DIA analysis of serum samples

Sample concentration was adjusted to 0.5 µg/ml in 3 % ACN, 0.1 % FA and retention time normalization peptides (iRT, Biognosys) were added at a 1:20 ratio. Peptide samples were analyzed in random order in data independent acquisition (DIA) mode on a hybrid quadrupole-orbitrap mass spectrometer (Q Exactive HF, Thermo Scientific) operated in line with a nano UHPLC system (Easy-nLC 1000, Thermo Scientific). For each sample 3 μl of purified peptide mixture was loaded onto a custom-made fused-silica column (150 mm x 75 um ID) filled with porous reverse-phase chromatography resin (Reprosil-Pur C18 AQ, 1.9 μm spherical silica particles, 120 Å pore diameter, M/N: r119.aq, Dr. Maisch GmbH, Ammerbuch, Germany). Bound peptides were eluted from the analytical column by running a linear gradient from 5 % to 35 % B in 120 min (Solvent A: 0.1% FA in water, solvent B: 0.1% FA in acetonitrile) at a linear flow rate of 300 nl/min. Eluted peptides were injected into the MS utilizing a nano ESI source (10 μm fused-silica spray emitter and Digital PicoView 565, O/N: DPV-550-565, New Objective, Woburn, MA). Mass spectra were recorded at a resolution of 60,000 (at 200 m/z) in profile mode using positive polarity. MS1 scans covering 350-1800 m/z were acquired using an AGC target of 3e6 and a maximum IT of 200 ms, and a centroid spectrum data type.

The acquired raw files were deposited at PRIDE under ... 

PRM assay development and characterization

A blank sample and 6 dilutions of peptides were included and measured in duplicates. Detection window consisted of five min before and after expected retention time for each peptide; these are provided in Table X. Response curves made of a serial dilution of heavy peptides in a digested serum matrix from a healthy beagle was used to determine the spiked-in concentrations. The analysis was conducted in Skyline (v. 4.2 [24]), using the top 5 predicted y fragment ions were extracted from the MS2 spectra to generate elution profiles, excluding y1 and y2 because of their small size. One peptide was monitored by extracted ion chromatogram (XIC) of 4 products due to the absence of a 5^th.

Validation of DIA data by PRM

Bound peptides were eluted from the analytical column by running a linear gradient from 5 % to 35 % B in 45 min (Solvent A: 0.1% FA in water, solvent B: 0.1% FA in acetonitrile) at a linear flow rate of 300 nl/min. MS1 scans were recorded at a resolution of 120,000 at a scan range of 350-1800 m/z in profile mode using positive polarity. We used AGC target of 3e6 and a maximum IT of 55 ms, and a centroid spectrum data type. Each MS1 scan was followed by 15 multiplexed fragment ion scans. Parallel reaction monitoring (PRM) scans were recorded at a resolution of 120,000 in profile mode using positive polarity. PRM scans were acquired as centroid spectrum data using an AGC target of 1e5; isolated precursors were fragmented using HCD at a NCE of 28. Further settings included maximum IT of 247 ms, isolation windows of 1.4 m/z. Fixed first mass was set at 100 m/z. The maximum injection time per PRM scan was selected by the instrument software to assure acquisition at a maximum scan speed of 3 Hz. Accordingly, the instrument cycle time was around 3.7 s at a chromatographic peak width of 15 s. The inclusion list encompassed 9 endogenous peptides, 9 heavy peptides (C-terminal R or K, carbamidomethyl C, z = .2 and .3; PEPotec, Thermo Scientific) listed in supplemental Table X), as well as the retention time normalization peptides (iRT).