications are going to be tested with specific cell lines or primary cells.

Cell viability

Cell proliferation was measured using colorimetric MTT assay that reflects the viability of the cells on the surface of fibrous materials. Once the cells adhere to the materials, the proliferation rate is accelerated which results in the increased absorbance of the seeded samples during the testing period. The greatest viability increase during culturing was observed on those materials based on PLLA at both the tested molecular weights (PLLA(46), PLLA(80)) as can be seen in the graph in Figure 6. With respect to electrospun PCL(36)_16, PCL(36)_22, PLCL(49), PCL(95) and PLCL(95), a slight decrease in cell viability was observed after a week of culturing followed by a slight increase after 11 days. These results suggest weaker cell adhesion as compared to the PLLA samples which can be attributed to surface wettability.

Figure 6.

Metabolic activity of the fibroblasts seeded on the electrospun materials after 3, 7 and 11 days measured by means of colorimetric MTT assay. Statistical analysis of the data was performed using two-way ANOVA with Boferonni multiple comparison. PLCL: polylactide and polycaprolactone; PLLA: poly-L-lactide; PCL: polycaprolactone.

Cells were cultured on tissue culture plastic as a control of cell growth. The proliferation rate of these control cells was much higher compared to tested materials reaching the absorbance after MTT assay higher than 1 (1.05 ± 0.10 after 3 days of culturing, 1.67 ± 0.05 after 7 days and 2.14 ± 0.05 after 11 days). Since these values are much higher than the absorbance of tested materials, the controls were excluded from the graph. However, cytotoxic effects were not been observed in any tested material since all of the surfaces were at least partially covered with cells after 11 days of culturing, as shown in Figure 7.

Figure 7.

Fluorescence microscopy images of 3T3 mouse fibroblasts stained with phalloidin-FITC (green) and DAPI (blue) after 11 days of culturing on the tested materials. PLLA: poly-L-lactide; PCL: polycaprolactone; PLA: poly(lactic acid); PLCL: polylactide and polycaprolactone.

Figure 8.

Cell quantities on the surface of the tested electrospun materials after 3, 7 and 11 days of culturing. PLLA: poly-L-lactide; PCL: polycaprolactone; PLA: poly(lactic acid); PLCL: polylactide and polycaprolactone.

The cells on the surface of the scaffold were quantified and the number of cells per unit area was plotted in the graph in Figure 8. The number of cells increased over time with respect to the PCL(95), PLLA(80) and PLCL(95) materials. An increase in cell quantity between days 3 and 7 was observed for the PCL(36)_16, PCL(36)_22, PLLA(46) and PLCL(49) materials which was inverse to the viability of the cells measured by means of MTT. Proliferation at different rates was observed for all the tested materials. At the end of the experiment (11 days following cell seeding), the highest concentration of cells was determined with respect to the electrospun PLLA(46) (861 ± 157 cells/1 mm²) and PLLA(80) (1028 ± 146 cells/1 mm²) and the lowest concentration of cells was observed on electrospun PLCL(49) with 546 ± 163 cells/1 mm², PCL(36)_16 with 568 ± 189 cells/1 mm², PCL(36)_22 with 705 ± 116 cells/1 mm² and PCL(95) with 700 ± 86 cells/1 mm². The PLCL(95) copolymer also supported cell proliferation exhibiting a final number of cells at the end of the experiment of 828 ± 118 cells/1 mm².

Fluorescence microscopic analysis was performed after 11 days of fibroblast culturing on the tested materials using the double staining procedure (see Figure 7). A confluent layer of cells was found on the electrospun PLLA(46) and the lowest degree of fibroblast surface coverage was observed on the electrospun PCL(36)_16 and PCL(95), thus correlating with the above-mentioned metabolic activity and cell quantification.

The most favourable results in terms of the highest proliferation rate and the degree of spreading on the surface of the fibrous scaffolds was observed with respect to the electrospun layers fabricated from PLLA, followed by PLCL. The metabolic MTT test revealed that the fibrous PCL layers exhibited the lowest proliferation rate and cell density. However, there is no significant difference between all tested polymers and all can be successfully used for tissue engineering applications. The main goal of this study was to determine the spinnability and degradation of the studied polymers. There are different needs depending on the specific use in tissue engineering. By combination of aforementioned materials, we can adjust properties of prepared scaffolds for specified applications and control the degradation and/or release of integrated drugs driven by diffusion and the degradation process itself.

