Abstract

There is increasing interest in aliphatic polyesters from lactones and lactides because of their biodegradability and biocompatibility. Among these compounds, poly(lactide), and poly(glycolide), poly(-caprolactone) and their copolymers are especially interesting because of their potential applications as biomedical materials. The aim of this study was to examine the properties of membranes of poly(L-co-D,L lactic acid) (PLDLA) with poly(caprolactone triol) (PCL-T) obtained by solvent evaporation. The blends were characterized by differential scanning calorimetry, dynamic mechanical analysis, Fourier transform infrared spectroscopy, and tensile strength tests. Based on the results of in vitro studies, PLDLA/PCL-T blends of 100/0 and 90/10 were implanted in subcutaneous tissue of Wistar rats for 1, 3, 7, 15, and 60 days to evaluate their biocompatibility. Histological analysis indicated that, although PCL-T-containing membranes caused a more prominent inflammatory reaction in the initial time intervals, by 60 days after implantation, the material was surrounded by dense, organized collagen with almost no inflammatory infiltrate.

1. Introduction

The biomaterials are meant to be implanted in the human body as constituents of devices that are designed to perform certain biological functions by substituting or repairing different tissues such as bone, cartilage or ligaments, and tendons and even by guiding bone repair when necessary. For this reason, polymers need to be biocompatible and bioresorbable in order to minimize adverse reactions in the organism [1]. The increasing therapeutic use of polymers as biomaterial implants makes it essential to evaluate the behavior of subcutaneous implants in vivo. This is generally done in rats, which provide an ideal model for analyzing the rate of neovascularization, the formation of collagen types I and III, and the consolidation of the conjunctive capsule around the polymer [2].

Poly(L-lactide) is characterized by a high content of crystalline, high strength particles, and prolonged period of degradation. Poly(Dl lactide) is in contrast characterized by lower strength and quicker rates of degradation. Combining both poly(L-lactide) and poly(D,L lactide) results in a single amorphous copolymer that has a structure that combines the best characteristics of poly(L-lactic acid) and poly(DL-lactic acid), that is, the mechanical properties of the former and the shorter degradation time of the latter; these properties have made PLDLA a compound of great relevance in the controlled release of drugs and the treatment of bone fractures [3]. In addition to these applications, PLDLA serves as substrates for regenerative medicine applications with special focus on skin tissue engineering [4]. Despite these advantages, biomaterials containing poly(L-co-D,L lactic acid) are generally extremely dense, that is, devoid of pores and very rigid, and have a high modulus of elasticity, low elongation, and are hydrophobic; these characteristics limit the use of this polymer in a variety of medical applications, particularly when tissue/implant interactions are an important factor.

Many studies have shown that one way to improve the elongation characteristics, reduce the hydrophobicity of the polymer, and facilitate its interaction with tissue is to incorporate a low molecular weight polymer that can act as a plasticizer in polymeric matrices [5]. Polycaprolactone triol, a low molecular weight aliphatic polyester obtained by ring opening of -caprolactone to yield three OH groups in the monomer molecule of the polymer chain terminal, has been used as a plasticizer.

The aim of this work was to study the thermal, mechanical, morphological, and biological properties of PLDLA membranes containing different concentrations of poly(caprolactone triol) (PCL-T) as a plasticizer and to assess their potential usefulness in vivo.

2. Experimental

2.1. Materials

Poly(L-co-D,L lactic acid) (PLDLA; 70 : 30; Mw = 282,700 g·mol−1) was synthesized via bulk ring-opening copolymerization of L-lactide and D,L lactide, using stannous octoate as the catalyst, essentially as described by Motta and Duek [6]. Poly(caprolactone triol) (PCL-T; average Mw = 900 g·mol−1), also known as 2-oxepanone, polymer with 2-ethyl-2-(hydroxymethyl)-1,3-propanodiol, was supplied by Solvay (CAPA 3091).

2.2. Membranes

PLDLA/PCL-T membranes were prepared by the solvent casting technique, in blends of 100/0, 90/10, 70/30, and 50/50% (mass/mass), respectively, PLDLA/PCL-T in dichloromethane (Merck KgaA, Darmstadt, Germany), at a concentration of 10% (m·vol−1). After complete dissolution of the copolymer, the solution was poured onto a glass plate and placed in a glass container that had been saturated with the solvent. After solvent evaporation, the membranes were dried under vacuum for 8 h and stored in a desiccator under vacuum until characterized.

