Synthesis and Modification of Functional Poly(lactide) Copolymers: Toward Biofunctional Materials

Abstract

A polylactide copolymer with pendant benzyloxy groups has been synthesized by the copolymerization of a benzyl-ether substituted monomer with lactide. Debenzylation of the polymer to provide pendant hydroxyl groups followed by modification with succinic anhydride affords the corresponding carboxylic acid functionalized copolymer that is amenable to standard carbodiimide coupling conditions to attach amine-containing biological molecules. An amino-substituted biotin derivative was coupled to the carboxyl functional groups of copolymer films as proof-of-concept. In a demonstration of the function of these new materials, an RGD-containing peptide sequence was tethered to copolymer films at various densities and was shown to enhance the adhesion of epithelial cells. This strategy provides the opportunity for the attachment of a variety of ligands, allowing for the fabrication of a versatile class of biodegradable, biocompatible materials.

Introduction

The use of biodegradable polyesters such as poly(lactic acid) (PLA), poly(glycolic acid), and polylactones in biomedical applications has increased significantly over the past decade.^1–4 These materials are used in various forms (films, fibers, foams) for a diverse set of applications ranging from drug delivery to sutures.^5–7 While the ester functional groups in the background contribute to a set of attractive physical properties and facilitate biodegradation, the absence of pendant functional groups limits the versatility of these materials. The incorporation of reactive sites that are amenable to modification with a wide variety of biologically relevant ligands would provide a host of opportunities to control cell adhesion and function.⁸ For example, a combination of synthetic biodegradable polymers with peptide ligands may provide hybrid scaffolds for use in tissue engineering with control over scaffold–cell interactions as well as targeted drug delivery.⁹ Specifically, the attachment of peptides with the RGD (Arg-Gly-Asp) sequence has been shown to improve the cytocompatibility and cellular attachment characteristics of a variety of materials.^10–13

The main strategy employed to incorporate pendant functional groups into PLA involves the copolymerization of cyclic lactide monomer with comonomers¹⁴ such as functional glycolides,^15–19 esteramides,^20–25 N-carboxyanhydrides,^26–29 and lactides.^16,30–33 While we³⁴ and others³⁵ have reported the successful copolymerization of lactide monomers containing protected functional groups followed by deprotection to provide PLA copolymers bearing new chemical functionality (alcohols, amines, and carboxylic acids), these materials have not been subjected to further postpolymerization modification.

Here we demonstrate the ring-opening copolymerization of cyclic lactide and a new dibenzyloxy-substituted lactide, followed by debenzylation to afford a hydroxy-substituted copolyester. Reaction of the functional copolymer with succinic anhydride provides a material that presents carboxylic acid functionality to allow standard carbodiimide coupling to attach amine-containing biological molecules. We chose this route to prepare carboxylic acid bearing copolylactides over the incorporation of a protected carboxylic acid containing monomer owing to the ease of monomer synthesis and previous difficulties in polymerizing the protected acid containing lactide.³⁴ We posited that incorporation of a small amount of functional comonomer (5 mol %) would be sufficient to allow for modification with biomolecules without adversely affecting the thermal properties of the copolymer. As proof-of-concept, succinylated films of our new functional PLA copolymer were treated with an amine-bearing biotin derivative and an enzyme-linked immunosorbent assay (ELISA) was used to demonstrate successful coupling. Given the wide interest in interactions between synthetic polymeric materials and biological matrices, we tethered an RGD-containing peptide sequence (GGRGDSPGGK) to the copolymer films. A fluorescence assay of succinylated films indicated that the amount of tethered RGD can be controlled by treatment with various concentrations of peptide. The function of these films was demonstrated by the enhanced adhesion of mammalian cells to RDG-modified PLA copolymers relative to unmodified PLA. This study clearly indicates that the functional groups of the polymer are accessible and can be used for the attachment of bioactive ligands to control surface-cell interactions. Thus, this presents a versatile approach to functionalizing PLA with biomolecules and provides the opportunity to mediate cellular adhesion and impart new biological functionality leading to tailorable biodegradable, biocompatible materials with numerous potential applications.

Experimental Section

Materials and Methods.

