1. Introduction
The development of electrospray (ESI) and matrix-assisted laser desorption/ ionization (MALDI) in the early 1990s built upon the advances made by fast atom bombardment (FAB). FAB allowed the analysis of biomolecules previously intractable to mass spectrometric analysis, without derivatization. The subsequent development of MALDI and ESI provided increased sensitivity while significantly increasing the molecular mass of proteins that could be accurately measured. Further refinement of both ionization techniques and the associated mass analyzers has provided modern, state-of-the-art instrumentation that can detect low levels of endogenous biological samples with high accuracy. In particular, the coupling of MALDI and ESI with mass spectrometers capable of MS/MS (see Article 10, Hybrid MS, Volume 5) has resulted in significant advances in protein identification and characterization. However, one of the limiting factors in this inherently sensitive technology is the preparation of samples for the mass spectrometer. This crucial step often defines the quality of results obtained. For example, one common cause of problems is the incomplete removal of buffers used in the biology laboratory. The presence of detergents, salts, and other common buffer components can adversely affect the performance of ESI and MALDI mass spectrometers. The presence of alkali metals in a sample can cause what should be a single component to appear as multiple adduct peaks containing varying numbers of metal ions.
The following section details the most commonly used sample preparation techniques in use in the laboratory today.
2. Sample preparation for MALDI
Matrix-assisted laser desorption/ionization is a soft ionization technique. In this technique, the analyte molecules are embedded in a matrix consisting of small organic molecules. The sample is irradiated with a laser pulse, which leads to rapid heating of the matrix molecules and results in a transition into the gas phase. Analyte molecules included in the matrix are also brought into the gas phase and are ionized, often through adduct formation. In its ionized form, the analyte can be separated according to its m/z ratio, using a variety of different mass analyzers.
In modern configurations, a time-of-flight (TOF) mass analyzer (see Article 7, Time-of-flight mass spectrometry, Volume 5) is most commonly used.
MALDI sample preparation of peptide and protein samples can be extremely straightforward, and as a result, MALDI has rapidly established itself as one of the standard techniques for rapid protein identification. In the years after the introduction of MALDI, the main focus was on the mass determination of intact proteins. The emphasis shifted rapidly to peptide analysis, with the invention of a new technique for protein identification termed peptide mass fingerprinting (PMF) (see Article 3, Tandem mass spectrometry database searching, Volume 5). In this approach, proteins are enzymatically digested, typically with trypsin, and the resultant peptides mass-measured. The monoisotopic masses obtained from the MALDI PMF experiment are then compared to a protein or nucleotide sequence database. The main sample preparation factors affecting the overall performance of MALDI experiments are choice of matrix, sample purity, solvents used, and the method of applying the matrix and sample. More specialized applications such as phospho-peptide analysis may require more rigorous sample cleanup and sample preparation routines, and these are dealt with later under electrospray sample preparation.
2.1. Sample cleanup
The best results in MALDI analysis are obtained from clean samples. Much can be done in the steps leading up to the MALDI analysis to reduce or even eliminate the need for sample cleanup. Common biological buffers, contaminants, and salts can be tolerated to the levels shown in Table 1. If the buffer/salt levels are too high, these can be reduced by a number of different methods. A widely used technique is binding peptides on reverse-phase material and then washing them with aqueous solutions, followed by elution in high organic. Alternatively, salts may be removed by dialysis. Washing of the sample by adding a droplet of water to a prepared MALDI sample spot and then drawing off the liquid and salt has also been shown to be a useful strategy. It should, however, always be borne in mind that sample cleaning may result in some loss or dilution of sample and if it can be avoided through careful choice of the sample preparation steps, then the best results will be obtained. A common source of sample contamination when a protein sample is being enzymatically digested is keratin. This originates from clothing and human skin, and it is often surprising to researchers when they first start working with MALDI just how sensitive the technique is and how readily keratin can be detected. This contamination can usually be avoided through the use of gloves and clean air enclosures.
2.2. Choice of matrix and solvent
There are numerous ways to prepare a given sample for analysis by MALDI, however, some broad guidelines can be made. The first choice that needs to be made is the choice of matrix. Over the years, a large number of matrices have been suggested, however, two have been used routinely for the analysis of peptides. The most commonly used matrix for peptide mass fingerprinting is a-cyano-4-hydroxycinnamic acid (CHCA). It is relatively easy to obtain fine matrix crystals with this matrix, which tends to give a more homogenous sample surface. A homogenous sample surface makes it easier to find a “hot spot” providing mass spectra containing the analyte of interest. It is also possible to get excellent sensitivity with this matrix. Several different concentration levels and solvents have been suggested for the preparation of CHCA, however, the most commonly used concentration is 10 mg mL-1. Despite this, experiments have shown that for low peptide concentrations, that is, subpicomol on target, matrix concentrations of 2mg mL-1 or lower work better. The solvent mix used depends on the evaporation rate desired; this can be very dependent on ambient humidity and temperature. A mixture of 1:1 acetonitrile:water containing 0.1% trifluoroacetic acid (TFA) generally works well, although the water can also be replaced by ethanol.
