Investigation of the MALDI Process Used to Characterize Self

Feb 17, 2009 - Tae Kyung Ha, Han Bin Oh, Jayong Chung, Tae Geol Lee and Sang Yun Han*. Center for Nano-Bio Convergence Research, Korea Research ...
0 downloads 0 Views 1MB Size
3692

Langmuir 2009, 25, 3692-3697

Investigation of the MALDI Process Used to Characterize Self-Assembled Monolayers of Alkanethiolates on Gold Tae Kyung Ha,†,‡ Han Bin Oh,‡ Jayong Chung,§ Tae Geol Lee,† and Sang Yun Han*,† Center for Nano-Bio ConVergence Research, Korea Research Institute of Standards and Science (KRISS), Daejeon 305-340, Republic of Korea, Department of Chemistry, Sogang UniVersity, Seoul 121-742 (200811036), Republic of Korea, and Department of Food and Nutrition, Kyung Hee UniVersity, Seoul 130-701, Republic of Korea ReceiVed NoVember 4, 2008. ReVised Manuscript ReceiVed December 29, 2008 In this work, we investigated the surface processes involved in matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS), which produce intact characteristic ions, typically in disulfide form, from self-assembled monolayers (SAMs) of alkanethiolates on gold. For the study, SAMs decorated with peptides and a THAP matrix were employed. Using two-laser MS, it was previously found that irradiation with a UV laser gave rise to the direct desorption of SAM molecules from alkanethiol SAMs on gold, producing disulfide species in vacuum. However, a closer examination of this study suggests that the MALDI process in which the matrix is used may not be the case. Instead, the results indicate that the treatment of the matrix solution is responsible for the characteristic ion formation in MALDI MS. We propose that the matrix solution dissolves alkanethiolate molecules from SAMs, leading to the generation of characteristic disulfide species in the solution. The disulfides are then cocrystallized with matrix molecules and subsequently detected by MALDI MS. Because MALDI MS is a powerful tool for biopolymers with high molecular weights, it has been successfully applied to SAMs presenting large biomolecules. This understanding of the MALDI process in the surface MS of alkanethiol SAMs on gold may allow advances in the biomolecular application of SAMs in combination with mass spectrometric analysis.

Introduction In this article, we explored the surface processes involved in matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS) of self-assembled monolayers (SAMs) of alkanethiolates on gold. SAMs of thiolates on metals are important components used in nanotechnology and biological applications.1 SAMs are widely used to provide various model biological surfaces. For example, SAMs are capable of immobilizing biological molecules such as peptides, carbohydrates, and antibodies in a structurally well-defined way, thereby facilitating the examination of biomolecular interactions and reactions on the monolayers.1-15 Futhermore, SAMs can be used as substrates for studying cell biology processes such as cell adhesion.1,16 SAMs can also be protein-resistant surfaces that prevent the nonspecific adsorption of proteins and other biological molecules * To whom correspondence should be addressed. Phone: +82-42-8685716. Fax: +82-42-868-5032. E-mail: [email protected]. † KRISS. ‡ Sogang University. § Kyung Hee University. (1) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. ReV. 2005, 105, 1103–1169. (2) Mrksich, M. ACS Nano 2008, 2, 7–18. (3) Lee, Y.-S.; Mrksich, M. Trends Biotechnol. 2002, 20, S14–S18. (4) Su, J.; Mrksich, M. Angew. Chem., Int. Ed. 2002, 41, 4715–4718. (5) Su, J.; Mrksich, M. Langmuir 2003, 19, 4867–4870. (6) Min, D.; Tang, W.; Mrksich, M. Nat. Biotechnol. 2004, 22, 717–723. (7) Min, D.; Su, J.; Mrksich, M. Angew. Chem., Int. Ed. 2004, 43, 5973–5977. (8) Min, D.; Yeo, W.; Mrksich, M. Anal. Chem. 2004, 76, 3923–3929. (9) Li, J.; Thiara, P.; Mrksich, M. Langmuir 2007, 23, 11826–11835. (10) Yeo, W.; Min, D.; Hsieh, R.; Greene, G.; Mrksich, M. Angew. Chem., Int. Ed. 2005, 44, 5480–5483. (11) Patrie, S.; Mrksich, M. Anal. Chem. 2007, 79, 5878–5887. (12) Marin, V.; Bayburt, T.; Sligar, S.; Mrksich, M. Angew. Chem., Int. Ed. 2007, 46, 8796–8798. (13) Griessera, H. J.; Kingshott, P.; McArthur, S. L.; McLean, K. M.; Kinsele, G. R.; Timmonse, R. B. Biomaterials 2004, 25, 4861–4875. (14) Evans-Nguyen, K. M.; Tao, S.; Zhu, H.; Cotter, R. J. Anal. Chem. 2008, 80, 1448–1458. (15) Tsubery, H.; Mrksich, M. Langmuir 2008, 24, 5433–5438. (16) Mrksich, M. Chem. Soc. ReV. 2000, 29, 267–273.

