ABRF Workshop: Mass Spectrometry--What's Available for Facilities?

John T. Stults
Genentech, Inc.

This workshop focused not on the fantastic things mass spectrometry can do, but rather on how to select an instrument from the myriad of choices now available. Mass spectrometry has become an essential tool in many aspects of protein and peptide characterization. For example, most recently, the ABRF Committee on Peptide Synthesis and Mass Spectrometry recommended that mass spectrometry, amino acid analysis, and HPLC, were the minimum analytical techniques that should be applied to the characterization of synthetic peptides. Much of the interest in mass spectrometry is due to the revolutionary improvements, in just the last 3-5 years, in the capabilities of the technique. Two new techniques, electrospray ionization and matrix-assisted laser desorption/ionization, now permit mass determinations on proteins in excess of 100 kDa, and frequently require subpicomole amounts of material. The 1992 Symposium of the Protein Society witnessed a new milestone in mass spectrometry. For the first time, several manufacturers displayed mass spectrometers that were designed for ease-of-use by protein chemists, and one even sported a price-tag under $100,000. Given the variety of different instruments now available from almost a dozen manufacturers, how does one make a choice?

There is no single instrument that is ideally suited to all applications. It is helpful to answer several questions before any further:

What types of analyses do you want to perform?

What measurement(s) will you do 90% of the time?

Will the instrument be dedicated to a single use or will it perform different types of analyses?

Will multiple instruments be available for different types of analyses?

How many samples will be analyzed per week?

What level of mass spectrometry expertise is available? What amount of effort can the principle operator devote to the instrument?

What level of performance is required--sensitivity, mass range, mass accuracy, resolving power?

How much space is available?

How much do you have to spend?

Although there are many important, wide-ranging uses of mass spectrometry, there are some things that mass spectrometry cannot presently do on a routine basis. The following is a compendium of corrected misconceptions about mass spectrometry:

* Mass spectrometry is not quantitative without standards

* Ionization efficiency is not identical for all samples

* MS/MS can not directly sequence proteins--just peptides from a purified protein

* MS/MS is not routine for 1 pmol or less of peptide

* sequencing by MS/MS is not a "black box" technique and cannot be easily done by an untrained technician

* sequencing by MS/MS has not replaced Edman degradation nor made it obsolete

* MS/MS does not provide complete sequence of every peptide

* sequence interpretation programs for MS/MS data do not always yield single, unambiguous results

* LC-MS does not require capillary HPLC

* DNA can not be sequenced by MS/MS

* Masses of large DNA (>80 mer) can not be measured accurately

This listing is intended not to dissuade anyone from using the technique, but rather to broaden its use by precluding unrealistic expectations.

The four most common ionization techniques are compared and contrasted in Table 1. Any of these techniques works well with peptides. Clearly, though, electrospray ionization and laser desorption have extended capabilities for higher sensitivity and much greater mass range.

(24k) The most common mass analyzers (time-of-flight, quadrupole, and sector) each have different characteristics, and their use may be dictated by the ionization technique. The price of an instrument is most often dictated by the analyzer. There is a reasonably good correlation between price and performance in mass spectrometers. Furthermore, there is a good correlation between performance and complexity. Other types of analyzers are available, most notably ion traps, and Fourier-transform ion cyclotron resonance instruments. Each of these techniques shows tremendous potential for the analysis of peptides and proteins. However, neither is presently commercially available in a configuration for routine use in a core facility.

What degree of operator sophistication or expertise is needed for mass spectrometry? The more complex techniques, and consequently those with the higher prices in Table 1, frequently require a well-trained, dedicated mass spectrometrist. On the other hand, the less complex instruments require considerably less sophistication in the user. A good comparison is the level of expertise required to operate a protein sequencer. A number of people in a laboratory can be quickly trained to operate it. However, there needs to be one person who can ensure that the instrument is operating properly, and who has the training and experience to keep the instrument in proper order, to spot problems, to fix the minor problems as they occur, to train others, and to act as a resource. That person can be a B. S. -level person who receives training from the manufacturer and who has sufficient time (e.g., 50% effort) both to become familiar with the operation and maintenance of the instrument, and to serve as the primary operator as described above.

