Anal. Chem. 1999, 71, 518-521
Use of Poly(tetrafluoroethylene)s as a Sample Support for the MALDI-TOF Analysis of DNA and Proteins K. C. Hung, H. Ding, and Baochuan Guo*
Department of Chemistry, Cleveland State University, Cleveland, Ohio 44115
Matrix-assisted laser desorption/ionization mass spectrometry of DNA and proteins, directly deposited on the poly(tetrafluoroethylene) (Teflon) surface, is demonstrated. For DNA analysis, this technique apparently produces a more homogeneous coverage of the matrix/ DNA over the sample surface. Moreover, it enhances the sensitivity and salt tolerance. As described here, this technique can also achieve an excellent mass resolution, similar to that observed using a metal probe for DNA up to 62mer. We also examined the use of Teflon as a sample support for protein analysis since Teflon has been used as a transfer membrane. Less than 25 fmol of myoglobin has been detected with this technique. In addition, effective MALDI-TOF analysis of salt-contaminated protein samples can also be accomplished by loading the protein sample onto Teflon, followed by steps of washing away salts, adding the matrix, and desorbing sample directly from Teflon. Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF) is a potential method for automated, low-cost, and high-accuracy DNA analysis. During the past several years, progress has been made in the MALDI-TOF analysis of DNA due to the discovery of a 3-hydroxypicolinic acid (3-HPA) matrix.1,2 This discovery made it possible to detect DNA up to 500 bases.3,4 Although several other matrixes have been subsequently discovered,5,6 3-HPA is still the most effective and widely used matrix for the MALDI-TOF analysis of DNA. The conventional MALDI sample preparation method uses a metal sample tip. The 3-HPA/DNA solution tends to spread to a large area on the metal surface as a result of strong interactions between metals and the polar solvents used. For example, 0.5 µL of the solution can spread to an area of 1.5 mm diameter. Moreover, the 3-HPA crystal film formed on the metal surface is * To whom correspondence should be addressed. E-mail: B.GUO@ POPMAIL.CSUOHIO.EDU. (1) Wu, K. J.; Steding, A.; Becker, C. H. Rapid Commun. Mass Spectrom. 1993, 7, 142. (2) Wu, K. J.; Shaler, T. A.; Becker, C. H. Anal. Chem. 1994, 66, 1637. (3) Tang, K.; Taranenko, N. I.; Allman, S. L.; Chang, Y. L.; Chen, C. H. Rapid Commun. Mass Spectrom. 1994, 8, 727. (4) Ross, P. L.; Belgrader, P. Anal. Chem. 1997, 69, 3966. (5) Zhu, Y. F.; Chung, C. N.; Taranenko, N. I.; Allman, S. L.; Martin, S. A.; Haff, L.; Chen, C. H. Rapid Commun. Mass Spectrom. 1996, 10, 383. (6) Hunter, J.; Hua, L.; Becker, C. H. Presented at the 45th ASMS Conference on Mass Spectrometry and Allied Topics, Palm Spring, CA, June, 1997.
518 Analytical Chemistry, Vol. 71, No. 2, January 15, 1999
highly heterogeneous, and 3-HPA tends to form crystals residing mainly at the rim of the sample drop, where sweet spots (producing good DNA signals) are often found. The formation of a large, heterogeneous 3-HPA film on the metal surface causes two potential problems. First, the DNA ion yield from MALDI depends on the absolute concentration on a particular desorption spot instead of the total amount of DNA loaded. The formation of a large sample film would inevitably reduce the DNA concentration on the desorption spots, thereby reducing the detection sensitivity. Second, the formation of a highly heterogeneous MALDI sample makes it difficult to locate the sweet spots from a large sample area, especially when low amounts of DNA are used. One of the potential advantages of MALDI-TOF is that it allows massive parallel analysis to increase throughput.7 However, this advantage is negated if long times are required to locate sweet spots. Recently, we discovered that the use of a Parafilm sample support improved the detection and ease of collecting high-quality spectra in the MALDI-TOF analysis of DNA.8 We attributed this improvement to the hydrophobic nature of Parafilm. The aqueous 3-HPA solution formed a small droplet on the hydrophobic Parafilm surface instead of quickly spreading over a large area as it did on metal surfaces. As a result, the sample film formed on a Parafilm surface was smaller and more homogeneous than that formed on a metal surface, thereby improving the performance of MALDI. Since the hydrophobic surface appears to play a key role in this improvement, it stands to reason that other hydrophobic