Mass Spectrometry and Proteomics

• Mass spectrometry analyses and identification of proteins, peptides, oligonucleotides, carbohydrates and small molecules
  • 2D LC separations of complex protein and/or peptide mixtures for proteomics – up to 5,000 to 7,000 confident protein identifications from a single sample
  • Protein expression analysis using iTRAQ/ICAT/TMT, SILAC or 15N methodology – relative quantitation of thousands of proteins across up to nine samples simultaneously
  • SWATH/MSall Data-Independent Acquisition for simultaneous identification and label-free quantitation of peptides/proteins/lipids/small molecules
  • Proteomic data analysis using Byonic, MSFragger, and Scaffold
  • MS Quantitation using Selective/Multiple Reaction Monitoring
    • Antibody-free accurate quantitation of protein levels in hundreds of samples
    • Accurate quantitation of small molecules, hormones, NTPS, drugs and metabolites (e.g., for pharmacokinetic studies, drug metabolism studies)
    • Accurate quantitation of cytokines, amino acids, etc. from microdialysates
• Accurate quantitation of cytokines, insulin, glucagon, etc., using MesoScale Discoveries (MSD) QuickPlex SQ120 Electrochemiluminescent Plate Reader

• ABI/MDS Sciex 5800 MALDI TOF-TOF mass spec
• ABSciex Selexion QTrap 6500+ mass spectrometer with ion mobility
• Bruker TimsTOF Flex mass spectrometer with ion mobility, MS Imaging capacities
• MesoScale Discoveries (MSD) QuickPlex SQ120 and Sector Imager 2400 Electrochemiluminescence Detector/Plate Reader
  • MSD’s electrochemiluminescence detector/plate readers (QuickPlex SQ120, Sector 2400 Imager) are located in Room C1732 of Penn State College of Medicine, and are a medium-throughput imaging detection system. They are capable of multiplexing in all spot formats and read 96-well plates. They enable the detection of biomarkers in single and multiplex formats and are capable of multiplexed, high-throughput screening and assay development. An ultra-low-noise CCD camera with custom-designed telecentric lenses allows for rapid detection in 96-well multi-array plates or 96-well multi-spot plates. The Discovery Workbench has software that can calculate concentrations and produce graphs and tables that can be used in presentations or publications. A complete listing of assay classifications and kits available from MSD can be found at mesoscale.com. Free on-site application support and training is available from MSD. Assistance with data analysis is also provided. For details, email Brian Gribble at bgribble@mesoscale.com.
• Bioinformatics: Byonic, MSFragger, ProteinPilot and MaxQuant advanced protein-MS identification software; Scaffold Mass Spec Quantitative Biostatistics Software; MultiQuant; Skyline
• HPLC/UPLC:
  • nanoElute UHPLC
  • Sciex Exion LC high pressure system
  • Shimadzu Prominence XR UFLC
  • Waters Acquity UPLC liquid chromatography systems
  • ÄKTA Prime Plus system
• Wyatt DAWN HELEOS light-scattering detection system for protein sizing

In addition to using classic gel spot densitometric (2D gels) or fluorescence (DIGE gel) measurements to find proteins whose levels differ between two or more samples, identification and quantitation of protein level changes can be done by using chemical isotope labeling methods.

In such methods, one tags all proteins from different samples with identical chemical tags that differ only in their distribution of heavy and light (non-radioactive) isotopes of the same atoms, e.g., 1H vs. 2H, 12C vs. 13C, 14N vs. 15N. Where possible, this can be done in living cells or animal models by feeding heavy vs. light isotope versions of various naturally incorporated amino acids (sometimes called SILAC methods); however, for many studies, such pre-labeling is not possible.

In such cases, various chemical isotope incorporation methods are available. The College of Medicine Mass Spectrometry and Proteomics Core primarly uses Isobaric Tags for Relative and Absolute Quantitation (iTRAQ) methods to perform many hundreds of large-scale analyses of complex samples, including human serum samples, animal and plant cell lysates and cell organelle preps. TMT 9 plex and older ICAT quantitative labeling methods are also available.

Such experiments typically identify and quantitate the relative amounts of 500 to 3,500 proteins from four to eight samples at once.

Standard sample separations for iTRAQ/TMT experiments done in the Mass Spectrometry facility (after the iTRAQ/TMT reagent labeling step) are identical to those for an LC-MALDI MudPit experiment; see the standard sample separations section of this page for details.

Complex analyses requiring MS/MS and MS3 fragmentation, or requiring electrospray-, APCI- or photo-ionization are performed on one of three mass spectrometry instruments.

The Sciex Triple Quad 6500+ LC-MS/MS system with Selexion (ion mobility) Technology in Room C3615 offers industry-best sensitivity, speed and performance for the most challenging methods. It is ideal for drug discovery, development and pharmacokinetics; metabolite identification; MRM/SRM quantitation; contaminant testing; and MRM/SRM-type proteomic applications, including post-translationally modified proteins discovery, protein identification and biomarker validation. This instrument also includes the possibility of MS3 analyses, as well as additional ionization modes possibly advantageous for analyses of some types of molecules (APCI or PhotoIonization).

For details on using these instruments, contact Dr. Dongxiao Sun at dsun@pennstatehealth.psu.edu or 717-531-0003, ext. 287146.

Interferences and contaminants encountered in modern mass spectrometry, Bernd O. Keller, Jie Sui, Alex B. Young and Randy M. Whittal Analytica Chimica Acta 627, Issue 1, 3 October 2008, 71-81, offers a guide to common contaminants and their sources.

Of particular note, in the supplemental data, a spreadsheet is provided that contains a searchable ion list of all compounds identified to date by the study team, including positive ion contaminants; negative ion contaminants; repeating unit contaminants; adducts, losses and replacements; solvent masses; salt and detergent tolerances for MALDI, ESI; references; and a glossary.

Explore the contaminant study

Transfer protocol suggested

The core recommends the VWR Fluorotrans 0.2 micron membrane by Pall. If you have an old lots of BioRad PVDF (pre-1997), you may still use that, but the newer BioRad PVDF membranes do not appear to have the same quality. The preferred size of the blot piece to put on the sequencer is anything smaller than 2mm square. Using the least amount of PVDF possible is advantageous, since the more PVDF in the reaction chamber, the higher the background will be. Submitting two pieces of PVDF will not be better since the ratio of PVDF/sample will stay the same.

The secret of a good gel: Let the gel polymerize over night. Prerun at 3mA constant current for two hours prior to loading the sample. (Unpolymerized acrylamide in gel can block the N-termini of proteins, making Edman Sequencing impossible; see a similar warning from Iowa State University here.)

If the protein is not too basic

Use CAPS 3-[cyclohexylamino]-1-propane sulfonic acid as the transfer buffer.

10X CAPS (100mM, PH11)

• Dissolve 22.13 g of CAPS in 900ml of DI water. Titrate with 2N NaOH to pH 11 and add DI water to 1000ml. Electroblotting buffer (1X stock buffer in 10% MeOH): Prepare two liters of buffer by mixing 200ml of 10X Caps buffer with 200ml of MeOH and 1600ml of DI water.
• Wet PVDF with MeOH for a few seconds and place membrane in a dish containing blotting buffer.
• Remove gel for the electrophoresis cell and soak in electroblotting buffer for five minutes.
• Assemble the sandwich and electroblot at either constant voltage of 50 volts at room temperature for 30 minutes or constant current of 500 mA for the same amount of time.
• Remove PVDF and rinse with DI water.

