Peptides up to approximately 50 amino acids in length are synthesized by the facility; both L- and D-amino acid stereoisomers can be incorporated. Synthesis is carried out on a polymer resin, beginning with the C-terminus and ending with the N-terminal residue, and all reactive or labile side chains are protected. The completed peptide chain is released from the resin by cleavage with TFA; most protecting groups leave during this reaction. Peptides are prepared in the reduced form, but internal disulfide bonds can be formed if needed.
Peptides are also available carrying the following modifications: biotinylation, acetylation, phosphorylation, glycosylation, methylation, and sulfation. Many others are possible but particular amino acid derivatives may have to be custom synthesized by an outside contractor before incorporation in a peptide. All manipulations are carried out by the facility staff.
Although optional, essentially all peptides are purified in the facility using preparative reversed phase HPLC; yields are generally around 10 to 20 mg, but larger amounts are available upon request. Also at the request of the investigator, peptides can be further purified in a second and third round of HPLC using the same or different columns; this results in substantial cost increase and decrease of yield.
Quality control is done in the facility by analytical HPLC, mass spectrometry, and by sequencing (if necessary). Facility staff will provide information on any anticipated sequence/composition related difficulties during synthesis and for solubilization.
Protein Identification
A reliable method to prepare proteins for mass spectrometric identification is by SDS-PAGE. Coomassie blue-stained proteins, while still inside the excised gel pieces, are readily digested with trypsin and the resulting peptides extracted. Peptide masses can then be quickly and accurately determined from unfractionated mixtures by matrix-assisted laser-desorption / ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS), and this at final concentrations below 10 femtomoles / µL. Such information, a peptide mass fingerprint enables protein identification by direct querying of sequence databases.
In MALDI-TOF MS (and TOF/TOF MS/MS), the sample is introduced as a solid crystal. The technique is relatively simple, very accurate and very sensitive. As few as five to six peptide masses (accurate to 30 to 50 parts per million -- i.e., 0.03 to 0.05 Da for a peptide of 1,000 Da) will suffice for mass fingerprinting, by asking the question: find me a protein in the database that contains predicted tryptic peptides with theoretical molecular masses that match 5 out of 5 (or even 4 out of 5) of the experimental peptide masses. In a best-case scenario, such query will yield a single, positive identification. In practice, the approach usually fails when:
(i) protein preparations are mixed (2 to 4 proteins in a single band may work; 5 and more is really difficult) or contaminated (e.g., keratin, immunoglobulin, GST-fusions, ... ) or
(ii) when the full-length sequence is not available in any database (only in case of unusual study organisms). TOF/TOF MS/MS analysis is typically used to confirm a known or predicted peptide structure, but can also be used for identification (see below, under ESl-MS/MS, for details or mass spectrometric sequencing).
Commercially available tandem mass spectrometers utilize electrospray (ESI), which involves introduction of the sample in liquid form, and at a very low flow (nanoL / min). ESI-MS/MS, as provided in our facility by a hybrid quadrupole-TOF mass analyzer, enables protein identification on the basis of fragmentation (mass) data derived from a single tryptic peptide. This peptide need not be purified before analysis as the MS/MS method allows to select peptides of any particular mass out of complex mixtures. Subsequent fragmentation results from high-speed collisions with gas molecules, which induce cleavages across the amide backbone of a peptide and provide ion series corresponding to the amino acid sequence.
The process is often referred to as sequencing, but it rarely ever yields a bona fide sequence (e.g., 10 to 20 residues) that could be used for a BLAST search, and then only if ample quantities are available. Rather, specialized database search algorithms are used that query GenBank with the results 'uninterpreted' MS/MS spectra. Identifications solely based on MS/MS data are certainly achievable but single peptide matches, even when statistically sound, must be interpreted with caution; two seems a dependable minimum.