Table 1 Common buffers and contaminants and the maximum concentrations tolerated in MALDI and ESI MS
| Contaminant | MALDI max. conc. | ESI max. conc. |
| Phosphate buffer | 20 mM | <2mM |
| Tris buffer | 50 mM | 5 mM |
| Detergents | 0.1% | 0.1% |
| SDS | 0.01% | 0.01% |
| Alkali metal salts | 1 M | - |
| Glycerol | 2% | - |
| NH4 bicarbonate | 30 mM | 50mM |
| Guanidine | 1M | 10mM |
| Sodium azide | 1% | - |
For protein analysis, sinapinic acid can be a very effective matrix. This matrix can be used in a dried droplet experiment and in thin-film preparations (see spotting methods). For both of these methods, a sinapinic acid solution of 10 mg mL-1 in 6:4 acetonitrile:0.1% TFA is used. The thin-film method additionally calls for a sinapinic acid solution of 10 mg mL-1 in acetone or acetonitrile. Sinapinic acid forms very homogenous sample spots with an even analyte distribution.
Another versatile matrix is 2,5-dihydroxybenzoic acid (DHB). This can be used for protein and peptide analysis but is also the matrix of choice for other compounds, for example, oligosaccharides. DHB tends to form large crystals, where the analyte is not evenly distributed and this leads to localized hot spots usually at the tips and edges of the crystals. A wide range of different solvent mixtures and matrix concentrations has been suggested for DHB. A typical example would be 20 mg mL-1 in 7:3 0.1% TFA:acetonitrile. Several other matrices and various derivations of the above matrices have also been described in the literature but these are not widely used.
2.3. Spotting methods
Several different spotting methods have been developed for MALDI samples. The methods mostly differ in the way the matrix solution is combined with the analyte solution, and they can have a drastic effect on the quality of the MALDI mass spectrum acquired.
The dried droplet method is a reliable and straightforward method. Roughly equal volumes of matrix and sample solution are mixed. This can be done prior to spotting in the sample tube, followed by spotting of a 00.5-2 |L droplet onto the target plate. This droplet is then allowed to dry either under normal lab conditions or at slightly reduced pressure. Alternatively, the matrix and analyte solutions can be combined directly on the target plate by first depositing a small volume of matrix (0.5-2 |L) and then subsequently adding a droplet of analyte solution. Mixing can then either be left to diffusion or can be aided by aspiration of the matrix/analyte droplet on the target plate using a pipette. Contact of the pipette tip with the sample holder should be avoided, as this can cause crystallization of the matrix.
The sandwich and thin-film methods are slightly more involved in comparison to the dried droplet method. However, these methods can be particularly effective for protein analysis, on-target desalting, and high-sensitivity work. In these methods, a thin film of matrix is deposited on the target plate and allowed to dry. Depositing a small volume of matrix in high organic, for example, acetone, produces the thin film. This thin film consists of very fine crystals. The sample solution added, either combined with matrix or on its own, and the solvent mix in the droplet that is added must be in a high aqueous environment to avoid dissolving the thin film.
A further derivation of this technique is the sandwich method in which a drop of matrix solution is applied after this procedure, covering the analyte with a matrix layer.
3. Sample preparation for electrospray
Electrospray ionization is a soft ionization technique. The sample is introduced into the electrospray ion source as a continuous liquid stream. As such, ESI is often directly coupled to on-line sample cleanup and separation techniques, such as liquid chromatography. This forms the basis of most sample preparation techniques for ESI. LC-ESI MS is compatible with a wide range of solvent compositions and ionizes a diverse range of compounds. It has found particular use in the area of biological mass spectrometry due to the ability of molecular species to accept multiple charges during the electrospray process. As the mass spectrometer separates according to mass/charge (m/z), high-molecular-weight compounds, such as large protein complexes, now appear lower down the m/z range, within the mass range of conventional mass analyzers.
Further refinements of electrospray for biological applications resulted in the development of nanoelectrospray (Wilm and Mann, 1994), where flow rates of the liquid stream maybe as low as a few nanoliters per minute. One of the limiting factors for analysis of proteins and peptides by electrospray or nanoelectrospray mass spectrometry is sample preparation. The presence of many biological buffers (e.g., HEPES, Urea) interfere with the ionization process and as such they should be avoided, or removed, prior to analysis.
