Mapping the Interaction Forces between TAR RNA and TAT Peptides

The complexation of the HIV transactivation response element (TAR) RNA with the viral regulatory protein TAT is of enormous interest for the design of...
0 downloads 0 Views 513KB Size
Download PDF
1768

Langmuir 2006, 22, 1768-1774

Mapping the Interaction Forces between TAR RNA and TAT Peptides on GaAs Surfaces Using Chemical Force Microscopy Youngnam Cho† and Albena Ivanisevic*,†,‡ Department of Chemistry and Weldon School of Biomedical Engineering, Purdue UniVersity, West Lafayette, Indiana 47907 ReceiVed October 8, 2005. In Final Form: NoVember 24, 2005 The complexation of the HIV transactivation response element (TAR) RNA with the viral regulatory protein TAT is of enormous interest for the design of new sensing and therapeutic strategies. In this work, we anchored TAT peptides on GaAs surfaces using microcontact printing. Atomic force microscopy was used to quantify the interaction between TAR RNA and model TAT peptide sequences. Different pH conditions were utilized in order to assess specific vs nonspecific interactions. AFM tips functionalized with TAR RNA molecules were used to collect adhesion maps that displayed stronger interaction with peptide sequences that contained a greater number of arginine residues. All of the studies consistently showed a pH dependence of the interaction between the surface bound peptides and the TAR RNA on the AFM tips. This work quantifies the TAR RNA/TAT peptide interaction after one of the molecules is anchored on a surface. The conclusions in this paper are consistent with previous work and demonstrate that cationic residues are responsible for the polyelectrolyte-like affinity of TAT peptides for TAR RNA.

Introduction The HIV TAT protein has been widely studied and is known for its specific binding interactions.1 Investigators have shown that transcription of HIV RNA is not possible without the interaction of the virally encoded TAT protein with a transcriptional activator-responsive element (TAR).2 The TAR RNA element is a 59 base stem-loop structure that is located at the 5′-end of all nascent HIV-1 transcripts. This element contains a six-nucleotide loop as well as a three-nucleotide pyrimidine bulge.3 Several studies have shown that the trinucleotide bulge is responsible for the high affinity and specific binding to the TAT protein.4 Quantifying the affinity of the TAT sequence for TAR RNA is of great interest and importance when trying to design new drugs and therapeutic strategies.5,6 Despite the useful information derived from NMR studies,7 there is a need to introduce new methods to study the structure and strength of the interaction between TAR RNA and the TAT protein.8 Such methods should give data under aqueous conditions and provide information at the molecular level. Inorganic surfaces such as III-V semiconductor substrates seem very unlikely candidates to provide an alternative platform to study biological interactions. However, recent studies have shown proof-of-concept experiments toward the creation of hybrid organic-inorganic devices that utilize peptides and III-V semiconductors.9,10 III-V materials have optical, electrical, and mechanical properties that are essential for fabricating novel * Corresponding author. E-mail: [email protected]. † Department of Chemistry. ‡ Weldon School of Biomedical Engineering. (1) Long, K. S.; Crothers, D. M. Biochemistry 1995, 34, 8885-8895. (2) Kikuta, E.; Aoki, S.; Kimura, E. J. Am. Chem. Soc. 2001, 123, 7911-7912. (3) Runyon, S. T.; Puglisi, J. D. J. Am. Chem. Soc. 2003, 125, 15704-15705. (4) Huq, I.; Wang, X.; Rana, T. M. Nat. Struct. Biol. 1994, 4, 881-882. (5) Thoren, P. E. G.; Persson, D.; Esbjorner, E. K.; Goksor, M.; Lincoln, P.; Norden, B. Biochemistry 2004, 43, 3471-3489. (6) Hariton-Gazal, E.; Feder, R.; Mor, A.; Graessmann, A.; Brack-Werner, R.; Jans, D.; Gilon, C.; Loyter, A. Biochemistry 2002, 41, 9208-9214. (7) Olsen, G. L.; Edwards, T. E.; Deka, P.; Varani, G.; Sigurdsson, S. T.; Drobny, G. P. Nucleic Acids Res. 2005, 33, 3447-3454. (8) Wang, Z.; Shah, K.; Rana, T. M. Biochemistry 2001, 40, 6458-6464. (9) Whaley, S. R.; English, D. S.; Hu, E. L.; Barbara, P. F.; Belcher, A. M. Nature 2000, 405, 665-668. (10) Goede, K.; Busch, P.; Grundmann, M. Nano Lett. 2004, 4, 2115-2120.

