Research ArticleMOLECULAR BIOLOGY

A molecular mechanism for the enzymatic methylation of nitrogen atoms within peptide bonds

See allHide authors and affiliations

Science Advances  24 Aug 2018:
Vol. 4, no. 8, eaat2720
DOI: 10.1126/sciadv.aat2720
  • Fig. 1 Omphalotin A and OphA.

    (A) The primary structure of OphA protein consists of a methyltransferase domain (orange), the clasp domain (green), and the C-terminal substrate peptide (blue). The dashed red line indicates the C termini of the OphAΔC18 and ΔC6 constructs. Omphalotin A derives from N to C self-methylation of the 9 of 12 possible amides within this sequence. (B) OphAΔC18 monomer shown as a cartoon with SAM cofactor shown as space fill (oxygen, red; nitrogen, blue; carbon, yellow; and sulfur, green). The N-terminal methyltransferase (orange) is connected to a C-terminal domain (green) forming a ring. An extended loop (in cyan and denoted by an asterisk) sits above the active site. (C) The dimer has a pseudo interlocking ring (catenane) arrangement, the other monomer is colored pink (N-terminal methyltransferase domain) and gray (C-terminal clasp domain). The clasp terminal domain wraps around the loops that cover the active site of the other monomer.

  • Fig. 2 Co-complexes of OphAΔC6.

    (A) Mass spectrometry (MS) analysis of typically digested peptides shows that OphAΔC6 can be isolated in a partially methylated form (after a 2-hour induction in E. coli), which, upon incubation with SAM in vitro [overnight (o/n) and after 3 days], becomes increasingly methylated. (B) The C-terminal substrate peptide (colored dark blue) in monomer A inserts into the active-site monomer B (and vice versa; the substrate peptide of monomer B is cyan). The dimer structure is otherwise colored as in Fig. 1C. The substrate peptide binding site forms a tunnel. The orientation is rotated 180° from Fig. 1C. (C) A zoomed-in view (i−3 to i+1 range) of simulated-annealing Fo-Fc omit electron density contoured at 2σ for the OphAΔC6-SAH/M structure. The electron density shows methylated amides. Full electron density for all structures is shown in fig. S3. Carbons are shown in cyan, oxygen in red, nitrogen in blue, and methylation sites in orange. (D) The OphAΔC6-SAH/M complex shows Ile410, which is not methylated, opposite SAH/M (denoted position i) and so represents a product complex. Residues that have already been methylated are N-terminal to position i and shown with an orange ball and stick. The substrate peptide (shown as sticks) makes a number of hydrogen bonds with conserved residues.

  • Fig. 3 Key residues for methylation.

    (A) MS analysis of typically digested peptides of OphAΔC6 mutants after purification. Y63F, Y66F, Q172A, and W400A variants show reduced methylation, while R72A, R72K, Y76F, Y98F, and Y98A mutants show no detectable methylations at all. The W400A mutation alters the mass of the tryptic peptide and is shown with a different mass scale. (B) MS analysis of typically digested peptides of in vitro assay of mutants with SAM. After 12 days of incubation, Y98A shows obvious single methylation, and R72A shows methylation at the detection limit, while Y76F shows no methylation. An increase in single methylation was observed for R72A in the presence of 200 mM guanidine. (C) The inactive mutant R72A has the substrate peptide (i−3 to i+3 range) bound at the active site with Trp400 opposite SAM at position i. This substrate peptide conformation cannot undergo methylation as the amide nitrogen points away from SAM and is denoted flipped or inactive. The carbon atoms of the substrate peptide are colored gray, other atoms are colored as above. (D) Inactive R72A (cyan) and Y76F (green) complexes with SAH have structures in which the substrate peptide reverts to the “active” conformation. The R72A-SAM structure with the flipped/inactive conformation is shown for comparison with the same color scheme as in Fig. 3C.

  • Fig. 4 Substrate peptide and sinefungin coordination in the active site.

