pubs.acs.org/Langmuir © 2010 American Chemical Society
Modification of Poly(lactic/glycolic acid) Surface by Chemical Attachment of Poly(ethylene glycol) Kiss,*,† E. Kutnyanszky,† and I. Bertoti‡ E. †
Laboratory of Interfaces and Nanostructures, Institute of Chemistry, E€ otv€ os Lor and University, Budapest 112, POB. 32, H-1518 Hungary and ‡Institute of Materials and Environmental Chemistry, Chemical Research Centre, Hungarian Academy of Sciences, Budapest, POB. 17, H-1525 Hungary Received September 8, 2009. Revised Manuscript Received December 14, 2009
Biodegradable polyesters such poly(lactic acid) and poly(lactic/glycolic acid) (PLGA) copolymers are preferred biomaterials and used among others as drug delivery systems, although their surface hydrophobicity limits their application. In this work, chemical modification of the PLGA surface was developed by coupling of either linear or starlike poly(ethylene glycol) (PEG) molecules via chemical bonds to the PLGA surface following amino functionalization as a first step to improve its biocompatibility. The chemical attachment was followed by detailed X-ray photoelectron spectroscopy (XPS) studies. It was shown that substantial modification can be achieved by linear PEG, but even higher surface coverage with hydrophilic groups can be obtained when the six-armed PEG is applied with the additional advantage of possible further funcionalization via free amino groups available on the surface of the latter. As a final goal, a significant increase of water wettability together with reduced protein adsorption was achieved on PEG-coupled PLGA surfaces.
Introduction Interfacial interactions with biological systems are of paramount importance in designing and application of biomaterials and, among others, of various drug delivery devices. The requirement of biocompatibility of such surfaces is twofold: while the nonspecific adsorption of biomolecules should be generally suppressed, specific adsorption and directed cell adhesion are to be enhanced. Polylactide and polyglycolide, owing to their biodegradability, represent an essential class of biomaterials for various biomedical applications including tissue engineering and colloidal drug delivery.1 The application of poly(lactic acid) (PLA) and its copolymers with glycolic acid (PLGA) is preferred, since their degradation can be adjusted by the composition and crystallinity. In addition, the degradation products are nontoxic and readily metabolized in the human body. Their surface hydrophobicity is a general drawback restraining broader application, together with the lack of functional groups on the surface preventing directed adhesion and targeted drug delivery. Recent works dealing with tissue scaffolds reported that such thermoplastic biodegradable polymers like PLA, PLGA, and PGA, accepted for in vivo use, all need some kind of activation techniques to modulate cell response.2-4 Effective pharmaceutical application also requires adjustment of surface properties of PLA and PLGA colloidal particles to make them compatible with the biological environment.5 Reduced protein adsorption and prolonged circulation time have been already demonstrated, *To whom correspondence should be addressed. Telephone: 36 1-3722500/1308. Fax: 36 1-372-2592. E-mail:
[email protected]. (1) Ratner, B. D.; Hoffman, A. S.; Schoen, F. J.; Lemons, J. E. Biomaterials Science, An Introduction to Materials in Medicine, 2nd ed.; Academic Press: San Diego, 2004. (2) Atthoff, B.; Hilborn, J. J. Biomed. Mater. Res., Part B 2007, 80B, 121. (3) Djordjevic, I.; Britcher, L. G.; Kumar, S. Appl. Surf. Sci. 2008, 254, 1929. (4) Wan, Y.; Qu, X.; Lu, J.; Zhu, C.; Wan, L.; Yang, J.; Bei, J.; Wang, S. Biomaterials 2004, 25, 4777. (5) Santander-Ortega, M. J.; Jodar-Reyes, A. B.; Csaba, N.; Bastos-Gonzalez, D.; Ortega-Vinuesa, J. L. J. Colloid Interface Sci. 2006, 302, 522.