devices and microelectromechanical systems (MEMS) such as microcantilever based platforms.11,12 Developing strategies to immobilize biomolecules on III-V surfaces is an important step toward the establishment of future biosensing and drug delivery approaches that rely on alternative detection platforms such as individually actuated micromechanical components.13 Alternative sensing approaches have already shown promise with respect to the investigation of the interaction of TAR RNA and TAT peptides.14 Thompson and co-workers have demonstrated the advantage of using an acoustic wave detection system to look at interactions of TAR RNA with drug molecules without the need for slow batch-based assays.15 We believe that new methodologies to modify III-V surfaces such as GaAs with TAT peptides16 can lead to chemically functionalized device architectures that are ready to be used in in vitro and in vivo experiments. However, before that becomes a reality, one needs to execute fundamental studies that will allow us to understand and quantify the interaction of TAT peptide molecules at the semiconductor interface. The knowledge acquired through such studies can become the basis for a reliable bio-detection platform. In this paper, we utilize chemical force microscopy to look at TAR RNA and TAT peptide interactions on functionalized GaAs surfaces. We used peptides rather than the whole protein since studies have shown that such synthetic peptides can bind to the TAR RNA with an affinity and specificity of the full length TAT protein.17 We describe a soft-lithography methodology to prepare a test GaAs surface using two different peptide fragments and characterize the substrate using atomic force microscopy (AFM), Fourier transform infrared reflectance absorption spectroscopy (FT-IRRAS), and X-ray photoelectron spectroscopy (XPS). Chemical force microscopy is used to quantify the interaction between areas of the surface patterned (11) Ukita, H. Opt. ReV. 1997, 4, 623-633. (12) Petroni, S.; Tripoli, G.; Vigna, B.; Vittorio, M. D.; Todaro, M. T.; Epifani, G.; Cingolani, R.; Passaseo, A. Appl. Phys. Lett. 2004, 85, 1039-1041. (13) Wang, X.; Bullen, D.; Zou, J.; Liu, C.; Mirkin, C. J. Vac. Sci. Technol. B 2004, 22, 2563-2567. (14) Tassew, N.; Thompson, M. Anal. Chem. 2002, 74, 5313-5320. (15) Tassew, N.; Thompson, M. Org. Biomol. Chem. 2003, 1, 3268-3270. (16) Cho, Y.; Ivanisevic, A. J. Phys. Chem. B 2005, 109, 12731-12737. (17) Long, K. S.; Crothers, D. M. Biochemistry 1999, 38, 10059-10069.

10.1021/la052729x CCC: $33.50 © 2006 American Chemical Society Published on Web 12/31/2005

Mapping Forces on GaAs Surfaces Scheme 1. Sequences of the Two Peptides Used in This Study

