To better characterize the signalling pathways activated during human liver transplantation, we aimed to identify phosphotyrosine (PY) containing proteins whose phosphorylation was regulated upon I/R. To this end, we used liver biopsies collected as described in Figure 1A. Protein extracts (I0, I10, I60 corresponding to 0, 10 and 60 min ischemia and R0, R10, R60 corresponding to 0, 10 and 60 min reperfusion, respectively) were prepared as described in the Methods section and processed for immunoblot analyses using anti-PY antibodies. This allowed us to obtain a global tyrosine phosphorylation pattern during I/R (Figure 1B). Although, no significant change was observed for the large majority of the proteins protein detected, a major protein with a molecular weight of approximately 50 kDa was subjected to a dramatic change in its tyrosine phosphorylation status during both phases of the transplantation. An important tyrosine phosphorylation increase during the ischemic phase and a consecutive decrease during reperfusion were observed. In order to identify PY-containing proteins during the two phases of the transplantation, proteins extracts from biopsies collected at 60 min ischemia (I60) and 60 min reperfusion (R60) were chromatographed on anti-PY antibodies coupled to agarose beads as schematically represented in Figure 2A. Proteins were then eluted, resolved by 1D SDS-PAGE and stained using G250 Coomassie Blue (Figure 2B). Noteworthy, a major Coomassie Blue stained band in the I60 tyrosine phosphorylated fraction was no longer detectable in the R60 fraction, indicating that this ~50 kDa band may contain PY proteins specific to the ischemic phase. This protein may most likely correspond to the major protein detected in Figure 1B. Each band visualized on the gel was then excised, trypsin digested and the resulting peptides analyzed using Liquid Chromatography – ElectroSpray Ionization – quadrupole Time of Flight – tandem Mass Spectrometry (LC-ESI-QToF) MS/MS as described in the Methods section. The proteins identified are listed in Tables 1 (I60) and 2 (R60) and represented after classification into functional families according to their GO (Gene Ontology) annotation (Figure 2C). We identified only 7 proteins in the 60 min ischemia fraction (Table 1). Interestingly, the ~50 kDa band disappearing between I60 and R60 was reproducibly identified as the SH₂/SH₃ containing adaptor protein Nck-1 (Table 1). Besides Nck-1, which represented the only signalling component of this fraction, the 6 other proteins identified were composed of abundant proteins involved in liver metabolic functions (energy metabolism and detoxification) and of blood proteins involved in transport (Table 1 and Figure 1C, left panel). These 6 proteins were also found in the 60 min reperfusion fraction (Tables 1 and 2), suggesting that they may either represent contaminants or be constitutively tyrosine phosphorylated. Interestingly, a larger data set of 37 proteins was identified for the 60 min reperfusion time point (Table 2 and Figure 1C, right panel). In addition to hepatocytes to proteins involved in liver metabolism (energy metabolism: 16% and detoxification: 8%) and blood proteins (blood coagulation: 5%, transport: 11%), a large proportion (33%) of these proteins were involved in protein synthesis (mainly ribosomal proteins). We also identified structural components of the cytoskeleton (16%) and proteins involved in nucleobase metabolism (8%), as well as 1 unknown protein. Although the computational prediction of serine, threonine and tyrosine phosphorylation using the NetPhos software indicated potential tyrosine phosphorylation sites for the large majority of the proteins identified (Table 1 and 2), only 3 of these proteins had been previously reported as tyrosine phosphorylated: the alpha chain of tubulin, vimentin and the 60S ribosomal protein, L8 (Table 2).

The SH₂/SH₃ containing adaptor Nck-1 has been linked to the regulation of multiple intracellular signal transduction events 10 with significant contribution to actin cytoskeleton remodelling. Consequently, Nck-1 could represent a critical target during I/R as actin cytoskeleton was previously reported to be subjected to major structural changes in livers subjected to I/R 11. Four peptides were specifically attributed to Nck-1 (underlined on Nck-1 sequence, Figure 3A) in the 60 min ischemia fraction. The MS/MS spectrum corresponding to a peptide specific of Nck-1 is presented for reference (Figure 3B). Remarkably, besides the fact that the band corresponding to Nck-1 could not be detected by Coomassie Blue staining of the anti-PY purified fractions after 60 min of reperfusion (Figure 2B), Nck-1 was no longer identified in the anti-PY binding fraction after 60 min of reperfusion (Table 2), thus suggesting that Nck-1 may be either specifically tyrosine phosphorylated during the ischemic phase or associated to a phosphotyrosine-containing complex through its SH2 domain.