Conclusion

Despite gel permeation chromatography revealing a number of differences between the molecular weight distributions provided by the producers and those as measured using GPC, two main distributions were finally selected and compared for all the polyesters. Three LA/CL monomer alternations for PLCL were tested for their spinnability via electrospinning. It was determined that if the amount of CL was higher than 20%, the copolymer could not be spun in the solvent used in our experiments. The use of statistical copolymers in the experiments led to the inhomogeneous distribution of the polar and nonpolar groups. The inability to electrospin these copolymers was probably due to higher interaction between the monomer-monomer groups than between the monomer-solvent groups according to the Flory–Huggins solution theory. In such cases, the polymers created individual globules rather than spatially distributed chains, thus preventing electrospinning. Semi-crystalline polymers create thinner fibres than their amorphous counterparts, which is probably due to the better alignment of the macromolecules in the fibre. The biodegradation studies proved that electrospun layers degrade principally as a result of the surface erosion mechanism and that their morphology influences the degradation process only to a minor degree. A more important parameter influencing fibrous degradation consists of molecular mass; however, no significant de-polymerisation was observed in the residual material following the conclusion of the degradation process. Cell seeding on the surface of the tested materials revealed that the fibroblasts attached to all the tested fibrous mats, thus proving the cytocompatibility of all the tested materials.

Footnotes

Acknowledgment

The authors would like to express their gratitude for financial support provided from the GAČR 17-02448S improved growth of human skin cells on biomimetic nanofibrous matrices for active wound healing project.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Lanza

Langer

Vacanti

. Principles of tissue engineering, 4th ed. London: Elsevier Science, 2014.

Fisher

Mikos

Bronzino

, et al. Tissue engineering: principles and practices, Boca Raton: CRC Press, Taylor & Francis, 2012.

Andrades

. Regenerative medicine and tissue engineering, London: InTech, 2013.

Wnek

Bowlin

. Encyclopedia of Biomaterials and Biomedical Engineering, 2nd edition. New York, USA: Marcel Dekker, 2008.

Elisseeff

. Scaffolding in Tissue Engineering, Boca Raton, USA: Taylor & Francis, 2006.

Barnes

Sell

Boland

, et al. Nanofiber technology: Designing the next generation of tissue engineering, scaffolds. Adv. Drug Deliv. Rev 2007; 59: 1413–33.

Jain KK. Drug Delivery System, Methods in Molecular Biology, Vol 1141, 2014.

Son

Kim

Yoo

. Therapeutic applications of electrospun nanofibers for drug delivery systems. Arch. Pharm. Res 2014; 37: 69–78.

Bacakova

Filova

Parizek

, et al. Modulation of cell adhesion, proliferation and differentiation on materials designed for body implants. Biotechnol. Adv 2011; 29: 739–67.

10.

Koski

Yim

Shivkumar

. Effect of molecular weight on fibrous PVA produced by electrospinning. Materials Letters 2004; 58: 493–7.

11.

Saleh

HE-DM

. Polyester, London: InTech, 2012.

12.

Thakur

. Handbook of polymers for pharmaceutical technologies, Beverly: Scrivener Publishing, Wiley, 2016.

13.

Andreas

Adam

. Handbook of biodegradable polymers, Weinheim: Wiley-VCH, 2011.

14.

Ceonzo

Gaynor

Shafferm

, et al. Polyglycolic acid-induced inflammation: Role of hydrolysis and resulting complement activation. Tissue Eng 2006; 12: 301–308.

15.

Woodruff

Hutmacher

. The return of a forgotten polymer—polycaprolactone in the 21st century. Prog Polym Sci 2010; 35: 1217–1256.

16.

Kotaki

, et al. Electrospun P(LLA-CL) nanofiber: A biomimetic extracellular matrix for smooth muscle cell and endothelial cell proliferation. Biomaterials 2004; 25: 1883–1890.

17.

Khatri

Jatoi

Ahmed

, et al. Cell adhesion behavior of poly(ɛ-caprolactone)/poly(L-lactic acid) nanofibers scaffold. Mater Lett 2016; 171: 178–181.

18.

Shin

Lee

Cho

, et al. Electrospun PLGA nanofibers scaffolds for articular cartilage reconstruction: Mechanical stability, degradation and cellular responses under mechanical stimulation in vitro. J Biomater Sci Polym Ed 2006; 17: 103–119.

19.

Pan

Ding

. Poly(lactide-co-glycolide) porous scaffolds for tissue engineering anf regenerative medicine. Interface Focus 2012; 2: 366–377.

20.

Gentile

Chiono

Carmagnola

, et al. AN overview of poly(lactic-co-glycolic) acid (PLGA)-based biomaterials for bone tissue engineering. Int J Mol Sci 2014; 15: 1640–1659.