with different peptide sequences and an AFM tip functionalized with TAR RNA. Specific vs nonspecific interactions are investigated under different pH conditions. The findings show an increased interaction between TAR RNA and the peptide sequence that is rich in arginines. Our results are consistent with the notion that clusters of cationic residues can provide polyelectrolyte-like affinity for RNA.18 Experimental Section Materials. Single-crystal silicon doped n-type [100] GaAs wafers with a doping ratio of (0.5∼2) × 1018 cm-3 were purchased from University Wafer, Inc (Boston, MA). 3′-Mercaptopropyl-methoxysilane (MTS) and N-succinimidyl 3-(2-pyridyldithio)-propionate (SPDP) were obtained from Sigma. Dimethyl sulfoxide was purchased from Aldrich. Phosphate-buffered saline (PBS) at pH ∼6.5 and TRIS EDTA buffer at pH ∼8.0 were purchased from Sigma. The two peptide sequences, Scheme 1, were synthesized by Bio-Synthesis, Lewisville, TX where they were purified by HPLC and analyzed by MALDI-TOF. The amine modified TAR RNA sequence (5′-NH2-(CH2)6-GGC CAG AUC UGA GCC UGG GAG CUC UCU GGC C-3′) and biotinylated TAR RNA were synthesized, purified, and analyzed by Integrated DNA Technologies, Coralville, IA. The biotin attachment was done at the 5′ end. Standard solidphase chemistry protocols were used in both cases. 20 nm nanogoldstreptavidin conjugates were obtained from Nanoprobe, Yaphank, NY and diluted according to the manufacturer’s instructions. Surface Preparation and Modification. The GaAs wafers were cleaned with acetone, pure ethanol, and deionized water and subsequently dried with nitrogen gas. All GaAs substrates were etched with concentrated HCl for 1 min to remove the native oxide layer and provide an arsenic rich surface. The two different types of peptides were immobilized on the GaAs surface using a PDMS stamp. The stamp was fabricated and inked according to literature protocols.19 The buffered concentrations of the peptide inking solutions were kept at 1 mM. The stamp was brought in contact with the GaAs surface for ∼ 5 s to generate the pattern. After stamping, each surface was rinsed with buffer and pure water multiple times and subsequently dried with a stream of nitrogen. AFM Tip Modification. AFM tips were terminated on TAR RNA molecules using a modified procedure originally reported by Wangsa-Wirawan et al.20 This procedure is given step by step in the Supporting Information. Briefly, all tips were cleaned with isopropyl alcohol, ethanol, and pure water and subsequently dried with N2 gas. The surface of the cleaned tips was oxidized by immersion in a piranha solution (3:1 concentrated H2SO4: 30% H2O2, v:v) for 15 min. Caution: piranha solutions react Violently with organic solVents and should be handled with extreme care! Silanization of the oxidized tips was performed with MTS. The silanized tips were extensively rinsed with ethanol and toluene solutions. The thiol terminated surface of the tips was reacted with a heterobifunctional cross-linker, SPDP, for 1 h. Subsequently, the TAR RNA sequence (5′-NH2-(CH2)6GGC CAG AUC UGA GCC UGG GAG CUC UCU GGC C-3′) was covalently attached to the tip by a nucleophilic attack of the amine group. The steps of this tip modification were verified by X-ray photoelectron spectroscopy (XPS) and are presented in the Supporting Information. In addition, the quality of the tip surface after the last step was evaluated by SEM (see the Supporting Information). (18) Gelman, M. A.; Richter, S.; Cao, H.; Umezawa, N.; Gellman, S. H.; Rana, T. M. Org. Lett. 2003, 5, 3563-3565. (19) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498-1511. (20) Wangsa-Wirawan, N. D.; Ikai, A.; O’Neill, B. K.; Middelberg, A. P. J. Biotechnol. Prog. 2001, 17, 963-969.

Langmuir, Vol. 22, No. 4, 2006 1769 Force Measurements and Data Analysis. Force measurements were performed with a Nanoscope IIIa atomic force microscope equipped with a fluid cell. All of the studies were done in forcevolume (FV) mode. Data were collected initially in air, and subsequently, the FV measurements were repeated in buffered solutions with pH of 4, 7, and 10. In all experiments, we used the trigger mode and tips purchased from Veeco Instruments (model # OTR8-105). The spring constants were determined by a method reported by Poggi et al.21 We used the same FV imaging conditions for experiments with clean and RNA terminated tips. The Z travel distance was 150 nm, and we collected 16 points per force curve. Individual FV images were collected at 128 points per line. The topography images collected simultaneously with the FV data had resolution of either 128 or 256 points per line. The FV data, which are a collection of numerous deflection vs distance curves, were used to generate adhesion maps. The adhesion data was calculated using a custom-written MatLab code. The data in the code was processed using well-documented literature approaches that analyzed the retraction curves and took into account the appropriate spring constant and sensitivity values.22,23 The data from the adhesion maps were used to tabulate histograms of measured adhesion forces, as well as to calculate the mean adhesion forces and standard deviations when different parts of the surface were surveyed. Fourier Transform-Infrared Reflection Absorption Spectroscopy (FT-IRRAS). FT-IRRAS measurements were collected on all patterned surfaces. To survey specific areas of the substrate we used an instrument equipped with a motorized XYZ translation stage and coupled to a Continuum IR microscope. The microscope part of the instrument was enclosed in a home-built chamber that allowed us to purge the area around the sample stage with nitrogen prior to any experiments. We purged the home-built chamber with nitrogen for 1 h to accomplish sufficient water exclusion. All of the studies reported in this paper were done in a single reflection mode with a Thermo Nicolet 670 FTIR spectrometer (Madison, WI). The reflectance studies were done with p-polarized light at an incident of 80° from the GaAs surface normal. The instrument utilized a narrow-band mercury-cadmium-telluride (MCT) detector that allowed us to collect the reflected light. All data analysis was performed using the OMNIC software purchased from the instrument supplier. The resolution of all spectra shown in this paper is 4 cm-1, and in all cases, we collected 1024 scans per spectrum. Data were collected and analyzed on each sample from two regions: 2750∼3100 and 1500∼1850 cm-1. Prior to collecting spectra from the patterned surfaces, we acquired spectra from the bare GaAs substrate. In our data analysis, we subtracted the spectra of the bare GaAs surface from the one obtained from areas containing monolayers of peptides. The surfaces that we used for the FT-IRRAS experiments were also characterized by XPS. X-ray Photoelectron Spectroscopy (XPS). XPS studies were done on a commercial instrument, Kratos Axis ULTRA X-ray photoelectron spectrometer. In all of the XPS studies, initial survey spectra were taken from 0 to 1100 eV with pass energy of 160 eV. Subsequently, high-resolution scans of regions of interest were obtained with a pass energy of 40 eV. The data are plotted using Kalaidagraph and are based on the intensity obtained from the instrument. The XPS data was primarily utilized to verify the chemical composition of the different surfaces used in this study and are shown in the Supporting Information. Quantitative analysis of the spectra, needed in order to confirm the AFM tip modification and GaAs patterning, was performed with XPS PEAK (version 4.1). Details regarding the fitting procedure and data interpretation were previously published.16,24,25 (21) Poggi, M. A.; McFarland, A. W.; Colton, J. S.; Bottomley, L. A. Anal. Chem. 2005, 77, 1192-1195. (22) Poggi, M. A.; Lillehei, P. T.; Bottomley, L. A. Chem. Mater. 2005, 17, 4289-4295. (23) Plassard, C.; Lesniewska, E.; Pochard, I.; Nonat, A. Langmuir 2005, 21, 7263-7270. (24) Cho, Y.; Ivanisevic, A. J. Phys. Chem. B 2004, 108, 15223-15228. (25) Cho, Y.; Ivanisevic, A. J. Phys. Chem. B 2005, 109, 6225-6232.

