Portal de Periódicos da CAPES

Solvent-Free Mass Spectrometry for Hydrophobic Peptide Sequence Analysis and Protein Conformation Studies

2005; Future Science Ltd; Volume: 39; Issue: 6 Linguagem: Inglês

10.2144/05396te01

1940-9818

Sarah Trimpin, Max L. Deinzer,

Metabolomics and Mass Spectrometry Studies

BioTechniquesVol. 39, No. 6 Techniques EssayOpen AccessSolvent-Free Mass Spectrometry for Hydrophobic Peptide Sequence Analysis and Protein Conformation StudiesSarah Trimpin & Max DeinzerSarah TrimpinOregon Health & Science Universit, PortlandOregon State University, Corvallis, OR, USA & Max DeinzerOregon State University, Corvallis, OR, USAPublished Online:30 May 2018https://doi.org/10.2144/05396TE01AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinkedInRedditEmail The breakthrough in soft ionization methods for biopolymers, electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI) (1,2), revolutionized mass spectrometry (MS). As a result of these effective bioanalytical technologies, the entire field of proteomics changed, and the use of mass spectrometers for protein identification by accurate mass measurements or de novo sequencing of the corresponding proteolytic peptides is now universally accepted. Bioanalytical techniques ideally aim for a maximum degree of unbiased analytical information hence, simplification of these techniques can become both more efficient and often more reliable. Such is the case with a new approach in the use of MALDI-MS.Mass SpectrometryThe primary sequences of peptides can be elucidated by fragment ion analysis, tandem mass spectrometry (MS/MS) analysis, and incorporating intelligent data interpretation programs. One of the major challenges in bioanalytical MS analysis has to do with the solubility of the protein or its peptides. If the peptide or protein binds to the wall of the tube due to low solubility, hydrophobicity, and/or precipitation, for example, the biooligomers or biopolymer will not be transferred with the rest of the sample to the MALDI plate. Membrane proteins and their peptides are often not characterized mainly because of their low solubility. Generally, ESI works better than MALDI for these applications.PrinciplesMALDI is a MS method that depends on a laser primarily to desorb an analyte homogenously distributed in a matrix bed. The time-of-flight (TOF) analyzer operates on the principle that when a temporally and spatially well-defined group of ions of differing mass-to-charge (m/z) ratios are accelerated at the same applied electric field and allowed to drift in a region of constant electric field, they will traverse this region in a time that depends on their masses or m/z ratios.Normally, the analyte is dispersed in a large molar excess of the matrix material (2), which is commonly a low molecular weight aromatic carboxylic acid with a chromophore that absorbs the incoming laser radiation at wavelength λ 337 nm. By isolating the analyte in a protective chemical environment, the matrix bed, the probability of ion formation of the analyte without fragmentation is enhanced.A homogenous sample mixture is placed on the MALDI plate, which is then introduced into the MALDI source. Short laser pulses, which are focused on the sample spot, cause the matrix and therefore the sample to volatilize and ionize. The ion formation process may occur at any time during the MALDI process; however, the exact mechanism is not entirely understood. Several mechanisms have been discussed including gas-phase photoionization, excited-state proton transfer, ion molecule reactions, and desorption of preformed analyte ions. Regardless of the mechanism, however, only a small portion of the analytes undergo ion formation. The short laser pulses that desorb the matrix and sample are synchronized to the detector and thus serve as the start point for the TOF measurement of the ions. The arrival of the ions at the end of the flight tube is detected and recorded by a high-speed recording device.A very important aspect of MALDI-MS analysis is the proper preparation of the sample that requires some type of analyte-matrix preorganization, which in the past has been associated with co-crystallization. Many variables influence the analyte-matrix integrity and the success of MALDI analysis including the concentration of the matrix and analyte, choice of matrix, sample