21.

Horakova

Mikes

Saman

, et al. The effect of ethylene oxide sterilization on electrospun vascular grafts made from biodegradable polyesters. Mater Sci Eng C: Appl Biomat 2018; 92: 132–142.

22.

Tara

Kurobe

Rocco

, et al. Well-organized neointima of large-pore poly(L-lactic acid) vascular graft coated with poly(L-lactic-co-ɛ-caprolactone) prevents calcific deposition compared to small-pore electrospun poly(L-lactic acid) graft in a mouse aortic implantation model. Atherosclerosis 2014; 237: 684–691.

23.

Jiang

Wang

, et al. Electrospun poly(L-lactide-co-ɛ-caprolactone)-collagen-chitosan vascular graft in a canine femoral artery model. J Mater Chem B 2015; 28: 5760–5768.

24.

Pan

Liu

Sun

, et al. Preparation and characterization of electrospun PLCL/poloxamer nanofibers and dextran/gelatin hydrogels for skin tissue engineering. PLoS One 2014; 9: e112885–e112885.

25.

Kijenska

Prabhakaran

Swieszkowski

, et al. Interaction of Schwann cells with laminin encapsulated PLCL core-shell nanofibers for nerve tissue engineering. Macromol Nanotechnol 2014; 50: 30–38.

26.

Zhao

Hanif

, et al. Dual delivery of FGF-2/CTGF from silk fibroin/PLCL-PEO coaxial fibers enhances MSC proliferation and fibrogenesis. Sci Rep 2017; 7: 8509–8509.

27.

Ero-Phillips

Jenkins

Stamboulis

. Tailoring crystallinity of electrospun plla fibres by control of electrospinning parameters. Polymers 2012; 4: 1331–1348.

28.

Lim

Tan

EPS

. Effects of crystalline morphology on the tensile properties of electrospun polymer nanofibers. Appl Phys Lett 2008; 92: 141908–141908.

29.

Koski

Yim

Shivkumar

. Effect of molecular weight on fibrous PVA produced by electrospinning. Mater Lett 2004; 58: 493–497.

30.

Hem

Ki-Taek

Hyun-Ju

, et al. Effect of polymer molecular weight on the fiber morphology of electrospun mats. J Colloid Interface Sci 2011; 364: 107–111.

31.

Liu

, et al. Controlling numbers and sizes of beads in electrospun nanofibers. Polym Int 2008; 57: 632–636.

32.

Eggers

. Nonlinear dynamics and breakup of free-surface flows. Rev Mod Phys 1997; 69: 865–929.

33.

Rubinstein

Colby

. Polymer physics, New York: Oxford University Press, 2003.

34.

Ijaz Shah

Yang

, et al. Properties of electrospun nanofibers of multi-block copolymers of [Poly-ɛ-caprolactone-b- poly(tetrahydrofuran-co-ɛ-caprolactone)]m synthesized by janus polymerization. Polymers 2017; 9: 559–559.

35.

Li Z and Wang C. One-dimensional nanostructures. In Springer Briefs in Materials. New York: Springer, 2013.

36.

Lim

Tan

EPS

. Effects of crystalline morphology on the tensile properties of electrospun polymer nanofibers. Appl Phys Lett 2008; 92: 141908–141908.

37.

Dong

Liao

Ngiam

, et al. Degradation behaviors of electrospun resorbable polyester nanofibers. Tissue Eng Part B Rev 2009; 15: 333–351.

38.

Bikiaris

. Nanocomposites of aliphatic polyesters: An overview of the effect of different nanofillers on enzymatic hydrolysis and biodegradation of polyesters. Polym Degrad Stab 2013; 98: 1908–1928.

39.

Gan

Zhong

, et al. Enzymatic degradation of poly(ɛ-caprolactone)/poly(DL-lactide) blends in phosphate buffer solution. Polymer 1999; 40: 2859–2862.

Comparison and characterization of different polyester nano/micro fibres for use in tissue engineering applications

Abstract

Keywords

Introduction

Materials and methods

Characterisation of selected polymers

Gel permeation chromatography

1H NMR spectroscopy

Crystallinity measurement

Electrospinnability of the polymers and subsequent characterisation

Electrospinning process and the morphology assessment of the resulting fibres

Degradation studies

In vitro testing

Results and discussion

Characterisation of the input polymers

Electrospinning process

Degradation studies

Molar mass changes

In vitro testing

Cell viability

Conclusion

Footnotes

Acknowledgment

Declaration of conflicting interests

Funding

References

¹H NMR spectroscopy