materials should also work well for the DNA analysis. We recently tested this hypothesis using polyethylenes and poly(tetrafluoroethylene)s (Teflon), both hydrophobic materials, as the sample support. It was found that both materials worked well for the DNA analysis, with Teflon working the best among the hydrophobic materials that we have tested. It was found that the Teflon sample support also worked well for the protein analysis. This finding could be potentially important to the protein analysis since Teflon has been used as a transfer membrane.9 Reporting the results using Teflon as a sample support for the MALDI-TOF analysis of DNA and proteins constitutes the focus of this article. (7) Little, D. P.; Cornish, T. J.; O’Donnell, M.; Braun, A.; Cotter, R. J.; Koster, H. Anal. Chem. 1997, 69, 4540. (8) Hung, K. C.; Rashidzadeh, H.; Wang, Y.; Guo, B. C. Anal. Chem. 1998, 70, 3088. (9) Burkhard, W. A.; Moyer, M. B.; Bailey, J. M.; Miller, G. C. Anal. Biochem. 1996, 236, 364. 10.1021/ac980824n CCC: $18.00
© 1999 American Chemical Society Published on Web 12/15/1998
EXPERIMENTAL SECTION Two MALDI-TOF systems were used in this study. Most of the works reported here were conducted using a homemade linear MALDI-TOF instrument operating on a continuous extraction mode. A detailed description of this instrument was given previously.10 Briefly, a Nd:YAG laser of 355 nm was used for MALDI. The two-stage acceleration voltages were set at 30 and 15 kV, respectively. A pulsed electric field was applied to the deflection plates to remove low-mass ions. An electron multiplier was used to detect ions. All mass spectra were produced in the positive ion mode and averaged over 10 shots. The work involving delayed extraction was performed using a Dynamo MALDI-TOF spectrometer (Thermo BioAnalysis, Santa Fe, NM). The synthetic DNA of 17-, 41-, 62-, and 85mers were obtained from National Biosciences, Inc. (Plymouth, MN). The 26mer was a gift from Perseptive Biosystems. The matrix 3-HPA, diammonium citrate, and the metal salts were obtained from Aldrich (Milwaukee, WI). Myoglobin and the matrix R-cyanohydroxycinnamic acid (CHCA) were from Sigma (St. Louis, MO). All samples were used without further purification. Teflon was obtained from a local hardware store. The procedure for preparation of a DNA sample tip on the Teflon surface was similar to that used in the Parafilm work.8 For the protein analysis, the myoglobin solution was prepared in a 0.1% TFA aqueous solution, and the CHCA solution (10 g/L) was prepared using the 50% CH3CN aqueous solution. The protein sample was first loaded on the Teflon probe, followed by the addition of the matrix solution. RESULTS AND DISCUSSION DNA Analysis. The characteristic of forming the sample film on Teflon is similar to that observed on Parafilm. Figure 1 shows the MALDI-TOF mass spectrum of a DNA 26mer obtained by loading the matrix solution to Teflon, followed by the addition of the 26mer solution to the dried matrix film. The total amount of DNA loaded was approximately 15 fmol. We found that the laser power threshold with the Teflon method was similar to that observed on metal surfaces. However, the use of Teflon improved the shot-to-shot and sample-to-sample reproducibility of the DNA ion signal, especially when small amounts of DNA were loaded. In addition, it was easy to locate sweet spots. This suggests that the use of a Teflon surface produces a more homogeneous coverage of the matrix/DNA over the sample surface. In addition, we found that the thickness of the Teflon support was not critical. The Teflon support also works well for larger DNA. For example, we were able to detect an 85mer at the level of 50 fmol with a Teflon surface. This detection limit was better than that obtained using the Parafilm and the stainless steel methods. It should be noted that this comparison was made on the basis of our evaluation of using the same DNA sample and instrumental conditions since the detection limits depend on the sample purity and the experimental conditions. The detection of larger DNA components in the presence of a large excess of smaller DNA components is an important consideration in the analysis of the PCR products, since a large molar excess of the primers may be present in PCR reaction products. Recently, it was found that a 28-fold molar excess of a 20mer (8.3 pmol loaded) eliminated the signal from a 72mer (300 (10) Chen, H.; Guo, B. C. Anal. Chem. 1997, 69, 4399.