history, such as exposure to strong ionic detergents or formic acid, hydrophobicity or hydrophilicity of the sample, and contaminants, just to mention a few.Solvent-Based Sample PreparationIn solvent-based MALDI-analysis (2), an appropriate solvent is used to dissolve the matrix that is subsequently mixed either in the sample tube or on the MALDI plate with the sample solution. Hence, homogenization and transfer of the MALDI sample is based on a solvent or solvent mixture. A variety of different techniques are employed: dried-droplet (2) and thin-layer (3) are the most common ones. The solution on the MALDI plate is dried, leaving behind the sample co-crystallized within the matrix. The homogenization of the sample with the matrix at this stage is complete. Examples of dissolved MALDI samples may include solutions or suspensions. Crucial to the success of the analysis is that an analyte- and matrix-compatible solvent or solvent mixture is used. With the dried-droplet deposition, the analyte and matrix are mixed first in the dissolved state and then deposited on the MALDI plate. By evaporating the solvent, crystallization takes place according to the mixture components and their physicochemical characteristics. In the thin-layer deposition, the matrix is dissolved in acetone, and the solution is delivered onto the MALDI plate, whereupon the solvent is evaporated. The aqueous analyte solution is then applied onto the crystallized matrix layer on the MALDI plate. Miscibility of the analyte and matrix molecules is usually sufficient only at the boundary of the two layers. The thin-layer sample deposition protocol is therefore an advanced dried-droplet deposition protocol that minimizes inhomogeneities between the analyte and matrix in the solid state.Solvent-Free Sample PreparationSolvent-free methods generally consist of two steps: (i) dry homogenization, carried out mostly by mechanical mixing of sample and matrix and (ii) transfer of the resulting powder to the MALDI plate. In solvent-free MALDI-MS, at no point is the solvent used to mix analyte and matrix or during their transfer onto the MALDI plate; in the case of a dissolved sample, the sample is dried to completeness prior to mixing it with the matrix. Hence, homogenization and transfer of the sample in these approaches is not based on the solvent; the aqueous medium the proteins were originally dissolved in is unimportant. MALDI analysis is therefore simplified because fewer combinations and issues of compatibility or solubility must be considered. Two general methods were previously used for homogenization: (i) grinding by mortar and pestle (4,5) and (ii) vigorous shaking by a ball mill (5,6). Ball-milling is carried out in a vessel, into which is added the matrix and sample to be ground, along with suitably sized balls preferably made of the same material as the vessel. Grinding takes place as a result of the interaction between the balls, particles, and grinding vessel wall. The technique works equally well on soft, medium-hard, and even extremely hard, brittle, and fibrous materials and still yields much finer particles than do other methods. Generally speaking, smaller balls and longer grinding times result in smaller particles. The loose powder sample is finally dry-transferred and, with a spatula, gently pressed to the MALDI plate (5,6). Another approach is to prepare a pressed pellet and attach it to the MALDI plate with double-sided adhesive tape (4,5). These are the two important examples of dry MALDI sample transfer. The important similarities and differences in the analytical results between these approaches are that the homogenization method does not have a significant effect, but the transfer method does (5). The loose powder approach, in contrast to the pellet variant, employs conventional MALDI conditions including a large excess of matrix (5) that permits true matrix assistance, as evident by its capacity for high molecular weight analysis, molecular weight 100 kDa (5), which has not been observed in other solvent-free approaches.Introduction of Solvent-Free Mini-Ball Mill MALDI-MSThe delay in the development of this method for biological samples is due mainly to the relatively high sample amounts required for solvent-free sample preparation. An efficient, low load sample preparation procedure was required to have any