1770 Langmuir, Vol. 22, No. 4, 2006

Cho and IVaniseVic

Results and Discussion The complexation of TAR RNA and the TAT protein is one of the most studied interactions. The basic question that researchers are trying to address is centered on the specificity and strength of binding of the arginine rich region of the TAT protein with the target RNA. Studies have shown that even a single arginine can facilitate the recognition event.26 However, the binding affinity can be maximized if additional positively charged amino acids are present on each side of the arginine residue. Therefore in our studies, we chose two test TAT peptide sequences, Scheme 1, and anchored them lithographically on GaAs. Both peptides were synthesized to contain a cysteine residue at their N-terminus which is necessary in order to form a covalent attachment27 with the arsenic rich surface of the semiconductor material used in our studies. Peptide A contains 15 amino acids and has six arginine residues. In addition, this sequence also contains two lysine residues that can be protonated at low pH. Peptide B contains seven amino acids and has only one arginine. None of the other residues in the peptide B sequence can be protonated at low pH. The interaction between TAT peptides and TAR RNA is also thought to be mediated by the conformations the two molecules can adapt.7 Because of this finding, we also had an additional reason for choosing the two peptides in Scheme 1. Our previous studies have shown that once anchored on GaAs surfaces these two peptides form monolayers with different properties.16 Peptide A forms more densely packed and ordered monolayers, whereas peptide B participates in monolayers with disordered alkyl chain packing. In the experiments described in this section, we validate that chemical force microscopy can be used to quantify the interaction of TAR RNA with the two test TAT peptide sequences. We carry out additional characterization to verify the composition of the lithographically patterned GaAs surface. Our studies directly compare the binding strength and specificity of the interaction of TAT peptides with TAR RNA. The experiments show that GaAs surfaces modified with biomolecules can be used as the basis for a sensitive platform to read-out biochemical events. Surface Preparation, Characterization, and Testing of Biological Activity. The test surfaces used in this study were fabricated by microcontact printing.28 The GaAs substrates were cleaned and “freed” of their oxide layer as described in the Experimental Section. The stamp was cleaned and inked according to literature protocols.29-31 We used a stamp with line patterns to first deliver peptide A on the surface and subsequently utilized a second stamp with identical line patterns to deliver peptide B on the substrate. The line features composed of each peptide were placed perpendicular to each other on the surface, Figure 1. The tapping mode height and phase AFM images revealed virtually no difference in the quality of the two patterns. To confirm the chemical composition of the each type of feature on the surface, we characterized the GaAs after the stamping using XPS and FT-IRRAS. The characterization techniques also aimed to assess any differences with respect to the arrangement of the two types of peptides on the semiconductor surface. All samples (26) Futaki, S.; Goto, S.; Suzuki, T.; Nakase, I.; Sugiura, Y. Curr. Top. Protein Peptide Sci. 2003, 4, 87-96. (27) Sheen, C. W.; Shi, J. X.; Martensson, J.; Parikh, A. N.; Allara, D. L. J. Am. Chem. Soc. 1992, 114, 1514-1515. (28) Xia, Y.; Rogers, J. A.; Paul, K. E.; Whitesides, G. M. Chem. ReV. 1999, 99, 1823-1848. (29) Bernard, A.; Delamarche, E.; Schmid, H.; Michel, B.; Bosshard, H. R.; Biebuyck, H. Langmuir 1998, 14, 2225-2229. (30) Bernard, A.; Renault, J. P.; Michel, B.; Bosshard, H. R.; Delamarche, E. AdV. Mater. 2000, 12, 1067-1070. (31) Kane, R. S.; Takayama, S.; Ostuni, E.; Ingber, D. E.; Whitesides, G. M. Biomaterials 1999, 20, 2363-2376.