Most of the discussion up to this point has focused on mass measurements of single samples. Another level of use is on-line coupling of a mass spectrometer with a separation device, most commonly HPLC. There are two common techniques, flow-FAB and electrospray ionization. Flow FAB has an upper mass range of ~3500 Da (its chief limitation), a limit of detection between 0.1- 1 pmol, an optimal flow rate of 0.5-5 ul/min, and it requires addition of 0.5-2 % glycerol to the sample stream, either mixed with the solvents or added post-column. ESI has a wider mass range, in excess of 100 kDa, but otherwise it has similar detection limits and optimal flow rates as flow-FAB. Some (but not all) ESI interfaces require the addition of an organic make-up or sheath flow for optimal operation. Both techniques require volatile buffer systems of low ionic strength. For most applications, ESI appears to be the more versatile technique. In many applications where sample quantity is not limiting, the optimal flow is easily achieved by simple flow splitting of the column effluent; the unused portion can be collected for other uses.

Thus, many methods that employ higher flow rates can be accommodated as long as the sample concentration is approx. 1-10 pmol/ml. Capillary HPLC is recommended when the sample is limited and one wants to deliver the entire sample to the mass spectrometer in the highest possible concentration.

CE-MS is also possible with either flow-FAB or ESI, and the same limitations apply as with LC-MS. Due to the lower flow rates delivered by CE, a makeup solvent is normally required. CE-MS is used much less routinely than LC-MS, but this appears to reflect the less frequent use of CE than any particular problems with CE-MS.

Tandem mass spectrometry (MS/MS) may be used for peptide sequencing. There are two main categories of MS/MS experiments, those that employ low-energy (10-100 eV) collisionally activated dissociation (CAD) and those that use high energy (1-10 keV) CAD. Low-energy CAD is most often achieved with triple quadrupole instruments, it typically requires 5-20 pmol of peptide, and it gives relatively simple fragmentation--typically at the amide bonds. It does not permit distinction of Ile/Leu, which have identical masses. When used with ESI, complete sequences are not always obtained. If one has a pure peptide, adequate MS/MS can often be achieved with ESI on a single quadrupole (or 2-sector magnetic instrument), due to the ability to stimulate fragmentation in the gas expansion region of the interface. High-energy CAD is performed with 2- or 4-sector instruments, it requires 10-500 pmol of peptide, and it generates more complex fragmentation. This extra fragmentation frequently permits distinction of lle/Leu, and is very helpful in identifying unusual post-translational modifications.

Of all the variety of techniques available, which is the best? As noted earlier, it really depends on the particular needs of each laboratory. Two generalizations can be made. Matrix-assisted laser desorption/ionization on a time-of-flight instrument is the easiest to use of the techniques, it is the most tolerant of buffers, and it has the highest "success" rate, i.e., it produces signal on most peptides and proteins. Electrospray ionization gives results on the widest variety of compounds, it is easily interfaced to HPLC, it can provide fragmentation for sequencing, and it has the highest potential throughput (with an autosampler). One's choice of instrument should involve conversations with others who have used both techniques, and include the results of one's own samples, analyzed by different manufacturers and with different types of instruments.

Recommended references:

1. J.A. McCloskey, Ed., Methods Enzymol. 1991, 193, 1-960.

2. S.A. Carr, M.E. Hemling, M.F. Bean, and G.D. Roberts, Anal. Chem., 1991, 63, 2802-2824.

3. A.L. Burlingame, T.A. Baillie, and D.H. Russel, Anal. Chem., 1992, 64, 467R-502R.

4. K. Biemann, Annu. Rev. Biochem, 1992, 61, 977-1010.

5. R.D. Smith, J.A. Loo, C.G. Edmonds, C.J.Barinaga, and H.R. Udseth, Anal. Chem., 1990, 62, 882-899.

6. F. Hillenkamp, M. Karas, R.C. Beavis, B.T. Chait, Anal. Chem., 1991, 63, 1193A-1203A.

7. M.E. Hemling, G.D. Roberts, W. Johnson, S.A. Carr, T.R. Covey, Biomed. Environ. Mass Spectrom., 1990, 19, 677-691.

Return to the The ABRF Home Page