Nck-1 has previously been reported to be tyrosine phosphorylated in response to treatment with growth factors 1213, but no information relative to the precise PY site is yet available in the literature. Noteworthy, the NetPhos software predicted four potential tyrosine phosphorylation sites within the Nck-1 sequence; two of them being located within MS identified non phosphorylated peptides (Figure 3C and red bold on Nck-1 sequence, Figure 3A). We experimentally confirmed that Nck-1 was indeed tyrosine phosphorylated by Nck-1 immunoprecipitation followed by immunoblot using anti-PY antibodies. A PY containing Nck-1 protein was detected as increasing during the first hour of ischemia, and decreasing during the first hour of reperfusion (Figure 3D, top panel). This was not due to major changes in the amount of Nck-1 protein immunoprecipitated as indicated by the blot using anti-Nck-1 antibodies (Figure 3D, bottom panel, and graph). This pattern further confirms that the ~50 kDa protein observed in Figure 1B may indeed represent Nck-1. This result also validates both Nck-1 tyrosine phosphorylation, as determined in our mass spectrometry analysis and the relevance of our approach to identify tyrosine phosphorylated proteins modulated during human liver transplantation.

To further characterize the functional relevance of Nck-1 tyrosine phosphorylation during human liver transplantation, we first assessed the expression levels of Nck-1 in the ischemic and reperfused livers using immunoblot analysis (Figure 4A). We observed an increased expression of Nck-1 during ischemia followed by a decrease during reperfusion. We then investigated the in situ distribution/localization of Nck-1 in liver during the different phases of the transplantation. Tissue biopsies were homogenized in the presence of 150 mM KCl. Particulate fractions were partitioned by centrifugation as described in the Methods section, and collected as P100 (insoluble pellet fraction) and S100 (soluble fraction). Interestingly, Nck-1 showed a decreased association with the insoluble components during the course of ischemia, whereas this pattern was reversed upon reperfusion (Figure 4B, upper panel). This observation correlates with the detection of increased amounts of this protein in the soluble fractions upon ischemia followed by a decrease upon reperfusion (Figure 4B, lower panel). These data suggest that Nck-1 is subjected to severe localization changes upon I/R in the liver. Immunohistochemical studies on liver sections upon I/R revealed that Nck-1 localized in hepatocytes' cytoplasm under basal conditions and concentrated at the cell periphery after 1 h of ischemia (Figure 4C, I60). This phenomenon appeared to be reversed upon reperfusion (Figure 4C, R60). Interestingly, Nck-1 followed the same redistribution that we previously observed for F-actin 11. Since Nck-1 is known to link tyrosine phosphorylation induced by extracellular signals to downstream regulators of actin dynamics 14, this protein may therefore serve to promote actin reorganization at hepatocytes' periphery upon I/R.

To better characterize the involvement of Nck-1 in actin remodelling in livers subjected to I/R, we aimed at identifying Nck-1 interaction partners during the course of transplantation. To this end, GST Nck-1 pull-downs were carried out using I0, I60, R0 and R60 protein extracts pre-cleared with GST (see Methods). Nck-1 binders were eluted, resolved by SDS-PAGE and the bands visualized on Coomassie Blue stained SDS-PAGE gels were processed for mass spectrometry analysis as represented in Figure 5A. No significant changes in proteins binding to GST Nck-1 could be detected by Coomassie Blue staining (Figure 5B). However, besides the bait proteins (Nck-1 and GST) and 9 proteins commonly identified for the 4 time points (Figure 4C), we also identified Nck-1 interacting proteins specific to I60 (11) or R0 (10) (see Additional File 1). Although using this approach, we did not find any of the proteins previously reported to interact directly with Nck-1, we identified actin as a major Nck-1 binding partner specific to the ischemic phase (after 60 min of ischemia -I60- and just before reperfusion -R0-, but not for the other time points). We believe that the absence of known Nck-1 interactants in our list may result from their generally low abundance which could then be masked by the presence of abundant liver metabolism proteins or by the fact that we are using an exogenous Nck-1 molecule instead of the endogenous as a trap. The presence of actin in association with Nck-1 during the ischemic phase, however, correlates with our immunohistochemistry data and, as a consequence, strengthens the hypothesis of Nck-1 involvement of in ischemia-induced actin dynamics.