Figure 1. AFM height (a) and phase (b) images of the patterned surface. The diagonal lines corresponding to peptide A and peptide B areas are labeled on the phase image only. The images were collected simultaneously with a scan rate of 3.0 Hz. The z scale is 5 nm in (a) and 30° of phase lag in (b).

used for the XPS and FT-IRRAS studies were prepared using GaAs substrates that contained a micron sized alignment marker that ensured that data was always collected from a specific region. Features with dimensions of at least ∼100 × 100 µm2 were used for this characterization. We utilized the scanning option of the XPS and FT-IRRAS instruments in order to obtain spectra from a specific location on the surface. XPS survey scans verified the presence of Ga, As, C, N, O, and S32,33 on surfaces patterned by both peptides. The highresolution spectra from each of these regions (see the Supporting Information) was deconvoluted in order to deduce the presence of various chemical species on the surface. When the carbon region was examined both types of patterns exhibited three distinct peaks at: 285 eV assigned to the hydrocarbon chains; 286 eV due to -C-N bonds; and 288 eV because of amide bonds.34 Examination of the N 1s core level spectra revealed the presence of amide, NH2, and NH3+ species.35 The sulfur region was carefully evaluated for all patterned surfaces in order to confirm that the peptides were covalently anchored on the GaAs surface. Both types of patterns showed evidence for the formation of thiolated species (As-S) based on peaks above 161 eV.36 We (32) Shaporenko, A.; Adlkofer, K.; Johansson, L. S. O.; Tanaka, M.; Zharnikov, M. Langmuir 2003, 19, 4992-4998. (33) Ye, S.; Li, G.; Noda, H.; Uosaki, K.; Osawa, M. Surf. Sci. 2003, 529, 163-170. (34) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L.; Tao, Y. T.; Parikh, A. N.; Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152-7167. (35) Petrovykh, D. Y.; Kimura-Suda, H.; Tarlov, M. J.; Whitman, L. J. Langmuir 2004, 20, 429-440. (36) Adlkofer, K.; Eck, W.; Grunze, M.; Tanaka, M. J. Phys. Chem. B 2003, 107, 587-591.

Mapping Forces on GaAs Surfaces

Langmuir, Vol. 22, No. 4, 2006 1771

detected no unbound sulfur on the surface or S-S species.37 This is in contrast to our previous studies where we studied the modification of GaAs with peptides via adsorption from solution. Based on the XPS data alone one can conclude that the two types of peptides are covalently anchored on the surface and when compared qualitatively they have similar chemical composition. FT-IRRAS was utilized to survey each area patterned by peptides A and B in order to compare the molecular arrangement of each species after the covalent attachment to the GaAs surface. Spectra were acquired from two different regions, Figure 2. Previous studies in the literature have used the position of υ(CH2) bands in order to determine if the packing of the alkyl chains is in a “liquidlike” or “crystalline-like” state.38 Monolayers with crystalline-like packing have υas(CH2) bands at 2920 cm-1 and υs(CH2) bands at 2850 cm-1.39 In the liquidlike state, the peaks corresponding to the υas(CH2) and υs(CH2) bands shift toward higher frequencies when compared to the crystalline-like state.34 Our results in the high-frequency region, Figure 2A, show that the areas patterned with peptide A have two strong methylene peaks at 2915 and 2845 cm-1. Surfaces functionalized with peptide B display υas(CH2) bands at 2921 cm-1 and υs(CH2) bands at 2852 cm-1. The FT-IRRAS data suggests that the monolayer composed of peptides A contains densely packed and crystalline-