Coomassie staining

• Saturate PVDF with MeOH for a few seconds.
• Stain PVDF with 0.1 percent coomassie Blue R250 in 40 percent MeOH/1 percent acetic acid for 30 seconds.
• Remove PVDF and destain with 50 percent MeOH.
• Rinse with water.

If the protein is very basic

Use a regular Tris-Glycine buffer and after staining wash the blot extensively with plain H2O.

Example Tris-Glycine buffer – 25 mM Tris, 190 mM glycine, 0.1 percent SDS, 20 percent methanol, pH 8.5

To make 1000 ml, dissolve 14.4 g of glycine, 3.0 g of Trizma base, and 1.0 g of SDS in 800 ml of deionized water. Add 200 ml of methanol, stir and degas for 20 min. This buffer should be prepared freshly.

Note that for basic proteins it is possible to need to use blotting membrane on both sides of the gel. Proteins that are more basic than the pH of the transfer buffer will be captured on the cathode side membrane. Those less basic will be captured on the anode side. Increasing the pH of the Tris-Glycine transfer buffer to pH 9.2 will make all proteins below pI approximately 9.2 transfer toward the anode electrode, etc.

For details on this protocol, email Anne Stanley at aes7@psu.edu.

Prior to labeling with iTRAQ or TMT reagents, it is necessary to ensure that samples contain equal amounts of total protein. Many groups quantitate their protein from each sample (e.g., by BCA or BioRad/Bradford assay) and then run a gel and silver stain the gel to verify that each sample appears to have the same amount of stained material – a normalization procedure in the ProteinPilot analysis software can compensate for some differences in initial protein amount, but it is obviously best to start with equivalent amounts of protein in each sample. Also, be sure that the sample does not contain, nor was prepared with, a solution containing any interfering substance as seen below in this protocol.

The core suggests doing steps 1 to 15 below with slightly more than the final amount of protein that will be labeled. (The original iTraq protocol calls for labeling 100ug of protein, which each iTraq reagent vial is sufficient for, but perfectly good iTraq results can be obtained labeling only 10 ug of protein, or 80 ug total for 8 samples.)

By digesting more than the final amount of protein to be labeled, it is possible to run a gel and stain it after step 15 to make certain that the sample was digested well. (Note that a gel cannot show whether a sample is completely digested, but can show if it is not well-digested.)

When digesting more than the final amount of protein to be labeled, adjust the final volume of solution from each sample used at step 15 so that each sample tube contains the same amount of protein (less than 10 ug of protein per sample can also be used, as long as the same amount of total protein is labeled from each sample; however, there may start to be small decreases in the total number of protein identifications with less input protein, so at least 10 ug protein per sample, and the same amount of total protein in each sample tube, is the ideal amount to use for each sample.)

Suggested products

• iTRAQ Multiplex (4-plex) Kit, Applied Biosystems #4352135
• iTRAQ Multiplex (8-plex) Kit, Applied Biosystems #4390811 (single kit) or #4390812 (five kits)
  • Note that the College of Medicine Mass Spectrometry Core orders the 8-plex kits in bulk to get a discount, so the least expensive way to get the 8-plex kits is to get them through the core. Investigators can order a Multiplex Buffer Kit (ABI #4381664), but note that the core does not recommend using all of the reagents in that kit in the optimized labeling protocols shown here (e.g., the core recommends using iodoacetamide instead of MMTS for an alkylating agent), and all of the reagents in that kit are common laboratory reagents whose concentrations are listed in the protocol below, so it is not necessary to purchase that buffer kit.
• TCEP (tris-(2-carboxyethyl) phosphine) reducing agent, Pierce #20490 or Sigma #C4706 (powder) or #646547 (solution)
  • For investigators making their own solution from powder, a 0.5 M solution in water, brought to pH 7.0 with NaOH, is what the Sigma solution is; the solution without buffering is approximately pH 2.5. That solution needs to be diluted to 110 mM (1 volume 0.5M stock, plus 3.55 volumes water) to be used to add 1 µl below).
• Iodoacetamide, Sigma, #A3221-10vL
• Sequencing Grade Modified Trypsin, Promega #V511

Protocol

1. To each sample containing 10 to 100ug of sample (almost dried completely), add 20µL Dissolution Buffer (Dissolution Buffer = 0.5 M triethylammoniumbicarbonate at pH 8.5, e.g., Sigma product 17902, T7408 or 90360 diluted to 0.5 M with water), i.e., a protein concentration of approximately 5 mg/mL = 5 µg/µl
2. Add 1µL of the Denaturant (2 percent SDS) and vortex
3. Add 1µL of 110 mM Reducing Reagent tris-(2-carboxyethyl) phosphine (TCEP) to each sample to make 5 mM TCEP concentration
4. Vortex, spin
5. Incubate the tubes at 60 degrees C for 1 hour
6. Spin the sample
7. Add 1µL of freshly prepared 84mM solution of iodoacetamide; note that this is not the MMTS included in iTRAQ kits
8. Vortex, spin
9. Incubate the tubes in the dark at room temperature for 30 minutes (wrap tubes in foil)
10. Reconstitute a vial of Promega Sequencing Grade trypsin w/21µL of resuspension buffer (50 mM acetic acid, supplied with the Promega trypsin)
11. Vortex, spin
12. Prepare a 1mg/ml solution of trypsin and add 10µl to each sample
13. Vortex, spin
14. Incubate samples at 48 degrees C overnight (three to four hours is probably sufficient, but note that, for any given time of incubation, trypsin efficiency is higher at 48 degrees C than at 37 degrees C, see e.g. Kinetic characterization of sequencing grade modified trypsin, Finehout EJ, Cantor JR, Lee KH. Proteomics. 2005 Jun;5(9):2319-21)
15. Spin samples
    • In order to maximize labeling efficiency, the volume of the sample digest must be less than 50µL (33µl MAX for 8Plex reagents). If the volume of the sample digest is greater than 50µL (33 µl), dry the sample in a centrifugal vacuum concentrator, then reconstitute with 30µL Dissolution Buffer.
16. Bring each vial of iTRAQ Reagent needed to room temperature
17. For 4Plex reagents, add 70µL of ethanol to each iTRAQ Reagent vial being used. For 8Plex reagents, add 50 µl of isopropanol to each iTRAQ Reagent vial being used. The percentage of isopropanol in the reaction tube below (step 19) must be at least 50 percent v/v.
18. Vortex each vial for 1 minute, then spin. Check the pH; if it is not at least 7.8 to 8.5, then add up to 5µl of Dissolution Buffer to get the pH at or above 7.8. Simultaneously add isopropanol as needed to keep the final concentration of isopropanol in step 19 below at least 60 percent v/v.
19. Transfer the contents of one iTRAQ Reagent vial to one sample tube
20. Vortex, spin
21. Incubate the tubes at room temperature for one hour (4Plex) or two hours (8Plex)
22. After one hour (4Plex) or two hours (8Plex), add 100µL of Milli-Q water to each tube to quench the iTRAQ reaction. Incubate at room temperature for 30 minutes.
23. Combine the contents of all iTRAQ Reagent-labeled sample tubes into one tube
24. Vortex, spin
25. Dry the tube containing all the combined iTRAQ mixes
26. Add 100µL of water to the tube.
27. Vortex to mix, spin
28. Dry sample completely
29. Repeat steps 26 to 28 two more times (a total of three times)

Once these steps are completed, the Mass Spec iTRAQ/TMT/SWATH Sample Submission Sheet found in One Drive (Penn State Access ID login required) must be submitted by emailing the completed form to Anne Stanley at aes7@psu.edu and submitting it through iLab. The Excel sample submission form template can also be downloaded directly from iLab during the sample analysis request process.