Benzyl-l-serine was purchased from Indofine Chemical Company. l-Lactide was purchased from Aldrich and recrystallized twice from dry EtOAc. Dry benzene was purchased from VWR and stored in a glovebox. Biotin-PEO-LC-amine was obtained from Pierce Chemical Company (product No. 21347). Newborn calf serum (NCS) was obtained from Hyclone Laboratories (product No. SH30118.03). Antibiotin antibody (BN-34) was obtained from Aldrich (product No. A6561). The fluorescent peptide sequence GGRGDSPGGK-FITC (the FITC was attached to the lysine residue) was synthesized by the Emory University Microchemical Facility.

IR spectra were recorded on a Bruker Vector 22 spectrometer. Melting points were determined using a Mel-Temp II apparatus. Molecular weight data was collected using three Waters Styragel columns (5 μm beads: HR 1, 100 Å; HR 3, 1000 Å; HR 4, 10000 Å) connected to a Waters 2690 Separations Module with a Waters 2487 refractive index detector. The eluant was THF (1 mL/min, 303 K) and the molecular weights were determined relative to six narrow polystyrene standards in THF. Thermal transitions were determined on a Mettler DSC 822. The sample sizes were approximately 4 mg and each sample was subjected to thermal cycles from 0 to 200 °C at a rate of 10 °C per minute. NMR spectra were recorded at room temperature using a Varian Mercury spectrometer (300 MHz) or a Bruker AMX spectrometer (400 MHz). Mass spectral analyses were performed on a VG Instruments 70SE mass spectrometer. The fluorescence intensity of the ELISA samples and RGD assays were recorded using a Perkin-Elmer HTS 7000 Plus Bio Assay Reader.

(S)-3-Benzyloxy-2-hydroxypropionic Acid.³⁶

The hydroxy acid was prepared according to the general method reported by Deechongkit et al.³⁴ A solution of NaNO₂ (10.6 g, 154 mmol) in H₂O (100 mL) was added over 3 h to a solution of O-benzyl-l-serine (20.0 g, 103 mmol) in 0.7 M aqueous solution of trifluoroacetic acid (200 mL). The mixture was stirred for 3 h, NaCl (20 g) was added, and the mixture was extracted with EtOAc (3 × 200 mL). The organic layers were combined and dried over MgSO₄. The solvent was removed, and the residue was dried under vacuum overnight to give (S)-hydroxy acid as a waxy yellow oil (19.52 g, 97%), which was used without further purification.

(S,S)-3,6-(Benzyloxymethyl)-1,4-dioxane-2,5-dione 2.

A mixture of (S)-3-benzyloxy-2-hydroxypropionic acid (10 g, 51 mmol) and p-toluenesulfonic acid monohydrate (0.2 g, 1.1 mmol) in toluene (1.5 L) was heated at reflux, and water was removed by azeotropic distillation using a Dean–Stark apparatus. After 20 d, the mixture was cooled and washed with H₂O (6 × 500 mL), dried over MgSO₄, and the solvent was removed. The residue was recrystallized from anhydrous Et₂O to give 2 as a white crystalline solid (1.78 g, 21%); mp 70.2–73.2 °C. ¹H NMR (300 MHz, CDCl₃) δ 7.25–7.35 (m, 10H), 5.13 (t, J = 4.1, −CH−, 2H), 4.49 (d, J = 11.8, diast. −OCH₂Ph, 2H), 4.44 (d, J = 11.6, diast. −OCH₂Ph, 2H), 3.89 (d, J = 4.1, −CH₂−, 4H). ¹³C NMR (100 MHz, CDCl₃) δ 163.4, 136.7, 128.5, 128.1, 128.0, 76.5, 73.7, 69.9. IR (thin film): v 3024 (weak, aromatic C–H), 2920, 1757, 1247, 1101 cm⁻¹. MS (EI, m/z, relative intensity): 265.1 (M-PhCH₂, 28%), 91.1 (tropylium, 100%). HRMS calcd for C₂₀H₁₉O₆ (M − 1), 355.11816; found, 355.11705 (Δ = 3.1 ppm).

Copolymerization of Lactide with Benzyl-Substituted Comonomer 2.

