Protein expression patterns adapt to various cues to meet the needs of an organism. The dynamicity of an organism’s proteome can therefore reveal information about an organism’s health. Proteome databases contain limited information regarding organisms outside of medicinal biology. The Uniprot human and mouse proteomes are extensively reviewed and ~50% of both proteomes include tissue specificity, while >99% of the rainbow trout proteome lacks tissue specificity. This study aimed to expand knowledge on the rainbow trout proteome with a focus on understanding the origin of blood plasma proteins. Blood, brain, heart, liver, kidney, and gills were collected from adult rainbow trout, plasma and tissue proteins were analyzed using liquid chromatography tandem mass spectrometry. Over 10,000 proteins were identified across all groups. Our data indicated that the majority of the plasma proteome is shared among multiple tissue types, though 4-7% of the plasma proteome is uniquely originated from each tissue (gill > heart > liver > kidney > brain).

Nineteen mature healthy rainbow trout were housed and acclimated for several weeks prior to analysis. Blood was sampled using caudal vein puncture and plasma was separated via centrifugation at 2000 RP M for 20 minutes at 4 degrees Celsius. Plasma was stored at -80 degrees Celsius until further analysis. Brain, liver, kidney, heart, and gills were collected from all nineteen fish and stored at -80 degrees Celsius until further analysis. Plasma and tissue samples were reduced using Tris(2-carboxyethyl) phosphine, alkylated using iodoacetamide, and digested using formic acid and heat. Non-targeted data dependent acquisition using LC-MS was used to detect and identify peptides. Samples were injected twice per sample. Spectrum Mill was used to analyze the spectral files to identify proteins: proteins were identified by search against the Uniprot Reference Proteome for rainbow trout (Proteome ID#UP000193380, downloaded January 2020). Proteins were manually accepted when the following criteria were met, (1) peptide score (quality match between the observed spectrum and theoretical spectrum) greater than 6 in at least one peptide, and (2) a %SPI (percent of the spectral intensity accounted for by the theoretical fragments) greater than 70%, both of which are recommended for validating results using an Agilent Q-TOF mass spectrometer. Once peptides were identified and sequenced using Spectrum Mill at the MS/MS level, a peptide library was built and used to quantify precursors at the MS1 level that did not trigger MS/MS using the data-dependent acquisition (DDA) workflow in Skyline 20.2 (MacCoss Lab Software), with a cut-off score of 0.9, 5-minute retention time window, and 5 missed cleavages with transition settings for TOF (Pino et al., 2020). The list of Uniprot protein identifiers and their respective spectral intensities were exported from Skyline using the MPP APR report tool, and then were uploaded into Mass Profiler Professional (MPP, Agilent technologies) to then combined exported as a .csv file. The list of identified rainbow trout protein accession numbers was blasted against the human Uniprot Reference Proteome (Proteome ID#UP000005640) to obtain human protein orthologs gene symbols for data consolidation (to remove duplicate proteins) and subsequent gene ontology classification. Metaboanalyst was used to analyze and normalize the data.

Plasma and tissue samples were removed from the -80ºC and placed on ice to gradually thaw. For tissues, 100 mg portions of each tissue type taken from the same lobe/section of each organ were measured and mixed with 500 mL of 100 mM triethylammonium bicarbonate (TEAB) and two stainless steel 5 mm diameter beads, then homogenized on a ball mill for 1 minute at 20 Hz (QIAGEN, TissueLyser II). Samples were centrifuged at 14,000 xg for 15 minutes at 4ºC. Tissue supernatants containing soluble proteins were collected and a portion was used to estimate protein concentration using a Qubit 4 Fluorometer. Approximately 1 mg of total protein concentration – ranging between 40-50 µL homogenate depending on tissue type – was transferred into low retention tubes for further sample preparation; 100 mM TEAB was added to homogenates with <50 µL to bring the final volume to 50 µL. Plasma was vortexed to obtain a uniform mixture, and 15 µL of each sample was transferred into low retention microcentrifuge tubes. Next, 35 µL of 100 mM TEAB buffer was added to each tube of plasma and then vortexed. Proteins in plasma and homogenates were then reduced using 2.65 µL of 100 mM tris(2-carboxyethyl) phosphine (TCEP) in 100 mM TEAB, vortexed, and incubated at room temperature for 45 minutes. Then, 2.8 µL of alkylating solution (200 mM iodoacetamide (IAA) in 100 mM TEAB was added, vortexed, and incubated in the dark at room temperature for 45 minutes. Following incubation, 50 µL of chemical digestion solution (20% formic acid v/v) was added and vortexed gently. Each tube was lid-locked and incubated at 115ºC for 30 minutes using a VWR 96 heating block. Next, samples were evaporated to ~20 µL using a centrifugal evaporator (Genevac miVac Quattro Concentrator) for ~40 minutes. Samples were re-suspended in 20 µL high-performance liquid chromatography (HPLC) buffer (95% H2O, 5% acetonitrile, 0.1% formic acid), and gently vortexed until dried pellets were completely reconstituted. Reconstituted samples were centrifuged at 10,000 xg for 10 minutes to precipitate any debris. Next, 20 µL of the sample supernatant and 2 µL of internal peptide standard (Sigma Aldrich, HPLC peptide standard mixture) were added to 2 mL screw thread HPLC vials containing 250 µL pp bottom spring inserts. Peptide solutions were stored at 4ºC until instrumental analysis. Reverse phase separation of each sample was completed using an Agilent 1260 Infinity Binary LC and Zorbax, 300SB-C18, 1.0 × 50 mm 3.5 μm column (Agilent Technologies Canada Inc., Mississauga, ON). Specifically, 2 µL of peptide solution from each sample was injected into the instrument and separated using reverse phase chromatography. Liquid chromatography was coupled to The Agilent 6545 Accurate-Mass Quadrupole Time-of-Flight (Q-TOF) to detect and identify peptides. Refer to Appendix A for detailed instrumental methods. Each run included a solvent blank, a BSA digest standard (Agilent Technologies Canada Inc., Mississauga, ON), and a peptide standard (Sigma Aldrich, HPLC peptide standard mixture, H2016), which were injected between every 10 samples to monitor baseline, carry-over, drift, sensitivity, and overall instrumental variation during the runtime. As well, we used the HPLC peptide as an internal standard, spiked into each sample, and used the average peak areas of two peptides to normalize across each block of samples within the run. Tissue and plasma samples were injected twice per each individual fish, and averaged together before statistical analyses.