0.5 cm³ biopsies were surgically removed in the course of 34 liver transplantation procedures performed within McGill University Health Centre with approval of the institution's ethic committee (ERB 05-003) as previously described 1116 and as schematically represented in Figure 1A. Briefly, ischemic liver biopsies were taken shortly before the donor liver clamping (I0), and 10 and 60 min (I10 and I60) after perfusion of the liver with cold organ preservation solution (Belzer^®). Reperfused liver biopsies were collected during re-implantation in the recipient patient, just before clamp removal (R0), and respectively 10 and 60 minutes after the blood flow has been re-established in the hepatic portal vein (R10 and R60). The average cold ischemia phase was 8 h 30 ± 2 h 15 whereas the average warm ischemia time was 48 min ± 26 min.

Proteins were extracted from each tissue sample by homogenization and solubilization in phosphate buffer saline (PBS) added with 1% TX100 1 mM NaF, 1 mM Na₃VO₄ and protease inhibitors (Complete^®, Roche Diagnostics, Laval, QC). Equal protein amounts from 3 independent ischemia or reperfusion liver biopsies were pooled for each time-point to reduce the variability between individual patients.

Protein extracts (1 mg) were incubated overnight at 4°C with mouse monoclonal anti-phosphotyrosine antibodies covalently coupled to agarose beads (PY20, Sigma, St Louis, MO) or rabbit anti Nck-1 immune serum, kindly provided by Dr Louise Larose, McGill University, Montreal, QC 31. Immune complexes were collected following incubation with protein G-Sepharose beads (GE HealthCare, Baie d'Urfé, QC) for 45 min at 4°C. After three to five washes in PBS, proteins were eluted using Laemmli sample buffer and resolved by 1D SDS-PAGE prior to mass spectrometry analysis or immunoblotting.

Following separation of the proteins purified as described above by 1D SDS-PAGE and Coomassie R-250 staining, each band was excised, transferred to a 96 wells tray, dehydrated with acetonitrile and washed by two cycles of 10 min in 100 mM (NH₄)₂CO₃ before the addition of an equal volume of acetonitrile. The destained gel slices were then treated for 30 min with 10 mM dithiothreitol to reduce cystinyl residues and for 20 min with 55 mM iodoacetamide to effect alkylation. After an additional round of (NH₄)₂CO₃ and acetonitrile washes, the slices were extracted with acetonitrile at 37°C. They were then incubated with trypsin (6 ng/μl in 50 mM (NH₄)₂CO₃) for 5 h at 37°C and the peptides were extracted in 1% formic acid/2% acetonitrile followed by two further extractions with additions of acetonitrile. All treatments were performed robotically using a MassPrep Workstation (Micromass, Manchester, UK). All extracts from a given gel slice were combined in the corresponding well of another 96 wells tray. This tray was transferred to the autosampler of a CapLC (Waters, Milford, MA) for mass spectrometry analysis on a QTOF2 (Micromass, Manchester, UK) upgraded with EPCAS™ (Embedded PC Acquisition System). A volume of 20 μL of sample was injected at a flowrate of 30 μL/min with pump C to a μ-Precolumn™ (LC Packings, Amsterdam, NL) filled with C18 Pepmap 100 (300 μm ID × 5 mm, 5 μm, 100 Å) held on a 10 port Cheminert^® valve (VICI Valco Canada, Brockville, ON). After 5 min of washing, the 10 port valve was actuated so the acetonitrile gradient from pump A and B eluted the peptides toward the PicoFrit column (New Objective, Woburn, MA) filled with BioBasic^® C18 stationary phase (75 μm ID, 10 cm, 5 μm, 300 Å). Gradient solvent was delivered at a flowrate of 1 μL/min and was splitted to 200 nL/min for the PicoFrit™ column with a splitting tee. Solvent A was water (formic acid 0,1%) and solvent B was acetonitrile (formic acid 0,1%). The linear gradient was set from 5% B to 40% in 20 min, from 40% to 70% in 5 min, from 70% to 95% in 5 min, held at 95% for 7 min and brought back to 5% in 10 min. The PicoFrit™ column was installed on a nanospray probe so that the spraying tip was near the sampling cone of the mass spectrometer. Voltage on the capillary was adjusted to get a nice plume during elution of peptides. Acquisitions were done in Data Directed Acquisition mode (DDA) while a 1 second survey scan was first done from 350 to 1600 m/z. The four most intense doubly and triply charged ions were selected to undergo MS/MS fragmentation in 1 second scans from 50 to 2000 m/z. The collision energies were determined automatically by the instrument based on the m/z values and charged states of the selected peptides. MS/MS fragmentation stopped when the total ion current was lower than 3 counts or up to a maximum acquisition time of 5 seconds, whatever came first.

