Comprehensive identification and modified-site mapping of S-nitrosylated targets in prostate epithelial cells

Abstract

Background: Although overexpression of nitric oxide synthases (NOSs) has been found associated with prostate diseases, the underlying mechanisms for NOS-related prostatic diseases remain unclear. One proposed mechanism is related to the S-nitrosylation of key regulatory proteins in cell-signaling pathways due to elevated levels of NO in the prostate. Thus, our primary objective was to identify S-nitrosylated targets in an immortalized normal prostate epithelial cell line, NPrEC.

Methodology/principal findings: We treated NPrEC with nitroso-cysteine and used the biotin switch technique followed by gel-based separation and mass spectrometry protein identification (using the LTQ-Orbitrap) to discover S-nitrosylated (SNO) proteins in the treated cells. In parallel, we adapted a peptide pull-down methodology to locate the site(s) of S-nitrosylation on the protein SNO targets identified by the first technique. This combined approach identified 116 SNO proteins and determined the sites of modification for 82 of them. Over 60% of these proteins belong to four functional groups: cell structure/cell motility/protein trafficking, protein folding/protein response/protein assembly, mRNA splicing/processing/transcriptional regulation, and metabolism. Western blot analysis validated a subset of targets related to disease development (proliferating cell nuclear antigen, maspin, integrin beta4, alpha-catenin, karyopherin [importin] beta1, and elongation factor 1A1). We analyzed the SNO sequences for their primary and secondary structures, solvent accessibility, and three-dimensional structural context. We found that about 80% of the SNO sites that can be mapped into resolved structures are buried, of which approximately half have charged amino acids in their three-dimensional neighborhood, and the other half residing within primarily hydrophobic pockets.

Conclusions/significance: We here identified 116 potential SNO targets and mapped their putative SNO sites in NPrEC. Elucidation of how this post-translational modification alters the function of these proteins should shed light on the role of NO in prostate pathologies. To our knowledge, this is the first report identifying SNO targets in prostate epithelial cells.

Figure 1. S-nitrosylation in NPrEC.

A) Normal prostate cells, NPrEC, were treated with or without 1 mM CysNO followed by a biotin BST. Protein extract (100 µg) was loaded onto an SDS-PAGE (10%). Western-blot analysis was carried out, and the membrane was probed with anti-biotin. B) After cell treatment (1 mM CysNO) and the biotin switch assay, neutravidin pull-down and SDS-PAGE (4–15%) was performed (3 mg was used for IP in each lane). Each lane (CNTL, CNTLΔNaAsc, CysNO, CysNOΔNaAsc) was divided into five portions (I–V) as indicated and subjected to tryptic digestion and mass spectrometry protein identification.

Figure 2. Identification of S-nitrosylated proteins.

A) Schematic of the workflow. Approach 1. Protein extracted from NPrEC treated with CysNO (1 mM) was subjected to a BST, and nitrosylated proteins were purified by Neutravidin pull-down, followed by SDS-PAGE. Nitrosylated proteins were identified by tryptic digestion and LC-MS/MS in the five gel bands (I–V) of lanes (CNTL, CNTLΔNaAsc, CysNO, CysNOΔNaAsc) according to the defined peptide selection criteria (XCorr 1.5, 2.0, 2.5, ΔCn>0.1, Sp>300, precursor <15 ppm, with <1% FP). After the protein-SNOs were identified, their sites of modification in peptide (peptide SNO) were identified in approach 2: after biotin switching, 1 mg of protein was digested and the nitrosylated peptides were pulled down and subjected to mass spectrometric site mapping, according to the same peptide selection criteria. B) Number of protein-SNO and peptide-SNO identification: Protein pull-downs from two independent experiments were performed to identify protein-SNOs, followed by the identification of the corresponding peptide-SNOs from peptide pull-downs. A total of 116 proteins were identified, with the sites of modification determined for 82.

Figure 3. Protein-SNO classification.

Proteins that were identified as nitrosylated are grouped according to the biological processes they belong to according to the Panther classification system and listed in Table S1A and S1B with their biotinylated peptides, if available.

Figure 4. Verification of nitrosylated proteins.

A) Western blot analysis of targets nitrosylated by CysNO. NPrEC were treated with 1 mM CysNO, and 1 mg of protein extract was subjected to BST. Biotinylated proteins were pulled down with Neutravidin beads, eluted with β-mercaptoethanol, concentrated and detected using western blot. 8% of total protein was loaded as input, and the same membrane was probed with respective antibodies with stripping and re-probing. B) Protein nitrosylation in iNOS expressing PC3 cells. PC3 cells were transiently transfected with an iNOS expressing plasmid. Cells were harvested 24 h post-transfection. 3.5 mg of total protein was subjected to BST, as described in Materials and Methods. After BST, the overall protein nitrosylation was assessed by western blot using an anti-biotin antibody. The membrane was stripped and re-probed with GAPDH to show equal loading. C). Endogenous nitrosylation of protein targets. Biotinylated proteins were pulled down with Neutravidin and analyzed by Western blot. Seventy µg was loaded as input, and the same membrane was probed with respective antibodies with stripping and re-probing.

Figure 5. Frequency of amino acids surrounding S-nitrosylated cysteines in 42 sites mapped into structurally resolved domains from 24 different proteins.

Hydrophobic residues are shown in black; charged residues in red and blue, respectively; and hydrophilic residues in magenta and green (for mixed-character residues).

Figure 6. Crystal structure of the representative protein-SNOs.

A) An example of a buried S-nitrosylation site in a crystal structure of the human protein disulfide-isomerase, PDIA3 (PDB code 3F8U), with positively charged (R280) and negatively charged (E216) residues in direct spatial proximity of the nitrosylated cysteine (C244). B) An example of an S-nitrosylation site in direct contact with a ligand and thus likely disrupting/affecting ligand binding upon nitrosylation. One of such examples is EGFR, which is shown here in complex with the monoclonal antibody inhibitor cetuximab (PDB structure 1YY9). Binding of the antibody partially occludes the ligand-binding site in EGFR and keeps it in an inactive conformation. As can be seen from the figure, the targeted cysteine (highlighted using red stick model) directly supports the protein interaction interface in EGFR, suggesting that S-nitrolysation may attenuate its interactions with ligands and inhibitors. C) An example of S-nitrosylation site(s) located within protein-protein (oligomerization) interfaces and that thus may disrupt/attenuate complex formation in PCNA and its function in DNA replication. Note that DNA, which occupies the central hole in the PCNA trimer, is not included in this complex (PDB structure 1VYJ). On the other hand, p21 peptides, which mediate regulatory interactions with CDK/cyclin complexes, are shown using cartoon models.