Anal. Chem. 2005, 77, 350-353
Correspondence
Radio Frequency Plasma Polymer Coatings for Affinity Capture MALDI Mass Spectrometry Meiling Li, Richard B. Timmons, and Gary R. Kinsel*
Department of Chemistry and Biochemistry, The University of Texas at Arlington, Arlington, Texas 76019-0065
Surface modification of MALDI probes is an attractive approach for combining bioaffinity isolation of targeted biomolecules with mass spectrometric analysis of the captured species. In this work, we demonstrate that a polymer thin film, produced by pulsed rf plasma polymerization of allylamine and deposited directly on a MALDI probe, can be subsequently biotinylated to develop a bioaffinity capture MALDI probe. The synthesis and characterization of the probe by XPS, FT-IR, and AFM is described, and the selective isolation of avidin from a three-component mixture of avidin, lysozyme, and cytochrome c is presented. These initial results offer encouragement for the further exploration of rf plasma polymer deposition as a novel approach for the development of onprobe affinity capture MALDI probes. The analysis of low concentrations of active biomolecules contained in complex biological samples is a constant challenge for bioanalytical chemistry. Bioaffinity isolation of a targeted biomolecule is a well-established approach to simplification of the analytical challenge,1 and matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS) offers one of the most sensitive approaches for biomolecular detection.2 Thus, the coupling of bioaffinity isolation with MALDI MS is expected to yield an extraordinarily powerful method for the analysis of targeted biomolecules in complex mixtures. Various approaches to coupling bioaffinity isolation with MALDI MS have been described.3-27 In general, these techniques may be classified into two categories, “off-probe” and “on-probe” * To whom correspondence should be addressed. E-mail:
[email protected]. Fax: 817-272-3808. (1) Turkova´, J. Bioaffinity Chromatography, 2nd ed.; Elsevier Science Publisher B.V.: New York, 1993; pp 1-27. (2) Keller, B.; Li, L. J. Am. Soc. Mass Spectrom. 2001, 12, 1055-1063. (3) Hutchens, T. W.; Yip, T. T. Rapid Commun. Mass Spectrom. 1993, 7, 576580. (4) Zhao, Y.; Kent, S. B. H.; Chait B. T. Anal. Chem. 1994, 66, 3723-3726. (5) Nelson, R. W.; Krone, R. K.; Williams, P. Anal. Chem. 1995, 67, 11531158. (6) Schriemer, D. C.; Li, L. Anal. Chem. 1995, 68, 3382-3387. (7) Papac, D. I.; Hoyes, J.; Tomer, K. B. Anal. Chem. 1994, 66, 2609-2613. (8) Schriemer, D. C.; Yalcin, T.; Li, L. Anal. Chem. 1998, 70, 1569-1575. (9) Hurst, G. B.; Buchanan, M. V.; Kennel, S. J. Anal. Chem. 1999, 71, 47274733. (10) Niederkofler, E. E.; Tubbs, K. A.; Nelson, R. W. Anal. Chem. 2001, 73, 3294-3299. (11) Brockman, A. H.; Orlando, R. Anal. Chem. 1995, 67, 4581-4585.
350 Analytical Chemistry, Vol. 77, No. 1, January 1, 2005
techniques. Off-probe techniques refer to methods in which the isolation of the biomolecule is accomplished in a chromatography column, in a pipet tip, or on agarose beads prior to deposition of the isolated target molecule on the MALDI MS probe.3-10 While these off-probe methods are demonstrably effective for MALDI MS analysis of targeted biomolecules, these approaches can lead to loss of sample during the biomolecule isolation and transfer steps, resulting in poorer limits of detection. On-probe techniques, of which SELDI is perhaps most widely recognized, generally refer to methods in which biomolecule isolation is accomplished directly on the MALDI MS probe.11-27 These on-probe approaches eliminate the need for transfer steps, potentially leading to lower limits of detection for targeted compounds, and may be somewhat simpler to integrate in automated MALDI MS analysis. Modifications to the MALDI probe surface for on-probe affinity capture of targeted biomolecules have taken a variety of approaches. Commercial polymers have been used both with and without chemical derivatization.14,15,23-26 Self-assembled monolayers on gold surfaces have also been used both as is and following chemical derivatization to incorporate lectin and dextran.11-13,16 Polymer thin films cast directly on the MALDI probe have also been modified to incorporate immobilized bioaffinity ligands.27 (12) Brockman, A. H.; Orlando, R. Rapid Commun. Mass Spectrom. 