on their surfaces.1,17 In generating such model biological surfaces, SAMs of alkanethiolates on gold are particularly versatile because reliable surface chemistry is available to tailor the surfaces for various applications.1,18-20 Alkanethiol SAMs on gold are known to be particularly amenable to MALDI MS.4,5 Because MS allows direct analyses of molecular entities of monolayers by their molecular masses,4,5,21-25 MALDI MS can be used to characterize alkanethiol SAMs on gold and the interfacial reactions on the monolayers.4-9 Additionally, MALDI MS is capable of monitoring the adsorption of proteins and oligonucleotides by specific interactions on biochemically tailored SAM surfaces, which makes it possible to interrogate biomolecular interactions such as protein-protein interactions in a label-free manner using biochips.3,10-15 These features of MALDI MS, particularly when applied to alkanethiol SAMs on gold, have been extensively shown by Mrksich and colleagues in their SAMDI (self-assembled monolayers for MALDI) applications.2 In SAMDI experiments using alkanethiol SAMs on gold (Au-S-R), MALDI has been found to produce characteristic (17) Ostuni, E.; Chapman, R. G.; Liang, M. N.; Meluleni, G.; Pier, G.; Ingber, D. E.; Whitesides, G. M. Langmuir 2001, 17, 6336–6343. (18) Houseman, B. T.; Gawalt, E. S.; Mrksich, M. Langmuir 2003, 19, 1522– 1531. (19) Smith, E. A.; Wanat, M. J.; Cheng, Y.; Barreira, S. V. P.; Frutos, A. G.; Corn, R. M. Langmuir 2001, 17, 2502–2507. (20) Lee, C.-Y.; Nguyen, P.-C. T.; Grainger, D. W.; Gamble, L. J.; Castner, D. G. Anal. Chem. 2007, 79, 4390–4400. (21) Hanley, L.; Kornienko, O.; Ada, E.; Fuoco, E.; Trevor, J. J. Mass. Spectrom. 1999, 34, 705–723. (22) Trevor, J. L.; Lykke, K. R.; Pellin, M. J.; Hanley, L. Langmuir 1998, 14, 1664–1673. (23) Trevor, J. L.; Mencer, D. E.; Lykke, K. R.; Pellin, M. J.; Hanley, L. Anal. Chem. 1997, 69, 4331–4338. (24) Leggett, G. J. In TOF-SIMS: Surface Analysis by Mass Spectrometry; Vickerman, J., Briggs, D., Eds.; IM Publications, SurfaceSpectra Limited: West Sussex, Manchester, U.K., 2001; pp 573-593. (25) Wolf, K.; Cole, D.; Bernasek, S. Anal. Chem. 2002, 74, 5009–5016.