like alkyl chains. The monolayer composed of peptide B displays methylene peaks that are shifted toward higher frequencies and has a more disordered alkyl chain packing compared to a monolayer of peptide A.40 These results are in good agreement with our previous work.16 We analyzed the low-frequency region, Figure 2B, to reveal further details with respect to the structure of the peptides on the patterned areas. GaAs surfaces modified with peptide A and B showed strong amide I peaks at 1657 and 1649 cm-1. These peaks are indicative of monolayers composed of polypeptides with alpha helical structure and interchain hydrogen bonding.41 The amide II peaks were located at ∼1560 cm-1 on surfaces patterned with peptide A, and at 1530∼1560 cm-1 on substrates modified with peptide B. The peak positions corresponding to the presence of amide II species are indicative of hydrogen bonding within both types of monolayers.42 In summary, the FT-IRRAS studies confirmed that the two types of peptides result in microcontact printed patterns with different types of packing on the GaAs surface. Biomolecules are anchored on surfaces for a variety of different detection platforms. One needs to verify that the biomolecules retain their biological activity despite the presence of a surface in such sensing strategies. Different approaches are used in the literature in order to obtain such a confirmation.43 One of the most popular methods involves the use of fluorescent markers on either the surface anchored functionality or the biomolecule utilized to test a specific interaction. Another methodology employs modified nanoparticles that can be localized on surface patterns due to the specific recognition between ligands on the particles and biomolecules on the surface. We chose to apply the nanoparticles-based methodology in order to verify that the TAT peptide patterns retain their biological activity. For this purpose, we worked with commercially available 20 nm gold nanoparticles coated with streptavidin. The complete scheme that describes this biological activity test was previously published,16 and therefore we briefly outline each step. The patterned surfaces were passivated with octadecanethiol in order to prevent nonspecific adsorption of nanoparticles. Subsequently a solution of biotinylated RNA was allowed to adsorb on the surface. The GaAs was washed from the RNA solution using buffer and dried with nitrogen. A drop of solution containing 20 nm streptavidin modified gold nanoparticles was placed on top of the surface and after 1 h washed with buffer and water. The evaluation of the specific interaction among the peptides on the surface, the biotinylated RNA and the streptavidin on the nanoparticles was done using AFM. The height of the patterned areas showed an increase of 17.60 (1.09 nm (see the Supporting Information) which is consistent with the diameter of the nanoparticles used. The experiment we described so far suggests that the peptides retain their biological activity after they are anchored on the surface. However, in our experiments we recorded aggregation of nanoparticles on the patterned areas and were unable to see quantitative differences with regard to the interaction of the TAR RNA with the two different peptides. Adhesion Force Measurements. Chemical force microscopy (CFM) was utilized to measure and compare the interaction between each surface anchored peptide sequence and TAR RNA. We initiated the studies by performing FV imaging with a clean AFM tip in air, Figure 3. The FV imaging was used to collect a topography image as well as a FV map comprised of numerous

(37) Shaporenko, A.; Adlkofer, K.; Johansson, L. S. O.; Ulman, A.; Grunze, M.; Tanaka, M.; Zharnikov, M. J. Phys. Chem. B 2004, 108, 17964-17972. (38) Jennings, G. K.; Yong, T.-H.; Munro, J. C.; Laibinis, P. E. J. Am. Chem. Soc. 2003, 125, 2950-2957. (39) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. J. Am. Chem. Soc. 1990, 112, 558-569.

(40) Pardo, L.; Wilson, W. C.; Boland, T. Langmuir 2003, 19, 1462-1466. (41) Enander, K.; Aili, D.; Baltzer, L.; Lundstrom, I.; Liedberg, B. Langmuir 2005, 21, 2480-2487. (42) Uvdal, K.; Vikinge, T. P. Langmuir 2001, 17, 2008-2012. (43) Demers, L. M.; Ginger, D. S.; Park, S. J.; Li, Z.; Chung, S. W.; Mirkin, C. A. Science 2002, 296, 1836-1839.

Figure 2. FT-IRRAS spectra of (a) the high-frequency region and (b) the low-frequency region of areas of the GaAs surface modified with the two different peptides using microcontact printing.

1772 Langmuir, Vol. 22, No. 4, 2006

Cho and IVaniseVic

Scheme 2. Schematic Representation of the AFM Tip Modification Chemistry Described in the Texta

a

The inset shows the expected interaction between a tip modified with RNA molecules and areas of the surface containing TAT peptides.

