Paper Menu >>
Journal Menu >>
J. Biomedical Science and Engineering, 2011, 4, 242-247 JBiSE doi:10.4236/jbise.2011.44033 Published Online April 2011 (http://www.SciRP.org/journal/jbise/). Published Online April 2011 in SciRes. http://www.scirp.org/journal/JBiSE Immobilization of antibodies on the self-assembled monolayer by antigen-binding site protection and immobilization kinetic control Myungok Yoon1, Hyun Jin Hwang2, Jeong Hee Kim3 1Department of Chemistry, Kyung Hee University, Seoul, Korea; 2R&D Center, Ahram Biosystems Inc., Seoul, Korea; 3Department of Oral Biochemistry, Kyung Hee University, Seoul, Korea. Email: hjhwang@ahrambio.com, jhkimh@khu.ac.kr Received 17 January 2011; revised 23 March 2011; accepted 28 March 2011. ABSTRACT The orientation of the biological molecule immobi- lized on a solid surface has been critical in devel- opment of various applications. In this study, ori- entation of antibody was retained by protecting the antigen-binding site of the antibody prior to immo- bilization to -functionalized mixed self-assembled monolayer (SAM) of 12-mercaptododecanoic acid and 1-heptanethiol. More importantly, the number of immobilization bonds formed between each an- tigen-binding site protected antibody molecule and the solid surface was controlled by optimizing the mole fraction of the activated carboxyl group of the linker molecules in the mixed SAM. The amount of antibody used in this study was approximately equivalent to the amount for one monolayer surface coverage. The resulting activity of protected immo- bilized antibody was about 10 fold higher than that of random immobilized antibody. Keywords: Antibody; Oriented Immobilization; Antigen-Binding Site Protection; Self-Assembled Monolayer; Kinetic Control 1. INTRODUCTION Interests in the efficient immobilizations of biomolecules such as enzymes, proteins, antibodies, and DNA [1-6] on the solid surfaces for biomedical and biotechnological applications have rapidly increased during the last two decades. Immunoassays, affinity chromatography, and DNA microarray [3,7,8] are all based on the immobiliza- tion of biomolecules on the solid phases for the purpose of clinical diagnostics, food industry and environmental monitoring [9-11]. Widely used methods for the attach- ment of antibodies to solid surfaces are physical adsorp- tion, covalent coupling, cross-linking, or entrapment in a gel network [4,12-14]. However, since these methods may result in decreased binding activity and selectivity of the antibodies after immobilization due to improper orientations, or denaturing of the antibodies, much effort have been recently put in the development of site-spe- cific immobilization of antibodies [3,15-18]. Among these are immobilizing protein A or G first to the solid surface followed by immobilization of antibodies [16,19]. In an- other method, azobenzene-containing polymers were used to control antibody orientation [20]. It was reported that the antigen-binding activities of immobilized Fab’ frag- ments of rabbit anti-human IgG with proper orientation were more than 2 fold increase than those with random orientation [15]. In this study, in order to maximize the natural bio- logical activity of an antibody after immobilization, the antigen-binding sites of the antibody were protected by incubating with its own antigen prior to immobilization. More importantly, the number of chemical bond formed between the antigen-binding site-protected antibody and the solid surface was optimized by kinetic control of the immobilization reaction (protected immobilization, PIM). The resulting activity of the antigen-binding site pro- tected immobilized antibody was significantly increased compared to that of randomly immobilized antibody. 2. MATERIALS AND METHODS 2.1. Chemicals and Reagents Chemicals for immobilization reaction are purchased from the following sources; (D,L)-thioctic acid (Aldrich, USA), 1-ethyl-3-[3-(dimethylamino)propyl]carbamide (EDC) (Sigma, USA), and N-hydroxysulfosuccinimide (sulfo- NHS) (Pierce, USA). Au-coated slides were purchased from EMF, USA. Mouse anti-DNA monoclonal antibody (IgM) recognizes both single- and double-stranded DNA M. Yoon et al. / J. Biomedical Science and Engineering 4 (2011) 242-247 Copyright © 2011 SciRes. JBiSE 243 was obtained from Roche Diagnostics, Germany. Taq polymerase, and deoxy nucleotide mixture (dNTP) were obtained from Takara, Japan. [-35S] d-ATP (1250 Ci/ mmole) and scintillation cocktail solution were pur- chased from NEN and ICN, USA, respectively. Univer- sal and reverse primers were synthesized from Bioneer, Korea. Other chemicals were purchased from Sigma, USA, or from other common sources. 