10.1021/la8036567 CCC: $40.75  2009 American Chemical Society Published on Web 02/17/2009

Self-Assembled Alkanethiolate Monolayers on Gold

Langmuir, Vol. 25, No. 6, 2009 3693

Scheme 1. Preparation of SAMs Containing Peptides and Their MALDI Mass Spectra

disulfide ions (R-S-S-R + Na+ or H+) or, in some cases, thiolate ions, providing direct information on the molecular composition of the SAMs.2,4,5 The production of noticeable characteristic ions has been exploited to interpret MALDI mass spectra straightforwardly. Thus, MALDI MS can be used for the rapid characterization of biochemical reactions occurring on monolayers and high-throughput screening using biochips.6-9 Despite numerous applications, however, little is known about how the MALDI process produces characteristic ions from alkanethiol SAMs on gold. Previously, Hanley and colleagues reported experimental results exhibiting such characteristic ion formation using twolaser MS.21-23 In their investigations, UV laser irradiation (337 nm) was found to lead to the direct desorption of intact SAM molecules from alkanethiol SAMs on gold (Au-S-R). The dominant reaction products were disulfides (R-S-S-R), and dimerization was suggested to occur on the surface during the UV-laser-induced desorption process rather than through recombination in the gas phase.22 In the experiments, neutral disulfides desorbed under vacuum were postionized by a second VUV laser (118 nm) and then mass analyzed. Very recently, a new scheme of cation-assisted laser desorption/ionization (CALDI) was suggested, in line with the findings of previous two-laser MS experiments. In the CALDI method, it was shown that cationization can be substituted for the otherwise expensive and less efficient postionization employed in the earlier twolaser MS scheme.26 From the earlier laser desorption/ionization experiments, Mrksich and colleagues reasoned that the treatment of monolayers with the energy-adsorbing matrices commonly used in MALDI MS for biopolymers would allow laser desorption/ionization MS combined with SAMs to be applied to a broader range of biochemical applications.2 Indeed, the use of matrices is very efficient at ionizing alkanethiol SAMs on gold with large biomolecules, as shown in many SAMDI applications, but the role of the matrix in MALDI MS for alkanethiol SAMs on gold is not fully understood yet, although an understanding of the mechanism is essential for further development of this surface MS for use in advanced applications such as quantitative assays using biochips. In this work, in an effort to deepen our understanding of MALDI MS for SAMs, we investigated the

Scheme 1 outlines the synthesis steps used to prepare the peptideimmobilized SAMs employed in the present study. The SAMs containing peptides were prepared by immobilizing cysteineterminated peptides in the mixed SAMs with maleimide functional groups.18 The maleimide functionality was introduced at the NH2 groups in the mixed SAMs with -OH and -NH2 terminals using a linker reagent, sulfosuccinimidyl 4-[N-maleimidomethyl]-cyclohexane-1-carboxylate (SSMCC, Thermo Scientific).19,20 AcLTTASLGSGC-NH2 (Pep1) and Ac-LRRASLGSGC-NH2 (Pep2) were custom synthesized (Peptron, Korea). The gold-coated substrate was cleaned with a superpiranha solution for 2 min, rinsed with deionized water and ethanol,27 and cut into ∼1 cm2 chips. SAMs were then grown on the gold substrate by immersing the chips in ethanol solutions of commercial SAM reagents (CosBiotech, Korea). For the mixed SAMs, the chips were treated in a 1:2 solution of thiol reagents (HS-(CH2)11-EG6-NH2 and HS-(CH2)11-EG3-OH in ethanol (2 mM total concentration)) for 12 h, producing mixed SAMs with -NH2 and -OH terminals in an estimated ratio of ∼1:5. Here, EG represents an ethylene glycol moiety (-CH2CH2O-). The prepared monolayers were rinsed with ethanol and dried under a stream of nitrogen. In the mixed SAMs, oligoethylene glycols serve as a background that reduces the nonspecific adsorption of peptides on monolayers and also provides sufficient spacing between bulky immobilized peptides. The maleimide functionality was implemented by treating the mixed SAMs with 20 mM SSMCC in PBS buffer (pH 7.2) for 30 min at room temperature. To immobilize peptides at the maleimide groups, the surface of the mixed SAMs with maleimide groups was further treated with 10 mM peptide in PBS buffer (pH 7.4) for 60 min at room temperature. After the coupling reaction, the SAMs presenting peptides were thoroughly washed with PBS buffer and deionized water several times and dried under a stream of nitrogen. For MALDI MS experiments, a THAP (2′,4′,6′-trihydroxyacetophenone, Fluka) matrix was chosen exclusively because it is the matrix that has been applied in most MALDI MS experiments characterizing SAMs presenting biomolecules.6-9 The matrix was treated by pipetting a 1 µL drop of THAP solution (typically, 5 mg/mL in methanol) onto the SAM surface. MALDI mass spectra were taken using a commercial MALDI mass spectrometer equipped with a 355 nm MALDI laser (Autoflex III, Bruker-Daltonics,

(26) Ha, T. K.; Lee, T. G.; Song, N. W.; Moon, D. W.; Han, S. Y. Anal. Chem. 2008, 80, 8526–8531.