2.3. Characterization of PLDLA/PCL-T Membranes
2.3.1. Dynamic Mechanical Analysis (DMA)

DMA was done with a model 242 dynamic mechanical analyzer (Netzsch). The samples were subjected to deformation (amplitude: 240 μm) at a constant frequency of 1.00 Hz over a temperature range of −100°C to 200°C and a heating rate of 5°C/min, using a traction system.

2.3.2. Differential Scanning Calorimetry (DSC)

Calorimetric measurements were done with a DSC 2920 calorimeter (TA-Instruments) in which 7–10 mg samples sealed in aluminum pans were heated from 25°C to 200°C at a rate of 10°C/min and maintained at the final temperature for 5 min prior to cooling to 25°C at the same rate. The samples were maintained at 25°C for 5 min and then reheated. All tests were done under a nitrogen atmosphere.

2.3.3. Fourier Transform Infrared (FTIR) Spectroscopy

The quality of the PLDLA/PCL-T blends was assessed by FTIR spectroscopy. The IR measurements were done on Perkin-Elmer Spectrum One Perkin-Elmer 983G infrared spectrophotometer, using the KBr pellet technique, over the frequency range of 400–4000 cm−1, with a total of 256 scans.

2.3.4. Surface Contact Angle of the Membrane

The contact angles were measured with a contact angle goniometer (Ramé-Hart 100-00) and drop shape analysis software (RHI 2001 Imaging Software) at room temperature. A drop of deionized water (0.2 mL) was placed on the sample and the images of the water on the sample were recorded using a digital camera at 0.5 second intervals. From the angle measured between the droplet baseline and the tangent at the water/air boundary, a contact angle was calculated as the average of the measured left and right contact angles. For statistical analysis, three contact angle measurements were obtained.

Contact angle measurements were also used to evaluate the surface energy of the samples. This was done by measuring the angle to a polar liquid (water) and nonpolar liquid (diiodomethane) using the method harmonic [7].

2.3.5. Tensile Strength Test

PLDLA/PCL-T membranes containing polymer blends of 100/0, 90/10, 70/30, and 50/50 (m/m) were cut into strips  mm wide and approximately  mm thick and the tensile strength was determined with a model 810 Testar II apparatus (Instron) at 50 mm·min−1 and a load capacity of 1 kN. The samples were tested five times at a fixed humidity (50%) and temperature (°C).

2.4. Studies In Vivo with PLDLA/PCL-T Membranes (Blends 100/0 and 90/10)
2.4.1. Animals

Six 3-month-old Wistar rats weighing ~300 g were used. The rats were supplied by the Vivarium of the Pontifical Catholic University of São Paulo, the Center for Biological and Medical Sciences Sorocaba, and were housed at °C on a 12 h light/dark cycle with food and water ad libitum. The animal experiments were approved by an institutional Committee for Ethics in Animal Use CEUA/PUC-SP.

2.4.2. Membrane Implantation

The PLDLA/PCL-T membranes (blends 100/0 and 90/10) were immersed in 70% ethanol and then vacuum-dried. The rats were randomly divided into two groups of three rats each, with one group receiving the 100/0 PLDLA/PCL-T membrane and the other receiving the 90/10 PLDLA/PCL-T membrane. The time intervals studied were 1, 3, 7, 15, 30, and 60 days after implantation.

The rats were anesthetized with a combination of ketamine and xylazine (50 mg/kg and 10 mg/kg, i.p., resp.). Subsequently, two lateral incisions approximately 1 cm long were made in the dorsal cervical region. A membrane disc 6 mm in diameter was inserted into the left incision while the right incision served as the control (sham operated, no membrane inserted) (Figure 1). One implant was placed in each rat. The incisions were then sutured with nylon 5-0 suture.

At each time interval after implantation, the rats were anesthetized with ketamine 10% (100 mg/Kg) and xylazine 2% (6 mg/Kg) and the tissues surrounding the implant were excised.