Figure 3. (a) Topography AFM image collected in FV mode using a clean tip. (b) An adhesion map obtained by analyzing a FV image composed of 64 × 64 Deflection vs Distance curves. The topography and FV images were collected simultaneously in air.

deflection vs distance curves. We evaluated an area where both types of patterns were present, Figure 3a. The data acquired through the FV image was used to create an adhesion map, Figure 3b. A total of 1024 deflection vs distance curves were analyzed in order to generate the adhesion map. The adhesion force data were calculated using a literature procedure that utilized the sensitivity and spring constant values of the cantilever as well as the maximum cantilever deflection values acquired by the FV image.22,23 A close examination of Figure 3 allows one to see

a cross pattern on the topography as well as the adhesion map. Statistical analysis of the individual adhesion force values extracted from areas where each type of peptide pattern was located showed no appreciable differences. However, one can see clear differences in adhesion values extracted from regions of the GaAs that were not covered with peptides vs regions with peptide features. Areas that are higher in the topography image due to the stamped peptides displayed low adhesion forces compared to the rest of the areas where higher adhesion forces were observed. This result is expected if one uses a clean tip that is not contaminated throughout the acquisition of the FV image. Larger adhesion forces are expected when the tip contacts the “hard” and clean GaAs surface.44 When the tip encounters areas patterned by the soft peptide structures, the tip can indent this material, can take longer to separate from the surface due to the larger contact area, and can result in smaller adhesion forces.22,45,46 The FV imaging in air with a clean tip allowed us to validate the experimental setup by recording the expected relationship between the adhesion forces on patterned vs clean areas of the GaAs surface. The specific interaction between TAR RNA and each peptide type on the GaAs surface was recorded in buffer solution using properly modified AFM tips and the FV imaging option. Tips were functionalized using a modified procedure adapted from Wangsa-Wirawan et al.20 The tip modification steps are outlined in Scheme 2. The surface was cleaned and terminated on thiol groups using a silane reagent. A commercially available crosslinker was utilized in order to facilitate the covalent attachment of the TAR RNA sequence which was synthesized to contain a spacer of six CH2 groups and NH2 functionality at the 5′ end. Each coupling step was verified using XPS (see the Supporting Information). The solution CFM measurements were done with tips modified with TAR RNA at room temperature. All experiments were done using the FV mode of the Nanoscope IIIa instrument. We performed studies in solutions buffered at pH 4, 7, and 10. Figure 4a shows representative force vs distance curves collected with a functionalized tip over an area patterned with peptide A. As the pH of the solution is decreased, the adhesion force between the tip and the patterned area increases. The data (44) Grinevich, O.; Mejiritski, A.; Neckers, D. C. Langmuir 1999, 15, 20772079. (45) Akhremitchev, B. B.; Mohney, B. K.; Marra, K. G.; Chapman, T. M.; Walker, G. C. Langmuir 1998, 14, 3976-3982. (46) Florin, E.; Moy, V. T.; Gaub, H. E. Science 1994, 264, 415-417.

Mapping Forces on GaAs Surfaces

Langmuir, Vol. 22, No. 4, 2006 1773 Scheme 3. Representation of the Expected Interaction of a Tip Modified with TAR RNA Molecules and the Peptide Terminated GaAs Surface at Different pH Ranges

Figure 4. (a) Representative retract force vs distance curves between a tip modified with RNA molecules and areas of the surface stamped with peptide A. Curves were collected at pH 4, 7, and 10; (b) Representative retract force vs distance curves between a tip modified with RNA molecules and areas of the surface stamped with peptide A (lower curve) and peptide B (upper curve). The curves were collected at pH 7.

shown on this figure was collected with the same tip over the same microcontact printed GaAs substrate. The pH dependence of the interaction between the TAR RNA and the peptide patterns was confirmed using multiple different tips and patterned surfaces. We also recorded larger adhesion values when the tip interacted with areas patterned with peptide A vs peptide B, Figure 4b. Figure 4 displays measurements of TAT peptides/TAR RNA unbinding forces. The data in this figure indicates that when individual binding interactions are broken, one can observe discrete events during the retraction of the AFM cantilever.47 Individual discrete breaking forces correspond to the breaking of an interaction between the TAR RNA molecules on the tip and the peptides on the surface.46 We quantified the magnitude of the TAR RNA/TAT peptide interaction by collecting a number of FV images over areas patterned with each peptide and analyzed the numerous force curves we obtained under the three pH values, Figure 5. The histograms show total adhesion values. The histograms, mean adhesion forces and standard deviations shown in Figure 5 were calculated using a custom-written MatLab code that utilized the data from the traction portion of the individual approach/retract curves that make up the FV image. Each FV image used for this analysis was composed of 64 × 64 force curves that were collected over an area of 500 × 500 nm2. We observed that the interaction between TAR RNA and peptide A is stronger compared to the interaction between TAR RNA and peptide B. The histograms clearly demonstrate a pH dependence of the interactions regardless of which peptide sequence one uses. One expects to observe a pH dependence due to the presence of cationic sites on the surface created by the existence of lysine and arginine residues in the peptide sequences. The number of charged species in the peptide A sequence is greater and is expected to contribute to the slightly (47) Wang, M. S.; Palmer, L. B.; Schwartz, J. D.; Razatos, A. Langmuir 2004, 20, 7753-7759.