(27) Min, H.; Park, J. W.; Shon, H.-K.; Moon, D. W.; Lee, T. G. Appl. Surf. Sci. . 2008, 255, 1025-1028.

MALDI process involved in the surface MS of alkanethiol SAMs on gold.

Experimental Details

3694 Langmuir, Vol. 25, No. 6, 2009

Ha et al.

Germany). All spectra were obtained in the reflectron mode for positive ions.

Results and Discussion MALDI MS of SAMs Presenting Peptides. Scheme 1 presents MALDI mass spectra taken for the SAMs with two different peptides, Pep1 (Ac-LTTASLGSGC-NH2) and Pep2 (AcLRRASLGSGC-NH2). The peptides are different in that Pep2 contains basic residues (arginine, R) offering protonation sites, whereas Pep1 does not. The MALDI mass spectra for the two SAMs display characteristic ion formation by MALDI MS of alkanethiol SAMs on gold. The mass spectrum of the Pep1-immobilized SAMs shows the production of characteristic ions for OH-terminated background SAMs (HO-EG3-(CH2)11-S-S-(CH2)11EG3-OH + Na+) as well as disulfide ions with an immobilized peptide at one end (HO-EG3-(CH2)11-S-S-(CH2)11EG6-NH-linker-Pep1 + Na+) at m/z values of 693.4 and 1993.1, respectively. In the case of SAMs with Pep2, MALDI produced characteristic ions similar to those in the Pep1immobilized SAMs case. One of the products was the disulfide ions corresponding to background SAMs observed at m/z ) 693.4. Disulfide ion HO-EG3-(CH2)11-S-S-(CH2)11EG6-NH-linker-Pep2 + H+ or Na+ containing Pep2 was also produced at m/z values of 2081.2 and 2103.2, respectively. The presence of basic arginine residues in Pep2 resulted in protonated disulfide products at m/z ) 2081.2. In the above two samples, no thiolate ions corresponding to either background SAMs or peptide-immobilized SAMs were observed. As shown in the two samples and in other studies, MALDI generally leads to a noticeable production of characteristic (disulfide) ions from alkanethiol SAMs on gold, offering straightforward monitoring of the molecular composition of monolayers.2,4-9 Effects of Matrix Solutions on MALDI Efficiency. In an effort to understand the MALDI process for bioconjugated alkanethiol SAMs on gold, we first examined how the choice of matrix solvent affected MALDI efficiency. The effects of ethanol and water as solvents in addition to methanol were examined in this study. The methanol solution of THAP (typically 3-5 mg/ mL) is the most widely used matrix solution in MALDI MS of monolayers.6-9 Ethanol is a general solvent for SAM growth using thiol reagents in solution. Water was also examined as a nonalcoholic matrix-dissolving solvent. The concentration of all matrix solutions was 5 mg/mL. Matrix solution was applied by dropping 1 µL of the solution onto the monolayers. Figure 1 shows the MALDI mass spectra obtained using the three matrix solutions. As seen in the mass spectra, both alcoholic solvents (a) methanol and (b) ethanol clearly produced the characteristic ions for Pep1-immobilized SAMs at m/z ) 1993.1 and those for background SAMs at m/z ) 693.4. However, as shown in Figure 1c, when water was employed as the solvent, the characteristic ions for Pep1-immobilized SAMs were not produced at all, although those for background SAMs at m/z ) 693.4 were still observed. In addition, when the matrix application was done by vacuum sublimation, which is a solvent-free condition, the MALDI mass spectrum taken from the matrixcoated SAM chip did not give any characteristic ion peak for the monolayer (Figure S2). In this study, it was learned that MALDI efficiency was greatly influenced by the choice of matrix solvent, which suggests that a certain process occurring in solution during matrix treatment may play an important role in the resulting MALDI process. Observations of Characteristic Ions from Matrix Crystals. The MALDI experiments were examined in more detail. Figure 2a shows the optical image for the Pep1-immobilized SAM

Figure 1. Effects of the matrix solution on MALDI MS of SAMs with Pep1. Mass spectra were obtained using (a) methanol, (b) ethanol, and (c) water as the matrix solvent. THAP was employed for the matrix.