1440 DOI: 10.1021/la903373g
however, for poly(ethylene glycol) (PEG)-covered polymeric nanoparticles.6-9 Techniques developed to immobilize PEG on PLA or PLGA surfaces were based so far on physical interactions including adsorption of PEG or PEG-containing block copolymers and blend formation.10-14 There is no report, to our best knowledge, on coupling of PEG by stable chemical bond to a PLGA surface resulting in a well-defined PEG-modified surface with desired composition and properties. In this work, we describe a novel way of chemical coupling of linear PEG and starlike PEG compounds to PLGA surfaces following amino functionalization as a first step. The primary and secondary amino groups were introduced to PLGA chains according to the controlled aminolysis described by Croll and co-workers.15 Activated linear PEG molecules were reacted with the surface amino groups directly, while the star-PEG was immobilized through a bifunctional glutaraldehyde serving as a mediating agent. The chemical composition of the surface layer was characterized by X-ray photoelectron spectroscopy (XPS), whereas the effect of surface modification was evaluated by wettability and protein adsorption measurements.
Experimental Section Poly(lactic/glycolic acid) (PLGA) (Mw: 40 000-75 000) with an 85/15 component ratio of lactide/glycolide was obtained from (6) Li, J. T.; Caldwell, K. D.; Rapoport, N. Langmuir 1994, 10, 4475. (7) Gref, R.; Minamitake, Y.; Peracchia, M. T.; Domb, A.; Trubetskoy, V.; Torchilin, V.; Langer, R. Pharm. Biotechnol. 1997, 10, 167. (8) Yuancai, D.; Si-Shen, F. J. Biomed. Mater. Res., Part A 2006, 78A, 12. (9) Farokhzad, O. C.; Cheng, J.; Teply, B. A.; Sherifi, I.; Jon, S.; Kantoff, Ph. W.; Richie, J. P.; Langer, R. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 6315. (10) Kingshott, P.; Thissen, H.; Griesser, H. J. Biomaterials 2002, 23, 2043. Bertoti, I.; Vargha-Butler, E. I. J. Colloid Interface Sci. 2002, 245, 91. (11) Kiss, E.; (12) Boillot, P.; Ulrich, N.; Sommer, F.; Duc, T. M.; Loeffler, I.-Ph.; Dellacherie, E. Int. J. Pham. 1999, 181, 159. (13) Shi, Q.; Ye, S.; Kristalyn, C.; Su, Y.; Jiang, Z.; Chen, Z. Langmuir 2008, 24, 7939. Dravetzky, K.; Hill, K.; Kutnyanszky, E.; Varga, A. J. Colloid (14) Kiss, E.; Interface Sci. 2008, 325, 337. (15) Croll, T. I.; O’Connor, A. J.; Stevens, G. W.; Cooper-White, J. J. Biomacromolecules 2004, 5, 463.
Published on Web 01/08/2010
Langmuir 2010, 26(3), 1440–1444
Kiss et al.
Letter
Figure 1. Coupling of linear activated PEG (lin-PEG) to amino functionalized PLGA surface.
Figure 2. Coupling of star-PEG molecule to aldehyde functionalized PLGA surface. Table 1. Advancing and Receding Water Contact Angles (ΘA and ΘR) with Standard Deviation (SD) Measured on the Polymer (PLGA) and Modified Polymer Surfacesa PLGA ΘA
aminated PLGA ΘR
ΘA
ΘR
lin-PEG-PLGA ΘA
ΘR
star-PEG-PLGA ΘA
ΘR
adsorbed PEG ΘA
ΘR
contact angle 74.4 58.3 70.4 49.5 68.2 36.0 69.8 37.7 75.1 54.8 SD 1.0 1.9 2.2 5.5 2.7 4.4 2.4 4.3 1.7 3.1 Θ 66.6 60.5 53.8 55.3 65.5 τ (mN/m) 28.5 35.4 42.4 40.9 30.1 a Data for PLGA surface with physically adsorbed PEO are also shown. Wetting tension (τ = γ cos Θ) was calculated from the average contact angle: cos Θ = (cos ΘA þ cos ΘR)/2.