1996, 10, 1688-1692. (13) Bundy, J.; Fenselau, C. Anal. Chem. 1999, 71, 1460-1463. (14) Liang, X.; Lubman, D. M. Anal. Chem. 1998, 70, 498-503. (15) Wang, H.; Tseng, K.; Lebrilla, C. B. Anal. Chem. 1999, 71, 2014-2020. (16) Neubert, H.; Jacoby, E.; Bansal, S.; Iles, R.; Cowan, D.; Kicman, A. T. Anal. Chem. 2002, 74, 3677-3683. (17) Tang, H.; Tornatore, P.; Weinberger, S. R. Mass Spectrom. Rev., 2004, 23, 34-44. (18) Merchant, M.; Weinberger, S. R. Electrophoresis 2000, 21, 1164-1167. (19) Petricoin, E. F.; Liotta, L. A. Trends Biotechnol. 2002, 20, S30-S34. (20) Issaq, H. J.; Veenstra, T. D.; Conrads, T. P.; Felschow, D. Biochem. Biophys. Res. Commun. 2002, 292, 587-592. (21) Roecken, C.; Ebert, M. A.; Roessner, A. Pathol., Res. Pract. 2004, 200, 69-82. (22) Clarke, W.; Zhang, Z.; Chan, D, W. Clin. Chem. Lab. Med. 2003, 41, 15621570. (23) Bundy, J. L.; Fenselau, C. Anal. Chem. 2001, 73, 751-757. (24) Blackledge, J. A.; Alexander A. J. Anal. Chem. 1995, 67, 843-848. (25) Worrall, T. A.; Cotter, R. J.; Woods, A. S. Anal. Chem. 1998, 70, 750-756. (26) Bai, J.; Qian, M. G.; Liang, X.; Lubman, D. M. Anal. Chem. 1995, 67, 17051710. (27) Hobbs, S.; Shi, G.; Bednarski, M. D. Bioconjugate Chem. 2003, 14, 526531. 10.1021/ac0488107 CCC: $30.25
© 2005 American Chemical Society Published on Web 12/02/2004
In this paper, we describe an alternate approach wherein a pulsed rf plasma28,29 is used to deposit a thin polymer film directly on the surface of a MALDI probe with subsequent covalent immobilization of a bioaffinity ligand. Plasma polymer deposition is well established as an efficient, all-dry method for the deposition of pinhole-free, conformal thin films having a broad spectrum of chemical functionality on a wide variety of substrates.28,29 Furthermore, pulsing of the rf plasma allows for systematic adjustment of the retention of the chemical functionality of the monomer used for polymer formation.30 Because of these attractive features, our efforts have focused on the development of pulsed rf plasma polymers for on-probe affinity capture (OPAC) MALDI mass spectrometry. Here, we show that a robust bioaffinity surface for the capture of targeted proteins directly on a MALDI-MS probe can be created through the attachment of biotin to a MALDI probe modified via pulsed rf plasma polymerization of allylamine. EXPERIMENTAL SECTION Materials. Allylamine was obtained from Aldrich (Milwaukee, WI). Low-density polyethylene (LDPE) was obtained from GoodFellow Inc. (Berwyn, PA). NHS-biotin was obtained from Molecular Biosciences, Inc. (Boulder, CO). Avidin was obtained from Sigma (St. Louis, MO). Anhydrous dimethyl sulfoxide (DMSO) was purchased from Fisher Chemical (Fairlawn, NJ). Formic acid was purchased from Spectrum Laboratory Products, Inc. (New Brunswick, NJ). Pulsed rf Plasma Modification of the MALDI Probe Surface. The pulsed rf plasma reactor is a custom-designed, laboratory-constructed instrument and has been described in detail elsewhere.30 Silicon wafers (for XPS and AFM characterization), NaCl disks (for FT-IR experiments), and 4.8-mm-diameter LDPE disks (for MALDI MS experiments), were cleaned with acetone and affixed to glass slides. The slides were then placed in the center of a 10 cm × 30 cm cylindrical glass reactor and treated with a 10-min pulsed argon plasma to provide additional surface cleaning. Next, the entire system was evacuated to background pressure (4 mTorr), and a constant flow of allylamine was introduced to the plasma reactor. Allylamine polymerization was carried out at a constant power input of 200 W using plasma duty cycles of 3 ms on/5 ms off for 3 min and then 3 ms on/45 ms off for 12 min. Finally, the system was evacuated to background pressure and held for 1 h before removing the samples from the plasma reactor. This final annealing step was taken because of the propensity of amine-containing plasma polymers to undergo surface air oxidation.31,32 Biotinylation of the Polymer-Modified MALDI Probe. Once created, the allylamine plasma polymer-modified surfaces were immediately subjected to subsequent chemical derivitization. A 5-µL aliquot of a 1 mg/mL solution of NHS-biotin in DMSO was deposited on the allylamine-modified MALDI probe and allowed to stand for 30-40 min at room temperature. Subsequently, the surface was washed with deionized water to remove unreacted NHS-biotin. (28) Savage, C. R.; Timmons, R. B.; Lin, J. W. Adv. Chem. Ser. 1993, 236, 745768. (29) Rinsch, C. L.; Chen, X.; Panchalingam, V.; Savage, C. R.; Wang, Y. H.; Eberhart, R. E.; Timmons, R. B. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 1995, 36 (1), 95-96. (30) Panchalingam, V.; Chen, X.; Savage, C. R.; Timmons, R. B.; Eberhart, R. C. J. Appl. Polym. Sci: Appl. Polym. Symp. 1994, 54, 123-141.