Samples should then be taken to the mass spectrometry core facility (Room C1734 in the College of Medicine) for analysis. The 1D-2D LC separations and mass spectrometry analyses will take six to eight days to complete once started.

Potential iTraq interfering substances

Recommended Alternative Detergent/Denaturants

From iTRAQ Reagents Protocol, Applied Biosystems

• SDS (0.05 percent)
• OG (octyl B-D-glucopyranoside) (0.1 percent)
• NP-40 (0.1 percent)
• Triton X-100 (0.1 percent)
• Tween 20 (0.1 percent)
• CHAPS (0.1 percent)
• Urea (less than 1M)
  • Note: When using urea, always use a fresh solution. When reducing a sample containing urea, incubate the tubes at 37 degrees C for one hour.

Recommended Alternative Buffers

From iTRAQ Reagents Protocol, Applied Biosystems

• BES
• BICINE
• Boric acid
• CHES
• DIPSO
• EPPS
• HEPBS
• HEPES
• HEPPSO
• MOBS
• MOPS
• Phosphate-buffered saline
• PIPES
• POPSO

This recipe is just one example of satisfactory protocols. This was done primarily for digesting approxmiately 100 nMoles of protein in a final volume of trypsin digestion reaction of 100 µl (protein solution about 10 percent of the volume, i.e., enough to keep cations like Na+ concentration only about 5 mM – above 40-50 mM concentration Na+ or K+ can inhibit subsequent mass spec).

If doing without reduction/alkylation of cysteines, this 100 µl final volume contains:

• 50 mM NH₄HCO₃, pH 8 (if digestion is for subsequent iTRAQ labeling, use triethylammonium bicarbonate, but note that iTRAQ digests should be alkylated; see the iTRAQ sample prep protocol elsewhere on this page for details)
• 10 percent v/v acetonitrile (10 µl)
• 0.1 µg Promega Gold or Sequencing grade modified (methylated) Trypsin (to digest 100 ug protein for iTRAQ, use 5.0 µg Trypsin)
• Incubate at least 3 hours at 48 degrees C (alternately, 16 to 18 hours at 37 degrees C)
• (Optional: Stop reaction by addition of 4 µl of glacial acetic acid)
• SpeedVac to dry down reaction completely to evaporate off NH₄HCO₃ and acetonitrile. Resuspend in 200 µl H₂O with vortexing.
• Repeat drying down three times total, but last time, dry down to leave approximately 9 to 10 µl instead of complete evaporation. (See iTRAQ sample prep protocol elsewhere on this page for different instructions for iTRAQ.)
• Add 1/9th volume of 1.0 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI, or 1.0 percent formic acid for samples that will be analyzed by ESI (Electrospray or Nanospray) instruments such as the ABSciex 5600 TripleTOF or QTrap 4000 or Waters Synapt HDMS to make final 0.1 percent TFA or formic acid concentration (can be done in mass spectrometry facility).

If doing the reactions with prior reduction/alkylation (which can improve digestion efficiency by preventing disulfide bonds formation, which in turn helps the protein to be fully unfolded and therefore having all tryptic sites exposed):

For iTRAQ samples, see the iTRAQ sample prep protocol elsewhere on this page; do not use these instructions.

• If the protein isn’t already in at least 2.5 mM DTT, add DTT or TCEP to the solution to 2.5 mM and incubate at 50 degrees C for at least 15 minutes.
• Then, in a final volume of 20 µl:
  • Approximately 100 nmole (approximately 4 µg) of protein in no more than 10 µl, to keep the DTT concentration down to avoid quenching the iodoacetamide reaction)
  • 50 mM NH₄HCO₃, pH 8
  • 10 mM iodoacetamide (e.g., add 0.8 µl of 250 mM iodoacetamide)
• Incubate in the dark for 30 minutes at 37 degrees C.
  • After 30 minutes, the reaction can be quenched somewhat (to avoid subsequent alkylation of the trypsin) by adding an additional 2.5 µl of 100 mM DTT and incubating for another 15 to 30 minutes at 37 degrees C. (This would make the DTT concentration in the 20+2.5 µl volume at least 11 mM, theoretically enough to stoichiometrically titrate all the 10 mM original concentration of iodoacetamide.) The College of Medicine has no direct evidence that this is necessary, and researchers have gotten very nice trypsin digestions without this quenching.
• Bring the volume of the protein alkylation reaction up to a final volume of 100 µl containing:
  • 50 mM NH₄HCO₃, pH 8
  • 10 percent v/v acetonitrile (10 µl)
  • 0.1 µg Promega Gold or Sequencing grade modified (methylated) Trypsin (basic rule of thumb is 1 µg trypsin for every 20 to 100 µg of protein)
• Incubate for more than 3 hours at 48 degrees C (alternately, 16 to 18 hours at 37 degrees C)
• (Optional: Stop reaction by addition of 4 µl of glacial acetic acid.)
• Dry down reaction completely to evaporate off NH₄HCO₃, and acetonitrile. (Lyophilization is better than SpeedVac if the option is available, but either works). Resuspend in 200 µl H₂O with vortexing.
• Repeat drying down three times total, but last time, dry down to leave approximately 10 µl instead of complete evaporation.
• Add 1/9th volume of 1.0 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI or 1.0 percent formic acid for samples that will be analyzed by ESI (Electrospray or Nanaospray) instruments such as the ABSciex QTrap 4000 or Waters Synapt HDMS to make final 0.1 percent TFA or formic acid concentration (can be done in mass spectrometry facility).

Although it has not been tested at the College of Medicine, there is also evidence that adding the acid-labile surfactant sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl)methoxyl]-1-propanesulfonate, marketed by Waters under the name RapiGest (p/n 186001860, 186001861, 186002123, 186002122) to digests will produce more complete digests, particularly in the case of more difficult proteins (see for example Rapid Commun. Mass Spectrom. 2004; 18: 822–824). There is also evidence that 5 percent trifluroethanol (TFE) may act as an excellent denaturant for digestion and subsequent mass spectrometrometry, replacing the 10 percent acetonitrile or other organics used in some protocols, and the urea used in others.

Note that multiple other protocols exist, all of which probably work fine for most digests, with only occasional individual proteins benefiting from one protocol vs. another.

1. Separation of the combined four or eight iTRAQ-labeled samples offline into 15 strong cation exchange fractions, using a 4.6 X 250 mm PolySULFOETHYL Aspartamide Strong Cation exchange column (PolyLC, Columbia, MD) with an ammonium formate gradient in 20 percent acetonitrile. See LC-MS/MS and Data Analysis Procedures elsewhere on this page for details.
2. SpeedVac drying and resuspension in H₂O of all fractions (three times) to remove all acetonitrile and ammonium formate.
3. Separation of each SCX fraction on a 15 cm Reprosil C18 nanoflow column on an Eksigent ChipLC system, and direct injection into the ABSciex 5600 TripleTOF mass spectrometer. (For MALDI analyses, separation of each SCX fraction on an LC-Tempo nanoflow separation and MALDI spotting system, using a Chromolith CapRod column C18 column (150 X 0.1 mm, Merck), into 370 MALDI spots on a stainless steel MALDI target plate, then adding 13 calibration spots to the same target plate).
4. Instrument is calibrated with an injection of beta-gal digest standard before each sample injection. (For MALDI analyses, update plate calibration and MS/MS default calibration for each plate as it goes into the Applied Biosystems 5800 MALDI TOF-TOF mass spectrometer).
5. A 250 ms parent scan is acquired, then up to 50 MS/MS spectra over 2.5 seconds, then repeated with the next LC eluant fraction. (For MALDI analyses, acquisition of 400 laser shots for MS spectra from each spot, then data-dependent acquisition of MS/MS spectra for each peptide mass, with the MS/MS spectra taken from the spot containing the largest MS peak representing each of the peptide peaks observed across the entire plate).
6. Combination of the MS/MS data from all 15 SCX fractions for a Paragon algorithm search (ProteinPilot software, against a concatenated normal and reversed database – SwissProt, NCBI, etc.). This search of a concatenated normal and reversed database allows estimation of the False Discovery Rate (or False Positive Rate) so that one can set the score limits for “positive identifications” to only accept identifications where the local False Discovery Rate estimate of the lowest ranking protein is 5 percent, and all higher ranking proteins have decreasingly lower probabilities of being false positives (See Calculating False Discovery Rates elsewhere on this page for details on the importance of using False Discovery Rate estimations and how they are calculated.
7. The resulting ProteinPilot .group files can be viewed with a Windows XP or later computer by installing a trial version of the ProteinPilot software. Once the trial period is over, you can continue to use the software as a viewer for .group files produced by licensed versions of the software.