2.2. Formation of the Mixed SAM on the Au Surface Au-coated glass slides (3 mm × 5 mm) were carefully cleaned with Piranha solution (30% H2O2: Concentrated H2SO4 = 1:3) for 15 - 30 sec and rinsed with d-H2O and then ethanol. The cleaned bare Au surface was soaked in 10 mM thioctic acid in ethanol for overnight rinsed with ethanol and dried. Mixed solution of 12-mercaptodode- canoic acid [HS(CH2)11COOH] and 1-heptanethiol [HS- (CH2)6CH3] was prepared in ethanol. Au-coated glass slides were incubated with the mixed SAM solution for 1 hr at room temperature and then rinsed with ethanol. The thiol groups were chemically adsorbed to the Au surface, thereby creating a mixed monolayer of 12-mer- captododecanoic acid and 1-heptanethiol. Then, Au- coated glass slides were immersed in 5 mM sulfo-NHS and 10 mM EDC in MES buffer (pH 6.0) for 1 hr to ac- tivate the carboxyl groups on the surface and rinsed with ethanol. 2.3. Radioactive Labeling of DNA by Polymerase Chain Reaction (PCR) Bacterial plasmid DNA, pBluescriptII KS(+) was used as a template DNA. Concentration of DNA was meas- ured by UV spectrophotometer (Pharmacia Biotech Ul- traspec 2000, USA). A typical polymerase chain reaction (PCR) mixture contained 200 ng of DNA, 0.4 M each of universal and reverse primer, 50 M of dNTP, 2.5 U of Taq polymerase and 0.1 vol. of 10x buffer in 100 l final volume. For labeling purpose, 2 l of [-35S] d- ATP was added to the reaction mixture. The reaction mixture was heated to 94˚C for 5 min. The PCR profile was 94˚C for 30 sec, 50˚C for 1 min, and 72˚C for 30 sec for 30 cycles, followed by 72˚C for 10 min. We always ran labeling reactions with the non-labeled standard con- trol reaction side by side. After PCR, an aliquot of the control reaction was analyzed on 1.2% agarose gel con- taining 0.5 g/ml ethidium bromide to confirm the gen- eration of PCR product. 2.4. Immobilization of Anti-DNA Antibody and Immunoassay In order to prepare protected antibody-DNA solution, approximately 0.54 pmol of anti-DNA antibody was incubated with approximately 1.07 pmol of labeled DNA for 1 hr at 37˚C in 0.1 M phosphate buffer (pH 7.4). Otherwise the concentration of antibody and DNA were indicated in the text. The activated Au surface was incu- bated with protected Antibody-DNA solution for 30 min at 37˚C, rinsed with a buffer of 1.0 M potassium phos- phate, pH 6.7. Randomly immobilized antibody was prepared by the same procedure described above except the incubation with labeled DNA. The antibody immobi- lized Au-coated glass slides were incubated with ap- proximately 2.0 pmol of labeled DNA for 2 hrs at RT and then washed with TBST (20 mM Tris, pH 7.8, 150 mM NaCl, and 0.05% Tween-20) buffer three times. The glass slides were dried and the -emission was measured with a scintillation counter (Wallac, system 1400, EG&G Co., Finland). 3. RESULTS AND DISCUSSION 3.1. Experimental Scheme Immobilization scheme of anti-DNA antibody is shown in Figure 1. In this experiment, Au surface was chosen as a solid phase since it has an advantage over polysac- Figure 1. Schematic representation of the steps of anti-DNA antibody immobilization to the mixed SAM on Au surface. PIM; protected immobilization, RIM; random immobilization. M. Yoon et al. / J. Biomedical Science and Engineering 4 (2011) 242-247 Copyright © 2011 SciRes. JBiSE 244 charides, polystyrene or silica which are most frequently used solid phases for the immobilization of antibodies; thiols form self-assembled monolayers (SAM) on Au surface spontaneously due to the formation of strong Au-S covalent bonds [15,21,22], which make follow-up reactions for modification of the surface functional groups easier. The mixed monolayer of 12-mercaptododecanoic acid [HS(CH2)11COOH] and 1-heptanethiol [HS(CH2)6- CH3] on Au surface was used in this work. Surface car- boxyl groups on the SAM