Specific sequence motifs direct the oxygenation and chlorination of tryptophan by myeloperoxidase

Abstract

Most studies of protein oxidation have typically focused on the reactivity of single amino acid side chains while ignoring the potential importance of adjacent sequences in directing the reaction pathway. We previously showed that hypochlorous acid (HOCl), a specific product of myeloperoxidase, inactivates matrilysin by modifying adjacent tryptophan and glycine (WG) residues in the catalytic domain. Here, we use model peptides that mimic the region of matrilysin involved in this reaction, VVWGTA, VVWATA, and the library VVWXTA, to determine whether specific sequence motifs are targeted for chlorination or oxygenation by myeloperoxidase. Our results demonstrate that HOCl generated by myeloperoxidase or activated neutrophils converts the peptide VVWGTA to a chlorinated product, WG+32(Cl). Tandem mass spectrometry in concert with high resolution 1H and two-dimensional NMR analysis revealed that the modification required cross-linking of the tryptophan to the amide of glycine followed by chlorination of the indole ring of tryptophan. In contrast, when glycine in the peptide was replaced with alanine, the major products were mono- and dioxygenated tryptophan residues. When the peptide library VVWXTA (where X represents all 20 common amino acids) was exposed to HOCl, only WG produced a high yield of the chloroindolenine derivative. However, when glycine was replaced by other amino acids, oxygenated tryptophan derivatives were the major products. Our observations indicate that WG may represent a specific sequence motif in proteins that is targeted for chlorination by myeloperoxidase.

Figure 1. Total ion chromatograms of VVWGTA (A) or VVWATA (B) exposed to H₂O₂, the myeloperoxidase-H₂O₂-chloride system, or HOCl

Peptide (20 µM) VVWGTA or VVWATA was incubated for 30 min at 37°C in buffer (PBS, pH 7.4) alone (Control) or supplemented with 100 µM H₂O₂, 100 µM H₂O₂ and 50 nM myeloperoxidase (MPO), or 40 µM HOCl. Reactions were initiated by adding oxidant, and terminated by adding L-methionine (1:10, mol/mol, oxidant/Met). The reaction mixture was analyzed by LC-ESI-MS. Ion chromatograms of precursor and product peptides were normalized to the ion current of VVWGTA or VVWATA incubated in PBS alone. M, peptide.

Figure 2. MS and MS/MS analysis of VVWGTA exposed to the complete myeloperoxidase-H₂O₂-Cl⁻ system

Reaction conditions were identical to those described in the legend to Figure 1. A. MS (left) and MS/MS (right) of M+16 products. B. MS (left) and MS/MS (right) of M+32 products. C. MS (left) and MS/MS (right) of M+32(Cl) product. Note that the ions of m/z 664.2 and 666.21 in the full scan mass spectrum of M+32(Cl) exhibit the isotopic ratio of a chlorinated compound (~3:1, the anticipated relative abundance of ³⁵Cl and ³⁷Cl).

Figure 3. Kinetics (A) and product yields (B and C) of VVWGTA and VVWATA oxidized by the complete myeloperoxidase-H₂O₂-Cl⁻ system

A peptide mixture of 20 µM VVWGTA and 20 µM VVWATA was incubated with 100 µM H₂O₂ and 50 nM myeloperoxidase (MPO) for the indicated time at 37°C in buffer (PBS, pH 7.4). Reaction mixtures were analyzed by LC-ESI-MS, and peptides were quantified by reconstructed ion chromatograms. The results of M+16 represent the sum of all isomers of M+16. Results represent the average and SD of four independent experiments.