Sample analysis requests are submitted using iLab at psu.corefacilities.org, choosing the “Mass Spectrometetry & Proteomics (COM)” Core and choosing the appropriate sample request type under the “Request Services” tab (Proteomics ID, Quantitative Proteomics, Small Molecule, etc.).

A free-to-obtain iLab login is required. The sample request process requires you to download an Excel file, fill out the required information, and then upload the completed Excel form back into iLab. Investigators should also email the completed Excel form to Dr. Bruce Stanley at bas12@psu.edu, as well as to Anne Stanley at aes7@psu.edu for proteomic samples or to Dr. Dongxiao Sun at dsun@pennstatehealth.psu.edu for small-molecule samples.

For details, contact Dr. Bruce Stanley at bas12@psu.edu or 717-531-5329.

The iTRAQ and non-iTRAQ sample submission forms are also available in OneDrive (Penn State Access ID login required).

If samples are quickly kept frozen after cell lysis, protease inhibitors may not be necessary; however, if you must use protease inhibitors during sample preparation procedures, please use the recommended protease inhibitors below. AVOID inhibitors such as AEBSF and others (see below) that

• inhibit trypsin or other proteolytic enzymes used to prepare proteins for mass spectrometry analysis or
• modify proteins covalently (as this changes their masses and makes mass spectrometry identificaiton more difficult).

Protein solutions that will be run on gels may be prepared with lysis buffers containing protease inhibitors that do not modify protein masses, since the protease inhibitors will not impact subsequent in-gel digestion and LC/MS/MS analysis (no inhibition of subsequent intentional protein degradation for mass spec prep.

Recommended Protease Inhibitor cocktail if the sample requires protease inhibitors

PMSF (100 mM), phenanthroline (1 mg/ml), benzamidine HCl (100 mM), pepstatin A (1mg/ml). Add about 1/1000th to 1/500th volume of that stock to the lysis buffers, then ideally remove by dialysis into ammonium bicarbonate (use TEAB instead of ammonium bicarbonate for iTRAQ analysis experiments) before submitting the protein for mass spec analysis (keep solution frozen after dialysis/removal of inhibitors).

Potential Problems with Protease inhibitors

Some protease inhibitors can affect mass spectrometry analyses and should be not be used unless absolutely required by the particular samples and experimental goals; please discuss with core staff.

For example, many protease inhibitor cocktails contain AEBSF (4-(2-aminoethyl)-benzenesulfonylfluoride, a serine-protease inhibitor that both covalently modifies proteins (shifting their peptide masses and therefore changing their detectability by mass spectrometry) and will inhibit trypsin used in mass spectrometry sample prep. (Adds a mass of 183 preferentially to tyrosines, and also to lysines, histidines, and N-terminal amines). While the trypsin inhibitory effects can be removed by gel separation or ZipTip or other pre-digestion cleanup steps, any covalent modifications on proteins in the sample would still remain.

TLCK and TPCK are also serine-protease inhibitors, so they can inhibit the activity of trypsin, chymotrypsin and other enzymes.

EDTA should not be used in sample preps with experimental goals of phosphorylation detection when IMAC purification will be used to enrich for phosphopeptides.

Importance of listing all sample treatments, kits and chemical exposures for proteomic samples

Both commercial protease inhibitor cocktails and commercial protein preparation kits may contain reagents that derivatize proteins, urea treatment in preparation for trypsin digestion can carbamylate peptides, and some reagents such as detergents can inhibit mass spec ionization and sensitivity. Because these derivatized peptides have different masses, it is crucial that all reagents, buffers, kits and chemical exposures used in sample preparation be listed on the Sample Submission Sheets, so that core staff can properly clean up and analyze the samples. Identification efficiency may suffer or even be lost completely if core staff are not made aware of all reagents and steps to which the samples have been exposed.

All publications, press releases or other documents that result from the utilization of any Penn State College of Medicine Institutional Research Resources including funding, tools, services or support are required to credit the core facility and associated RRID for each core used. Use of an instruments in the Mass Spectrometry and Proteomics core should include the following:

Mass Spectrometry and Proteomics Core (RRID:SCR_017831) services and instruments used in this project were funded, in part, by the Pennsylvania State University College of Medicine via the Office of the Vice Dean of Research and Graduate Students and the Pennsylvania Department of Health using Tobacco Settlement Funds (CURE). The content is solely the responsibility of the authors and does not necessarily represent the official views of the University or College of Medicine. The Pennsylvania Department of Health specifically disclaims responsibility for any analyses, interpretations or conclusions.

Due to the greater sensitivity of the Bruker TimsTOF Flex mass spectrometer, 2D-LC separations are not usually required, as a 1D separation gives deep coverage. However, as requested for still greater depth of coverage, or for analyses with the Sciex TripleTOF 5600+, initial SCX fractionation is performed as described above. (ERLIC and high-pH RP separations are also used as first-dimension separations for some sample types.) Calibration and validation of Bruker instrument performance is performed weekly by injecting 100 ng K562 digest (Promega) and requiring at least 3,000 confident protein IDs from a 30-minute gradient elution from the nanoElute UHPLC.

For analysis with the TripleTOF 5600+, the second-dimension separation by low pH reverse phase nanoflow LC is performed by having 2mg of each SCX fraction autoinjected from a NanoLC AS-2 Autosampler (ABSciex/Eksigent) into an NanoLC-Ultra-2D Plus HPLC (ABSciex/Eksigent) using a 10 µl injector loop. Trap and elute mode is used to separate each SCX fraction using the microfluidics on a cHiPLC Nanoflex system equipped with a Trap Column (200 µm x 0.5 mm ChromXP 3C18-CL 120Å resin) and a separation column (75 µm x 15 cm ChromXP 3C18-CL 120Å resin). Buffer C was degassed 0.1 percent formic acid in water, and Buffer D was degassed 0.1 percent formic acid in acetonitrile. After loading the trap column with Buffer C 95 percent Buffer D 5 percent, elution and nanospray into the mass spec source was accomplished with the following 185 minute gradient (shorter gradients from 30 to 120 minutes are used for less complex samples), in which most peptides elute between 15 and 110 minutes: Buffer D continuing at 5 percent (0 to 1 minutes following sample injection), 5 percent to 35 percent Buffer D (1 to 155 minutes), 35 percent to 85 percent Buffer D (155 to 157 minutes), then isocratic 85 percent Buffer D (157 to 165 minutes), 85 percent to 5 percent Buffer D (165 to 166 min), then isocratic at the original 5 percent Buffer D start conditions (166 to 185 minutes) to re-equilibrate for the next injection.