meaningful impact for the analysis of peptides and proteins that are almost always in very short supply. Substantial modifications to the loose powder method resulted in the mini-ball mill (MBM) approach (7), which is most efficient, requiring the least amount of sample and preparation time. Disposable containers (0.2-mL PCR tubes) are used, thereby avoiding carry-over. Mixing of ana-lyte and matrix powder is performed with a mini-bead beater, a device that is commonly used in biological laboratories for cell homogenization or cellular cracking. If drying of the sample is involved, it is generally freeze-dried in the plastic tube, after which three metal beads (1.2 mm diameter) and matrix powder are added. The capped tube is placed in the original container that comes with the bead beater and placed in the two-dimensional (2-D) MBM shaker. The two axes of motion provide for efficient and thorough grinding, and the best results are obtained by applying a minimum amount of the fine powderto the MALDI plate with a spatula. The powder is then gently pressed to the plate spot with the flat edge of spatula producing a thin film and maximum area coverage. A gas stream from a container of dif luoroethane is used for dusting off the excess powder and other dust particles, after which the MALDI plate is introduced into the chamber for analysis. Picomole sample amounts can be handled conveniently, and only 30 s of MBM homogenization is needed. Another advantage with the method is that matrix purity is not as critical as it is with methods that use solvents. The MBM approach easily allows for peptide mapping and other biochemical manipulations prior to sample preparation. For example, with a TOF-TOF instrument, the detection of the tryptic peptides of cytochrome c in the low-picomole range was performed by MALDI-MS; MS/MS measurements were performed for a selected ion in the medium- to high-picomole range with a m/z ratio of 964.6 (7). Accessibility of high molecular weight proteins via the MBM approach has been demonstrated with the successful analysis of bovine serum albumin, molecular weight 66 kDa (7). But the most important advantage of any of these dry MALDI methods is their capacity for analyzing insoluble biomolecules and that the application of the MBM approach should significantly improve the quality of data in any analysis of proteins and peptides where the use of solvents creates difficulties, as for hydrophobic and solubility limited peptides and membrane proteins.β-Amyloid PeptidesThe role of cerebral amyloid β (Aβ) accumulation in common forms of Alzheimer disease has been widely studied and discussed. Despite the extensive amount of research that has been carried out, amyloid-precursor protein (APP) processing and Aβ accumulation is poorly understood. Mature APP is metabolized by two competing pathways, the α-secretase pathway generating sAPPα and C83, and the β-secretase pathway generating sAPPβ. Both C-terminal fragments (C83 and C99) are substrates for γ-secretase generating the APP intracellular domain, and respectively, the secreted peptides p3 and Aβ. The latter is also known as β-amyloid peptide 1–42, which has the lowest solubility of any of the amyloid peptides; the low solubility of peptide 1–42 is believed to be a key property in the progression of the disease. Aβ aggregates to oligomers, which apparently are among the most potent neurotoxins. Aβ oligomers subsequently form the final stage, senile plaques, which are relatively low in neurotoxicity.With the solvent-free MBM procedure (Reference 8 and unpublished data), the β-amyloid 1–11 could be analyzed with a detection sensitivity in the high fmol range for both MS and MS/MS. Mixtures of soluble β-amyloid 1–11, the non-β-amyloid component (NAC) peptide of Alzheimer Disease, and insoluble peptides 1–42 and 33–42 were prepared in molar ratios 3:1:1:1, 1:1:1:1, and 1:3:3:3. Analysis of the peptide mixture 1:1:1:1 demonstrates the relative performance of the two MALDI methods; the molecular ions of the mixture (Figure 1) and the fragment ions of the individual peptides (Figure 2) show that the soluble peptide 1–11 and NAC were detectable by both solvent-based and MBM-MALDI-MS, but the insoluble component, peptide 1–42, could only be detected by the latter, and it was detected in all three mixtures. The hydrophobic low molecular