Sigma (Germany). PLGA films of a few micrometer thicknesses were prepared on glass or silicon substrates by solvent casting from a dichloromethane solution of 0.1 wt/vol %. Aminolysis of the surface was conducted by immersion into aqueous solution of ethylene diamine at a concentration of 0.05 M that was then allowed to react at 20 °C for 30 or 60 min according to the method described by Croll et al.15 For modification of this aminolyzed PLGA surface, two types of activated poly(ethylene glycol) were used. One is an activated linear PEG, that is, O-[2-(N-succinimidyloxicarbonyl)-ethyl]O0 -methylpolyethylene glycol (lin-PEG) (Mw: 5000) obtained from Fluka (Hungary). The other one is an amino terminated six-arm poly(ethylene glycol) with a dipentaerythritol core (star-PEG) (Mw: 12 000) purchased from Polymersource (Canada) introduced recently for surface modification.16-18 Lin-PEG was coupled directly to the surface amine groups in phosphate buffer (pH = 7.4) solution (10 mg/mL) at room temperature allowing 45 min interaction (Figure 1). Before coupling of star-PEG, the aminated PLGA films were treated with glutaraldehyde (0.5 vol %) in phosphate buffer solution (pH = 7.4) at room temperature for 3 h. Then NaCNBH3 reducing agent was added to the solution and allowed to react for 20 min. For the coupling of star-PEG, the modified PLGA film was immersed into the phosphate buffer solution (pH = 7.4) of star-PEG (10 mg/mL) at room temperature for 2 h (Figure 2). After a standard rinsing procedure (washing the samples one by one in fresh 50 mL of distilled water for 5 min three times), the films were dried overnight before further investigation. The hydrophobic/hydrophilic nature of the unmodified PLGA film and the chemically modified surfaces was characterized by wettability measurements. Initially, a 10 μL drop of water was (16) Irvine, D. J.; Mayes, A. M.; Satija, S. K.; Barker, J. G.; Sofia-Allgor, S. J.; Griffith, L. G. J. Biomed. Mater. Res. 1998, 40, 498. (17) Groll, J.; Ameringer, T.; Spatz, J. P.; Moeller, M. Langmuir 2005, 21, 1991. (18) Lensen, M. C.; Mela, P.; Mourran, A.; Groll, J.; Heuts, J.; Rong, H.; Moeller, M. Langmuir 2007, 23, 7841.
Langmuir 2010, 26(3), 1440–1444
Table 2. Chemical Composition (atom %) of the Surface Layer of PLGA, Aminated PLGA, and PEG-Coupled Samples Determined by XPS
PLGA aminated PLGA lin-PEG-PLGA star-PEG-PLGA
C 1s (284.8 eV)
O 1s (533.0 eV)
N 1s (400.0 eV)
O 1s/ C 1s
59.4 60.3 64.1 72.2
40.6 39.2 35.4 26.9
0.5 0.5 0.9
0.68 0.65 0.55 0.37
placed on the polymer surface, and the advancing and receding angles were determined goniometrically (Dataphysics, OCA15, Germany) by increasing and decreasing the volume of the drop using a motor-driven Hamilton micropipet. For the contact angle determination, 8-12 measurements were made on each sample and a minimum of three films of each composition were measured. The chemical composition of the surface layer was determined by X-ray photoelectron spectroscopy (XPS; Kratos XSAM800type spectrometer, using MgKR1,2 radiation). Data processing was performed by applying the Kratos “VISION 2000” program. The overview spectra were taken between 50 and 1300 eV with an energy step of 0.5 eV, while the detailed spectra of the peaks of interest (C 1s, O 1s, and N 1s) were recorded with an energy step of 0.1 eV. The overlapping peaks were resolved by the peak synthesis method, applying Gaussian peak components of equal widths after Shirley-type background subtraction. In order to evaluate protein adsorption properties of the PEGcoupled and reference films, they were immersed into a bovine serum albumin (BSA) solution with a concentration of 1.2 mg/mL for 30 min. Following rinsing and drying, the samples were analyzed by XPS. Morphological characterization of the untreated and surface modified PLGA was performed by atomic force microscopy (AFM) in contact mode (PSIA XE-100, Park Sci. South Korea). DOI: 10.1021/la903373g
1441
Letter
Kiss et al.