A solution of twice-recrystallized lactide (3.8 g, 27 mmol), comonomer 2 (0.50 g, 1.4 mmol), and SnOct₂ (34 μL of a 0.8 M stock solution in benzene) in benzene (75 mL) was prepared in a drybox and freeze-dried overnight. The mixture was heated to 140 °C for 10 h and poured into MeOH (1 L). The precipitate was collected by filtration and dried under vacuum overnight to give protected copolymer 3 as a white solid (3.69 g, 85%). ¹H NMR (300 MHz, CDCl₃) δ 7.20–7.40 (10H, −C₆H₅), 5.3–5.5 (2H, −CH− of benzyloxy-substituted repeat unit), 5.0–5.3 (−CH− of LA unit), 4.4–4.6 (4H, −OCH₂C₆H₅), 3.8–4.0 (4H, −CH₂OCH₂C₆H₅), 1.4–1.8 (−CH₃ of LA unit). ¹³C NMR (100 MHz, CDCl₃) δ 169.6, 137.4, 128.4, 127.8, 127.7, 73.4, 69.0, 68.4, 16.6. IR (thin film) 3016 (weak, aromatic C–H), 2991, 1748, 1178, 1080 cm⁻¹.

Deprotection of Copolymer 3 to Afford Hydroxyl-Substituted Polymer 4.

Protected copolymer 3 (1.82 g) was dissolved in a 4:1 (v/v) mixture of EtOAc and CH₂Cl₂ and an equal mass of Pd(OH)₂ was added. The solution was placed on a hydrogenator and charged to 50 PSI of H₂ for 48 h. The catalyst was removed by filtration and the solvent was removed to give functional copolymer 4 (1.50 g, 82%). ¹H NMR (300 MHz, CDCl₃) δ 5.0–5.4 (−CH− of functional repeat unit and LA unit), 4.1–4.3 (−CH₂OH of functional repeat unit), 3.9–4.1 (−CH₂OH of functional repeat unit), 1.4–1.8 (−CH₃ of LA unit). ¹³C NMR (100 MHz, CDCl₃) δ 169.6, 69.0, 67.0, 16.6. IR (thin film) 2993, 2943, 1748, 1179, 1082 cm⁻¹.

Modification of Hydroxyl-Bearing Copolymer 4 with Succinic Anhydride.

Functional copolymer 4 (1.50 g) was treated with succinic anhydride (0.517 g, 5.17 mmol; 5 equiv. of succinic anhydride per mol of hydroxyl side chain) and pyridine (0.5 mL, 6 mmol; 6 equiv. per mol of hydroxyl side chain) in CH₂Cl₂ (90 mL). The reaction mixture was stirred for 10 d, the solvent was removed, and the polymer was precipitated into MeOH (250 mL) to give carboxylic acid-functionalized copolymer 5 (1.2 g, 80%). ¹H NMR (300 MHz, CDCl₃) δ 5.3–5.5 (2H, −CH− of succinate repeat unit), 5.0–5.3 (−CH− of LA unit), 4.4–4.8 (4H, −CH₂OCOCH₂CH₂COOH), 2.6–2.8 (8H, −CH₂OCOCH₂CH₂− COOH), 1.4–1.8 (−CH₃ of LA). ¹³C NMR (100 MHz, CDCl₃) δ 169.6, 69.0, 28.7, 16.6. IR (thin film) 2995, 2958, 1753, 1710 (weak, −CO₂H), 1182, 1085 cm⁻¹.

Film Formation.

Films were spin-coated at 2200 rpm on glass discs by applying several drops of a 25 mg/mL solution of copolymer 5 in a 1:1 (v/v) mixture of toluene and CH₂Cl₂. The glass slides containing the films were placed into the wells of a 24-well plastic plate.

Biotin Coupling.

Biotin stock solution (600 μL of a solution containing 16 mg biotin-PEO-LC-amine and 0.9 mg EDC in 14 mL of a 0.1 M aqueous N-morpholinoethane sulfonic acid buffer solution adjusted to a pH of 5.5 using NaOH) was added to each well and the samples were incubated for 2.5 h at room temperature. The biotin solution was removed from the wells by aspiration, and the samples were sequentially incubated in phosphate-buffered saline (PBS; no Ca/Mg) for 1 h, a 0.1% (w/v) solution of sodium dodecyl sulfate (SDS) in PBS (no Ca/Mg) for 5 min, and DI water at 25 °C to remove any physically adsorbed biotin.

RGD-FITC Coupling.