Figure 2. Optical images of matrix-treated SAMs with Pep1 in the mass spectrometer (a) before and (b) after the MALDI mass spectrum was taken. (c) MALDI mass spectrum obtained from the region on the matrix rim indicated by the dotted circle. The inset shows the color image of the SAM chip. THAP in methanol (5 mg/mL) was used for matrix treatment.

surface on which a 1 µL drop of THAP solution (5 mg/mL in methanol) was applied and dried, resulting in a rim of matrix crystals on the surface. In the preliminary test for the monolayer, it was found that the characteristic ions for Pep1-immobilized

Self-Assembled Alkanethiolate Monolayers on Gold

Figure 3. MALDI mass spectrum for matrix crystals collected from the surface of SAMs with Pep1 using conductive adhesive tape.

SAMs at m/z ) 1993.1 were produced when the rim was irradiated with a UV laser. Laser irradiation outside of the rim region rarely produced the characteristic ion signals. A separate imaging MS experiment led to a clearer observation of the same phenomena. In the mass imaging experiment, the matrix treatment was done by dipping the entire SAM chip in the matrix solution, which gave larger and thicker matrix crystals on the surface. The resulting mass image mapped for the ion peak at m/z ) 1993.1 revealed that the characteristic ions were produced only from the parts of the SAM surface that were covered with matrix crystals (Figure S1). Therefore, we focused on the production of characteristic ions from crystal regions. A MALDI mass spectrum was obtained at low laser power from the rim region marked by the dotted circle in Figure 2a, and ion signals were accumulated by scanning the MALDI laser over the region. As shown in Figure 2c, a good mass spectrum that exhibits characteristic ions for both background SAMs and Pep1-immobilized SAMs at m/z values of 693.4 and 1993.1, respectively, was obtained. Interestingly, even after obtaining a good mass spectrum, the bare surface of the monolayer was not exposed (Figure 2b). The SAM surface from which MALDI ions were produced was still covered with matrix crystals whose presence would only hinder MALDI ions from escaping from the SAM surface. It is therefore unlikely that the characteristic ions originated directly from the SAM surface. Instead, the MALDI ions may be produced from the matrix crystals that the MALDI laser actually interrogated. To examine this possibility, a more direct experiment was performed. To examine possible ion formation from matrix crystals, a part of the crystal rim was collected using conductive adhesive tape, as illustrated in Figure 3a. Tiny matrix crystals were successfully sampled on the tape and then loaded into a MALDI mass spectrometer. The MALDI mass spectrum for the collected crystals is shown in Figure 3b. We observed the formation of both characteristic disulfide ions for background SAMs and Pep1-immobilized SAMs at m/z values of 693.4 and

