A hugely successful ABRF symposium was held on the evening of April 14 at the Second European Symposium of the Protein Society in Cambridge, UK. The stimulating program was organised by Alastair Aitken. The venue for the meeting was the newly opened Biochemistry Department of Cambridge University; the workshop had the honour of being the first meeting to use the new facilities. Some 170 delegates came to hear talks on "New Techniques in Protein Mass Spectrometry", and it was standing room only in the 150-seat lecture theatre. Unfortunately since Carol Robinson was taken ill just before the meeting her scheduled appearance was cancelled, but the other speakers adjusted their talks to compensate and the feedback from the audience was very positive. The after-talk socialising went with a real buzz, and a great deal of interest was shown in ABRF and the Research Committee posters by many attendees who were previously unaware of ABRF and its activities. As a result, several individuals and companies have expressed a wish to join ABRF. A summary of the meeting is given below.
After Alastair Aitken welcomed the audience and introduced the speakers, Len Packman described the growth of ABRF over the last 10 years and its mission, membership, and Research Committee activities. Details and examples of the use of the ABRF's electronic communications service activities was accompanied by many delegates hastily writing down the addresses. Talk of the ABRF's electronic mailing list was greeted with interest and the supply of newsletters soon disappeared into the delegates' bags.
Darryl Pappin talked about "Developments in Sequencing by Mass Spectrometry". His laboratory routinely identifies proteins from two-dimensional gels and has developed this technique to such an extent that Edman sequencing is now used only occasionally. A layered strategy for analysis is used and has the following steps:
The described digestion and extraction protocol requires no membrane pretreatment or destaining, and samples are analysed directly from the formic acid:ethanol extract without chromatographic clean-up. Typically, only 2 to 5% of the digest is required for analysis. Search discrimination is greatly improved by the use of an orthogonal measurement. This may include the use of a second digest enzyme or repeated MALDI analysis of the digest mixture following quantitative esterification to give acidic residue composition, a straightforward procedure that leads to a 14 Da increase in mass for each esterified site. The use of MALDI-TOF analysis as the first screen permits rapid analysis of multiple samples before committing to more labor-intensive (and more expensive) mass spectroscopic analysis. Individual samples require only 10 minutes for analysis, with database searches completed in 3 to 5 seconds.
If the sample remains an unknown the residual digest mixture (>90% remaining) is:
1) reacted with a pyridine-NHS ester which modifies amine groups;
2) purified by single-step reverse-phase capillary HPLC; and
3) analysed using a triple quadrupole tandem mass spectrometer fitted with a nanoelectrospray source.
This offers routine MS/MS sensitivity at 50 fmol/µl, with a practical limit of 10-15 fmol/µl. These values are well below what may be achieved with routine Edman sequencing. Because injection of a 1 to 2 µl sample can take up to an hour, there is ample time to select and fragment peaks to derive sequence information. With the use of the pyridine tag, the pattern of ions is simplified as the b-series is enhanced. Partial sequences can be rapidly deduced and combined with mass and cleavage enzyme specificity to give extremely rapid matches with known sequences (both whole protein and EST sequences). If matches are still not found, then more peptides are fragmented for the assembly of long continuous sequence stretches for DNA probe design. The use of the pyridine tags permits facile identification of b- and y-type ions, allowing complete sequences of 12 to 20 residues to be assembled from a single spectrum.
Pappin concluded by addressing the issue of stringency of searching databases, noting the importance of increased mass accuracy, orthogonal digestion, and esterification. Such additional information can be used in a search, allowing the identification of a protein from a single peptide in some cases and from only 4 to 5 peptides in the general case.
Brian Chait gave a talk on "Probing higher order structures of proteins with mass spectrometry" and described how the approach could be used to study molecular interactions. The method is based on inferring structural information from determinations of protection against enzymatic proteolysis, as governed by solvent accessibility and protein flexibility, using MALDI-TOF-MS to rapidly read out the proteolytic fragment masses as a function of time. If it is desired to probe the interaction of a protein with its molecular partners, proteolysis is carried out in the absence and presence of the interacting partners. A comparison of the difference in the resulting proteolysis patterns provides information on conformational changes that occur upon binding and regions of the molecules that are protectedduring binding. To illustrate the method, Chait described a study of the solution structure of the transcription factor Max, a member of the basic/helix-loop-helix/zipper family of DNA-binding proteins (1).
