Langmuir 2009, 25, 865-872
865
Adsorbed r-Helical Diblock Copolypeptides: Molecular Organization, Structural Properties, and Interactions Bayu Atmaja,† Jennifer N. Cha,‡,§ and Curtis W. Frank*,† Department of Chemical Engineering, Stanford UniVersity, Stanford, California 94305, and IBM Almaden Research Center, San Jose, California ReceiVed June 23, 2008. ReVised Manuscript ReceiVed September 29, 2008 In this work, we have developed 11-mercaptoundecanoic acid (MUA)-polypeptide “bilayer” systems by adsorbing poly(diethylene glycol-L-lysine)-poly(L-lysine) (PEGLL-PLL) diblock copolypeptide molecules of various architectures onto MUA-functionalized gold substrates. An objective of our present work is to use the PEGLL-PLL/MUA bilayer as a model system for studying the interfacial phenomena that occur when PEGLL-PLL molecules interact with carboxylic acid (COOH) moieties of nanoparticle ligands. Specifically, we have elucidated the nature of the interactions between the PEGLL-PLL and COOH moieties as well as the resulting polypeptide conformation and organization, using a combination of surface techniquessgrazing-incidence IR spectroscopy, ellipsometry, and contact angle. We have also thoroughly characterized other film properties such as the packing and graft density of the polypeptide molecules as a function of the PEGLL-PLL architecture. From the IR data, the adsorption process occurs primarily by means of electrostatic interaction between the protonated PLL residues (pKa ≈ 10.6) and carboxylate moieties of the MUA self-assembled monolayer (SAM) (pKa ≈ 6) that is enhanced by H-bonding. The PLL block is thought to adopt a random-coil (extended) conformation, while the PEGLL block that is not interacting with the MUA molecules is found to adopt an R-helical conformation with an average tilt angle of ∼60°. The PEGLL-PLL molecules have also been deduced to form a heterogeneous film and adopt liquidlike/disordered packing on the surface. The average contact angle of the MUA-polypeptide bilayer systems is ∼40°, which implies that the diethylene glycol (EG2) side chains of the PEGLL residues may be oriented somewhat toward the surface normal. From ellipsometry measurements, it is found that PEGLLx-PLLy molecules with a longer R-helical block are associated with a lower graft density on the MUA surface compared to those with a shorter R-helical block. This observation may be attributed to the greater repulsionssteric and H-bonding effectssthat is imposed by the EG2 side chains found on and projected area occupied by the longer PEGLL block. The bilayer systems have been found to be extremely stable over a 2-week period with no changes in the contact angle, thickness, polypeptide tilt angle, or conformation. Beyond that, there is a gradual decrease in the thickness and increase in the contact angle of the bilayer that could be attributed to the oxidation of the MUA SAM molecules.
Introduction In this work, we present a simple method to modify a gold (Au) substrate with poly(diethylene glycol-L-lysine)-poly(L-lysine) (PEGLL-PLL) diblock copolypeptide (Figure 1) that generates a “bilayer” structure (Figure 2). Specifically, the Au surface is first functionalized with 11-mercaptoundecanoic acid self-assembled monolayer (MUA SAM) prior to adsorption of the PEGLL-PLL molecules. The presence of multiple diethylene glycol (EG2) side chains along the PEGLL block and its welldefined secondary structure1-3 may offer superior nonfouling properties similar to those observed for grafted PEG brushes and OEG-containing molecules on surfaces. The use of poly(L-lysine) (PLL) as one of the blocks in the diblock copolypeptide molecule is desirable because its positively charged side chains impart the functionality to “anchor” the molecule onto the negatively charged surface simply by means of electrostatic attraction. In addition, the amine residues of the PLL block can be reacted with various * To whom correspondence should be addressed at 381 North-South Mall, Rm 111, Stauffer III, Stanford, CA 94305. Phone (650) 723-4573; fax (650) 723-9780; e-mail
[email protected]. † Stanford University. ‡ IBM Almaden Research Center. § Current Address: Department of Nanoengineering and Materials Science, 9500 Gilman Drive, M/C 0448, University of California, San Diego, La Jolla, CÁ 92093-0448.
