Protein Engineering in Ribosomally Synthesized and Post-translationally Modified Peptides (RiPPs)

Abstract

Ribosomally synthesized and post-translationally modified peptides (RiPPs) make up a rapidly growing superfamily of natural products. RiPPs exhibit an extraordinary range of structures, but they all begin as gene-encoded precursor peptides that are linear chains of amino acids produced by ribosomes. Given the gene-encoded nature of RiPP precursor peptides, the toolbox of protein engineering can be directly applied to these precursors. This Perspective will discuss examples of site-directed mutagenesis, noncanonical amino acid mutagenesis, and the construction and screening of combinatorial libraries as applied to RiPPs. These studies have led to important insights into the biosynthesis and bioactivity of RiPPs and the reengineering of RiPPs for entirely new functions.

Figure 1:

Biosynthesis and examples of RiPPs. Top: schematic of RiPPs biosynthesis. A ribosomally synthesized precursor peptide comprised of leader (blue) and core (green) peptide segments (and sometimes follower segments) is acted on by enzymes (yellow) to install modifications such as macrocyclization and sidechain alterations (red residues). Bottom: examples of RiPPs discussed here. Lanthipeptides include thioether linkages and dehydroamino acids (Dha and Dhb). Thiopeptides are cyclized via a pyridine/piperidine linkage and include thiazoles (Thz) and dehydroamino acids. Cyanobactins are head-to-tail cyclized, and include thiazoles and prenylation. Lasso peptides contain an N-terminal isopeptide bonded macrocycle through which the C-terminus threads.

Figure 2:

Protein engineering techniques applied to RiPPs. Top: site-directed mutagenesis of a gene encoding a RiPP precursor peptide. Middle: Creation of a gene library of RiPP precursors via saturation mutagenesis. Yellow, blue, and burgundy positions correspond to degenerate (i.e., mixed) codons. Bottom: non-canonical amino acid (ncAA) incorporation into RiPP precursors. Site specific incorporation, in which a 21^st amino acid is added to the genetic code via amber suppression is depicted.

Figure 3:

Non-canonical amino acids in RiPP engineering. A: Structure of microcin J25 (MccJ25, sticks) bound to E. coli RNA polymerase (RNAP, space-filling). The precise positioning of MccJ25 within RNAP was enabled by the incorporation of p-bromophenylalanine (pBrF) into MccJ25. The orange mesh shows the anomalous bromine signal from pBrF. Image derived from PDB file 6N60. B: The cyanobactin patellin 2 was engineered with multiple para- substituted phenylalanine analogs.

Figure 4:

Engineering a lanthipeptide into a protein-protein interaction inhibitor. A: Native sequence of the lanthipeptide mProcA2.8. Positions colored maize were subjected to saturation mutagenesis to create a ~10⁹ member library. B: Bacterial reverse two-hybrid screen to identify library members capable of inhibiting the interaction between two proteins, schematized here as A and B. Proteins A and B are fused to transcription factors that bind to transcription factor binding sites (TFBSs). Successful disruption of the interaction between A and B allows for cell growth. C: Hit from the library screen XY3–3. Note that all 10 positions have been modified relative to the native lanthipeptide.

Figure 5:

Examples of RiPP enzyme and pathway engineering. A: Proper maturation of RiPPs can occur when the leader peptide is provided in trans or in cis. Left panel: native presentation of the leader peptide, covalently fused to the core peptide. Middle panel: leader peptide provided separately in excess. Right panel: fusion of the leader peptide to the maturation enzyme leads to more efficient maturation than when the leader peptide is provided in trans. B: Chimeric RiPPs can be generated by mixing two different RiPP enzymes and by generating a chimeric leader peptide that is recognized by both enzymes.