Langmuir, Vol. 25, No. 6, 2009 3695

1993.1, respectively. This finding clearly demonstrates that the characteristic MALDI ions can be produced from matrix crystals. MALDI versus UV-Laser-Induced Direct Desorption. As a control experiment, we conducted a comparative study of ionization behaviors in MALDI and CALDI of alkanethiol SAMs on gold. The CALDI method is a matrix-free laser desorption/ ionization scheme for alkanethiol SAMs on gold that was developed on the basis of previous findings in two-laser MS investigations.26 In earlier two-laser MS investigations, UV laser irradiation (337 nm) induced the direct desorption of intact alkanethiolate molecules from monolayers, typically producing disulfide products during the desorption process. The desorbed disulfide molecules under vacuum were further postionized by a VUV laser (118 nm) and were then subject to mass analysis.21-23 The CALDI method is a strategy that supplies the charge required for mass analysis by cationization of SAM molecules prior to laser-induced desorption. It was shown that this is possible by simply treating monolayers with a salt-containing solution such as NaI(aq); this substitutes for the positioning step in two-laser MS. Therefore, the CALDI method can be considered to be a measure of direct desorption by UV laser irradiation from alkanethiol SAMs on gold. By comparing ionization behaviors with (MALDI) and without matrix assistance (CALDI), we attempted to gain further insight into the role of the matrix in laser desorption/ionization. Figure 4 shows the CALDI and MALDI mass spectra obtained for the SAMs with Pep1 and Pep2. For the CALDI MS experiments, a SAM chip was soaked in a 10 mM solution of NaI(aq) for 10 min and dried under ambient conditions. The CALDI mass spectrum was then taken using a MALDI mass spectrometer without any further treatment such as matrix application. After the CALDI measurement, the SAM chip was taken out and thoroughly rinsed with deionized water until CALDI ion signals were no longer observed. A drop of matrix solution (5 mg/mL THAP in 100% methanol) was applied to the same SAM chip and dried. The MALDI mass spectrum was then obtained from the SAM chip. Figure 4a shows the CALDI mass spectrum for the Pep1-immobilized SAMs. As shown in the spectrum, although the characteristic ions for background SAMs at m/z ) 693.4 were produced, the formation of ions for Pep1immobilized SAMs at m/z ) 1993.1 was not observed. However, both characteristic ions at m/z values of 693.4 and 1993.1 were produced from the same SAM chip in the following MALDI MS experiment (Figure 4b). Another experiment using the SAMs with Pep2 gave exactly the same results (Figure 4c,d). As reflected in the above CALDI results, the direct desorption of large peptideimmobilized SAM molecules only by UV laser irradiation seems to be quite inefficient, whereas the use of a matrix in the MALDI method results in good desorption/ionization behavior for the large SAM molecules. This indicates that direct desorption alone cannot fully account for the MALDI ionization. In other words, the matrix assistance in MALDI indeed plays a critical role in the ionization process. For MALDI MS of biopolymers, analyte molecules must be cocrystallized with excess matrix molecules to produce MALDI ions upon UV laser irradiation. In the case of MALDI MS of SAMs, however, it is quite unlikely that a SAM molecule in the monolayer will be fully cocrystallized with excess of matrix molecules because of the close-packing nature of self-assembly. In addition, the matrix molecules may not penetrate down to the Au-S bonds where actual desorption photochemistry should be initiated. Thus, it would also be difficult for matrix molecules to sensitize desorption photochemistry. Likewise, the known mechanism for MALDI MS of biopolymers may not be applicable

3696 Langmuir, Vol. 25, No. 6, 2009

Ha et al.

Figure 4. (a) CALDI and (b) MALDI mass spectra for SAMs with Pep1 and (c) CALDI and (d) MALDI mass spectra for SAMs with Pep2.

Figure 5. MALDI mass spectra for (a) the recovered methanol from the surface of SAMs with Pep1 and (b) the SAMs after the methanol test. MALDI mass spectra for (c) the recovered water and (d) the SAM surface after the water test.

to this case of self-assembled monolayers. Therefore, a new mechanism other than UV-laser-induced direct desorption or conventional MALDI of biopolymers must account for the observed phenomena for MALDI of alkanethiol SAMs on gold. Formation of Characteristic Disulfide Ions in Solution. The formation of disulfide ions in the MALDI process also requires an explanation; these disulfide ions may be viewed as products of direct photodesorption. However, as the above results indicate, UV laser-induced desorption is not capable of producing disulfide ions for SAMs with large biomolecules. In fact, it was shown that the solvent employed in the matrix application had an effect on the resulting MALDI behavior (Figure 1). To determine if the solution chemistry during matrix treatment is related to the behavior of MALDI, we further investigated the reaction of solvent with SAM molecules in the monolayers.