    (A) The simulated annealing Fo-Fc omit electron density (3σ) for sinefungin in OphAΔC6 Y63F and W400A complex. The HPLC analysis in fig. S3B shows some residual SAH remains. Carbon atoms are shown in yellow, sulfur in dark yellow, oxygen in red, and nitrogen in blue. (B) The OphAΔC6 W400A sinefungin structure represents a substrate complex and shows an active arrangement of the substrate peptide. The same color scheme and perspective as in Fig. 2D are used. (C) In both OphAΔC6 W400A (and Y63F) sinefungin complex structures, there is 3.1 to 3.3 Å hydrogen bond between the nitrogen atoms of sinefungin and the substrate amide. The arrangement of atoms is consistent with the SN2 attack by the amide on the cofactor. The same color scheme and perspective as in Fig. 4B are used. (D) Replacing SAH with SAM (in silico) shows that SAM would create van der Waal clashes with the amide; this extremely close contact is predicted to stabilize the SN2 transition state. (E) The inactive mutant R72A-SAM complex, where the anchoring interactions between substrate and enzyme are disrupted, avoids this steric clash. The enzyme does not significantly rearrange to tolerate this change in substrate conformation.

  • Fig. 5 Mechanistic calculations and the role of water.

    (A) Structure of the simplified model used in the QM calculations, optimized with the B3LYP functional (59, 60). Water molecules observed in the MD simulation were considered. SAM is shown in yellow, the substrate in cyan, and key residues in the binding pocket in pink. Apolar hydrogens are omitted for clarity. (B) Energy profile of the proposed two-step reaction obtained from the QM calculations. Results using the B3LYP functional (59, 60) are shown in black, and results using the TPSSH functional are shown in orange. QM calculations require a defined location of the proton in the intermediate state for convergence. The deprotonated carboxyl group of SAM was used as the ultimate destination; however, other routes are plausible. Perfect SN2 alignment of the methyl group of SAM was not observed in the model system and level of theory used.

  • Fig. 6 Proposed mechanism of methylation.

    (A) The conserved water molecule acts as base to remove the proton from the substrate to generate the imidate. The imidate is stabilized by hydrogen bonds to Tyr66 and Tyr76. Arg72 could stabilize tyrosinate and thus proton-transfer from Tyr76 to yield imidic acid (shown bracketed), which is essentially the formation of the imidic acid tautomer. SAM is shown in simplified form; the chemical structure is given in fig. S4A. (B) The displacement of SAH with SAM would create an impossible steric clash with the newly methylated amide; we propose that these clashes drive the required movement of the substrate (a 180° rotation and one residue translation). We suggest that the flipped conformation seen in Fig. 3C resembles an intermediate in the translocation process.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/8/eaat2720/DC1

    Fig. S1. Structural analysis of OphA.

    Fig. S2. pH rate profile of OphAΔC6-2h at different pH values (7.0 to 10.0).

    Fig. S3. Experimental data for complexes.

    Fig. S4. Superimposition of OphAΔC6 structures.

    Fig. S5. Sequence alignment of the methyltransferase domain of OphA (Ompol1_2087).

    Fig. S6. Mass analysis of all mutants discussed in the manuscript.

    Fig. S7. All in vivo inactive mutants have similar inactive confirmations.

    Fig. S8. MD simulations and QM calculations.

    Fig. S9. Kinetic isotope effect and solvent viscosity effect studies of OphAΔC6-2h.

    Fig. S10. Alternate register for substrate peptide.

    Table S1. Crystallographic data.

    Table S2. Occupancy of SAM/SFG and SAH in OphA variants.

    References (7175)

  • Supplementary Materials

    This PDF file includes:

    • Fig. S1. Structural analysis of OphA.
    • Fig. S2. pH rate profile of OphAΔC6-2h at different pH values (7.0 to 10.0).
    • Fig. S3. Experimental data for complexes.
    • Fig. S4. Superimposition of OphAΔC6 structures.
    • Fig. S5. Sequence alignment of the methyltransferase domain of OphA (Ompol1_2087).
    • Fig. S6. Mass analysis of all mutants discussed in the manuscript.
    • Fig. S7. All in vivo inactive mutants have similar inactive confirmations.
    • Fig. S8. MD simulations and QM calculations.
    • Fig. S9. Kinetic isotope effect and solvent viscosity effect studies of OphAΔC6-2h.
    • Fig. S10. Alternate register for substrate peptide.
    • Table S1. Crystallographic data.
    • Table S2. Occupancy of SAM/SFG and SAH in OphA variants.
    • References (7175)

    Download PDF

    Files in this Data Supplement:

Navigate This Article