(1) Rixman, M. A.; Dean, D.; Ortiz, C. Langmuir 2003, 19, 9357–9372. (2) Harder, P.; Grunze, M.; Dahint, R.; Whitesides, G. M.; Laibinis, P. E. J. Phys. Chem. B 1998, 102, 426–436. (3) Morra, M. J. Biomater. Sci., Polym. Ed. 2000, 11, 547–569.
Figure 1. (a) Chemical structures of poly(diethylene glycol-L-lysine)poly(L-lysine) diblock copolypeptide (PEGLL-PLL) (top) and 11mercaptoundecanoic acid (MUA) (bottom). Nε moieties of the poly(Llysine) backbone are functionalized with diethylene glycol (EG2) side chains to synthesize the PEGLL block. (b) Schematic showing the secondary structure of the PEGLL-PLL molecule. In aqueous solution at pH ≈ 8.2, the PLL block is positively charged and adopts a random coil conformation; the PEGLL block is hydrophilic and uncharged and adopts an R-helical conformation represented by a rigid rod.6,7 The multiple EG2 side chains along the length of each R-helix are envisioned to constitute a “sheath” around the PEGLL block.
molecules in order to impart additional functionalities into and/ or alter the properties of the system.4,5 Furthermore, it is a flexible method that is suitable for many different charged substrates such as metal oxides, SAM-functionalized surfaces, and silica. (4) Langer, R.; Tirrell, D. A. Nature 2004, 428, 487–492. (5) Barrera, D. A.; Zylstra, E.; Lansbury, P. T.; Langer, R. J. Am. Chem. Soc. 1993, 115, 11010–11011.
10.1021/la801973x CCC: $40.75 2009 American Chemical Society Published on Web 12/24/2008
866 Langmuir, Vol. 25, No. 2, 2009
Figure 2. (a) Schematics of the MUA-polypeptide bilayer formed on Au surface. The alkyl chains of the MUA SAM form a relatively ordered packing on the Au surface with tilt angle ∼30°.8-10 Upon adsorption, the PLL block is thought to adopt an extended (random coil) conformation while the PEGLL block is found to adopt an R-helical conformation (b) Depiction of molecular interactions that exist within the MUA-polypeptide bilayer. (---) H-bonding. Electrostatic interaction that is stabilized by H-bonding occurs between the carboxylate moieties of the MUA and the protonated PLL side groups. H-bonding also occurs between the remaining terminal groups of the MUA that are protonated and the PLL side groups.
It is important to note, however, that the evaluation of this MUA-polypeptide bilayer system as a nonfouling surface is reserved for future investigation and will not be addressed here. One objective of our work is to use the PEGLL-PLL/MUA bilayer as a model system for understanding the interfacial phenomena that occur when PEGLL-PLL molecules interact with carboxylic acid (COOH) moieties of nanoparticle citrate ligands. In our previous work, we have hypothesized that specific noncovalent interactions between the PEGLL-PLL molecules and COOH functional groups are responsible for the PEGLLPLL/nanoparticle ultimate assembly into supramolecular structures.7 Here, we intend to use the bilayer system as a model to elucidate and hence verify our previous hypotheses on the molecular origin of the interactions between PEGLL-PLL molecules and COOH moieties as well as the molecular organization and conformation adopted by the former as a result of the interactions. SAMs on surfaces have been extensively used as versatile model systems for understanding interactions such as H-bonding and electrostatic that occur in complex interfacial phenomena.11-13 In particular, we have chosen MUA (6) Yu, M.; Nowak, A. P.; Deming, T. J.; Pochan, D. J. J. Am. Chem. Soc. 1999, 121, 12210–12211. (7) Atmaja, B.; Cha, J. N.; Marshall, A.; Frank, C. W. Langmuir doi: 10.1021/ la801848d. (8) Smith, E. L.; Alves, C. A.; Anderegg, J. W.; Porter, M. D.; Siperko, L. M. Langmuir 1992, 8, 2707–2714. (9) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321–335. (10) Yam, C. M.; Zheng, L.; Salmain, M.; Pradier, C. M.; Marcus, P.; Jaouen, G. Colloids Surf., B 2001, 21, 317–327. (11) Sun, L.; Crooks, R. M.; Ricco, A. J. Langmuir 1993, 9, 1775–1780. (12) Sun, L.; Kepley, L. J.; Crooks, R. M. Langmuir 1992, 8, 2101–2103.