Eluate is delivered into the ABSciex 5600 TripleTOF mass spectrometer with a NanoSpray III source and using a 10 mm id nanospray tip (New Objective, Woburn, MA).

Mass Spectrometer settings used vary depending on optimized conditions on each day, but typical values are curtain gas = 25, Gas1 = 4 to 6, Gas2 = 0, an ionspray floating voltage around 2,200, and a rolling collision energy voltage was used for CID fragmentation for MS/MS spectra acquisitions. Each cycle consisted of a TOF-MS spectrum acquisition for 250 ms (mass range 400 to 1,250 Da), followed by information-dependant acquisition of up to 50 MS/MS spectra (50 ms each) of MS peaks above intensity 150 (TOF mass range 65 to 1,600 Da) with a charge state between 2 and 5, taking 2.8 seconds total per full cycle. Once MS/MS fragment spectra were acquired for a particular mass, that mass was dynamically excluded for six seconds. Mass spectrometer recalibration was performed using a known beta-galactosidase digest prior to analysis of each fraction. Full instrument optimization was also performed at least once a week.

The combined MS and MS/MS spectra from all SCX or other first-dimension fractions are analyzed by
Byonic software (proteinmetrics.com); MSFragger (Alexey Nesvizhskii lab); or for studies prior to 2021 using ProteinPilot 5.0 software (Revision 4895) using the Paragon and Pro Group algorithms – Shilov IV, Seymour SL, Patel AA, Loboda A, Tang WH, Keating SP, Hunter CL, Nuwaysir LM, Schaeffer DA. The Paragon Algorithm, a next-generation search engine that uses sequence temperature values and feature probabilities to identify peptides from tandem mass spectra, Mol Cell Proteomics, 2007 Sep;6(9):1638-55. Searches are commonlhy done against complete Uniprot (EMBL) or RefSeq (NCBI) databases concatenated to a reversed sequence Decoy database derived from the same database, plus a list of 536 common lab contaminants.

Protein identifications are accepted if they have an estimated Local False Discovery Rate of less than 5 percent, which is a more stringent criterion than the often-used 1 percent Global False Discovery Rate. The Local False Discovery Rate estimation for each protein is calculated somewhat differently in each algorithm, but is based generally on the accumulations of Decoy database hits. For example, the Protein Pilot analyses used for years used the Proteomics System Performance Evaluation Pipeline (PSPEP) algorithm (Tang, W.H., Shilov, I.V., and Seymour, S.L. A Non-linear Fitting Method for Determining Local False Discovery Rates from Decoy Database Searches, Journal of Proteome Research 2008 Sep;7(9):3661-7. PMID: 18700793).

The resulting search results files with ID Confidence assignments can either be viewed as Excel exports of the search results, or the “raw” analysis results can be examined interactively using the free Byonic viewer (see “Byonic viewer download” section). MSFragger results are all Excel files. Previous ProteinPilot .group files can be viewed with a Windows XP or later computer by licensing and installing a free trial version of the ProteinPilot Software. Once the trial period is over, the software continues to work as a viewer for .group files produced by licensed versions of the software.

For the previously used ProteinPilot analyses, the preset Thorough (iTRAQ or Identification) Search settings are used, and identifications must have a ProteinPilot Unused Score greater than 1.3 (greater than 95 percent confidence interval) in order to be accepted.

In addition, all protein IDs accepted must have a “Local False Discovery Rate” estimation of no higher than 5 percent, as calculated from the slope of the accumulated Decoy database hits by the PSPEP (Proteomics System Performance Evaluation Pipeline ) program by Sean Seymour and colleagues (Tang, W.H., Shilov, I.V., and Seymour, S.L. A Non-linear Fitting Method for Determining Local False Discovery Rates from Decoy Database Searches, Journal of Proteome Research 2008 Sep;7(9):3661-7. PMID: 18700793).

Note that this Local or “Instantaneous” FDR estimate is much more stringent than less than 0.05 or 95 percent confidence scores in Mascot, Sequest, ProteinPilot or the aggregate False Discovery Rate estimations (number of Decoy database identifications/total identifications at any chosen threshold score) commonly used in the literature, and combined with the ProGroup algorithm included in ProteinPilot gives a very conservative and fully MIAPE-compliant list of proteins identified (i.e., Mascot and other lists of “proteins identified at less than 0.05” will produce more numerous “significant” IDs from the same data, but those larger lists are highly likely to contain many more False Positive IDs).

For additional discussion of False Discovery Rates and their estimation, please see Calculating False Discovery Rates.

For iTRAQ and LC-MudPit experiments analyzed with ProteinPilot, the core recommends accepting all protein identifications with a local estimated FDR of 5 percent or lower.

For statistical analysis of quantitative iTRAQ/TMT/Label-free experiments, the core uses Scaffold Q+S software (Proteome Software, Seattle, Wash.) and non-parametric statistics (Mann-Whitney test for quantitative differences between two groups; Kruskal-Wallis test for quantitative differences amongst more than two groups, with the Benjamini-Hochberg multitesting correction).

Most search engine attempts to provide an estimate of how likely a particular protein identification is to be the result of random matching, rather than a “true” ID. In the search engines that the core primarily uses (Byonic, MSFragger and previously ProteinPilot), the statistics on this are calculated by the strength of the match of a peptide ms/ms spectrum match, with the basic principle that the better the “score,” the less likely the identification is to have arisen by a random match.

In “probabilistic” search engines, the correlation between the score and the actual p-score is related to the size of the database searched (the number of possible sequence matches in the search space), which grows larger both as the number of database entries increases – e.g., searching against all mammalian sequences vs. searching against only mouse sequences – and as one increases the number of possible variable modifications considered (if each of 1,000 tryptic fragments in a database can either have or not have particular modification, then there are 2,000 masses in the database rather than 1,000). The larger the database searched, the stronger/higher the score for a match has to be in order to be considered “significant,” i.e., not arising by chance.

At a practical level, this means that no more than two to four variable modifications can be searched as possibilities in Mascot searches – attempting to include more possibilities makes it increasingly difficult to get significant matches for the peptides (modified or unmodified) which are actually in the sample.

Any search algorithm which assigns statistical values to the matches it identifies (peptide MS/MS spectra to proteins) attempts essentially to calculate the probability of a random match, but no algorithm does this perfectly; therefore, most lists of identified proteins contain more false positives than the 5 percent one would expect by using a less than 0.05 or 95 percent confidence cutoff. For this reason, many groups now advocate the use of Decoy Database searches (either reversed or randomized versions of the same Forward/Normal database used for searching), presumably containing no real sequences, with the assumption that the number of identification of Decoy (not real) peptides or proteins at a particular score threshold accurately estimates the number of false identifications from the Forward/Normal database (see for example Elias, J. E., et al., Comparative evaluation of mass spectrometry platforms used in large-scale proteomics investigations, Nature Methods 2 667-675 (2005)), and these Decoy database methods generally provide a more stringent estimation of False Discovery Rates (FDR) than the individual search algorithms’ internal estimates.

There are at least two ways to use the “Decoy” hits to calculate the False Discovery Rate, however. The most commonly used one, called the “Global False Discovery Rate” (or sometimes “Aggregate False Discovery Rate”), uses the formula [100 * (2 * number of Decoy Database Identifications)]/Total Identifications to define the FDR at any cutoff score chosen – for example, if one had a list of 200 identified proteins with Mascot scores ranging from 389 down to 57, and this list contained four identifications from the Decoy database, then the FDR estimate using the Global method for the set of IDs with Mascot scores 57 or higher would be [100 * (2 * 4)]/200 = 4 percent.