Films were treated with 2 mM EDC and 5 mM NHS in a 0.1 M aqueous 2-(N-morpho)-ethanesulfonic acid and 0.5 M NaCl solution (pH 5.5) for 30 min at room temperature to activate the carboxylic acids to amidation. Surfaces were subsequently immersed in a 20 mM solution of 2-mercaptoethanol in deionized water to quench any unreacted EDC,³⁷ leaving activated NHS esters on the PLA films. Films were then exposed to various concentrations of the fluorescent peptide GGRGDSPGGK-FITC in PBS for 30 min at room temperature in the dark. The free amines on this peptide were covalently tethered to the films using the surface NHS esters as leaving groups. The peptide solution was removed from the wells by aspiration, and the samples were incubated in 20 mM glycine for 10 min to quench any unreacted NHS esters. After carefully transferring the slides to a new set of wells, the films were rinsed three times with PBS to remove weakly bound, physically adsorbed RGD-FITC, and the discs were subsequently incubated in PBS for 30 min to further reduce nonspecific peptide adsorption. The solution was removed by aspiration, fresh PBS was added, and the fluorescence intensity was measured and quantified.

ELISA.

The glass discs were incubated in the plastic wells (24-well plastic plates obtained from Corning, product No. 3526) in newborn calf serum (NCS) for 1 h at 37 °C to block the surfaces and washed three times with DI water. A solution of antibiotin alkaline-phosphatase conjugated antibody (Sigma-Aldrich; 300 μL of a solution composed of 7 μL EDTA, 33 μL of a 10% NaN₃ solution, 3 μL of a 1% heat-denatured bovine serum albumin solution in PBS, 10 mL DI water, and 5 μL BN-34) was added to each well and the samples were incubated for 1 h at 37 °C. The glass discs were transferred to a new set of wells to minimize the response caused by the adsorption of antibody onto the wells. The dishes were washed five times with DI water, incubated in NCS for 10 min at 25 °C, and rinsed again five times with DI water. A solution of methylumbelliferyl phosphate (300 μL of a solution composed of 7.5 mL of DI water, 2 mL of 5× diluted diethanolamine, 500 μL of 0.l M NaHCO₃, and 0.6 mg methylumbelliferyl phosphate) was added to each well and the samples were incubated at 37 °C for 1 h in the dark. A 50 μL aliquot from each well was transferred into the corresponding well of a black Dynex 96-well plate, and the fluorescence intensity was recorded.

Cell Adhesion Assay.

A wash assay was used to measure relative adhesion of cells adhered to peptide-functionalized and unfunctionalized film surfaces. MDCK cells were labeled with 4 mM calcein-AM in a 2 mM dextrose-PBS solution for 20 min and resuspended in α-MEM media (1% penicillin-streptomyocin). Cells were seeded onto films at a density of 200 cells/mm² and allowed to adhere at 37 °C for 1 h. Wells were then subjected to three gentle washes with media to detach loosely bound and unbound cells. After a gentle aspiration, postwash fluorescence was quantified using an HTS 7000 Plus plate reader. The postwash fluorescence levels correlate with the number of adhered cells.

Statistics.

Data are reported as mean ± standard error. Results were analyzed by one-way ANOVA using SYSTAT 8.0 (SPSS). If treatment level differences were determined to be significant, pairwise comparisons were performed using a Tukey posthoc test. A 95% confidence level was considered significant.

Results and Discussion

Synthesis of Protected Functional Monomer.

We have previously reported that benzyl ether-protected lactide 1 is subject to copolymerization and reductive debenzylation to afford hydroxyl-bearing PLA copolymers.³⁴ In the current study, we chose to adopt a different strategy whereby we prepared the symmetrically disubstituted lactide monomer 2. This makes use of the robustness of the benzyl ether protecting group and provides a protected functional monomer in fewer synthetic steps. It also provides straightforward access to a single enantiomer of the protected functional monomer. Commercially available O-benzyl-L-serine was converted to (S)-3-(benzyloxy)-2-hydroxypropanoic acid by diazotization (NaNO₂, TFA) followed by hydrolysis (Figure 1).³⁶ Optimization of the reaction conditions and simple extraction of the product from the aqueous reaction mixture into ethyl acetate gave near quantitative yields of product with full retention of the stereochemistry.