3.1. Purification and concentration by reverse phase
A common strategy in the analysis of protein and peptide samples by mass spectrometry is to improve the quality of the data acquired by using reversed phase (RP) chromatography. This approach allows peptides or proteins to be separated on the basis of their hydrophobicity, and is often used “on-line”, coupling liquid chromatography directly to an electrospray MS instrument. This approach has been used widely for the analysis of oligonucleotides, intact proteins, and proteolytic digests of proteins. The RP HPLC separation requires an ion-pairing reagent, and with traditional protein and peptide separation, 0.1-1% trifluoroacetic acid (TFA) is used. The sensitivity of ESI ionization is reduced by approximately a factor of 5 if TFA is present in the liquid stream, and as such alternative stationary phases have been produced by a variety of manufacturers that provide excellent separations using 0.1 -1% formic acid as the ion pair. This provides an increase in ESI sensitivity compared to the TFA-based separation. In addition, as ESI is a concentration-dependent ionization technique, the use of capillary and nanoscale chromatography has lead to significant advances in ESI sensitivity.
For peptide analysis, the HPLC is often coupled to a mass spectrometer capable of data-directed switching between the MS and MS/MS modes. Hundreds of MS/MS spectra can be acquired from complex peptide samples in a fully automated fashion, resulting in the structural characterization of many species in a single analysis. For example, protein identification can be achieved via data bank searching of the ESI-MS/MS, providing qualitative information on the proteins that are present and identification of significant numbers of proteins, including low copy number proteins, from a single LC-MS/MS experiment.
The use of microcolumns for ESI sample preparation of in-gel digested samples for nano-ESI was first described by Wilm and Mann (1996). A small amount of reverse-phase material was packed into the tip of a nanospray needle and used only once, a new column being made for each sample. This provides a crude separation using step elution to desalt and concentrate analytes prior to analysis, and this approach has been shown to provide impressive levels of sensitivity. While this approach was first described for ESI, it has also been widely adopted for MALDI (Gobom et al., 1999). Disposable pipette tips packed with reverse-phase material have become commercially available from numerous sources and are suited to robotic sample preparation.
3.2. Strong cation exchange (SCX)
If a complex peptide or protein mixture is to be investigated, then a fractionation step prior to analysis via ESI RPLC-MS is advantageous. The use of SCX material provides an orthogonal separation mechanism to RP, separating peptides or proteins, on the basis of their inherent or carried charge. It is therefore possible to prefractionate the peptides, prior to further separation by reverse-phase on a C18 column. This provides additional separation power, or peak capacity. A further derivation of this approach has seen the use of a biphasic column, where both SCX and RP material are sequentially packed into a nanoscale HPLC column. This allows two-dimensional LC separations to be performed on-line to the mass spectrometer (Link et al., 1999).
3.3. Hydrophilic interaction liquid chromatography (HILIC)
HILIC is a versatile, effective alternative to ion exchange and reverse-phase chro-matography for the increased retention and separation of polar peptides. The technique permits direct coupling via LC-MS. An additional advantage of HILIC is that it works best where RP works worst – with polar solutes that are poorly retained on RP material. HILIC has been used successfully for the analysis of phosphopeptides, carbohydrates, peptide digests, and polar lipids.
HILIC separates compounds by eluting mostly organic mobile phase over a neutral hydrophilic stationary phase. This results in solutes eluting in order of increasing hydrophilicity – the inverse of reverse phase.
3.4. Purification using graphite columns
The use of reverse-phase material to desalt and concentrate proteolytic digests prior to mass spectrometry is widespread, as discussed previously. However, this does not allow for detection of small or hydrophilic peptides, or peptides altered in hydrophilicity. The use of graphite powder, in a microcolumn format, allows hydrophilic species to be retained efficiently on the graphite (Larsen et al., 2002). These can then be eluted and analyzed by mass spectrometry.
3.5. Immobilized metal-affinity chromatography (IMAC)
Despite the ability of mass spectrometry to analyze peptide components in complex mixtures, the detection of endogenous levels of phosphopeptides in a peptide mixture can be problematical. This is partly due to the low ionization efficiency of the phosphorylated species, and also due to the low stoichiometry of phospho-rylation. Analysis of phosphorylation sites in complex mixtures can be facilitated through the use of affinity methods, which isolate and enrich phosphopeptides from samples prior to mass spectrometry. A popular method for isolating and enriching involves using trivalent metal ions, such as Fe3+, Ga3+, or Ni3+ that are bound to a chromatographic support. This technique is referred to as immobilized metal-affinity chromatography (IMAC). Primarily, IMAC experiments are performed in an off-line manner, with microcolumns (Stensballe et al., 2001), although on-line studies have also been reported.