weight peptide 33–42 was sometimes detectable by solvent-based MALDI (e.g., in the 1:1:1:1 mixture; Figure 1, Inset 1), that also allowed fragment ion analysis (Figure 2, II); the MBM analysis always detected peptide 33–42. The results illustrate the greater reliability and potential utility of MBM-MALDI-MS, as solubility is not the limiting factor it is with solvent-based methods.Figure 1. Mass spectra of a molar 1:1:1:1 model peptide mixture with inherently different solubilities of the components: two insoluble (in aqueous solution), hydrophobic peptides β-amyloid 33–42 (MWtheor915.2 Da) and 1–42 (MWtheor4515.1 Da), as well as two soluble (in aqueous solution) hydrophilic β-amyloid 1–11 (MWtheor1325.3 Da) and non-β-amyloid component (NAC) of Alzheimer Disease (MWtheor3260.6 Da) obtained by (A) solvent-based MALDI-MS and (B) solvent-free MBM-MALDI-MS analysis of all four peptides showing excellent quality spectra and improved quantitation.Solvent-based MALDI (A) shows only the two hydrophilic peptides and the low molecular weight (MW) hydrophobic peptide, but fails to detect the higher molecular weight hydrophobic peptide (Inset 4). The hydrophobic peptide (Inset 1) has a much greater tendency to attach metal cations than the hydrophilic peptides, which prefer proton attachment (Inset 2 and 3) independent of the method used (A or B). MBM-MALDI-MS, mini-ball mill matrix-assisted laser desorption/ionization mass spectrometry; m/z, mass-to-charge ratio.Figure 2. Tandem mass spectrometry (MS/MS) spectra of a molar 1:1:1:1 model peptide mixture with inherently different solubilities of the components: two insoluble (in aqueous solution), hydrophobic peptides β-amyloid 33–42 (MWtheor915.2 Da) and 1–42 (MWtheor4515.1 Da), as well as two soluble (in aqueous solution) hydrophilic β-amyloid 1–11 (MWtheor1325.3 Da) and non-β-amyloid component (NAC) of Alzheimer Disease (MWtheor3260.6 Da).(A) Solvent-based MALDI-MS; (B) solvent-free MBM MALDI-MS. (Inset 1) peptide 33–42; (Inset 2) peptide 1–11; (Inset 3) peptide NAC; (Inset 4) peptide (1–42). MBM-MALDI-MS/MS (B) shows spectra for all four peptides with good fragment ion intensities. Solvent-based MALDI-MS/MS (A) yields spectra for the two hydrophilic peptides and the low molecular weight (MW) hydrophobic peptide, but fails to detect the higher molecular weight hydrophobic peptide (Inset 4). The quality of the MS/MS spectra of peptide 33–42 (Inset 1) and peptide 1–11 (Inset 2) are equally good, the MS/MS spectra of hydrophilic peptide NAC is much better with the solvent-based method than with the solvent-free method (Inset 4); we rationalize the differences on the basis of proton attachment to the peptide, which is generally easier obtained with the solvent-based analysis. MBM-MALDI-MS, mini-ball mill matrix-assisted laser desorption/ionization mass spectrometry; m/z, mass-to-charge ratio.All four peptides in mixtures 3:1:1:1 and 1:3:3:3, irrespective of solubility, were correctly recognized by application of the Mascot database search routine to the results from automated MALDI-MS and MS/MS analysis when analyzed by MBM, but data from the solvent-based method yielded hits only for the two soluble β-amyloid peptides, peptide 1–11, and NAC. The hydrophobic peptide 33–42 (Figure 1, Inset 1), has a much greater tendency to attach sodium and potassium ions than the hydrophilic peptides (e.g., peptide 1–11) (Figure 1, Inset 2), as detected by both methods. However, this is a critical issue, as MALDI in general is highly sensitive to the presence of salts, which has led to desalting procedures in solvent-based methods that are also applied in the MBM approach. Desalting is especially important for enhanced MS/MS fragmentation, since protonated peptides dissociate better than sodiated species, as observed for hydrophobic peptides in general (Reference 8 and unpublished data). This procedure does not interfere with the important analyte-matrix arrangement of the MBM sample, if the solvent manipulation step is handled carefully (unpublished data). Sequence coverage of the tryptic peptide 1–42 was improved to 100% when the MBM sample was desalted and analyzed by the MALDI analysis, as compared to 67% using traditional solvent-based MALDI-MS.BacteriorhodopsinFurther evaluation of the MBM-MALDI-MS method was conducted with bacteriorhodopsin, an α-helical integral