Figure 1.

Synthesis of protected functional monomer 2.

The cyclocondensation of the (S)-hydroxy acid was catalyzed by p-toluenesulfonic acid, and water was removed from the reaction mixture using a Dean–Stark trap (20 d). The condensation reaction also provides linear oligomers and conversion to the cyclic dimer was typically limited to about 40%, which is comparable to conversions previously reported for selfcondensation reactions for the preparation of analogous substituted glycolides.³⁸ Attempts to separate 2 from the linear oligomers by column chromatography on silica gel led to ring opening. However, recrystallization proved effective in affording pure protected functional monomer 2. This resulted in lower yields (21%) than those found in the literature for the synthesis of alkyl-substituted lactides using the same method.^30,33 However, while analogous uncyclized alkyl-substituted oligomers can be thermally cracked at high temperature in the presence of ZnO (a good transesterification catalyst) and vacuum distilled to give cyclic dimer,³⁰ our attempts to transform the benzyloxy-substituted oligomers in this fashion led to decomposition arising from the high temperatures necessitated by the low volatility of 2. In relation to the preparation of monosubstituted benzyl ether-protected lactide 1, this yield is acceptable, as the synthetic method has fewer reaction steps, results in the formation of a single diastereomer of monomer 2, and yields a dibenzyloxy-protected lactide monomer with two potential functional groups per monomer.

¹H NMR analysis was used to verify the stereochemistry of protected functional monomer 2. The methine protons of 2 appear as a triplet at 5.14 ppm (J = 4.1 Hz for coupling of the methine hydrogen to the neighboring pair of diastereotopic hydrogen atoms) corresponding to a single (S,S) stereoisomer (Figure 2a). In a separate experiment, when a racemic mixture of 3-(benzyloxy)-2-hydroxypropanoic acid (prepared from rac O-benzyl-serine) was cyclodimerized, two overlapping triplets were observed in the ¹H NMR spectrum, one at 5.14 ppm (J = 4.1 Hz) corresponding to the (S,S) and (R,R) enantiomers, and the other at 5.12 ppm (J = 2.8 Hz) corresponding to the meso (R,S) compound (Figure 2b). An X-ray structure of the benzyl-protected functional (S,S) monomer 2 is provided in Figure 2c, which shows a pseudochair conformation of the 1,4-dioxane-2,5-dione ring with the benzyloxymethyl side chains occupying equatorial positions.

Figure 2.

¹H NMR analysis and X-ray diffraction of cyclic disubstituted dimers: (a) ¹H NMR spectrum of the methine protons of (S,S) isomer 2; (b) ¹H NMR spectrum of a diastereomeric mixture of cyclic dimers obtained from rac-hydroxy acid; and (c) X-ray crystal structure of 2.

Copolymerization, Deprotection, and Modification with Succinic Anhydride.

A 5:95 mixture of protected monomer 2 and L-lactide monomer, (S,S)-3,6-dimethyl-1,4-dioxane-2,5-dione), was copolymerized using stannous octoate (SnOct₂) as the catalyst and residual water as the initiator (Figure 3).³⁴ The reaction mixture was heated to 140 °C to facilitate the ring-opening of monomer 2, as higher temperatures are often required to obtain high conversions in the polymerization of substituted lactides.^{33,38 1}H NMR analysis of the crude copolymerization mixture showed high conversion of both cyclic monomers to linear polyester (~90% by integration of the PLA peaks relative to lactide monomer). The reaction mixture was poured into methanol to precipitate the polymer, which was filtered and dried under vacuum. ¹H NMR spectroscopy of the resulting material verified that the copolymer (3) included 5% of protected monomer 2. The agreement between the amount of protect comonomer and lactide incorpotated into the polymer and present in the original polymerization mixture confirms that comonomer 2 polymerized with an equally high conversion as the lactide monomer (although the small amount of unreacted monomer 2 in the crude reaction mixture could not be determined directly due to overlap of the signals with those for polymer). While it is expected that the substituted protected lactide comonomer has a lower reactivity than lactide itself,³⁹ the effect of this on the ratio of comonomers incorporated would only be relevant at low conversions. In the current context, we performed polymerizations to high conversion with a resulting copolymer with a composition that matched the ratio of monomers in the reaction mixture. Gel permeation chromatography (GPC, using polystyrene calibration standards) indicated an average molecular weight of 1.5 × 10⁴ g/mol. This value is comparable to molecular weights found in the literature for the homopolymerization of alkyl-substituted lactides³³ and the copolymerization of protected functional monosubstituted lactides with lactide monomer.^34,35 Though molecular weights up to 7.7 × 10⁴ g/mol could be obtained, the molecular weight of this particular sample could be attributed to the presence of trace amounts of impurities in the monomer or catalyst acting as initiators.