Figure 1. MALDI-TOF mass spectrum of a 26mer obtained using a Teflon sample support. The total amount of the 26mer loaded on the probe was about 15 fmol.
fmol) when a metal surface was used.11 We have demonstrated the detection of 300 fmol of a 62mer in the presence of a 28-fold molar excess of a 17mer when a Parafilm surface was used.8 We found that the use of a Teflon surface could further improve detection. Figure 2 demonstrates the effectiveness of the Teflon method. The spectrum shown in Figure 2 was obtained by loading 120 fmol of the 62mer along with a 28-fold excess of a 17mer (3.4 pmol). In general, the absolute signal level of the 62mer decreased in the presence of an excess of the 26mer, but we were still able to observe good 62mer signals under the conditions used. One other issue related to the DNA analysis is the significant decrease in the peak intensities of the larger DNA components when a DNA mixture is analyzed. It appeared that Teflon worked as well as Parafilm in terms of improving the detection of larger DNA in a DNA mixture. In the same way, this improvement is attributed to the effectiveness of a hydrophobic surface in improving the detection of larger DNA.8 In the previous work, we found that the use of Parafilm could improve the salt tolerance. The same improvement was also found for Teflon. For example, we were able to detect an 85mer in the presence of over 200 mM NaCl or 2 mM MgCl2 without a significant loss of the DNA signal and without an increase of laser power. It should be noted that, although the use of Teflon increases salt tolerance, it could not improve mass resolution. High-resolution spectra obtained using delayed extraction revealed that metals could still attach to DNA in the presence of metal salts to a large degree. As pointed out in our previous paper,8 there may be a “saturation” effect when a hydrophobic surface (11) Hurst, G. B.; Doktycz, M. J.; Vass, A. A.; Buchanan, M. V. Rapid Commun. Mass Spectrom. 1996, 10, 377.
Analytical Chemistry, Vol. 71, No. 2, January 15, 1999
519
Figure 2. MALDI-TOF mass spectra of a 62mer (120 fmol loaded) obtained in the presence of a 28-fold excess of a 17mer (3.4 pmol). Other unlabeled peaks correspond to larger cluster ions of the 17mer.
was used. In other words, when salts are present in the DNA solution, only a portion of the salts could be doped into the matrix crystals. Once the salt concentrations in the matrix crystals reach a “saturation” level, no additional salts would be doped into the crystals, even if extra amounts of salts are present. At the saturation level, the salts would still reduce the mass resolution but would not significantly adversely affect the signal level. All the results reported above were obtained with the homemade instrument operated under continuous extraction conditions, and thereby the mass resolution was poorer. It has been shown that the use of delayed extraction could significantly improve the mass resolution, especially when the DNA size was less than 70 bases.12-15 One of the potential applications of MALDI-TOF was for a high-throughput detection of mutations or polymorphisms of DNA.16,17 This application requires a good mass resolution. A mass resolution of better than 500 has been achieved with the metal sample surface. An interesting question is whether the use of Teflon can achieve a similar mass resolution with delayed extraction. We examined this issue using DNA ranging from 26- to 85mers. It was found that an excellent mass resolution could be obtained using both Teflon and stainless steel for DNA of 2662mers under the delayed extraction conditions. For example, (12) Brown, R. S.; Lennon, J. J. Anal. Chem. 1995, 67, 1998. (13) Whittal, R. M.; Li, L. Anal. Chem. 1995, 67, 1950. (14) Roskey, M. Y.; Juhasz, P.; Smirnov, I. P.; Takach, E. J.; Martin, S. A.; Haff, L. A. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 4724. (15) Colby, S. M.; King, T. B.; Reilly, J. P. Rapid Commun. Mass Spectrom. 1994, 8, 865. (16) Braun, A.; Little, D. P.; Koster, H. Clin. Chem. 1997, 43, 1151. (17) Monforte, J. A.; Becker, C. H. Nature Med. 1997, 3, 360.
520 Analytical Chemistry, Vol. 71, No. 2, January 15, 1999
Figure 3. High-resolution MALDI-TOF spectra of a 41mer obtained using delayed extraction on the sample support of (a) stainless steel and (b) Teflon, respectively. The peaks labeled by ( correspond to the doubly charged 41mer.
