Structure-based redesign of docking domain interactions modulates the product spectrum of a rhabdopeptide-synthesizing NRPS

Abstract

Several peptides in clinical use are derived from non-ribosomal peptide synthetases (NRPS). In these systems multiple NRPS subunits interact with each other in a specific linear order mediated by specific docking domains (DDs), whose structures are not known yet, to synthesize well-defined peptide products. In contrast to classical NRPSs, single-module NRPS subunits responsible for the generation of rhabdopeptide/xenortide-like peptides (RXPs) can act in different order depending on subunit stoichiometry thereby producing peptide libraries. To define the basis for their unusual interaction patterns, we determine the structures of all N-terminal DDs (^NDDs) as well as of an ^NDD-^CDD complex and characterize all putative DD interactions thermodynamically for such a system. Key amino acid residues for DD interactions are identified that upon their exchange change the DD affinity and result in predictable changes in peptide production. Recognition rules for DD interactions are identified that also operate in other megasynthase complexes.

Fig. 1

RXP NRPSs (Kj12ABC) from Xenorhabdus stockiae KJ12.1 and selected RXPs found in this strain. a Overview of the domain organization of Kj12ABC (C condensation, A adenylation, MT methyltransferase, T thiolation, C_term terminal condensation domain, ^NDD N-terminal docking domain, ^CDD C-terminal docking domain). b Structures of selected RXPs derived from the Kj12ABC system, showing differences in size and methylation patterns. The simplified structure nomenclature used within other figures is also shown (V Val, mV N-methylated Val, PEA phenylethylamine)

Fig. 2

N-terminal docking domains have the same three-dimensional structure. a Structure-based sequence alignment of the TubC-^NDD and the N-terminal docking domains of Kj12ABC. Identical residues are highlighted with dark gray boxes and residues with similar chemical properties are shown in light gray boxes. Key residues for DD interactions are shown in red. The secondary structure based on structural information for TubC-^NDD and Kj12C-^NDD is indicated above and below the sequence. b Solution structure bundle of the 19 lowest energy conformers and the regularized mean structure for Kj12C-^NDD. c Overlay of cartoon representations of the energy minimized mean structures of Kj12A-^NDD (blue), Kj12B-^NDD (green), and Kj12C-^NDD (magenta). d Electrostatic surface potentials of Kj12C-^NDD mapped on the solvent-accessible surface in the same orientation as in b (left) with negatively charged surface areas colored in red, positively charged areas coloured in blue, and white areas corresponding to hydrophobic surfaces. e Overlay of the energy-minimized mean structure of Kj12C-^NDD with one monomer of the TubC-^NDD dimer

Fig. 3

Docking domain interaction. a Sequence alignment of Kj12A-^CDD and Kj12B-^CDD used in this study. Identical residues are highlighted with dark gray boxes and residues with similar chemical properties are shown in light gray boxes. Key residues for DD interactions are shown in red. b Overlay of the ¹H,¹⁵N-HSQC spectrum of 100 µM ¹⁵N-labeled Kj12C-^NDD in the absence (black) and presence of increasing amounts of unlabeled Kj12B-^CDD. The molar ratios of the two docking domains are 1:0.25 (yellow), 1:1 (orange), 1:2.5 (dark orange), and 1:5 (red). c Histogram of chemical shift changes vs. sequence for Kj12C-^NDD upon addition of a five-fold molar excess of unlabelled Kj12B-^CDD with the secondary structure depicted above. d Chemical shift changes for Kj12C-^NDD upon addition of unlabeled Kj12B-^CDD mapped onto the structure of Kj12C-^NDD. e K_d values for all DD interactions in the Kj12ABC NRPS as measured by ITC

Fig. 4

Structure of ^NDD-^CDD complex. a Schematic representation of the N- and C-terminal docking domain linker construct used in this study with Kj12C-^NDD in magenta, the 12 amino acid long Gly-Ser linker in gray, and Kj12B-^CDD (amino acids 1545–1568 from Kj12B corresponding to residues 75–99 in the complex construct Kj12C-^NDD–12xGS–Kj12B-^CDD) in green. b Solution structure bundle of Kj12C-^NDD-Kj12B ^CDD linker construct with colour coding as in a. c Detailed view of the ^NDD-^CDD interaction of β-sheet 2 of Kj12C-^NDD (magenta) with the last 5 amino acids of Kj12B-^CDD (green). The charged residues forming salt bridges between the two docking domains are shown in stick representation. d Schematic representation of the “recognition rules” for the interaction of β2 of Kj12C-^NDD (magenta) and β3 of Kj12B-^CDD (green) in the complex. Positively and negatively charged residues are shown as blue and red circles, respectively, and hydrophobic residues as white circles. K_d values were determined by ITC titration experiments with synthetic peptides of Kj12B-^CDD_short (Supplementary Table 2 and 4) carrying individual variations of all five residues of β3

Fig. 5

Optimization of ^CDD or ^NDDs in Kj12ABC system for the production of longer RXPs. Additionally, the ITC thermograms and the derived binding curves for titrations between optimized ^CDD or ^NDDs variants are shown. a Kj12A-^CDD was optimized by two amino acid exchanges, E1169R and H1171E, on Kj12A-^CDD. Co-expression of natural Kj12B with optimized Kj12C led to increased production of longer RXPs. Red lines (I) represent RXP production in the modified system, black lines (II) represent RXP production in the natural Kj12BC. Solid lines indicate fully methylated Val (mV) RXPs, dashed lines indicate RXPs containing only one non-methylated Val. b Optimization of Kj12B-^NDD for tighter interaction with Kj12B-^CDD via three amino acid exchanges, K26Q, K24R and K28E, on Kj12B-^NDD. A red line (I) represents RXP production in the optimized system, a black line (II) represents RXP production in the natural Kj12BC. Only fully methylated RXPs are shown. c In addition to amino acid exchanges in b, Kj12C-^NDD was additionally modified via two amino acid exchanges, Q26K and E28A, to reduce the interaction with Kj12B-^CDD and allowing better interaction between Kj12B-^NDD and Kj12B-^CDD. No RXPs were detected after co-expression of natural Kj12B and modified Kj12C-^NDD probably due to very weak affinities. A red line (I) represents RXP production in the optimized system, a black line (II) represents RXP production in natural Kj12BC. Only fully methylated RXPs are shown. x Axis, numbers of amino acid residues in RXPs (RXP length). y Axis, production of the corresponding RXPs relative to the most abundant derivative set to 100%