Figure 3.

Copolymer synthesis, deprotection, and modification with succinic anhydride.

The benzyl protecting groups of copolymer 3 were removed by hydrogenolysis over Pd(OH)₂ (50 PSI of H₂ in a 4:1 (v:v) mixture of EtOAc and CH₂Cl₂) to give functional copolymer 4 with pendent hydroxyl groups. Complete removal of the protecting group was verified by ¹H NMR spectroscopy by the disappearance of the peaks for the phenyl protons at ~7.3 ppm and the appearance of separate one-proton signals at δ 4.0 and 4.2 ppm for the diastereotopic protons of the methylene group (Figure 4B, peaks labeled d). The deprotection of copolymer 3 was also confirmed by the disappearance of the aromatic C–H stretching vibration at 3016 cm⁻¹ in the infrared (IR) spectrum. While the polymer is soluble in EtOAc alone, repeated attempts at reductive debenzylation in the absence of added CH₂Cl₂ failed to go to completion.

Figure 4.

¹H NMR spectra of copolymers: (A) benzyloxy-substituted copolymer 3; (B) deprotected hydroxymethyl bearing copolymer 4; (C) succinylated copolymer 5.

The hydroxymethyl-substituted copolymer 4 was treated with succinic anhydride to give the acid functionalized copolymer, 5. The progress of the succinylation reaction was monitored by ¹H NMR spectroscopy. The protons of the methylene group of the hydroxymethyl-substituted copolymer 4 shift downfield to δ 4.6 ppm upon esterification, with simultaneous formation of peaks attributed to the succinate monoester at 2.7 ppm (Figure 4C, peaks d and f, respectively). The succinylation was also confirmed by the appearance of a weak peak at 1710 cm⁻¹ in the IR spectrum for the C=O stretching vibration carboxylic acid group. GPC analysis (Table 1) indicated that the amount of polymer degradation of the polyester backbone throughout the deprotection and succinylation steps was minimal, as observed by a slight decrease in the molecular weight. NMR spectroscopy confirmed the preservation of the polymer structure and that the polymers had not broken down into monomeric components. The copolymers were also characterized by differential scanning calorimetry (DSC; Table 1). Copolymers 3–5 displayed similar glass transition temperatures and lower melting temperatures than literature values for semicrystalline PLA homopolymers derived from the ring-opening polymerization of L-lactide.⁴⁰ This shows that the low level incorporation of difunctional comonomer (5 mol %) does not significantly affect the glass transition or melting point of the PLA copolymer. This is in contrast to our previous report where greater amounts (25%) of monofunctional monomer 1 incorporated into the polymer backbone resulted in a significant decrease in the T_g.³⁴ Copolymer 3 displays a strong cold crystallization on first heating, and copolymer 4 was completely amorphous after the first heating and cooling cycle.

Table 1.

GPC and DSC Characterization of Copolymers 3–5

copolymer	Mn (kg/mol)	Mw (kg/mol)	PDI	Tg (°C)	Tm (°C)	ΔHm (J/g)
3	15	33	2.2	56	149	9
4	14	28	2.0	57	152	49
5	11	21	1.9	61	152	25

Biotin Coupling.