were activated using 1-ethyl -3(3-dimethylaminopropyl)carbodiimide (EDC) and sulfo- N-hydroxysuccinimide (sulfo-NHS) to form sulfo-NHS esters. This coupling reaction was performed in 2-(N- morpholino)ethane sulfonic acid (MES) buffer at pH 6.0 since it is reported that sulfo-NHS ester has longer life- time at lower pH [23]. Then, anti-DNA antibodies were reacted to be immobilized to the -functionalized SAM through amide bonds. In order to preserve the natural activity of the anti- body after immobilization, the antigen-binding site of antibody were protected before immobilization to the Au surface by reactions with its antigen, DNA first to form antigen-antibody complexes followed by the reactions with sulfo-NHS esters (Protected immobilization, PIM). By this way the active sites are excluded from the sub- sequent immobilization reaction, thus contribute for the antibody to retain the proper orientation after immobili- zation. For random immobilization (RIM), antibody was immobilized as described above without protection of the antigen-binding sites of the antibody. 3.2. Kinetic Control and Protection of Antigen-Binding Site Increased the Activity of Immobilized Antibody We used the immobilization scheme presented in Figure 1 to immobilize antibodies on the SAM formed on Au surface. Before immobilization, the amount of the anti- body required to cover the Au surface to a monolayer was calculated and approximately 0.54 pmol of anti- DNA antibody was used. The number of surface car- boxyl group involved in the cross-linking of the antibody to the SMA was also considered. Due to the steric re- quirement of large bio-molecules, high concentration of surface carboxyl group was found to rather decrease the activity of immobilized biomolecule [24]. Therefore, considering the size of anti-DNA antibody (8.5 nm × 14.5 nm) [2], mixed monolayer of 12-mercaptodode- canoic acid and 1-heptanethiol was employed in our ex- periment instead of using pure monolayer of 12-mer- captododecanoic acid. Therefore, by controlling the mole fraction of 12-mercaptododecanoic acid in the SAM, the number of the carboxyl group involved in the immobili- zation of antibodies can be controlled, subsequently the number of antibody immobilized on the surface is con- trolled. In order to protect the antigen-binding site of the an- tibody, a 65 bp double stranded DNA (ds-DNA) labeled with 35S was prepared by polymerase chain reaction (PCR) and about 1.07 pmole of the labeled ds-DNA was used for protection of the two antigen-binding sites in each antibody (Antibody:DNA ≈ 1:2). For RIM, prein- cubation of the antibody with the labeled DNA step was excluded. After immobilization, glass slides were incu- bated with its labeled form of antigen, 35S-labeled ds- DNA. The activities of the immobilized antibodies were measured by counting -emission from antigen-antibody complexes which were formed by incubating immobi- lized antibodies with 35S-labeled DNA. Radioimmuno- assay is very sensitive to a very small amount of 35S- labeled DNA, thus enables us to measure a very small amount of immobilized antibody on the surface. The concentration of carboxyl group in the SAM was varied from 0% to 100% and the activity of immobilized antibody was measured as described above. The PIM antibodies preserved their activity much better than the RIM antibodies resulting -emission from these films are larger throughout the carboxyl group ratio used in this study (Figure 2(a)). It is very interesting to note that the maximum activities of immobilized antibodies occur at low surface carboxyl concentration of 5% (Figures 2 (a) and (b)). The activity of the immobilized antibodies by PIM method at 5% of carboxyl group was approxi- mately 10 times higher than the activity of RIM antibody (Figure 2(b)). In this experiment we labeled PIM antibodies with 35S and then immobilization was performed. After these treatments, part of antibodies would lose their activity through kinds of modifications or other unknown me- chanisms. If antibodies have been labeled by 35S, iso- topes on inactivated antibodies would not be all released from the Au surface, causing false conclusions. However, it seems that these effects are negligible when we com- pared the radioactivity acquired