There are important limitations to the global/aggregate calculation, however, most importantly that the estimated FDR of 4 percent applies to the whole set of IDs – it is intuitively obvious that the IDs with scores in the 57 to 75 range are much more likely to be false positives than those with scores in the 300 to 357 range. Thus, if the average probability of a false positive in the whole set of 200 proteins is 5 percent and the proteins at the top of the list have much lower than 5 percent probability of being false positives, the entries at the bottom of the list are likely to have much higher than 5 percent probability of being false positives.

The Proteomics System Performance Evaluation Pipeline (PSPEP) algorithm (Tang, W.H., Shilov, I.V., and Seymour, S.L. A Non-linear Fitting Method for Determining Local False Discovery Rates from Decoy Database Searches, Journal of Proteome Research 2008 Sep;7(9):3661-7, PMID: 18700793) is one example of how to improve on the “Global” or “Aggregate” FDR estimation, by using the slope of the accumulated Decoy database hits vs. total identifications to calculate the Local FDR (also called “Instantaneous FDR”) for each individual protein identification. This means that with a PSPEP-analyzed list of identified proteins, where there are 684 proteins with an instantaneous FDR of 5 percent or less, based on the rate of accumulation of Decoy database hits, the last protein on that list has an estimated 5 percent probability of being a false positive, and other proteins higher up that list will have decreasing estimated probabilities of being false positives.

The tables below show a portion of the PSPEP analyses from a few older large iTRAQ datasets (one human, one mouse, one rat dataset) from different groups at the College of Medicine, and one can easily see how much more stringent the Local/Instantaneous FDR is compared to the more frequently used Global/Aggregate FDR estimate, i.e., there are far fewer proteins (approximately 60 to 80 percent) included at a 1 percent or 5 percent instantaneous FDR cutoff than there are at the same 1 percent and 5 percent FDR cutoffs using the aggregate calculation method.

Similar analyses are also performed by PSPEP at the peptide level, with similar relative changes using the Local/Instantaneous vs. Global/Aggregate estimates. Thus, the use of the 5 percent “Local” FDR estimate as a cutoff actually provides a more stringent criterion than using a 1 percent “Global” FDR cutoff.

Protein-Level False Discovery Rate Analysis

Sample 1

Sample 2

Sample 3

Peptide-Level False Discovery Rate Analysis

• Small amounts of buffers (e.g. less than 50mM phosphate, Tris, NaCl) can be tolerated in MALDI-TOF samples; much less in samples for Electrospray ionization (ESI or Nanospray).
• Avoid detergents and glycerol. If detergents must be used, octyl glucoside is the best choice for compatibility with mass spectrometry. Trifuoroethanol (1 to 10 percent) is also a good chaotropic agent (toxic) that may be sufficient as a detergent replacer.
• Sample concentrations should be 1 to 10 µM, but only a few µl are needed; the sensitivity of the techniques is usually such that peaks at 0.1 to 10 pmols/µl in a simple mix can be detected easily, with sensitivities possible down to high attomole amounts of material with special preparation.
• However, peak suppression effects from buffer components or other peaks, as well as differences in the chemical nature of each peptide sequence, mean that not all peaks in a complex mix will necessarily be detected.

Lack of careful sample preparation is the biggest barrier to successful protein identification, particularly contamination of samples with keratins from skin and hair (work in a hood if possible; wear gloves and cover the head); contamination from bacterial/mold growth in old solutions (make fresh solutions up, including gel destaining solutions, etc.); and exposure to detergents (do not reuse laboratory containers for sample preparation).

Mass spectroscopy can detect low femtomole or even attomole amounts of material, and the more contaminant peaks there are, the less likely it is that the existing algorithms can correctly identify the protein of interest from the surrounding noise. More complete lists are available in the Methods notebook in the College of Medicine mass spectrometry facility.

A number of similar protocols with different details have been successfully used at various mass spectrometry facilities, but the procedure below has been used successfully at the College of Medicine to identify unknown proteins from gels, and represents a synthesis of multiple published methods and experimental determinations of optimal conditions. See in particular Systematic Analysis of Peptide Recoveries from In-Gel Digestions for Protein Identifications in Proteome Studies, Kaye D. Speicher, Olivera Kolbas, Sandra Harper and David W. Speicher, but note that College of Medicine investigators have subsequently found that for optimal sample cleanup with ZipTip SCX tips, complete evaporation and 3X resuspension in 200 µl H₂O to get rid of NH₄HCO₃ gives much better overall MS peak spectra, even though the Speicher paper shows that complete evaporation gives lower radioactive peptide yields compared to partial evaporation.

It can be used for either bands from 1D gels (which are almost certain to have multiple proteins contained in it, complicating subsequent analysis, although the mass spectrometry facility can identify even hundreds of proteins from single 1D gel bands), or spots from 2D gels (preferred, since it cuts down – but doesn’t eliminate – the problem of multiple co-electrophoresing proteins, and provides additional information about pI which is significant in subsequent database searching).

Note that investigators must provide material for MALDI-TOF or NanaoSpray analysis from both a sample band/spot and an identically-treated negative control band/spot (no protein, or unrelated protein band/spot) to help eliminate potential artifact peaks.

It is also recommended (but not required) that investigators provide an identically-treated positive control such as BSA or another known protein which stains to approximately the same level as the unknown protein of interest.

• 50 percent acetonitrile (AcN), plus 0.1 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI TOF-TOF (5800) or 0.1 percent formic acid for samples that will be analyzed by ESI (electrospray, nanospray, i.e., on the Sciex 5600 TripleTOF, QTrap or Waters Synapt)
• Appropriate gel stain
• 200 mM ammonium bicarbonate (NH₄HCO₃) pH 8 mixed 1:1 with Acetonitrile (AcN) to give final concentration of 100 mM ammonium bicarbonate in 50 percent can
• 600 µL of 25 mM ammonium bicarbonate, pH 8.0
• 20 to 50 µl of 0.02 µg/µl of Promega Sequencing grade modified trypsin in 10 percent AcN, 40 mM NH₄HCO₃ pH 8
  • 0.1 percent w/v n-octylglucoside (1-O-n-Octyl-beta-D-glucopyranoside), 0.1 percent TFA or formic acid can be added to improve digestion, elution and/or mass spectrometry response, but are not absolutely necessary
• 2 mM TCEP (Tris(2-carboxyethyl)phosphine, Sigma #C4706) in 25 mM ammonium bicarbonate (pH 8.0) – only needed if reducing/alkylating
  Alternate reagent – use 10 mM DTT, 25 mM ammonium bicarbonate (pH 8.0) – only needed if reducing/alkylating
• 100 µL of 20 mM iodoacetamide in 25 mM ammonium bicarbonate (pH 8.0) – only needed if reducing/alkylating

While highly successful results have been obtained on a routine basis for reducing/alkylating samples both before and after electrophoresis, the College of Medicine highly recommends reduction/alkylation before electrophoresis. This is normally done with most 2D gel methods before the first dimension anyway, but even with 1D gels, doing the reduction/alklyation steps before putting samples onto gel will reduce the amount of open-tube sample manipulation of gel spots after the spots are cut out, and therefore reduce the likelihood of keratin contamination of samples.