The pendant carboxylic acids of 5 provide the opportunity to use standard carbodiimide chemistry for attachment of amines. As proof of concept, we coupled an amine-substituted biotin derivative to the copolymer films (Figure 5). Films of succinylated copolymer 5 were prepared by spin-coating a solution (25 mg/mL) of the copolymer in a 1:1 (v:v) mixture of toluene and dichloromethane onto glass discs. The discs were then placed into wells in a 24-count polystyrene plate and subjected to conditions for amidation. To verify the attachment of biotin, we performed an ELISA using an alkaline phosphatase enzyme-conjugated antibiotin antibody. After extensive rinsing to remove weakly adhered antibody, a solution of 4-methylumbelliferyl phosphate was added and the fluorescence resulting from the formation of the methylumbelliferone anion was quantified and used as a relative measure of the amount of biotin. Films treated with the amine-substituted biotin in the presence of EDC show increased fluorescence relative to control samples which had been treated with the biotin-amine in the absence of EDC (p < 0.005), or treated with a carboxylic acid substituted biotin (with [p < 0.008] in the presence of ECD and [p < 0.003] in the absence of EDC; Figure 6). The ELISA method is not quantitative: the fluorescence intensity is not linearly related to the density of antigen (biotin). This arises, in part, from the relative size of the antibody and antigen, whereby the fluorescence intensity will saturate at levels far below 5% biotin incorporation. The slightly elevated fluorescence levels of the sample treated with the biotin-carboxylic acid and EDC relative to the other control experiments can be attributed to coupling to small amounts of hydroxyl side chains still present in the sample of succinylated copolymer 5 (which in this case had failed to react completely with succinic anhydride).

Figure 5.

Attachment of an amine-substituted biotin derivative to copolymer films.

Figure 6.

ELISA determination of relative density of amine-substituted biotin on succinylated PLA copolymer: Relative fluorescence indicating biotin attachment to films of copolymer 5 exposed to biotin-amine and biotin-carboxylic acid derivatives with and without EDC.

RGD–Peptide Coupling and Control of Cell Adhesion.

We also treated the succinylated polymer films with an RGD-containing peptide (GGRGDSPGGK) conjugated to a fluorescein derivative (FITC) to demonstrate control over the density of tethered bioactive ligand and its subsequent effect on surface–cell interactions. The green fluorescence of the FITC was quantified to determine the effect of concentration of the peptide in standard NHS/EDC coupling reactions on the density of immobilized peptide. Films coupled with RGD-FITC showed increased levels of fluorescence (Figure 7). Films exhibited increased fluorescence when treated with higher concentrations of RGD-FITC and low fluorescence in the absence of EDC activation (data point at 20 μg/mL RGD-FITC). This clearly demonstrates the ability to control tethered surface density of bioactive RGD ligand by varying coating concentration.

Figure 7.

Relative fluorescence of succinylated PLA copolymer films 5 exposed to RGD at various concentrations with and without EDC (lower data point at 20 μg/mL).

Functionalization of the copolymer films with peptides containing the RGD sequence, which is a ubiquitous cell adhesion-promoting sequence, should offer control over cell adhesion. To validate this assertion, we subjected both RGD-functionalized and unfunctionalized copolymer films to a simple adhesion assay using epithelial MDCK cells in serum-free media. Films of polymers were exposed for 1 h to MDCK cells labeled with calcein-AM, a membrane-permeable green fluorescent-dye. Gentle washes with media were performed to dislodge weakly bound cells and postwash fluorescence was recorded to determine the density of cells adhered to the surface. More cells adhere to the RGD-tethered films than the unfunctionalized films (p < 0.008), and adhesion is dependent on RGD surface density (p < 0.04; Figure 8). Taken together, these data demonstrate the ability to functionalize PLA films with bioactive ligands and subsequently promote cell adhesion.

Figure 8.

Relative fluorescence indicating cell attachment to films of copolymer 5 exposed to RGD at various concentrations with and without EDC.

Conclusions

In conclusion, we have synthesized a new protected functional lactide monomer and successfully copolymerized it with lactide to give random copolymers that contain protected functional groups along the backbone. The copolymers were deprotected and used in the ring-opening of succinic anhydride, showing that the functional groups are accessible and amenable further modification. Films of the carboxylic acid functionalized copolymer were treated with a biotin-amine derivative and the coupling was verified using ELISA, showing that the carboxylic acid-functionalized copolymers can be modified with amine-terminated biomolecules. Use of a fluorescently labeled RGD-containing peptide sequence allowed us to verify control over the amount of peptide immobilized onto the succinylated copolymer films. Finally, a cell assay was performed which showed enhanced cell adhesion to RGD-containing films. This provides a general strategy whereby a variety of biomolecules can be attached to functionalized PLA copolymers using standard peptide coupling protocols, thereby leading to novel materials with numerous potential biomedical applications.