from pre-labeled PIM antibodies and not pre-labeled RIM antibodies at higher concentration of carboxyl group (25% - 100%) which showed very low radioactivity in both PIM and RIM antibodies (Figure 2(a)). In the coupling reaction of antibody with sulfo-NHS ester, either the amino group on the Fc region of the an- tibody, or the one on the Fab’ fragment near the anti- gen-binding site can react to form an amide bond; the former will preserve the native structure of the antibody and the latter may lose the native structure of the anti- body. When the antibody was reacted randomly, it was found that control over the orientation of the immobi- lized antibody was difficult, and this random orientation M. Yoon et al. / J. Biomedical Science and Engineering 4 (2011) 242-247 Copyright © 2011 SciRes. JBiSE 245 (a) (b) Figure 2. The activity of immobilized anti-DNA antibody to the mixed SAM on Au surface. The activi- ties of immobilized antibodies by PIM method (●) or RIM method (○) were obtained as a function of the mole fraction of 12-mercaptododecanoic acid (a). The activity of the immobilized antibody either by PIM or RIM method at 5% of 12-mercaptododecanoic acid was compared (b). A mixed SAM of 12-mercap- tododecanoic acid and 1-hepthanethiol was used to introduce the carboxyl groups as the reactive group for immobilization. The protection ratio of antibody to antigen was 1:2. of the antibody results in the loss of the activity of the immobilized antibody. However, when the antigen bind- ing sites of the antibody were protected before immobi- lization, the activity of the PIM antibody was signifi- cantly increased compared to that of RIM antibody re- sulting -emission from these films are much larger as shown in Figure 2. In addition to the protection of antigen-binding sites, the increased activity at lower carboxyl group concentra- tion can be explained in terms of number of bonds formed between antibody and the supporting surface. Since there are multiple reaction groups exist on the surface of the antibody as well as the supporting surface, multiple immobilization bonds can be formed between the antibody and the supporting surface. Such non-spe- cific formation of multiple bonds in various region of the antibody can induce structural change and destruction of the biologically active molecule upon immobilization, thereby causing substantial reduction of the antigen- binding activity of the antibody. It seems that it is critical to minimize the number of bonds formed between the antibody and the surface. Our results revealed that about 5% of 12-mercaptododecnoic acid is appropriate to form a minimal number of covalent bond between the anti- body and the surface (Figure 2(a)). At higher concentra- tion of carboxyl groups, where multiple immobilization bonds were expected to formed, the activity of PIM an- tibody was dramatically reduced and it was similar to that of RIM antibody. This observation supports that the multiple bond formation cause the destruction of anti- body’s natural structure. Considering the maximum density of thiol groups on Au is about 0.5 nm [21,25], there will be about 100 - 200 thiol molecules under the antibody to be immobilized. Since our results showed the maximum activity of the immobilized antibody was acquired with about 5% of reactive group in the SAM on the Au surface. Theoreti- cally, 5% of the reactive group can produce at maximum of 5 immobilization bonds. Considering that in many available reaction conditions, especially in aqueous solu- tion, the reaction probability of the reaction group is substantially lower than 100%. Therefore, our results suggest that it is likely that only 1 immobilization bond or at most a few bonds are formed between the antibody and the SAM. 3.3. Activity of I mmobilized Antibody as a Function of the Concentration of the Antibody Used and the Protection Ratio We prepared different amount of antibodies ranged from 0.07 pmol to 0.68 pmol which is sub- to near-monolayer concentration of antibody. Prepared antibodies were an- tigen-binding site protected and immobilized on Au sur- face as described above. The mole fraction of the 12- mercaptododecanoic acid used to introduce carboxyl reaction group on the Au surface with respect to the total