2.4.3. Histological Analysis

The material surrounding the tissues was washed in saline solution and immediately fixed in 10% paraformaldehyde. The specimens were dehydrated in an increasing alcohol series, embedded in a paraffin (Histosec, Merck) and sectioned (4 μm) with a high-precision microtome (Leica RM 2245). The sections were subsequently mounted and stained with hematoxylin and eosin (H&E) and with Masson’s trichrome (for 30- and 60-day intervals). The stained sections were examined with a Nikon Eclipse E800 light microscope and images were captured using a Nikon FDX 35 digital camera attached to the microscope.

3. Results and Discussion

3.1. Mechanical and Physicochemical Properties

DMA was done on membranes containing PLDLA/PCL-T blends of 100/0, 90/10, 70/30, and 50/50. The α-relaxation associated with the glass transition () of the amorphous phases of PLDLA and PCL-T was determined from curves of loss of modulus versus temperature ( versus ), as described in ASTMD 4065-2001. The versus curves showed the presence of two glass transitions () typical of the presence of PLDLA and PCL-T (Figure 2(a)). The values for PLDLA decreased significantly in membranes containing 10% and 30% PCL-T (Table 1), indicating that the components were partially miscible. These findings agreed with those of Meier et al. (2004) [8] who used PCL-T as a plasticizer in cellulose acetate films and observed that the of cellulose acetate decreased as the PCL-T content increased.

The 50/50 blend (Figure 2(a)) showed behavior indicative of immiscibility (phase separation) of the components and this was confirmed by the values, which were very close to those of pure polymers; this immiscibility probably reflected the presence of excess plasticizer. Experiments with PLLA/PEG blends found that a plasticizer concentration > 20% caused phase separation [9].

Figure 2(a) shows that there was secondary β-type relaxation in PLDLA/PCL-T blend 100/0 (◆) caused by the mobility of the methyl group linked to its main chain. This relaxation disappeared in the presence of PCL-T, probably because of the steric hindrance caused by PCL-T or overlap in the melting temperature of PCL-T.

Figure 2(b) shows a drop in the storage modulus () for blends 90/10 and 70/30, whereas blend 50/50 had a ​​higher value than blends 90/10 and 70/30 at the start, but then decreased thereafter. The storage modulus of pure PLDLA declined at 37°C relative to the of this polymer. For blends 90/10, 70/30, and 50/50, the of the PLDLA phase was 31, 28, and 37, respectively, while the corresponding values for PCL-T were −63, −66, and −59°C (Figure 2(b)).

3.2. Calculation of the Interaction Parameter (χ12) for PLDLA/PCL-T Blends

The interaction parameters of the polymer blends were calculated based on the obtained by DMA. According to the Flory-Huggins theory, for a system to show complete miscibility, the interaction parameter (χ12) should be <0.5; in this case only the [10]. The χ12 for PLDLA/PCL-T blends was calculated using the Fox equation as follows: where = vitreous transition temperature of the blend (PLDLA/PCL-T), = composition of component 1 (PLDLA) in the blend, = vitreous transition temperature of component 1 (PLDLA in the 100/0 blend), = composition of component of 2 (PCL-T) in the blend, and = vitreous transition temperature of component 2 (PCL-T in the 0/100 blend).

Based on the content of PLDLA and PCL-T, (2) [11] was used to obtain the interaction parameter of the blend (χ12), the values of which are shown in Table 2: where = degree of polymerization of PLDLA = 1780.67 and = degree of polymerization of PCL-T = 2.63.

Table 2 shows that the addition of 10%, 30%, and 50% PCL-T to the blends resulted in χ12 values of 0.6, 0.5, and 0.8, respectively. The decrease in χ12 for the 70/30 blend may be directly related to the higher concentration of polar components (plasticizer) that interact with the ester groups of PLDLA. The 50/50 blend had the highest χ12 compared to the other blends and was again indicative of the immiscibility of the system.

3.3. Differential Scanning Calorimetry (DSC)

The DSC thermograms obtained in the second heating for PLDLA/PCL-T blends 100/0, 90/10, 70/30, and 50/50 are shown in Figure 3 and the results are summarized in Table 3.