How to Submit a Protein for Identification
Protein bands are cut from the gels. They should be reasonably homogeneous and in ≥200 femtomole quantities each. Mixed proteins are not only harder to analyze, they also create a misleading picture. A band estimated to contain 200 femtomoles, but with 4 different proteins in it, provides only 50 femtomoles on average per protein for analysis. The rule of thumb is that any protein band over 50 kDa should be visible by coomassie staining on a minigel. Coomassie is preferable anyway as silver has adverse effects on subsequent sample handling. Smaller proteins may not be visible in these low quantities and should be stained with silver, but only as lightly as possible and with a modified silver stain that does not involve aldehyde based fixing (protocol available at the facility).
Mass spectrometry is a concentration sensitive technique, highly limited in how much volume can be introduced for analysis (~0.5 µL for MALDI). Therefore, the gel piece(s) should also be kept to a minimal volume to avoid unwanted dilution, and lower concentrations, later on. Ideally, gel volume should not exceed 20 to 25 µL (i.e., a single 1 x 2 x 10 mm piece). Facility staff will gladly assist with cutting the bands if so requested.
CAUTION: Keratin Contamination. Please be very careful with bare hands, loose hair, dirty glass, and plasticware. Even a trace of keratin will preclude identification!
Polypeptide Mass Analysis (other than protein identification)
Molecular masses of polypeptides (partially) purified by investigators can be determined with a mass accuracy of 0.01 percent or better using MALDI-TOF. Samples are to be submitted in solution and in a concentration of 1 picomole / µL or more with the following restrictions. The ideal sample solution is in 0.1 to 1 percent formic acid or TFA in 30 to 50 percent acetonitrile; under no circumstances should the sample contain buffers and salts (and certainly not Na or K ions), ionic and zwitterionic detergents, glycerol, and should not be alkaline. Investigators wishing to have their polypeptides mass analyzed should contact the facility for advice on sample preparation before submitting their samples.
Facility staff will carry out all measurements and, if required, assist the investigator with computer-aided matching of the polypeptide mass data to specific sections of the parent protein (or an ORF), taking into account various common post-translational modifications.
Mapping of post-translational modifications, using mass spectrometric analysis of proteolytic fragments, is not offered as a routine service at this point. The level of difficulty, and the associated time and effort required, of such analyses is highly dependent on the protein sequence and the fraction that is modified. Feasibility studies may be conducted at the discretion of the manager with the understanding that the project may be terminated immediately thereafter depending on the likelihood, or absence thereof, of facile and timely completion on the entire study.
Polypeptide Chemical Sequencing
Proteins and large polypeptides can be successfully sequenced at the ≥5 picomole level by the microchemistry facility for approximately 10 to 25 cycles. Larger amounts will allow sequencing runs of over 40 cycles. However, certain restrictions apply. Polypeptides must be prepared under conditions that will not interfere with the sequencing chemistry and instrumentation. Sample requirements and restrictions are fairly well-defined and will be discussed beforehand with those scientists requesting assistance and service. Practical advice is given, and proven strategies for final purification are suggested. When the protein is prepared by gel electrophoresis and electroblotting onto a polyvinyl difluoride membrane, approximately 5 to 10 picomoles will be needed. In the latter case, sample characteristics are near optimal and the quality of the data greatly improved. As a rule, small to midsized peptides generated and purified outside the facility will not be analyzed, except by prior arrangement and discretion of the manager.
Radio-sequencing of proteins (solution or blotted onto PVDF; same restrictions as for 'regular' sequencing apply) is provided as a service by the facility. At the present time, only two type of labeled proteins will be accepted for analysis. Those labeled metabolically, or during in vitro translation, with C14, H3, or S35 containing amino acids; and those acetylated and methylated side-chains (not the alpha-amine!), and containing C14 or H3 label in the modifying group. Note that no typical amino acid identifications are being made during radio-sequencing; only scintillation counting of the cyclically removed amino acids and derivatives is done. Thus, to be useful, the sequence of the polypeptides must be known in advance.