spectra a and b of Figure 3 display the mass spectra of a DNA 41mer using the stainless steel and Teflon surfaces, respectively. Both spectra were generated using delayed extraction. The total amounts of DNA loaded on each sample were approximately 3.5 pmol. It should be noted that the DNA concentrations used in this experiment were somewhat higher than the detection limits since it was much easier to produce the best resolution spectra if the DNA concentrations were higher. It is seen from Figure 3 that a mass resolution of better than 700 was obtained using both materials. The unlabeled low-mass peaks in Figure 3 correspond to smaller DNA molecules present in the 41mer sample, and they were produced during the synthesis of the 41mer. A mass resolution of better than 500 was also obtained for a DNA 62mer using both Teflon and stainless steel, but no significant improvement in mass resolution was seen when an 85mer was analyzed. We first suspected that the 85mer sample might contain more metal salts, leading to poorer resolution. However, analysis of a mixture of 41- and 85mers showed an excellent resolution for the 41mer but not for the 85mer. This indicated that the amount of salts contained in the 85mer sample was not significantly different from that present in the other samples used. Thus, an important question is raised concerning what factors led to the poorer resolution problem when the 85mer was analyzed. A study examining this issue is currently under way in our laboratory. Protein Analysis. Several groups have used a number of different membrane materials for the purpose of replacing metals for the protein analysis.18-21 The aim of these works was to explore the possibility of integrating MALDI-TOF to SDS-PAGE separa-
Figure 4. MALDI-TOF mass spectra of myoglobin obtained in the presence of 1 M sodium acetate using the Teflon sample support (a) without washing and (b) after washing using the 70% methanol aqueous solution.
tion. The materials studied include polyvinylidene, nitrocellulose, nylon, and polyethylenes. Teflon has been used as a transfer membrane material for the protein analysis.9 An interesting question is if the MALDI-TOF analysis of proteins can be directly performed on a Teflon sample surface. If so, Teflon could thus be potentially a good transfer membrane for integrating MALDITOF to SDS-PAGE. Hence, a study was performed to evaluate the effectiveness of the Teflon method in protein analysis with myoglobin as the protein sample. A sample preparation procedure different from the one we have been using for DNA was utilized to prepare the (18) Vestling, M. M.; Fenselau, C. Anal. Chem. 1994, 66, 471. (19) Strupat, K.; Karas, M.; Hillenkamp, F.; Eckerskorn, C.; Lottspeich, F. Anal. Chem. 1994, 66, 464. (20) Loo, R. O.; Stevenson, T. I.; Mitchell, C.; Loo, J. A.; Andrews, P. C. Anal. Chem. 1996, 68, 1910. (21) Blackledge, A.; Alexander, A. J. Anal. Chem. 1995, 67, 843. (22) Worrall, T. A.; Cotter, R. J.; Woods, A. S. Anal. Chem. 1998, 70, 750. (23) Brockman, A. H.; Dodd, B. S.; Orlando, R. Anal. Chem. 1997, 69, 4716.
protein sample since we are interested in using Teflon as a transfer membrane for SDS-PAGE separation. The protein solution (instead of the matrix solution) was spotted on Teflon first, followed by addition of the matrix solution to the Teflon. In general, the performance of the Teflon method was comparable to that observed with stainless steel in terms of the detection limits, reproducibility, and mass resolution when the matrix CHCA was used. In addition, no increase in laser power was needed for MALDI. For example, we have been able to detect less than 25 fmol of myoglobin with the Teflon method. This suggests that Teflon has a potential use as a transfer membrane for SDS-PAGE separation, which could then be expediently analyzed by the direct MALDI-TOF detection of the electroblotted proteins from Teflon. One of the problems encountered in preparation of samples for the MALDI-TOF analysis of proteins is the presence of salts in the sample. Recently, several groups have used the hydrophobic surfaces to remove salts on the sample probe.22,23 In that approach, the contaminated peptide and protein sample is loaded on a hydrophobic surface, followed by washing the contaminants and adding the matrix. The strong hydrophobic interactions allow effective removal of the contaminants from the sample probe without the need for laborious further purification and associated sample loss. We found that the use of a sample probe having a Teflon surface can also achieve a similar effect. Figure 4 demonstrates this effect. A total of 1 pmol of myoglobin (1 µL) in the presence of 1 M sodium acetate was first loaded to the Teflon. No signals were observed without washing using either a Teflon or stainless steel sample probe. However, after the dried sample was washed two or three times using 70% methanol aqueous solution, followed by the addition of the CHCA solution, strong myoglobin signals were observed from the Teflon surface, but not from the stainless steel surface using this simple washing procedure. It appears that, as in the cases of other hydrophobic materials, salts are less strongly adsorbed on the Teflon surface. They can thus be washed away from the sample probe, thereby leaving purer proteins at the probe surface, which leads to better signals from MALDI. ACKNOWLEDGMENT The authors acknowledge helpful discussion with Dr. S. Wang during the course of this work and the help of Dr. David Anderson in the preparation of this manuscript. This research is supported by NIH Grants Nos. HG01437 and HG01815.
Received for review July 27, 1998. Accepted November 9, 1998. AC980824N
Analytical Chemistry, Vol. 71, No. 2, January 15, 1999
521