The MS/MS data were peaklisted (ProteinLynx, Micromass, UK) and submitted to a local Mascot database search software (Matrix Science, UK) for search analysis against the NCBI mammalian non-redundant database with a confidence level of 95% or greater. Specific and shared peptides with an equal or greater score than the identity score were kept and recorded for each band. Proteins for which at least one specific peptide (i.e. peptide found specifically in their cognate protein) was identified with a significant Mascot score in the two replicates were retained.

Proteins were extracted from each tissue sample by homogenization in 150 mM KCl, 10 mM Tris-HCl, pH 7.5, 2.5 mM MgOAc, 4 mM imidazole pH 7.4 added with 1 mM NaF, 1 mM Na₃VO₄ and protease inhibitors (Complete^®, Roche Diagnostics, Laval, QC) and pre-cleared. Insoluble fractions (pellet – P100) were then partitioned from soluble fraction (supernatant – S100) following centrifugation at 100,000 rpm for 30 min in a TLA-100.2 rotor (Beckman Coulter, Fullerton, CA). Immunoblot analyses were carried out as described previously 32.

Liver tissues sections (8 μm thick) on glass slides were fixed in 3.7% formaldehyde. Tissue sections were peroxidase-immunostained with Nck-1 antibody using the Envision+^® rabbit kit (Dako, Mississauga, ON) according to the manufacturer's instructions. Sections were then counterstained using hematoxylin (Vector, Burlington, ON) then dehydrated and mounted using Permount (Fisher, Nepean, ON).

BL21 bacteria expressing pGEX-2TK Nck-1 GST 33 were sonicated 3 × 30 s on ice and lysed in PBS added with 1% TX100, 1 mM NaF, 1 mM Na₃VO₄ and protease inhibitors (Complete^®, Roche Diagnostics, Laval, QC) for 30 min on ice. After 30 min centrifugation at 4°C, fusion proteins were purified with glutathione-Sepharose beads (GE HealthCare, Baie d'Urfé, QC) according to manufacturer instructions. Beads were then incubated with 1 mg of protein extract pre-cleared with GST as described previously 34, collected, washed three times with PBS and resuspended in Laemmli sample buffer prior protein separation on 1D SDS-PAGE and mass spectrometry analysis.

Immunoblots analyses were performed on 3 pools of three livers (9 independent livers) and in two replicates. Immunostainings were carried out on 6 independent livers and mass spectrometry analyses were performed on 2 pools of three livers (6 independent livers). Immunoblot data were analysed and quantified using the FluorChem software (Alpha Innotech, San Leandro, CA). Data are expressed as means ± SD (Standard Deviation). Comparisons between the 0 time point (control) and the 10 and 60 min time point (treatment) were performed using a Wilcoxon signed rank test. A value of p < 0.05 was considered as significant and marked with an asterisk.