Figure 4. Chlorination of VVWGTA by activated human neutrophils

VVWGTA (1 µM) was incubated with neutrophils (10⁶/mL) in buffer A (magnesium-, calcium-, phenol-, and bicarbonate-free Hank's balanced salt solution, pH 7.4) for 1 h at 37°C. Neutrophils were activated with 200 nM phorbol myristate acetate (PMA), and the supernatant of the cells was incubated overnight at 37°C. Where indicated, PMA or cells was omitted or 1 mM L-methionine (L-met), 1 mM azide, or 100 nM catalase was included in the buffer. Production of the chlorinated peptide M+32(Cl) was monitored by LC-ESI-MS with detection of ions of m/z 664.2 on reconstructed ion chromatograms.

Figure 5. UV analysis of chlorinated VVWGTA

Peptide VVWGTA (20 µM) was incubated for 30 min at 37°C in buffer supplemented with 40 µM HOCl. Reactions were initiated by adding oxidant and terminated by adding L-methionine (1:10, mol/mol, oxidant/Met). The reaction mixture was analyzed by HPLC equipped with an on-line diode array spectrophotometer. Note that M+32(Cl) exhibits an absorption maximum at 320 nm, suggesting that the number of conjugated double bonds had increased relative to the Trp containing precursor peptide.

Figure 6. High-resolution ¹H-NMR and total correlated spectroscopy analysis of chlorinated VVWGTA

A, ¹H-NMR analysis of the precursor peptide VVWGTA. B, ¹H-NMR analysis of chlorinated VVWGTA (M+32(Cl)). C, Total correlated spectroscopy analysis of chlorinated VVWGTA. VVWGTA (20 µM) was incubated for 30 min at 37°C in buffer supplemented with 40 µM HOCl. Following the addition of methionine, M+32(Cl) was isolated from the reaction mixture by reverse-phase HPLC, dried under vacuum, and subjected to NMR analysis.

Figure 7. Mass spectrometric analysis of the peptide library VVWXTA after exposure to the complete myeloperoxidase-H₂O₂-Cl⁻ system

A, Full-scan mass spectrum of VVWXTA (X = all 20 common amino acids). Peptide library (20 µM total peptide; ~1 pmol/µL of each peptide) solubilized in 0.2% formic acid, 20% CH₃CN was continuously infused (3 µL/min) into the ion trap mass spectrometer. Single letters indicate the anticipated amino acid at position X in each peptide. Note that peaks of material expected to contain isobaric amino acids (I, L) exhibited ~2-fold more ion current, suggesting that the mixture included peptides containing both amino acids. B, Oxidation of the peptide library by myeloperoxidase. VVWXTA (100 µM final concentration) was incubated with 100 µM H₂O₂ and 50 nM myeloperoxidase for 1 h at 37°C in buffer (PBS, pH 7.4). The reaction was terminated by adding L-methionine (1:10, mol/mol, oxidant/Met), and the reaction mixture was analyzed by LC-ESI-MS. Peptide loss was quantified from the ratio of the ion current of each peptide in the oxidized or control reaction mixtures relative to that of the internal standard VVGWVTA, which was added to the reaction mixture after reaction was terminated. Results represent the average and SD of 6 independent experiments.

Figure 8. Detection of chlorinated peptides in the library VVWXTA exposed to HOCl

Peptide library VVWXTA (100 µM final concentration) was oxidized with 200 µM HOCl as described in the legend to Figure 8. Chlorinated peptides were detected by LC-ESI-MS with monitoring of reconstructed ion chromatograms (RIC). (A) Material with the anticipated m/z of a chlorinated peptide was detected for WG (m/z 664.2), WA (m/z 678.2), and WS (m/z 694.2). The identities of the peptides were confirmed by MS/MS analysis. (B) Relative yield of chlorinated peptides WG, WA, and WS. The peptide library was oxidized with the indicated mol ratio of HOCl and analyzed by LC-ESI-MS. Relative yield was assessed by determining the ratio of the ion currents of the chlorinated and precursor peptides and was normalized to WG+32 (Cl). Results represent the average and SD of 3 independent experiments.

Scheme 1

Proposed reaction pathway for the formation of WG-4

Scheme 2

Proposed reaction pathways for the oxygenation and chlorination of tryptophan