moles of the thiol molecules was 5%. The activity of the immobilized antibody was compared and plotted. As shown in Figure 3, the activity of immobilized antibody was linearly proportional to the concentration of anti- body used in this experiment. The activity of the immobilized anti-DNA antibody was measured at different protection ratio from 1: 0.0625 to 1:4 (Figure 4). The mole fraction of the M. Yoon et al. / J. Biomedical Science and Engineering 4 (2011) 242-247 Copyright © 2011 SciRes. JBiSE 246 Figure 3. Antibody concentration dependence of the ac- tivity of antigen-binding site protected antibody after immobilization to Au-coated slide glass. The concentra- tion of antibody used was from sub- to near-monolayer to cover the immobilization surface. PIM method was used at 5% of 12-mercaptododecanoic acid. The ratio of anti-DNA antibody to DNA was 1:2. Radioactivity of PIM antibody on Au surface (3 × 5 mm2) was measured and plotted. Figure 4. The activity of antibody immobilized to the SAM on Au surface as a function of antigen-binding site protection ratio. The antibody concentration was 0.54 pmol and the protection ratio of anti-DNA antibody to DNA was ranged from 1:0.0625 to 1:4. The activity of immobilized antibodies by PIM method (●) or RIM method (○) were depicted. 12-mercaptododecanoic acid in the SAM was 5%. The activity of the immobilized antibody was increased as the protection ratio increased. PIM antibodies revealed much higher binding activity compared to RIM antibod- ies throughout the protection ratio used in this experi- ment. The saturation phenomenon was observed in the PIM case when the molar ratio of the anti-DNA antibody to the ds-DNA used for protection was in the range of 1:1 ~ 1:2. Since there are two antigen-binding sites for each antibody, it is very reasonable that the binding ac- tivity of immobilized antibody reaches maximum when the protection ratio increased from 1 to 2. This data support the previous results described above that the antigen-binding sites were protected by formation of the antigen- antibody complex. In conclusion, we report on the development of a novel method to immobilize an antibody on the solid surface by using a PIM method and kinetically control- ling the number of chemical bond formed between the protected antibody and the solid surface. The resulting antigen-binding activity of protected immobilized anti- body was about 10 fold higher than that of random im- mobilized antibody. This method could have wide ap- plication in production of various bio-chips. REFERENCES [1] Zhen, G., Eggli, V., Vörös, J., Zammaretti, P., Textor, M., Glockshuber, R. and Kuennemann, E. (2004) Immobili- zation of the enzyme beta-lactamase on biotin- derivat- ized poly(L-lysine)-g-poly(ethylene glycol)-coated sen- sor chips: A study on oriented attachment and surface ac- tivity by enzyme kinetics and in situ optical sensing. Langmuir, 20, 10464-10473. doi:10.1021/la0482812 [2] Browning-Kelly, M.E., Wadu-Mesthrige, K., Hari, V. and Liu, G.Y. (1997) Atomic force microscopic study of spe- cific antigen/antibody binding. Langmuir, 13, 343-350. doi:10.1021/la960918x [3] Torrance, L., Ziegler, A., Pittman, H., Paterson, M., Toth, R. and Eagleston, I. (2006) Oriented immobilization of engineered single-chain antibodies to develop biosensors for virus detection. Journal of Virological Methods, 134, 164-170. doi:10.1016/j.jviromet.2005.12.012 [4] López-Gallego, F., Betancor, L., Mateo, C., Hidalgo, A., Alonso-Morales, N., Dellamora-Oritz, G., Guisan, J.M. and Fernandez-Lafuente, R. (2005) Enzyme stabilization by glutaraldehyde crosslinking of adsorbed proteins on animated supports. Journal of Biotechnology, 119, 70-75. doi:10.1016/j.jbiotec.2005.05.021 [5] Lu, B., Xie, J., Lu, C., Wu, C. and Wei, Y. (1995) Ori- ented immobilization of Fab’ fragments on silica surfaces. Analytical Chemistry, 67, 83-87. doi:10.1021/ac00097a014 [6] Zhao, X., Nampalli, S., Serino, A.J. and Kumar, S. (2001) Immobilization of oligonucleotides with multiple anchors to microchips. Nu cleic Acids Research , 29, 955-959. doi:10.1093/nar/29.4.955 [7] Markoglou, N. and Wainer, I.W. (2001) Synthesis and characterization of an immobilized phenylethanolamine N-methyltransferase liquid chromatographic stationary