Table 1. Percent Atom Content of the Allyl Amine Plasma Polymer-Modified MALDI Probe before and after Biotinylation as Determined by XPS
before biotinylation after biotinylation
carbon (%)
oxygen (%)
nitrogen (%)
sulfur (%)
69.17 68.90
7.25 9.75
23.57 20.11
0.00 1.24
Surface Analysis via XPS, FT-IR, and AFM. XPS analyses of the probe surfaces were performed using a Perkin-Elmer PSI 5000 series instrument. The 1486.8-eV Al KR X-ray source impinged on the surfaces at a 45° angle. Charge buildup on the surfaces was neutralized using an electron flood gun during recording of the XPS spectra. Calibration of the XPS spectra was accomplished by centering the lowest binding energy component of the C1s signal at 284.6 eV. FT-IR analysis was performed using a Bruker Vector 22 instrument to characterize the surfaces as deposited on sodium chloride disks. AFM analysis was performed using a Quesant Q-scope 250 instrument operated in intermittent contact mode to characterize the surfaces as deposited on silicon wafers. Sample Preparation and MALDI MS Analysis. A stock solution containing avidin, lysozyme, and cytochrome c, each at a concentration of 1 mg/mL, was prepared in phosphate-buffered saline (PBS). To test the on-probe biospecific isolation of avidin, a 3-µL aliquot of the protein mixture was applied to the biotinmodified MALDI probe and allowed to stand for ∼15 min. Subsequently, the protein solution was aspirated away using a pipet and the surface was washed several times using 3-µL aliquots of PBS. Following the washing procedure, a 3-µL aliquot of formic acid (10%) and a 3-µL aliquot of the MALDI matrix solution (Rcyano-4-hydroxycinnamic acid, 15 mg/mL in methanol) were applied. After thorough drying, the sample was subjected to MALDI mass spectrometric analysis on a laboratory-constructed linear time-of-flight mass spectrometer, described in detail previously.33 RESULTS AND DISCUSSION The OPAC MALDI probe was characterized at all stages of synthesis by XPS, FT-IR, MALDI MS, and AFM to confirm the chemical modifications imparted. Table 1 lists the percent atom content of the MALDI probe surface, as determined by XPS, after allylamine plasma polymer deposition and following reaction of this surface with NHS-biotin. The XPS spectrum of the allylamine plasma polymer-modified surface reveals the deposition of a polymer containing a large amount of nitrogen, consistent with previous studies of plasma polymerization of allylamine.29 Following biotinylation of this surface, a new peak at 158 eV, which may be attributed to the S2p peak, is clearly observed in the XPS spectrum and accounts for 1.24% of the surface atom content. The uptake of sulfur by the plasma polymer may be taken as an indication of the successful attachment of NHS-biotin to the MALDI probe surface. (31) Gengenbach, T. R.; Chatelier, R. C.; Griesser, H. J. Surf. Interface Anal. 1996, 24, 271-281. (32) Gengenbach, T. R.; Chatelier, R. C.; Griesser, H. J. Surf. Interface Anal. 1996, 24, 611-619. (33) Walker, A. K.; Wu, Y.; Nelson, K. D.; Timmons, R. B.; Kinsel, G. R. Anal. Chem. 1999, 71, 2014-2020.
Analytical Chemistry, Vol. 77, No. 1, January 1, 2005
351
Figure 1. FT-IR spectra of the allylamine plasma polymer, as deposited on NaCl disks, (A) before and (B) after biotinylation, and (C) a spectrum of pure NHS-biotin for reference.
Figure 3. AFM micrographs of the allylamine plasma polymermodified silicon wafer (A) before and (B) after biotinylation.
Figure 2. MALDI mass spectra obtained from the OPAC MALDI probe (A) before and (B) after biotinylation.