First, the reaction of methanol, a typical solvent for the THAP matrix solution, was examined with respect to Pep1-immobilized SAMs. Pure methanol (2 µL) was dropped onto the SAM surface and was collected after a few seconds using the same pipet. The recollected solvent was loaded onto a MALDI target plate. Because the amount of collected solvent was small, the procedure was repeated two more times on the same SAM chip. Then, to analyze the reaction products contained in the recovered solvent, the MALDI mass spectrum for the solvent collected on the plate was recorded using a THAP solution (5 mg/mL in methanol). For comparison, the MALDI spectrum was obtained from the same SAM surface used for the test as well. As shown in Figure 5a,b, the two MALDI mass spectra for the recovered solvent and the SAM surface display essentially the same features. Both spectra exhibit disulfide ions representing the molecular com-

Self-Assembled Alkanethiolate Monolayers on Gold

positions of the mixed SAMs presenting Pep1 for background SAMs, mixed SAMs with -OH and -NH2 terminals, and Pep1immobilized SAMs at m/z values of 693.4, 824.5, and 1993.1. As clearly seen in Figure 5a, brief contact of methanol with the SAM surface instantaneously produced all characteristic ions in the solvent, which are probably the products of solvent extraction. We also examined the reaction of water solvent. As shown in Figure 1c, water was a poor matrix solvent for MALDI of SAMs with large biomolecules. The THAP matrix solution with water solvent was not able to produce MALDI ions for SAMs with large peptides. The same procedure used for the methanol study was applied in the MALDI experiment for the water case. As shown in Figure 5c, no characteristic ions, neither the disulfide ions for background SAMs nor the ions for SAMs with Pep1, were found in the recovered water solvent. However, in the following MALDI MS of the same SAM surface using the THAP matrix (5 mg/mL in methanol), the characteristic disulfide ions for the SAMs with Pep1 were all produced (Figure 5d), indicating that the sample was good. The results are consistent with the MALDI MS results presented in Figure 1c; the disulfide ions for background SAMs observed in Figure 1c are suspected to be products of the CALDI process. The growth of thiol SAMs on gold in solution is a dynamic process.1,28,29 The SAM molecules in monolayers can be exchanged with free thiols in the solution and can also be released back into the solution. Our results indicate that the disulfide ions are not necessarily the products of photochemical reactions but may result from solvent extraction, probably during the application of matrix solution to the sample. Disulfide formation is likely due to a low-energy pathway for desorption in solution, whose products are later detected as characteristic species by MALDI MS. Therefore, MALDI MS of alkanethiol SAMs on gold results in characteristic alkanethiolate species that are produced in the matrix solution during matrix application and are then cocrys(28) Schlenoff, J. B.; Li, M.; Ly, H. J. Am. Chem. Soc. 1995, 117, 12528– 12536. (29) Schwartz, D. K. Annu. ReV. Phys. Chem. 2001, 52, 107–137.

Langmuir, Vol. 25, No. 6, 2009 3697

tallized with matrix molecules on the SAM surface. MALDI MS is thus capable of characterizing self-assembled monolayers containing biomolecules with high molecular masses.

Conclusions In this study, we investigated the MALDI process for alkanethiol SAMs on gold. MALDI MS is employed to characterize SAMs of alkanethiolates on gold. However, to date, its mechanism has not been explored in any detail. A previous study using two-laser MS reported that UV laser irradiation induced direct desorption of SAM molecules from alkanethiol SAMs on gold, resulting in the formation of characteristic disulfide species under vacuum. However, we propose that treatment of the sample with matrix solution is the reason for the characteristic ion formation in MALDI. The matrix solution appears to be able to dissolve alkanethiolate molecules from SAMs, thereby producing characteristic disulfide species in the solution. The disulfides are then cocrystallized with matrix molecules and detected by subsequent MALDI MS. This study therefore deepens our understanding of the MALDI process for characterizing alkanethiol SAMs on gold, which may allow further development in biochemical applications using SAMs with mass spectrometric characterization. Acknowledgment. This work was supported by KRCF through the project of “Development of Characterization Techniques for Nano-materials Safety” and the Bio-signal Analysis Technology Innovation Program (M106450100002-06N4501-00210) of MEST via KOSEF. T.K.H. thanks the second stage Brain Korea 21 (BK21) program for its scholarship support. Supporting Information Available: MALDI mass image for SAMs presenting Pep1, MALDI experiment using matrix application by vacuum sublimation, and summary of characteristic ions observed in this work. This material is available free of charge via the Internet at http://pubs.acs.org. LA8036567