Photochemical Labeling of Proteins

An aqueous solution of lysozyme (160 μM) in 25 mM ammonium acetate or 20 mM Tris/150 mM NaCl was mixed with an equal volume of a 20 mM solution of the diazirine labelling agent (1 or 2) in 20 mM Tris/150 mM NaCl, to give a final label concentration of 10 mM. The mixture was left equilibrating for 5 min at room temperature. Aliquots (6 μl) of this solution were placed in ‘11 mm PP vial crimp/snap 250 ul' sample vials (Thermo Fisher Scientific). To avoid enzymatic cleavage of the pentasaccharide substrate, NAG₅ was added to the mixture, to a final concentration of 80 μM, just before snap-freezing the samples in liquid nitrogen (77 K). Freezing was found to be important for efficient and reproducible labelling. For lysozyme footprinting in the absence of substrate, the NAG₅ solution was substituted by an equal volume of buffer. The labelling reaction was initiated by irradiation (16 s) of the mixture using a Spectra Physics Explorer 349 laser (an actively Q-switched Nd:YLF laser operating at 349 nm, with a repetition frequency of 1,000 Hz, and a pulse energy of 125 μJ, Newport, Didcot, UK). The laser beam was directed into the open top of the sample vial using a small 45° mirror.
Photochemical labelling of C335A USP5. A solution of C335A USP5, Lys48 di-ubiquitin (30 μM each, 10 mM ammonium acetate, or 20 mM Tris/150 mM NaCl) and diazirine 2 (in 20 mM Tris/150 mM NaCl, to a final concentration of 10 mM) was snap-frozen and irradiated at 349 nm as described for lysozyme.
Labelling efficiency. Solutions of calmodulin, myoglobin, cytochrome c, ubiquitin or melittin (70 μM in 20 mM Tris/150 mM NaCl) and photoleucine 1 (100 mM in 20 mM Tris/150 mM NaCl) or diazrine 2 (10 mM in 20 mM Tris/150 NaCl) were snap-frozen and irradiated at 349 nm for 4 s in the case of diazirine 2 (except calmodulin, which required only 1 s), and 15 s in the case of photoleucine 1. Samples were diluted to 5 μM with mobile phase A (H₂O/acetonitrile 95/5+0.1% formic acid) and analysed directly by LC-MS.
Protein digestion. Following separation by SDS–polyacrylamide gel electrophoresis (SDS–PAGE; 15% acrylamide), protein bands were excised, reduced (DTT, 10 mM in 100 mM ammonium bicarbonate), alkylated (iodoacetamide, 55 mM in 10 mM ammonium bicarbonate) and digested at 37 °C with either trypsin (1:15 protease/protein ratio in 10 mM ammonium bicarbonate) or AspN (1:70 protease/protein ratio in 5 mM Tris-HCl) for 16 h (ref. 39 (link)). Formic acid (0.5 μl) was added to the supernatant (50 μl) to inactivate the proteases. The resulting solutions were analysed by LC/MS without further dilution.
LC-MS analysis. Measurement of intact mono-ubiquitin, lysozyme or protein digests was carried out on a Dionex Ultimate 3000 Nano LC (Dionex, Camberley, UK) system using a commercially available C4 trapping column (Dionex) and a custom packed WP C4 column (5μm, 100 Å, Phenomenex, Macclesfield, UK) with a Picofrit tip (75 μM i.d. × 150 mm, new objective, supplied by Aquilant, Basingstoke, UK). The mobile phases A and B consisted of 95/5 water/acetonitrile (v/v) and 5/95 water/acetonitrile (v/v), respectively, and both contained 0.1% formic acid. The samples (1 μl) were injected in load-trapping mode. Peptides were eluted using a 30 min linear gradient of mobile phase B from 0–70% at a flow rate of 0.3 μl min⁻¹ followed by column equilibration. The high-performance liquid chromatography system was coupled to a Thermo Scientific (Hemel Hempstead, UK) LTQ FT Ultra mass spectrometer equipped with a commercial nanoelectrospray ionization source. A 1.7 kV voltage was applied to a coated PicoTip emitter (New Objective). The capillary temperature was set at 275 °C, with inner capillary voltage value set on 37 V and tube lens value of 145 V. Spectra were acquired in positive ion mode over a 400–2,000 m/z range at a nominal resolution of 100,000 (at m/z=400). The instrument was controlled by Xcalibur software (Thermo Fisher). Identification of USP5 peptides was conducted using an automated data-dependent acquisition mode, followed by manual examination of the raw data. For data-dependent acquisition the four most intense ions for each scan were isolated within a window of 8 Th and subjected to collision induced dissociation (CID) using a nominal energy of 35.0. Signals with +1 charge state were rejected. The data were searched against a custom database including the C335A USP5 sequence using Bioworks software (Thermo Fisher Scientific).