There was no notable difference in the values for the first and second heating of membranes containing pure PLDLA (100/0 blend); there was also no melting transition phase in these curves, indicating that the material was amorphous, in case the membranes containing PCL-T has been a fall in the value of compared to pure PLDLA, which is a consequence of the increased flexibility of the material. The membranes of the blends 70/30 and 50/50 showed similar behavior in their curves. The and crystallization temperature of PCL-T were well defined.

3.4. Fourier Transform Infrared (FTIR) Spectroscopy

The FITR spectra in Figure 4 show all the characteristic groups of the copolymer poly(L-D,L-lactide). The spectrum for the PLDLA membrane showed absorption bands (, cm−1) at 2997–2965 (CH2, CH3), 1747 (C=O), 1360–1450 (CH3), 1266 (C–O bond in COO), and 750 (CH) characteristic of this material [12]. PCL-T showed bands similar to those of PLDLA that differed only in the axial asymmetrical deformation of fading in the 2995 cm−1 OH stretch and the appearance of a peak in the terminal region of the polymer chains (around 3467 cm−1).

The FTIR spectra for PLDLA/PCL-T blends revealed a shift in the OH stretch in blends 90/10 and 70/30 at a wavelength of 3500 cm−1 with respect to pure PCL-T (3467 cm−1); this shift can be assigned to links (H-bridges) between PLDLA and PCL-T. The 50/50 blend showed a band shift in the OH stretch to a shorter wavelength and this change was probably associated with excess plasticizer that could result in competition between the polymer/linked plasticizer, plasticizer/plasticizer.

3.5. Contact Angles

Measurement of the contact angle between water and a film surface is one of the easiest ways to characterize the hydrophilicity/hydrophobicity of a film. Good hydrophilicity is beneficial for biocompatibility and flow of nutritional liquid in vivo [13]. Cells adhere to and spread easily on substrates that have moderate hydrophilicity compared to very hydrophilic or hydrophobic substrates [14]. For the present analysis, the values (limits) for defining hydrophilic/hydrophobic interactions were a contact angle of 65° and a water adhesion tension of 30 dynes·cm−1, as described by Lim et al. (2004) [15].

The water adhesion tension was calculated as where γ = water surface tension (72.8 dynes·cm−1) and θ = contact angle [15].

Surfaces with contact angles < 65° and high values of were defined as hydrophilic, whereas surfaces high contact angles and low values of , that is, low wettability, were considered to be hydrophobic. Table 4 shows the contact angles, surface energy, and water adhesion tension for PLDLA/PCL-T membranes. The values of were calculated according to (3).

Based on the values shown in Table 4, the surface of PLDLA/PCL-T 100/0 membranes was hydrophobic and the addition of PCL-T made this more hydrophilic, as shown by the decrease in the contact angle. This effect was expected because of the presence of polar groups in PCL-T that increase the interaction with water, increasing the surface energy and water adhesion tension. Figure 5 shows the wettability of the membrane surface of different blends of PLDLA/PCL-T with deionized water; as can be seen, wettability increased proportionally to the amount of PCL-T, in agreement with the results in Table 4.

3.6. Tensile Strength Tests

Table 5 and Figure 6 show the relationship between tensile properties and the PCL-T content of PLDLA/PCL-T membranes (100/0, 90/10, 70/30, and 50/50 blends). PLDLA showed the typical characteristics of a vitreous polymer, with a modulus of elasticity of  MPa, resistance to traction of  MPa, and deformation at rupture of % (Table 5). The addition of PCL-T significantly reduced the modulus of elasticity () (Table 5), in agreement with the findings of Meier et al. (2004) [8] in their study of the effects of PCL-T on cellulose acetate.

The presence of PCL-T significantly altered the deformation at rupture, indicating an increase in flexibility (Table 5, Figure 6); the 90/10 blend showed the greatest deformation at rupture, with a value of % compared to pure PLDLA, which had a value of %. The higher value of the 50/50 blend reflected the ability of the plasticizer to dissipate energy at fracture interfaces, thereby delaying rupture of the membrane [9]. With regard to the 70/30 blend, one possible explanation for the unexpectedly higher modulus values obtained for this membrane compared to the 90/10 membrane could be the proximity between the temperature at which the test was done (°C) and the (28°C) of PLDLA containing 30% PCL-T.