Protein crystallographers continually seek improved strategies to obtain crystals that diffract to high resolution. In order to obtain high resolution structures it is important to have the assurance of working with compact, well-defined proteins. This generally accepted notion presumes that restricting the degrees of conformational motion in a protein increases the chances that it will crystallize into well-ordered lattices. A precisely defined compact "folding domain" also results in a protein with reduced tendency towards aggregation. Chait described how the combination of proteolysis with mass spectrometry can provide information about the compact folding domains of proteins with unprecedented speed and accuracy (2). As an example, he described work performed on a pair of transcriptional activators that were undergoing crystallization trials. The proteins were dTAF42 and dTAF62 (factors associated with the Drosophila TATA-box-binding proteins), portions of which share strong similarity to histone proteins H3 and H4 and which (like H3 and H4) associate strongly in solution to form a heterotetrameric complex. Initial constructs of dTAF42 and dTAF62 used for the crystallization trials (designed based on sequence alignment with H3 and H4) yielded crystals that diffracted weakly (4 to 5 Angstrom resolution). It was believed that some of the termini of the constructs were flexible, reducing the propensity for forming well-ordered crystal lattices. Limited proteolysis on the TAF constructs in combination with MALDI-TOF-MS confirmed that three of the four termini were indeed flexible and provided results that accurately defined the proteolysis resistant "core" domains. Using this information, new constructs were designed that quickly yielded a dTAF42/dTAF62 complex that diffracted to better than 1.4 Angstrom resolution. The resulting structure represents the first high-resolution view of the histone fold, showing that the tetrameric TAF complex strongly resembles the (H3/H4)2 hetrotetrameric core of the histone octamer (3).
Chait stressed the importance of using mass spectrometry to check all stages of the crystallization process, especially when co-crystals were being formed, as it is important to check that the crystal contains both the expected components. As an example, he described the limited proteolysis and co-crystallization of the conserved core of the Nef protein from HIV-1 with the SH3 domain of a Src family tyrosine kinase (4).
Finally, Chait discussed the use of mass spectrometry to elucidate the protein components of a large biological machine the nuclear pore complex of yeast through which all molecular traffic passes between the cytoplasm and the interior of the nucleus. The whole multisubunit complex (containing perhaps 50 different proteins) has an estimated mass of approximately 60 MDa and is about 100 nm in diameter with a 25 nm aqueous channel. Because the protein components are not all amenable to separation by two-dimensional electrophoresis, fractionation is carried out using reverse-phase HPLC and hydroxyapatite chromatography, each followed by SDS gel electrophoresis of the fractions. The proteins are then identified by in-gel trypsin digestion followed by MALDI-TOF-MS peptide mapping or MALDI-ion trap tandem MS with a custom instrument (5). The latter instrument offers fragmentation capabilities of tryptic peptides that are selective and highly informative. Identification of the proteins is performed using protein database searching strategies that are publicly available on the PROWL web site (http://chait-sgi.rockefeller.edu). This strategy is being used in an effort to identify all the proteins present in the yeast nuclear pore complex. Ultimately, it is hoped the positions of each protein in the complex will be mapped by techniques such as immuno-electron microscopy.
References
1. Cohen, S.L, Ferre-D'Amare, A.R., Burley, S.K. and Chait, B.T. (1995) Protein Science 4, 1088-1099.
2. Cohen, S.L. (1996) Structure 4, 1013-1016.
3. Xie, X., Kokubo, T., Cohen, S.L., Mirzam, U.A, Hoffman, A., Chait, B.T., Roeder, R.G., Nakatani, Y. and Burley, S.K. (1996) Nature 380, 316-322.
4. Lee, C-H., Saksela, K., Mirza, U.A., Chait, B.T. and Kuriyan, J. (1996) Cell 85, 931-942.
5. Qin, J., Steenvoorden, R.J.J.M., Chait, B.T. (1996) Anal. Chem. 68, 1784-1791.
Len Packman may be contacted at the Department of Biochemistry, Cambridge University, Tennis Court Road, Cambridge CB2 1QW, England, Tel: (+44)1223-333639, Fax: (+44)1223-333345, E-mail: lcp2@mole.bio.cam.ac.uk.
Return to the The ABRF Home Page