higher adhesion forces we recorded. An additional reason for the differences we observed can be found in the orientation of the monolayers on surface that was evaluated and discussed in the previous subsection. The variable orientation of the molecules within the monolayer can also be responsible for the two peaks in the histograms of Figure 5, parts d and f. We verified that the adhesion values we recorded are due to the interaction between TAR RNA and the peptides on the surface. For this purpose, we measured the interaction of a clean tip with a peptide pattern using the three different buffer conditions, Figure 6. We observed very little interaction and no pH dependence. Taken in sum, our results can be explained by the composition of the peptide sequence used and are consistent with other studies that have examined TAT peptide behavior.4 The side chains of lysine and arginine are responsible for the pH dependence, Scheme 3. At low pH values, these side chains are protonated, and at high pH conditions, they are not charged. Previous studies in the literature have reported the necessity for charged species in order for the TAR RNA binding to take place.7 Calnan et al. have studied the specific role of the number of arginine residues in the peptide sequence and have concluded that the presence of a single arginine is sufficient.48 Our investigations demonstrate that the same is true after the peptides are anchored on the semiconductor material. Moreover, our quantitative analysis showed that there is a slight increase in the binding interaction when a sequence with a larger number of argenines is used. The results also confirm that the interaction with a peptide sequence that contains a single argenine residue is significant and enough to facilitate proper recognition events.

Summary In this paper, we illustrate the use of chemical force microscopy to quantify the specific interaction between TAR RNA and TAT peptides immobilized on GaAs surfaces. Two different TAT peptides were lithographically anchored on the semiconductor material using microcontact printing. The composition of the patterns was characterized by XPS and FT-IRRAS. Adhesion force maps collected with a TAR RNA modified AFM tips showed stronger interaction with a peptide sequence that contained a greater number of arginine residues. The studies also demonstrated (48) Calnan, B. J.; Tidor, B.; Biancalana, S.; Hudson, D.; Frankel, A. D. Science 1991, 252, 1167-1171.

1774 Langmuir, Vol. 22, No. 4, 2006

Cho and IVaniseVic

Figure 5. Histograms of the adhesion force between an AFM tip functionalized with RNA molecules and areas of the surface modified with peptide A (parts a-c) and peptide B (parts d-f). The experiments were performed with the same tip at pH 4, 7, and 10.

Acknowledgment. This work was supported by the Bindley Biosciences Center at Purdue University and by NASA under Award No. NCC 2-1363. Any opinions, findings, and conclusions expressed in this material are those of the authors and do not necessarily reflect the views of the National Aeronautics and Space Administration. The authors acknowledge experimental help from Dr. Richard Haasch (UIUC) to carry out the XPS characterization. All XPS experiments were performed at the Center for Microanalysis of Materials, UIUC, which is partially supported by U.S. Department of Energy under Grant DEFG0296-ER45439. Figure 6. Comparison between the adhesion forces measured with a clean tip and a RNA modified tip over areas of the GaAs surface modified with peptide A. The experiments were performed with the same tips at pH 4, 7, and 10.

that the interaction between the surface bound peptides and the TAR RNA is dependent on pH. Our investigations are consistent with previous studies that reported evidence that cationic residues are responsible for the polyelectrolyte-like affinity of the TAT peptides for TAR RNA.18

Supporting Information Available: Additional data regarding the characterization of the tip modification chemistry by XPS and SEM, evaluation of the GaAs surface stamping with each peptide using XPS, and AFM images showing the localization of Au nanoparticles on peptide A patterns after the surface was exposed to biotylated RNA and streptavidin conjugated Au nanoparticles. This material is available free of charge via the Internet at http://pubs.acs.org. LA052729X