Even when sending gel bands/spots to the mass spectrometry facility for digestion, investigators should follow the appropriate Summary Procedure through the step of rinsing with acetonitrile and air-drying in a hood. If an investigator is sending the gel slices for digestion, or needs a stopping point, this is a good spot to stop.

Resulting peak MWs from the mass spectrum, and/or masses from ms/ms fragment spectra (usually using ms/ms fragmentation of the top five to 10 peaks at most) can then be used to search for corresponding proteins using various publicly available protein mass spectrometry search engines.

At the College of Medicine, searching is done almost exclusively using the Paragon Algorithm contained in the ProteinPilot software package; however, there are other search engines available for those who wish to perform additional searches of data on their own which can be found through an online search.

This is typically done before the first dimension on 2D gels, for example.

1. Prepare gel approximately one day ahead of time to allow maximal polymerization.
2. Wash clear microfuge tubes with 50 percent acetonitrile (AcN), plus 0.1 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI or 0.1 percent formic acid for samples that will be analyzed by ESI (Electrospray or Nanaospray) instruments such as the ABSciex QTrap 4000 or Waters Synapt HDMS.
3. Reduce sample in reducing buffer (final concentration 25 mM Tris, 0.5 percent SDS, 1 mM tris[2carboxyethyl] phosphine hydrochloride [TCEP-HCl], pH 8.0) for approximately 10 minutes at 37 degrees C. (While the College of Medicine has not yet tried this, Pierce now sells agarose-linked TCEP, Pierce #77712. If used, investigators can remove the TCEP before step 4, which will increase the efficiency of alkylation.)
4. Alkylate sample by adding pH 8.0 iodoacetamide to 5 mM (final concentration). If the eventual gel loading volume will tolerate it, add this in such a way as to reduce the previous TCEP concentration to 0.1 to 0.2 mM.
5. Incubate in the dark for approximately 15 minutes at 37 degrees C, then add 5X gel loading buffer and incubate an additional approximately 15 minutes at 37 degrees C in the dark.
6. Load samples and run gel, then stain (Epicocconone-based stains like LavaPurple or Deep Purple, Sypro Ruby, Krypton, Colloidal Coomassie Blue G250, Ruthenium II or non-formaldehyde Silver Stain using carbohydrazide instead of formaldehyde for reduction/development give overall best results (see About the mechanism of interference of silver staining with peptide mass spectrometry, Sophie Richert, Sylvie Luche, Mireille Chevallet, Alain Van Dorsselaer, Emmanuelle Leize-Wagner and Thierry Rabilloud, Proteomics 2004(4),909–916, for details) followed by Coomassie R250, Sypro Red, Sypro Orange, Silver Stain Plus, negative Zinc or copper staining at lesser sensitivity or increased interference with subsequent MALDI-TOF. A newer comparison suggests that Deep Purple may perform better that SyproRuby, at least with subsequent LC/MS/MS detection (Rapid Commun. Mass Spectrom. 22: 881–886, 2008).
   • Do not use glutaraldehyde or formaldehyde-containing silver stains. Silver stains with low formaldehyde content, e.g., Silver Stain Plus, can be used but they are suboptimal and give weaker MS signals than gel spots coming from the preferred stains above; however, the College of Medicine has identified thousands of proteins from such Silver Stain Plus-stained spots.
7. Cut out bands/spots of interest at the margin of detectable stain, measure/approximate gel slice volume, and put in pre-washed 500 µl microfuge tube.
8. Destain two times with 200 µl of 200 mM ammonium bicarbonate (NH₄HCO₃) pH 8 mixed 1:1 with Acetonitrile (AcN) to give final concentration in 200µl of 100 mM ammonium bicarbonate in 50 percent AcN, for 45 minutes at 37 degrees C.
9. After second destain, remove liquid, then agitate the gel band for approximately 10 minutes in 100 percent acetonitrile. Pour off liquid, and let the gel slice dry out in a covered, clean box in a hood. (Investigators can SpeedVac instead, but are much more likely to introduce keratin or other contaminants from room air or the SpeedVac itself). If sending the gel slices for digestion, or if a stopping point is needed, this is a good spot to stop.
10. Rehydrate gel slice in 20 µl or 1.5 X original gel slice volume (whichever is greater) of 0.02 µg/µl of Promega Sequencing grade modified trypsin in 10 percent AcN, 40 mM NH₄HCO₃ pH 8; 0.1 percent w/v noctylglucoside (1-O-n-Octyl-beta-D-glucopyranoside) for one hour at room temperature or on ice to allow the concentrated trypsin to diffuse into the gel slice.
11. Remove any trypsin-containing liquid that hasn’t been absorbed into the gel slice, then cover the gel slice with 50 µl 10 percent AcN, 40 mM NH₄HCO₃ pH 8; (for optimal results, include 0.1 percent w/v n-octylglucoside (1-O-nOctyl-beta-D-glucopyranoside) in the 50µl as well), and incubate with agitation at least three hours at 48 degress C – longer is OK if more convenient. (Alternately, incubate 16 to 18 hours at 37 degrees C, although note that, for any given time of incubation, trypsin efficiency is higher at 48 degrees C, see e.g. Kinetic characterization of sequencing grade modified trypsin, Finehout EJ, Cantor JR, Lee KH.Proteomics. 2005 Jun;5(9):2319-21.)
12. Remove the supernatant and put in fresh pre-washed 0.5 ml tube (extract 1).</li<>
13. Optional: Add 50µl of 0.1 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI; or 0.1 percent formic acid for samples that will be analyzed by ESI (Electrospray or Nanaospray) instruments such as the ABSciex QTrap 4000 or Waters Synapt HDMS to gel slice; incubate one hour at 37 degrees C. Remove supernatant (extract 2) and combine with extract 1 in 0.5 ml tube. (This may extract approximately 15 percent more digested material from the sample).
14. SpeedVac to dryness. Resuspend in approximately 200 µl H₂O, then SpeedVac to dryness again. Repeat this three times, and Speedvac the final resuspension to approximately 10 µl; a good practice is to mark 10 µl on the 0.5 ml microfuge tube used to estimate 10 µl. This procedure removes the NH₄HCO₃ (and AcN), which can interfere with subsequent binding to the strong cation exchange ZipTip SCX tips used for sample cleanup. Note that this is a change from the previously recommended “Dry only to 25 percent original volume” instructions from the Speicher article cited above. Note also that using SpeedVac after the trypsin digestion like this may still introduce keratin and other contaminants into the sample, but since these protein contaminants will not be digested, they are unlikely to interfere with subsequent mass spectrometry analyses of tryptic peptides.
15. ZipTip concentrate and clean digest. Normally, this is done in the facility, so investigators can just provide the sample from Step 12 above to facility staff.

Advantages: Pre-alkylation prevents (slow) alkylation by exposure to acrylamide and cross-linkers, thus preventing a mixed mass population for specific fragments (some modified by acrylamide +71, some by iodoacetamide +57).

Disadvantage: Since the protein sample is not fixed in a gel slice, the TCEP cannot be removed before adding iodoacetamide, and TCEP (or DTT) over approximately 0.1 mM can inhibit subsequent alkylation by iodoacetamide. (The agarose-linked TCEP, Pierce #77712, would theoretically overcome this.)

Notes: Other reducing agents (1 mM tributylphosphine, 10 mM DTT) have been used successfully elsewhere, and other alkylating agents can be substituted if +57 addition may confound subsequent analysis of other adducts (e.g., iodoacetic acid, vinylpyridine, or even more extensive exposure to unpolymerized acylamide.)