membrane protein (Reference 8 and unpublished data) that presents special problems. The study was focused on the degree to which the molecular weight could be determined and on the ability of the technique to map out peptides from this solubility restricted protein. A desalting procedure for the MBM sample directly on the MALDI plate was found necessary to minimize complexation of the protein with sodium and other cations (unpublished data). The superior results in comparison to the solvent-based procedure are apparent (Figure 3) (8). At least equally good results were obtained for the tryptic peptide mapping analysis comparing the MBM analysis with the traditional solvent-based method. Overall, the solvent-free MALDI approach can be viewed as having become a mature method for protein/peptide analyses.Figure 3. Mass spectra of bacteriorhodopsin by (A) solvent-based MALDI-MS and (B) solvent-free MALDI-MS after the MBM sample was washed.The signal intensities for the singly and doubly protonated membrane protein are similar in the spectra obtained by the two MALDI methods. The signal width (B) in the MBM-MALDI spectrum of the protein is much narrower than (A) in the solvent-based spectrum, which results from metal ion and possibly matrix molecule(s) adductions to the protein. A small shoulder is observable in (B) at mass-to-charge ratio (m/z) of about 28500, z = 1, and m/z of about 14250, z = 2. We assume that this small signal is a chemically modified variant of bacteriorhodopsin. MALDI-MS, matrix-assisted laser desorption/ionization mass spectrometry; MBM, mini-ball mill.Theoretical AspectsInvestigations of theoretical aspects of the MALDI process derived from solvent-free sample preparations have been pursued extensively (Reference 9 and unpublished data). It has been concluded that the incorporation of analyte in matrix crystals is not helpful for MALDI analysis but obstructive, since it is exactly thecrystallinity of the matrix that makes the underlying process energetically more difficult, thereby requiring increased laser power, which can be a disadvantage (9). In the analysis of hydro-phobic peptides, it has been demonstrated that solvent-based analysis fails not only due to possible solubility limitations but also to insufficient ion production (unpublished data). Although as described above, the presence of salts is a disadvantage even in dry MALDI, as ion formation was promoted in dry MALDI analysis for the solvent-based MALDI-inactive hydrophobic tryptic peptide, T17-42, and even peptide, P35-42, of 1–42 by competitive metal adduction, which is a great success. Overall, the MALDI process is more effective, or softer, because of decreased crystallinity that results in improved contact between the analyte and the matrix (9) as well as with sufficient analyte-appropriate ion formation (e.g., protonation of peptides and proteins) (unpublished data). It is well known that ESI-MS is the superior method for hydrophobic peptides and MALDI-MS for aromatic and hydrophilic peptides. These methods, in fact, have been proposed to be complementary, but with the solvent-free approach, a greater range of peptides and proteins should be analyzable by just MALDI analysis.ConclusionThe uniqueness and widespread accessibility to MALDI mass spectrometers makes the solvent-free approach particularly important as it expands the capabilities for mass spectrometric analysis to peptides and proteins that previously may have been better suited for ESI-MS. Most likely, this solvent-free MALDI approach will be important for accurate mass measurements of solubility limited hydrophobic peptides and membrane proteins. This capability on its own has tremendous significance. Other aspects improved by the solvent-free MALDI method, include quantitative information on mixtures, hydrogen/deuterium (H/D) exchange studies, analysis of membrane proteins (Reference 8 and unpublished data), and especially a broadened repertoire in proteomics (10). To accommodate this latter area, however, it will be essential to develop a more effective and efficient solvent-free homogenization and sample transfer approach to the mass spectrometer, mainly allowing for lower sample amounts, typically determined by gel bands or 2-D spots. Solvent-free sample preparation directly on the MALDI plate (11) seems to lead in the right direction.AcknowledgmentsThe MS Core of the Environmental Health Science Center (EHSC) and grants from the National Institutes of Health/National Institute of Environmental Health Sciences (NIH/NIEHS) (ES10338; ES00210; ES00040) were used in support of this work. We thank Elisabeth Barofsky, Dr. Mike Hare, and Brian Arbogast for technical assistance.References1. Fenn, J.B., M. Mann, C.K. Meng, S.F. Wong, and C.M. Whitehouse. 