Atmaja et al.
as the SAM molecule because the monolayer has been shown to be robust and relatively stable; moreover, its organization and structure have been well-studied.8,9,14-16 In addition, the pKa of its COOH terminal group is reported to range between ∼6 and 6.5,17,18 which is roughly the pKa (∼6.4) of the COOH moiety of the citrate ligand that has been implicated in the interaction with the PEGLL-PLL molecule.7,19 As such, we can use the MUA monolayer as a mimic for the COOH moieties of the citrate ligands. In this work, we also intend to thoroughly characterize the film propertiessfor example, packing, contact angle, and graft densitysas a function of the PEGLLx-PLLy architecture at varying values of x and y. This information could be useful for our future work to develop a correlation between the molecular structure of the film and its ability to control protein adsorption. A variety of surface techniques have been used to obtain comprehensive information on the equilibrium molecular structure and properties of the polypeptide film; that is, we do not intend to investigate the kinetics of the adsorption of the PEGLLPLL molecules (see Materials and Methods). Specifically, grazing-incidence infrared (IR) spectroscopy is used to investigate the polypeptide conformation and orientation (tilt angle) as well as the molecular interactions between MUA and PEGLL-PLL molecules. Contact-angle measurements are useful in elucidating the packing of polypeptide chains and surface energies of the films. In addition, ellipsometry measurements are used to monitor changes in the graft density of polypeptide molecules as a function of diblock copolypeptide concentration and architecture. These measurements are repeated at somewhat regular intervals over a period of a month in order to determine the stability of the film over time.
Materials and Methods Preparation of Gold Substrates. In this work, silicon wafers (Silicon Quest International) were used as the substrates. These substrates were first degreased with acetone and then cleaned rigorously with piranha solution: the wafers were immersed in a mixture of 3:1 concentrated H2SO4/30% H2O2 for 10 min while it was hot. They were then rinsed thoroughly with Milli-Q water and dried with N2. Thermal evaporation of Au (200 nm) using high-purity Au pellets (99.999% purity, Kurt J. Lesker) was performed immediately after the cleaning procedure described above. A Cr layer (2 nm) was evaporated prior to the Au deposition to improve adhesion. The evaporated substrates were then cooled to ∼45 °C prior to their removal from the evaporation chamber. Functionalization of 11-Mercaptoundecanoic Acid Self-Assembled Monolayer. Upon removal from the evaporator, the Aucoated substrate was immediately immersed in a solution of 1 mM 11-mercaptoundecanoic (MUA, Sigma-Aldrich) for ∼10-12 h. The SAM-functionalized substrate was then rinsed with copious amounts of ethanol (200 proof, Gold Shield) followed by Milli-Q water. Note that the pH of the Milli-Q water was measured to range between 6.2 and 7.2. It was then dried thoroughly with N2 and stored in a desiccator until it was modified further with polypeptides. Syntheses and Characterization of Polypeptides: Poly(diethylene glycol-L-lysine)-Poly(L-lysine) and Poly(L-lysine). We followed the protocol for synthesis of the PEGLLx-PLLy diblock copolypeptide as described previously.7 Poly(L-lysine) homopolymer (13) Nath, N.; Hyun, J.; Ma, H.; Chilkoti, A. Surf. Sci. 2004, 570, 98–110. (14) Chidsey, C. E. D.; Loiacono, D. N. Langmuir 1990, 6, 682–691. (15) Jordan, C. E.; Frey, B. L.; Kornguth, S.; Corn, R. M. Langmuir 1994, 10, 3642–3648. (16) Hutt, D. A.; Leggett, G. J. J. Phys. Chem. 1996, 100, 6657–6662. (17) Bain, C. D.; Whitesides, G. M. Langmuir 1989, 5, 1370–1378. (18) Jordan, C. E.; Corn, R. M. Anal. Chem. 1997, 69, 1449–1456. (19) Cha, J. N.; Birkedal, H.; Euliss, L. E.; Bartl, M. H.; Wong, M. S.; Deming, T. J.; Stucky, G. D. J. Am. Chem. Soc. 2003, 125, 8285–8289.
