Large-scale identification of N-glycosylated peptides using lectin-mediated affinity capture, glycosylation site–specific stable isotope tagging, and LC/MS.

 Dissolve the protein source materials (whole organisms, tissues, cells, organelles or other biological materials) in extraction buffer (we usually use 5–10 mL extraction buffer for 1 g material) using the appropriate equipment (e.g., sonicator, Teflon homogenizer, Waring blender, French press, Polytron). Remove insoluble materials, if any, by centrifugation at 10,000g for 30 min at room temperature (20°C).

 Quantify protein concentration by Bradford assay using human serum albumin as a standard.

 Add an aliquot of 100 mg mL^-1 DTT to the sample solution at DTT:protein = 1:1 (w/w), and incubate with gentle shaking for 2 h at room temperature to reduce disulfide bonds under nitrogen gas atmosphere. For a small-volume sample, perform the reduction in a sealed tube under N₂ gas.

 For S-carbamoylmethylation of cysteine residues, add iodoacetamide to the protein solution (2.5:1 w/w protein), mix and leave the mixture in the dark for 2 h at room temperature.

 Dialyze the reaction mixture at 4°C against 100 volumes of 10 mM HEPES-NaOH buffer, pH 7.5, to remove salts and excess reagents, changing the buffer three times for every 2 h and then leave overnight. For a small-volume sample, precipitate the protein by acetone, trichloroacetic acid, or methanol/chloroform precipitation method.

 Transfer the dialyzed mixture to a plastic conical tube. Add trypsin (1:50 w/w protein) or Lys-C protease (1:100 w/w protein) to the solution and digest the proteins overnight at 37°C (see Comment).

 Add an aliquot of 0.5 M PMSF in EtOH into the mixture at a final concentration of 1 mM to stop digestion and to protect the lectin column. Remove precipitates, if any, by centrifugation.

 Load the peptide mixture onto the lectin column equilibrated with a buffer appropriate for each lectin column (listed in Table 1). Take care not to overload the column — a typical lectin column with 4.6 mM ID × 150 mM can separate - 400 μg glycopeptides. Recover the flow-through fraction.

 Wash the column with the binding buffer until the absorption of the effluent at 280 nm is <0.1 (about 10- column -volumes of buffer is necessary).

 Elute glycopeptides with an elution buffer containing a mono/oligosaccharide appropriate for the lectin column (e.g., 0.2 M methyl α-D-mannopyranoside for a Con A column, see Table 1).

 Re-equilibrate the lectin column with five -column- volumes of binding buffer.

 Reload the flow-through fraction onto the lectin column and recollect the glycopeptide fraction under the same conditions as above. Repeat the chromatography three to five times until the absorption of the glycopeptide peak at 280 nm is <0.1. Combine the glycopeptide fractions.

 Add 3 volumes of MeCN containing 0.1% TFA into the glycopeptide solution eluted from the lectin column (or protease digest). Remove precipitates, if any, by centrifugation (see Comment).

 Immediately load the sample solution onto an Amide-80 column, which has been equilibrated with HILIC solvent B, and wash the column with the HILIC solvent B until the absorption of the effluent at 220 nm is <0.2.

 Elute glycopeptides with HILIC solvent A, monitoring the eluate at 220 nm. Samples can be stored at -80°C for several days.

 Transfer -200 μL of the HILIC eluate into a 500-μL microtube and evaporate to dryness using a centrifugal vacuum concentrator. Add another 200-μL aliquot of the HILIC eluate into the same microtube and evaporate. Repeat the process to dry up the total eluate (usually 1–2 mL) (see Comment).

 Dissolve glycopeptides in PNGase buffer (10–20 μg/50–100 μL) prepared with H₂¹⁸O.

 Add PNGase, dissolved in H₂¹⁸O at 1 mU uL^–1, at a final concentration of 1 mU 10 μg–1 glycopeptides and incubate the mixture at 37°C overnight in a sealed microtube.

 Stop the reaction by acidifying the solution to -pH 2 by the addition of 1% formic acid (-5 μL per 100 μL reaction volume).

 Analyze the ¹⁸O-labeled peptide mixture using an LC system connected to an electrospray-ionization tandem mass spectrometer. See Comment for details.

 Search the MS/MS spectra using a nonredundant protein database, such as Refseq and SwissProtKB, using Mascot algorithms for peptide sequence identification with the following parameters: fixed modification of carbamoylmethylation on Cys, variable modifications of deamidation on Asn (plus an incorporation of ¹⁸O = +3 Da) and pyroglutamination on N-terminal Gln.

 Identify the formerly N-glycosylated peptides. Export the result to csv file and process the data with Microsoft Excel. We first selected the peptides with rank 1 and an expectation value <0.05 as “identified peptides”.

 Inspect the data set to determine whether the ¹⁸O-labeled peptide has one or more consensus sequences for N-linked glycosylation (i.e., Asn-Xaa-Ser/Thr, where Xaa is any amino acid except Pro). If the candidate peptide has an Asn-Lys or Asn-Arg sequence at the C terminus, examine if the residue following the Lys/Arg in the protein sequence might be Ser or Thr, which meets the consensus tripeptide sequence.