1989. Electrospray ionization for mass spectrometry of large biomolecules. Science 246:64–71.Crossref, Medline, CAS, Google Scholar2. Karas, M. and F. Hillenkamp. 1988. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60:2299–2301.Crossref, Medline, CAS, Google Scholar3. Vorm, O., P. Roepstorff, and M. Mann. 1994. Improved resolution and very high sensitivity in MALDI TOF of matrix surfaces made by fast evaporation. Anal. Chem. 66:3281–3287.Crossref, CAS, Google Scholar4. Skelton, R., F. Dubois, and R. Zenobi. 2000. A MALDI sample preparation method suitable for insoluble polymers. Anal. Chem. 72:1707–1710.Crossref, Medline, CAS, Google Scholar5. Trimpin, S., A. Rouhanipour, R. Az, H.J. Räder, and K. Müllen. 2001. New aspects in matrix assisted laser desorption/ionization time-of-flight mass spectrometry: a universal solvent-free sample preparation. Rapid Commun. Mass Spectrom. 15:1364–1373.Crossref, Medline, CAS, Google Scholar6. Przybilla, L.M., J.D. Brand, K. Yoshimura, H.J. Räder, and K. Müllen. 2000. Characterization of large synthetic polycyclic aromatic hydrocarbons by MALDI-and LD-TOF mass spectrometry. Anal. Chem. 72:4591–4597.Crossref, Medline, CAS, Google Scholar7. Trimpin, S. and M.L. Deinzer. 2005. Solvent-free MALDI-MS for the analysis of biological samples via a mini-ball mill approach. J. Am. Mass Spectrom. 16:542–547.Crossref, Medline, CAS, Google Scholar8. Trimpin, S., P.S. Spencer, and M.L. Deinzer. 2005. Solvent-free MALDI-MS for the analysis of hydrophobic peptides and membrane proteins via the mini-ball mill approach. Proceedings of the 53rd ASMS Conference on Mass Spectrometry and Allied Topics, San Antonio, TX, USA.Google Scholar9. Trimpin, S., H.J. Räder, and K. Müllen. In press. Investigations of theoretical principles for MALDI-MS derived from solvent-free sample preparation part I preorganisation. Int. J. Mass Spectrom. Google Scholar10. Trimpin, S., P.S. Spencer, and M.L. Deinzer. 2005. Evaluation of solvent-free MALDI-MS for the analysis of proteins via the mini-ball mill approach. Mol. Cell. Proteomics 4(suppl 1):311.Google Scholar11. Hanton, S.D. and D.M. Parees. 2005. Extending the solvent-free MALDI sample preparation method. J. Am. Mass Spectrom. 16:90–93.Crossref, Medline, CAS, Google ScholarFiguresReferencesRelatedDetailsCited ByRecent developments in pre-treatment and analytical techniques for synthetic polymers by MALDI-TOF mass spectrometry1 January 2020 | Analytical Methods, Vol. 12, No. 48Total Solvent-Free Analysis, Charge Remote Fragmentation, and Structures of Highly Charged Laserspray Ions Using IMS-MS20 January 2011Amyloid-β Peptide Is a Substrate of the Human 20S Proteasome25 August 2010 | ACS Chemical Neuroscience, Vol. 1, No. 10Solvent-Free MALDI Sample Preparation6 November 2009Analysis of insoluble proteinsSarah Trimpin & Bill Brizzard25 April 2018 | BioTechniques, Vol. 46, No. 6Analysis of insoluble proteinsSarah Trimpin & Bill Brizzard25 April 2018 | BioTechniques, Vol. 46, No. 5Curent Situation Regarding Utilization of the International Classification of Functioning, Disability and Health by Physical TherapistsRigakuryoho Kagaku, Vol. 23, No. 2Multisample preparation methods for the solvent-free MALDI-MS analysis of synthetic polymers1 March 2007 | Journal of the American Society for Mass Spectrometry, Vol. 18, No. 3Current literature in mass spectrometry1 January 2006 | Journal of Mass Spectrometry, Vol. 41, No. 9 Vol. 39, No. 6 Follow us on social media for the latest updates Metrics History Published online 30 May 2018 Published in print December 2005 Information© 2005 Author(s)AcknowledgmentsThe MS Core of the Environmental Health Science Center (EHSC) and grants from the National Institutes of Health/National Institute of Environmental Health Sciences (NIH/NIEHS) (ES10338; ES00210; ES00040) were used in support of this work. We thank Elisabeth Barofsky, Dr. Mike Hare, and Brian Arbogast for technical assistance.PDF download