Arrangements and Properties of Adsorbed Polypeptides (Mw ) 15 000, average degree of polymerization ) 72) was purchased from Sigma-Aldrich and used directly. Adsorption of Polypeptides onto MUA-Functionalized Surface. Adsorption of the PEGLLx-PLLy molecule onto the MUA SAM results from interactions between the lysine residues and COOH moieties of the MUA molecules, while the PEGLL residues do not interact directly with the SAM (see Discussion). Therefore, all polypeptide concentrations are reported in terms of the concentrations of lysine residues present in the bulk solutions. This way, by using the same concentration of lysine residues in various PEGLLx-PLLy solutions, we have essentially maintained a constant driving force for diblock copolypeptide molecules to occupy the available binding sites (SAM functional groups) on the substrate. In this case, any differences in the observed film structure and properties can be attributed to factors that stem from the differences in the diblock copolypeptide architecture, that is, lengths of the PEGLL (x) and PLL blocks (y). Polypeptide was first dissolved in Milli-Q water to make a 1-2 mg/mL solution, and it was then diluted with Milli-Q water to achieve the desired concentration of the lysine residues in a final volume of 3.5 mL. For example, a target concentration of 1.3 mM lysine residues (final volume of 3.5 mL) for the PEGLL11-PLL90 (Mw ) 26 000) system will require a total amount of ∼1.3 mg of the polypeptide. If we start with a 1 mg/mL polypeptide solution, we need 1.3 mL of this solution, which is then diluted with water to make up a final volume of 3.5 mL. The pH of this polypeptide solution was then adjusted with 0.1 M NaOH (J.T. Baker) to a value between 8.2 and 8.3. This range of pH values corresponds to that at which supramolecular assembly between the PEGLL-PLL molecules and nanoparticles occurs (see Introduction). As mentioned above, in this work, we do not intend to investigate the kinetics of adsorption of polypeptide molecules. We instead focus on the equilibrium properties of the polypeptide film, which is presumably achieved when the maximum surface coverage is reached for a particular bulk concentration. From the work on the adsorption of PLL homopolymers onto MUA-functionalized Au substrates at pH 8.5, it is determined that full monolayer coverage of the PLL has been achieved when the exposure time to PLL solution is 30 min and the PLL molecular weight is g14 000.15,18,20 In these experiments, the bulk concentration of PLL solution was fixed at either 0.7 or 1 mM lysine residues. Considering the relatively large molecular weights (g24 000) of PEGLL-PLL molecules and the high polypeptide concentrations (g1.4 mM lysine residues) that we employ here, it is then likely that the maximum graft densities have been achieved when the SAM-functionalized substrates are exposed to the polypeptide solutions for g30 min. Indeed, when the adsorption time was varied between 30 and 50 min for either of the architectures, PEGLL11-PLL90 and PEGLL42-PLL47, each with a bulk concentration of 1.4 mM lysine residues, we found that the thickness of the PEGLL-PLL film remained unchanged, as determined by ellipsometry. Accordingly, in this work, the SAM-functionalized substrate was immersed in the polypeptide solution for ∼40 min in order to ensure that equilibrium is achieved for each bilayer system. The MUA-polypeptide bilayer sample was then rinsed with copious amounts of Milli-Q water and dried thoroughly with a N2 gas stream. It was immediately placed under vacuum in a desiccator and kept there for 12-18 h before it was used in infrared (IR), ellipsometry, and contact-angle measurements. Reproducibility of Measurements. For each PEGLLx-PLLy system adsorbed at a particular bulk concentration, measurements were made over at least three different samples; the thickness, contact angle, and tilt angle values reported below are averages over these samples. For uncertainties in the thickness and contact angle values for a particular sample, please refer to the experimental description for the appropriate technique below. Grazing-Incidence Reflection-Absorption Infrared Spectroscopy. IR spectra were acquired on a Thermo Electron’s Nicolet 6700 optical spectrometer equipped with a grazing-incidence apparatus (SMART-SAGA, grazing angle of 80°) and a liquid N2(20) Frey, B. L.; Jordan, C. E.; Kornguth, S.; Corn, R. M. Anal. Chem. 1995, 67, 4452–4457.