phase. Analytical Biochemistry, 288, 83-88. doi:10.1006/abio.2000.4884 [8] White, K.P., Rifkin, S.A., Hurban, P. and Hogness, D.S. (1999) Microarray analysis of drosophila development during metamorphosis. Science, 286, 2179-2184. doi:10.1126/science.286.5447.2179 M. Yoon et al. / J. Biomedical Science and Engineering 4 (2011) 242-247 Copyright © 2011 SciRes. JBiSE 247 [9] Ward, K. (2006) Microarray technology in obstetrics and gynecology: A guide for clinicians. American Journal of Obstetrics and Gynecology, 195, 364-372. doi:10.1016/j.ajog.2005.12.014 [10] Roy, S. and Sen, C.K. (2006) cDNA microarray screen- ing in food safety. Toxicology, 221, 128-133. doi:10.1016/j.tox.2005.12.025 [11] Lettieri, T. (2006) Recent application of DNA microarray technology to toxicology and ectoxicology. Environmen- tal Health Perspectives, 114, 4-9. [12] Giacomelli, C.E., Vermeer, A.W.P. and Norde, W. (2000) Adsorption of immunoglobulin G on core-shell latex par- ticles precoated with chaps. Journal of Colloid and In- terface Science, 231, 283-288. doi:10.1006/jcis.2000.7159 [13] Charelier, R.C., Gengenbach T.R., Griesser, H.J., Brig- ham-Burke, M. and O’Shannessy, D.J. (1995) A general method to recondition and reuse BIAcore sensor chips fouled with covalently immobilized protein/peptide. Ana- lytical Biochemistry, 229, 112-118. doi:10.1006/abio.1995.1386 [14] Charles, P.T., Goldman E.R., Rangasammy, J.G., schauer C.L., Chen, M.S. and Taitt, C.R. (2004) Fabrication and characterization of 3D hydrogel microarrays to measure antigenicity and antibody functionality for biosensor ap- plication. Biosensors and Bioelectronics, 20, 753-764. doi:10.1016/j.bios.2004.04.007 [15] Bonroy, K., Frederix, F., Reekmans, G., Dewolf, E., De Palma, R., Borghs, G., Declerck, P. and Goddeeris, B. (2006) Comparison of random and oriented immobiliza- tion of antibody fragments on mixed self-assembled mo- nolayers. Journal of Immunological Methods, 312, 167- 181. doi:10.1016/j.jim.2006.03.007 [16] Tanaka, G., Funabashi, H., Mie, M. and Kobatake, E. (2006) Fabrication of an antibody microwell array with self-adhering antibody binding protein. Analytical Bio- chemistry, 350, 298-303. doi:10.1016/j.ab.2005.12.034 [17] Ha, T.H., Jung, S.O., Lee, J.M., Lee, K.Y., Lee, Y., Park, J.S. and Chung, B.H. (2007) Oriented immobilization of antibodies with GST-fused multiple Fc-specific B-domains on a gold surface. Analytical Chemistry, 79, 546-556. doi:10.1021/ac061639+ [18] Lo, Y.S., Nam, D.H., So, H.M., Chang, H., Kim, J.J., Kim Y.H. and Lee, J.O. (2009) Oriented-immobilization of antibody fragments in Ni-decorated single-walled car- bon nanotube devices. ACS Nano, 3, 3649-3655. doi:10.1021/nn900540a [19] Liu, F., Dubey, M., Takahashi, H., Castner, D.G. and Grainer, D.W. (2010) Immobilized antibody orientation analysis using secondary ion mass spectrometry and fluo- rescence imaging of affinity-generated pattern. Analytical Chemistry, 82, 2947-2958. doi:10.1021/ac902964q [20] Mouri, M., Ikawa, T., Narita, M., Hoshino, F. and Wata- nabe, O. (2010) Orientation control of photo-immobi- lized antibodies on the surface of azobnzene-containing polymers by the introduction of functional groups. Mi- cromolecular Bioscience, 10, 612-620. [21] Ulman, A. (1996) Formation and structure of self-as- sembled monolayers. Chemical Reviews, 96, 1533-1554. doi:10.1021/cr9502357 [22] Prime, K.L. and Whitesides, G.M. (1991) Self-assembled organic monolayers: Model system for studying adsorp- tion of proteins at surfaces. Science, 252, 1164-1167. doi:10.1126/science.252.5009.1164 [23] Grebarek, Z. and Gergely, J. (1990) Zero-length cross- linking procedure with the use of active esters. Analytical Biochemistry, 185, 131-135. doi:10.1016/0003-2697(90)90267-D [24] Mittler-Neher, S., Spinke, J., Liley, M., Nelles, G., Weis- ser, M., Back, R., Wenz, G. and Knoll, W. (1995) Spec- troscopic and surface-analytical characterization of self- assembled layers on Au. Biosensors and Bioelectronics, 10, 903-916. doi:10.1016/0956-5663(95)99228-D [25] Alves, C.A., Smith, E.L. and Porter M.D. (1992) Atomic scale imaging of alkanethiolate monolayers at gold sur- faces with atomic force microscopy. Journal of the Ame- rican Chemical Society, 114, 1222-1227. doi:10.1021/ja00030a015 |