The proportion of labelling at the peptide level was determined by integrating the signals for each labelled and unlabelled peptide ion. Partial, or even complete chromatographic separation was seen for some labelling isomers (based on the position of the label along the peptide chain), and spectra were combined across the entire peak or set of peaks to ensure that the quantities of all labelled forms were included in the subsequent calculations. CID experiments for residue-level labelling identification/quantitation were performed at a nominal energy of 15.0. Each manually selected labelled precursor ion was isolated within a window of 10.0 Th (although the relatively wide isolation window lead in one case to co-isolation of another peptide, this did not impact upon the precision of the quantitation, and the associated improvement in total signal was found to be beneficial) and the activation time was set at 30 ms with an activation Q-value of 0.250. The scans for each labelled peptide were combined to give an average spectrum containing labelled and unlabelled fragments.
Data analysis: the fractional modification per peptide, and the average number of labels per residue were determined using the approach described by Jumper et al.24 (link) (see Supplementary Methods) and corroborated using quantification by parallel reaction monitoring. Protein structures were displayed using PYMOL.
Lysozyme activity assay. Cell suspensions of Micrococcus lysodeikticus ATCC (Sigma-Aldrich, Poole, UK, 1 ml, 0.01% in 60 mM potassium phosphate buffer, pH 6.2) were equilibrated in a Multiskan Go microplate spectrophotometer (Thermo Scientific) at 25 °C for 5 min with or without the presence of 10 mM diazirine 2. A solution of lysozyme (50 μl, 0.1 mg ml⁻¹) was added to each suspension, and the decrease in absorbance at 450 nm was measured every 60 s for a total of 10 min. As a control, a cell suspension without the addition of either lysozyme or 2 was similarly monitored.
USP5 activity assay. The deubiquitination assay was performed under conditions (1:4 enzyme: substrate molar ratio) previously used by Komander et al.40 (link) to uncover poly-Ub linkage selectivity of different deubiquitinating enzymes. Briefly, USP5 (Boston Biochem, Cambridge, MA) was diluted to 0.2 mg ml⁻¹ in 10 × deubiquitinating buffer (500 mM Tris (pH 7.5), 1500, mM NaCl, 100 mM DTT), to yield a 1 × deubiquitinating buffer and pre-incubated at room temperature for 10 min. 10 μl of the pre-incubated USP5, in the presence or absence of 3 μl of aryldiazirine 2 (100 mM), was then mixed with 3 μg of K63-linked tetraubiquitin (Boston Biochem, Cambridge, MA), 3 μl of 10 × deubiquitinating buffer and made up to 30 μl with water. 6 μl aliquots were removed at the indicated time points and quenched via the addition of 9 μl of SDS–PAGE gel application buffer (0.15 M Tris, 8 M Urea, 2.5% (w/v) SDS, 20% v/v glycerol, 10% (w/v) 2-mercaptoethanol, 3% (w/v) DTT, 0.1% (w/v) bromophenol blue, pH 6.8 (HCl)). 50% of each sample was resolved on a 5–20% SDS–PAGE, western blotted and visualized with anti-ubiquitin (VU-1, Life sensors, Malvern, PA) for the presence of poly-Ub disassembly. Densitometric analysis of the indicated tetraubiquitin band (Supplementary Fig. 17A) was completed using ImageJ software41 (link), to quantify the disappearance of tetraubiquitin over the time course in the presence and absence of aryldiazirine 2.