3.7. Histological Analysis of Implanted PLDLA/PCL-T Membranes (100/0 and 90/10 Blends)
3.7.1. Hematoxylin-Eosin Staining

The inflammatory response on the first day after implantation was greater at the implanted sites than at control sites. Membranes containing PLDLA/PCL-T blend 100/0 showed fibrin deposition with polymorphonuclear neutrophils, as well as occasional eosinophils and mast cells (Figure 7(a)). A similar situation was observed for PLDLA/PCL-T blend 90/10 (Figure 7(b)). In all cases, the inflammatory infiltrate in the controls was markedly less than that at the implanted sites (Figure 7(c)).

At three days after implantation, sites implanted with the 100/0 blend showed an area delimited by organized fibrin deposition, with congested vessels and the presence of mononuclear cells, but no giant multinucleated cells. Sites implanted with the 90/10 blend also showed the fibrin “capsule,” but it was not as organized and there were frequently macrophages around the implant, in addition to congestion and edema. The control sites of both groups had a similar appearance, that is, granulated tissues with a chronic inflammatory infiltrate containing macrophages, plasmocytes, and mast cells.

At 15 days after implantation with the 100/0 blend, there was a well-defined capsule delimited by dense collagen and an internal surface containing a few mononuclear cells and hardly any giant cells (Figure 8(a)). At sites implanted with the 90/10 blend, the capsule occasionally contained giant foreign body cells, in addition to mononuclear cells (Figure 8(b)). All control sites showed complete healing with no significant alterations.

At 60 days after implantation, sites implanted with the 100/0 blend showed thick fibrosis around the capsule and practically no inflammatory infiltrate except for a small number of mononuclear cells in the interstitium (Figure 9). Sites implanted with the 90/10 blend showed a similar capsular organization to that seen with the 100/0 blend but with a discrete chronic inflammatory infiltrate; there were no giant multinucleated cells. The control sites showed no marked alterations. In contrast to our findings, previous studies of PLLA membranes implanted subcutaneously in rats reporting the presence of giant multinucleated cells up to 180 days after the implantation reported that poly(DL-lactic acid) without plasticizer caused a chronic inflammatory response involving giant cells, macrophages with phagocytosed material, and fibroblast proliferation [5, 16]. Macrophages have an important role in tissue repair through their ability to produce growth factors and mediators responsible for fibroblast proliferation and collagen synthesis [17].

3.7.2. Staining with Masson’s Trichrome

The histological alterations observed with specific staining (Masson’s trichrome) confirmed those observed with general staining (HE). In the case of the 100/0 blend, the capsule was practically devoid of an inflammatory infiltrate, while, in the case of the 90/10 blend, the collagen was looser and there was granulous tissue containing macrophages.

At 60 days after implantation, for both of the blends tested, there was well-formed, dense collagen that was practically devoid of an inflammatory infiltrate. No giant multinucleated cells were observed at any of the sites. For the 100/0 blend, the capsule was well organized with mature collagen and no inflammatory infiltrate (Figure 9(a)), while for the 90/10 blend the collagen was more loosely organized and there were few mononuclear cells (Figure 9(b)).

Although this in vivo analysis was only a preliminary investigation, our results indicate that PLDLA membranes containing PCL-T are potentially useful, particularly since these polymers are nonallergic, nontoxic, and flexible; the latter characteristic in particular facilitates the handling of these membranes during surgical procedures.

4. Conclusions

In this study, we have shown that the addition of PCL-T to PLDLA membranes markedly affected the mechanical, thermal, morphological, physicochemical, and biological properties of the resulting membrane, especially increasing the flexibility of PLDLA. The addition of PCL-T at concentrations of 10% and 30% resulted in partially miscible systems while the addition of 50% resulted in immiscibility, probably because of the excess of plasticizer.

The analysis of PLDLA/PCL-T membranes in vivo indicated satisfactory polymer/tissue interaction. By 60 days after implantation, both 100/0 and 90/10 blends resulted in dense collagen deposition that was more loosely organized with the latter blend, probably because of the slower degradation.

These results indicate that the use of PLDLA PCL-T membranes in medical devices could be advantageous, particularly where there is no need for long-term implants, for example, in supporting cell cultures, in dressing skin ulcers, and in guided regeneration in periodontics.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.