Successful mass spectrometry determinations have also been done without any alkylation, and without adding AcN or n-ocylglucoside to the trypsin digest mix; the above simply represents a “best” protocol that is theoretically optimized and has worked well for the College of Medicine.

The reduction/alkylation steps below can be skipped, as they are not always necessary to perform in order to identify a protein from a gel spot; however, the reduction/alkylation will tend to increase the efficiency of trypsin digestion somewhat and will sometimes contribute to the identification of additional peptides from the protein spot.

1. Prepare gel approximately one day ahead of time to allow maximal polymerization.
2. Wash clear microfuge tubes with 50 percent acetonitrile (AcN), plus 0.1 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI or 0.1 percent formic acid for samples that will be analyzed by ESI (Electrospray or Nanaospray) instruments such as the ABSciex QTrap 4000 or Waters Synapt HDMS.
3. Load samples and run gel, then stain (Epicocconone-based stains like LavaPurple or Deep Purple, Sypro Ruby, Krypton, Colloidal Coomassie Blue G250, Ruthenium II or non-formaldehyde Silver Stain using carbohydrazide instead of formaldehyde for reduction/development give overall best results (see About the mechanism of interference of silver staining with peptide mass spectrometry, Sophie Richert, Sylvie Luche, Mireille Chevallet, Alain Van Dorsselaer, Emmanuelle Leize-Wagner and Thierry Rabilloud, Proteomics 2004(4),909–916, for details) followed by Coomassie R250, Sypro Red, Sypro Orange, Silver Stain Plus, negative Zinc or copper staining at lesser sensitivity or increased interference with subsequent MALDI-TOF. A newer comparison suggests that Deep Purple may perform better that SyproRuby, at least with subsequent LC/MS/MS detection (Rapid Commun. Mass Spectrom. 22: 881–886, 2008).
   • Do not use glutaraldehyde or formaldehyde-containing silver stains. Silver stains with low formaldehyde content, e.g., Silver Stain Plus, can be used but they are suboptimal and give weaker MS signals than gel spots coming from the preferred stains above; however, the College of Medicine has identified thousands of proteins from such Silver Stain Plus-stained spots.
4. Cut out bands/spots of interest at the margin of detectable stain, measure/approximate gel slice volume, and put in pre-washed 500 µl microfuge tube.
5. Destain two times with 200 µl of 200 mM ammonium bicarbonate (NH₄HCO₃) pH 8 mixed 1:1 with Acetonitrile (AcN) to give final concentration in 200µl of 100 mM ammonium bicarbonate in 50 percent AcN, for 45 minutes at 37 degrees C.
   • After second destain, remove liquid, then agitate the gel band for approximately 10 minutes in 100 percent acetonitrile. Pour off liquid, and let the gel slice dry out in a covered, clean box in a hood. (Investigators can SpeedVac instead, but are much more likely to introduce keratin or other contaminants from room air or the SpeedVac itself). If sending the gel slices for digestion, or if a stopping point is needed, this is a good spot to stop.
6. Reduce: Add 100 μL of 2 mM TCEP (Tris(2-­‐carboxyethyl)phosphine, Sigma #C4706) in 25 mM ammonium bicarbonate (pH 8.0) to the dried gel and incubate 15 minutes at 37 degrees C with agitation; remove supernatant. Alternate procedure: use 10 mM DTT, 25 mM ammonium bicarbonate (pH 8.0) for 15 minutes at 37 degrees C.
7. Alkylate: Add 100 μL of 20 mM iodoacetamide in 25 mM ammonium bicarbonate (pH 8.0) and incubate for 30 minutes at 37 degrees C in the dark. Discard supernatant; wash gel band three times with 200 μL of 25 mM ammonium bicarbonate for 15 minutes, each with agitation, then dry gel slice completely in SpeedVac.
8. Rehydrate gel slice in 20 µl or 1.5 X original gel slice volume (whichever is greater) of 0.02 µg/µl of Promega Sequencing grade modified trypsin in 10 percent AcN, 40 mM NH₄HCO₃ pH 8; 0.1 percent w/v noctylglucoside (1-O-n-Octyl-beta-D-glucopyranoside) for one hour at room temperature or on ice to allow the concentrated trypsin to diffuse into the gel slice.
9. Remove any trypsin-containing liquid that hasn’t been absorbed into the gel slice, then cover the gel slice with 50 µl 10 percent AcN, 40 mM NH₄HCO₃ pH 8; (for optimal results, include 0.1 percent w/v n-octylglucoside (1-O-nOctyl-beta-D-glucopyranoside) in the 50µl as well), and incubate with agitation at least three hours at 48 degress C – longer is OK if more convenient. (Alternately, incubate 16 to 18 hours at 37 degrees C, although note that, for any given time of incubation, trypsin efficiency is higher at 48 degrees C, see e.g. Kinetic characterization of sequencing grade modified trypsin, Finehout EJ, Cantor JR, Lee KH.Proteomics. 2005 Jun;5(9):2319-21.)
10. Remove the supernatant and put in fresh pre-washed 0.5 ml tube (extract 1).</li<>
11. Optional: Add 50µl of 0.1 percent trifluoroacetic acid (TFA) for samples that will be analyzed by MALDI; or 0.1 percent formic acid for samples that will be analyzed by ESI (Electrospray or Nanaospray) instruments such as the ABSciex QTrap 4000 or Waters Synapt HDMS to gel slice; incubate one hour at 37 degrees C. Remove supernatant (extract 2) and combine with extract 1 in 0.5 ml tube. (This may extract approximately 15 percent more digested material from the sample).
12. SpeedVac to dryness. Resuspend in approximately 200 µl H₂O, then SpeedVac to dryness again. Repeat this three times, and Speedvac the final resuspension to approximately 10 µl; a good practice is to mark 10 µl on the 0.5 ml microfuge tube used to estimate 10 µl. This procedure removes the NH₄HCO₃ (and AcN), which can interfere with subsequent binding to the strong cation exchange ZipTip SCX tips used for sample cleanup. Note that this is a change from the previously recommended “Dry only to 25 percent original volume” instructions from the Speicher article cited above. Note also that using SpeedVac after the trypsin digestion like this may still introduce keratin and other contaminants into the sample, but since these protein contaminants will not be digested, they are unlikely to interfere with subsequent mass spectrometry analyses of tryptic peptides.
13. ZipTip concentrate and clean digest. Normally, this is done in the facility, so investigators can just provide the sample from Step 12 above to facility staff.

Advantages: Since the protein sample is fixed in the gel slice, the TCEP can be removed before adding iodoacetamide, thus insuring optimal reduction and subsequent alkylation.

Disadvantage: Exposure to unpolymerized acrylamide/bis-acrylamide (some which always remains even in polymerized gels) causes slow alkylation of cysteines (+71 adduct), which thus produces a mixed population of Cys- and modified-Cys-containing fragments and decreases the power of subsequent mass spectrometry analysis (both decreased peak sizes and a more complex mix to deconvolute.)

Notes: Other reducing agents (1 mM tributylphosphine, 10 mM DTT) have been used successfully elsewhere, and other alkylating agents can be substituted if +57 addition may confound subsequent analysis of other adducts (e.g., iodoacetic acid, vinylpyridine, or even more extensive exposure to unpolymerized acylamide.)

Successful mass spectrometry determinations have also been done without any alkylation, and without adding AcN or n-ocylglucoside to the trypsin digest mix; the above simply represents a “best” protocol that is theoretically optimized and has worked well for the College of Medicine.

There is no charge for initial consultations and feasibility discussions.

For details on fees and options, contact Dr. Anne Stanley at astanley@psu.edu.