Figure 3. Carbon XPS signal of lin-PEG-coupled PLGA (a) and star-PEG-coupled PLGA (b) surfaces. Synthetic C 1s components correspond to aliphatic carbon at 284.8 eV, the carbon connected to and constituting the ester group of PLGA at 286.8 and 288.9 eV, and also the etheric carbon of PEG (filled) at 286.1 eV. Table 3. Various Components of C 1s and O 1s Peaks (in atom %) of XPS Spectra Obtained for PLGA, Aminated PLGA, and PEG-Coupled Samples
PLGA aminated PLGA lin-PEG-coupled star-PEG-coupled
C-H (284.8 eV)
C-O (286.8 eV)
CdOO (288.90 eV)
20.0 22.1 28.5 36.1
19.7 19.1 15.1 12.8
19.7 19.1 15.1 12.8
C-OPEG (286.1 eV)
OdC (532.1 eV)
O-C (533.5 eV)
5.7 10.5
20.4 20.0 18.7 14.4
20.2 19.2 16.7 12.5
Figure 4. N 1s XPS signal of PLGA and star-PEG-coupled surface following BSA adsorption (the signal for lin-PEG-coupled surface similar to the star one is not shown for clarity).
Results and Discussion The hydrophobic/hydrophilic character of the pristine and modified PLGA polymer layers was determined by water contact angle measurements, and the results are summarized in Table 1. 1442 DOI: 10.1021/la903373g
According to the data, the amination already resulted in a small increase in the wettability of the PLGA surface, while coupling of PEG compounds decreased substantially the hydrophobicity of the polymer surface. The advancing contact angles show a small but significant decrease, while the change in the receding values is Langmuir 2010, 26(3), 1440–1444
Kiss et al.
Letter
Figure 5. Three-dimensional images of PLGA (a) and PEG-coupled PLGA (b) surfaces before and after BSA adsorption (c and d) obtained by contact mode atomic force microscopy. Typical cross section profiles are included to show the different vertical scales.
more pronounced, indicating that the surface is enriched in hydrophilic domains. Indeed, the hydrophilic character of the PEG chains is involved in the hydration process at water contact.16 The effect of this interaction is detected more sensitively by the receding angles.19 The increase of surface hydrophilicity can also be expressed by the increase of the wetting tension (τ = γ cos Θ) calculated from the contact angles (cos Θ = (cos ΘA þ cos ΘR)/2) and surface tension of water (γ). For the PEG-modified PLGA surface, the wetting tension was equal to or greater than 40 mN/m irrespective of the type of PEG applied, exceeding the required value declared for hydrophilic and nonadsorbing biomaterials.20 Wettability measurements revealed the difference between the cases when the PEG was immobilized by chemical coupling and when the PEG molecules were just physically adsorbed on the surface. Contact angles measured on PLGA samples treated under the same conditions as it were coupled but using not activated (19) Johnson, R. E., Jr.; Dettre, R. H. In Contact Angle Wettability and Adhesion; Fowkes, F. M., Ed.; Advances in Chemistry Series; American Chemical Society: Washington, DC, 1964; Vol 43, p 112. (20) Vogler, E. A. In Wettability; Berg, J., Ed.; Surfactant Science Series; Marcel Dekker: New York, 1993; Vol. 49, pp 183-250.
Langmuir 2010, 26(3), 1440–1444
PEG are included in the last column of Table 1. The comparison with data obtained for chemically modified PLGA showed that the adsorbed PEG would not affect ΘA and may only slightly decrease ΘR. This latter change is negligible compared with the measured beneficial influence of the chemically coupled PEG. The chemical composition of the polymer surface layers was determined by XPS, recording the overview and detailed spectra at the carbon, oxygen, nitrogen, and silicon regions as well. The overview spectra indicated only the presence of carbon, oxygen, and nitrogen originating from the polymer films: no signal from the glass substrates could be detected, which means that the polymer always formed a continuous film with a thickness exceeding at least 15 nm (Table 2). The oxygen/carbon atomic ratio calculated from the chemical composition is expected to be 0.71 and 0.50 for the PLGA and pristine PEG, respectively. The O/C ratio obtained from the XPS data is very close to the expected values for untreated PLGA, and its decreasing trend is also reasonable for the PEG-modified surfaces (Table 2). The deconvolution of carbon and oxygen peaks provided further evidence of the unambiguous presence of PEG and its content in the surface layer. We have characterized the C1s XP DOI: 10.1021/la903373g
1443
Letter