Langmuir, Vol. 25, No. 2, 2009 867 cooled MCT detector. Prior to measurements, the SMART-SAGA chamber was continuously purged with dry air for ∼30 min, and the polypeptide was dried under vacuum overnight in a desiccator. The sample was further dried with a N2 stream before it was placed on the SMART-SAGA that held it in position for data acquisition. Each spectrum was an average over 600 scans and was acquired at a resolution of 4 cm-1 while being continuously purged with dry air. Background spectrum was acquired right before the sample spectrum on a coated (2 nm Cr, 200 nm Au) Si wafer, of which the evaporation was done in the same batch as that used for the sample. By use of an empirically determined factor, a water spectrum was subtracted from the sample spectrum to remove any residual water vapor features observed in the latter spectrum. No baseline correction and smoothing were performed on the acquired spectra before the peak positions and heights were determined. The tilt angle of the R-helical block of the diblock copolypeptide was calculated on the basis of the ratio of intensities of the amide I/amide II peaks, which were estimated from the corresponding peak heights. The equation used to calculate the tilt angle is as follows:21-23
2[(3 cos2 γ - 1)/2] [(3 cos2 θ1 - 1)/2] + 1 I1 )C I2 2[(3 cos2 γ - 1)/2] [(3 cos2 θ - 1)/2] + 1
(1)
2
where I1 and I2 represent the intensities of the amide I and II peaks, respectively; γ is the tilt angle of the R-helix axis from the surface normal; θ1 (39°) and θ2 (75°) are literature values of the angles of the transition moments of amides I and II, respectively, measured from the helix axis; and C is a proportionality constant. In this case, C is taken to be 1.5, which is a value commonly used in the literature.21,22,24 Ellipsometry. Ellipsometry measurements were made on a Gaertner model L116C with a wavelength of 632.8 nm and an incident angle of 70°. Optical constants of the Au substrate and thickness values corresponding to the different layerssMUA SAM and polypeptideswere computed with the Gaertner ellipsometer measurement program (GEMP). The average value of the set of optical constants (n and k) for a freshly evaporated Au substrate was determined by making measurements at five different spots on the surface prior to its functionalization with the MUA SAM. Immediately following adsorption of MUA SAM onto that particular substrate, this average value of the set of Au optical constants was used to determine the thickness of the SAM with the following properties: n ) 1.45 and k ) 0.8,14,20 Based on this value of the SAM thickness, the polypeptide thickness was then determined with n ) 1.5 and 1.5215,20 for the adsorbed PEGLLx-PLLy and PLL homopolymer systems, respectively. The refractive index of the former, 1.50, is taken as the average of the bulk refractive indexes of poly(ethylene glycol), which is reported to range between 1.47 and 1.48,25,26 and polylysine. For each SAM or MUA-polypeptide bilayer sample, more than five measurements were taken at different spots on the surface for which their values are