Further support for the successful biotinylation of the allylamine plasma polymer-modified MALDI probe is found in the FT-IR spectra obtained for modified and unmodified allylamine plasma polymer films deposited on NaCl disks and for free NHSbiotin (see Figure 1). The FT-IR spectrum of free NHS-biotin (Figure 1C) shows multiple peaks at 1702, 1731, and 1748 cm-1 from the ester and amide C-O linkages. The FT-IR spectrum of NHS-biotin immobilized on the allylamine plasma polymer surface (Figure 1B) is very similar to that of the free NHS-biotin but includes only the carbonyl peak at 1704 cm-1. This result supports covalent attachment of NHS-biotin to the allylamine plasma polymer via amide bond formation. Figure 2 shows the MALDI mass spectra obtained by depositing the MALDI matrix on the plasma polymer-coated probe before and after biotinylation. Before biotinylation (Figure 2A), prominent peaks in the low-mass region that may be assigned to matrix ions 352 Analytical Chemistry, Vol. 77, No. 1, January 1, 2005
are observed as well as a distribution of low-abundance peaks centered at m/z 400 that may be ions derived from the plasma polymer. After biotinylation (Figure 2B), only peaks attributable to the MALDI matrix are observed in the low-mass region and no ions that may be associated with noncovalently attached NHSbiotin are detected. Finally, AFM images taken before and after biotinylation of the allylamine plasma polymer show a clear change in surface morphology (see Figure 3). The average roughness of the surface, as indicated by the rms z-deviation, is ∼4.4 nm before biotinylation and increases to ∼73.6 nm after biotinylation. Figure 4 shows the results of testing of the bioaffinity capture capabilities of the biotinylated allylamine plasma polymer OPAC MALDI probe. Figure 4A is a conventional MALDI mass spectrum of a mixture of avidin, lysozyme, and cytochrome c obtained from sample deposited on an unmodified stainless steel MALDI probe. The mass spectrum shows peaks corresponding to the avidin monomer at 15 938 Da, lysozyme at 14 313 Da, and cytochrome c at 12 232 Da, as well as peaks at half these masses resulting from the doubly charged species. Interestingly, in solution avidin exists as a noncovalent tetramer having a mass of 63 752 Da. The observation of the avidin monomer in these MALDI mass spectra is likely due to the strong denaturing properties of the formic acid and methanol solvents used in the MALDI sample preparation. Figure 4B shows the MALDI mass spectrum of the same protein mixture deposited on the biotinylated MALDI probe surface and following washing to remove nonspecifically bound proteins. Only the signal associated with the avidin monomer is
cytochrome c are clearly observed in the MALDI mass spectrum of the wash solution, as well as a small signal for the avidin monomer. This observation of the avidin monomer signal in the wash solution mass spectrum is likely due to saturation of the biotin binding sites on the OPAC MALDI probe.
Figure 4. MALDI mass spectra of a mixture of avidin, lysozyme, and cytochrome c (A) as deposited on a stainless steel probe, (B) as deposited on the biotinylated OPAC MALDI probe and after washing, and (C) the washing solution as deposited on a stainless steel probe.
observed in the MALDI mass spectrum demonstrating the highly specific binding affinity of the biotinylated allylamine plasma polymer-modified probe. Furthermore, consistent with the results shown in Figure 2B, the lower mass region of the spectrum only contains ions that may be associated with the MALDI matrix. To confirm that the lysozyme and cytochrome c are removed in the washing procedure a MALDI mass spectrum was acquired of the wash solution as deposited on a stainless steel probe surface and is shown in Figure 4C. MALDI ion signals for the lysozyme and
CONCLUSION Radio frequency plasma polymer deposition offers a new approach for the modification of MALDI probes to allow on-probe affinity capture of targeted biomolecules. Radio frequency plasma polymerization is well suited to this application as it allows the direct deposition of polymer films having a wide array of chemical functionality, for use as is or following subsequent chemical modification, in a simple one-step process. The results presented in this study confirm that a plasma polymer-modified MALDI probe can be subsequently derivitized to produce a highly selective bioaffinity capture MALDI probe. Indeed, numerous other biomolecule binding motifs could be developed using the plasma polymer deposition approach, either through the direct deposition of polymers having affinities for various chemical classes of biomolecules or through subsequent derivitization of the plasma polymers to produce bioselective OPAC MALDI probes. The ease of probe modification and effectiveness of bioaffinity isolation offer encouragement for further study of rf plasma polymers as a means for the development of OPAC MALDI probes. ACKNOWLEDGMENT The support of the National Science Foundation (CHE0317073) is gratefully acknowledged. Received for review August 11, 2004. Accepted November 15, 2004. AC0488107
Analytical Chemistry, Vol. 77, No. 1, January 1, 2005
353