spectra of different PLGA earlier in detail, assigning signals at 288.9 and 286.8 eV to the carbon in the carboxylic group (CdOO) and to the neighboring carbon in the chain (C-O) of the polymer, respectively.11,21 The C-H component, serving also as charge reference, was set at 284.8 eV. These are in good agreement with the most reliable published values.22 Performing the peak fitting procedure accurately, by fixing the positions and by applying equal widths of the components, we found that the C-O carbon signal originating from the PEG ether units, named as C-OPEG, appeared at a lower binding energy than the C-O carbon related to the carboxylic group. This well detectable 0.7 eV difference allowed to us distinguish unambiguously the (C-O) signal of the PLGA from that of the PEG in the surface layer as demonstrated in Figure 3. In accordance with these results, the chemical state and, consequently, the binding energy values of the O 1s peak components were also different in the carboxyl group of the polymer (OdC-O) and in the ether group of PEG. The chemical shift between these components was smaller, thus making the deconvolution somewhat uncertain. The contribution of the O 1s signal of PEG is causing a difference of the otherwise equal intensities of the oxygen peak components of PLGA. The results of XPS analysis are summarized in Table 3. The chemical composition of the PLGA was reasonably well reflected in quantitative XPS analysis: the ratio of the three structurally different carbon atoms and that of the corresponding oxygen atoms was obtained very close to 1:1:1:1:1, reproducing the atomic ratio in the structural units of the polymer. The presence of the PEG in the surface layer is demonstrated by the appearance of C-OPEG components with 5.7 and 10.5 atom % for lin-PEG- and star-PEG-coupled surfaces, respectively. The surface coverage scaling with the amount of PEG can be represented by the ratio of monomer units in the surface layer: nPEG/ nPLGA. This value is equal to 0.2 for linear PEG and 0.4 in the case of star-PEG, taking into account the fact that two carbon atoms are in each monomer unit of PEG. In addition to higher surface coverage achieved by star-PEG, these molecules offer an advantage by providing a high number of functional amine groups on the surface, allowing further coupling which may be beneficial for pharmaceutical application. Lam, C. N. C.; Duc, T. M.; Vargha-Butler, E. I. Prog. Colloid (21) Kiss, E.; Polym. Sci. 2001, 117, 167. (22) Beamson, G.; Briggs, D. High resolution XPS of Organic Polymers; The Scienta ESCA300 Database; John Wiley and Sons, Ltd.: Chichester, England, 1992.
1444 DOI: 10.1021/la903373g
Kiss et al.
The adsorption of protein on the polymer surfaces was detected by the intense N 1s signals depicted in Figure 4. The amount of protein adsorbed onto the unmodified PLGA is represented by 9.1% nitrogen in the surface layer. The hydrophilization of PLGA by PEG coupling resulted in considerable suppression of protein adsorption, 4.75% and 4.54% nitrogen was detected for lin- and star-PEG-modified samples, respectively. The actual intensity difference is even higher for the case of star-PEG, if considering the contribution of its amine groups (0.9%). AFM images (Figure 5) showed that the polymer surfaces are rather smooth with a roughness (Ra calculated for the whole area of several images) in the range of 0.1 nm. The aminolysis and further chemical reactions did not affect significantly the surface topography, manifesting that the modification reactions are restricted to the outmost surface molecules (Figure 5a and b). A somewhat higher roughness was detected after adsorption of BSA, typically 0.3 nm. The increase of the surface roughness of the PEG-coupled PLGA samples detected after BSA treatment was smaller (Figure 5d), indicating that a reduced amount of protein was adsorbed in this case.
Conclusions Summarizing the above results, it was shown that both linear and star-PEG could be covalently coupled to the almost atomically smooth PLGA film surfaces by well-defined soft chemical routes. This coupling led to a considerable increase in the hydrophilic character and wetting tension of the surface. The XPS results supported the chemical attachment of the PEG chains. Quantitative analysis proved that a significant amount of PEG could be bound to PLGA by applying linear PEG but an even higher quantity can be attached when the PLGA surface is reacted with star-PEG. In line with the attachment of hydrophilic groups, significantly reduced protein adsorption was measured on the PEG treated PLGA film surfaces as compared to the untreated one. As an additional benefit, the application of starPEG offers a number of amine groups on the surface allowing attachment of “directing” molecules for potential drug carrier applications. Acknowledgment. Financial support of the Hungarian Research Foundation (OTKA K68120 and K60197) and National Office for Research and Technology (GVOP 2004/99) is acknowledged.
Langmuir 2010, 26(3), 1440–1444
