Combating Antimicrobial Resistance With New-To-Nature Lanthipeptides Created by Genetic Code Expansion

Abstract

Due to the rapid emergence of multi-resistant bacterial strains in recent decades, the commercially available effective antibiotics are becoming increasingly limited. On the other hand, widespread antimicrobial peptides (AMPs) such as the lantibiotic nisin has been used worldwide for more than 40 years without the appearance of significant bacterial resistance. Lantibiotics are ribosomally synthesized antimicrobials generated by posttranslational modifications. Their biotechnological production is of particular interest to redesign natural scaffolds improving their pharmaceutical properties, which has great potential for therapeutic use in human medicine and other areas. However, conventional protein engineering methods are limited to 20 canonical amino acids prescribed by the genetic code. Therefore, the expansion of the genetic code as the most advanced approach in Synthetic Biology allows the addition of new amino acid building blocks (non-canonical amino acids, ncAAs) during protein translation. We now have solid proof-of-principle evidence that bioexpression with these novel building blocks enabled lantibiotics with chemical properties transcending those produced by natural evolution. The unique scaffolds with novel structural and functional properties are the result of this bioengineering. Here we will critically examine and evaluate the use of the expanded genetic code and its alternatives in lantibiotics research over the last 7 years. We anticipate that Synthetic Biology, using engineered lantibiotics and even more complex scaffolds will be a promising tool to address an urgent problem of antibiotic resistance, especially in a class of multi-drug resistant microbes known as superbugs.

Keywords: RiPPs; chemoselectivity; genetic code expansion; lantibiotics; nisin; non-canonical amino acids; superbugs; synthetic biology.

FIGURE 1

Classification of AMPs on the basis of various criteria. (A) Based on gene encoding pattern (Data-mining strategy). (B) Significant cellular or thermodynamic criteria. (C) Subdivision of positively charged (cationic) AMPs. RiPPs are structurally diverse but exclusively synthesized by ribosomes. This figure was compiled from a variety of sources (Porto et al., 2017; Koehbach and Craik, 2019).

FIGURE 2

Configuration of the target membrane environment leading to the initial electrostatic and hydrophobic interactions of the AMPs on the bacterial membrane. Initial molecular event of the effect of the AMPs leads to the different scenarios that can be described using transmembrane pore and non-pore models. The carpet model, without pores: AMPs aggregate on the bilayer surface resulting in detergent-like disintegration, whereby the membrane is fragmented into micelles. Conversely, in the “barrel stave” model, membrane pores are formed by interactions between the hydrophobic surface of the pore and the acyl chains of the lipid core of the bilayer. The “toroidal pore” model (wormhole) combines the effects described by both the barrel stave and the carpet model. For a more comprehensive overview see Sun et al. (2018). However, it should be noted that the different mechanisms of AMPs action presented here do not cover the entire range of ribosomally produced and post-translationally modified peptides (RiPPs, vide infra), since not all members of this group damage membranes or form pores, but have different targets and mechanisms of action.

FIGURE 3

Nisin is one of the best-studied ribosomally synthesized, pore-forming, cationic, antimicrobial peptides. (A) Sequence comparison of three nisin variants from L. lactis and marking the important regions of nisin targeted for bioengineering by classical molecular biology approaches (Chatterjee et al., 2005); (B) two-steps mode of action of nisin.

FIGURE 4

The radial arrangement of the standard genetic code in RNA format (A) with the side chains of 20 canonical amino acids (cAA). The standard amino acid repertoire of the genetic code can be modified or expanded (B) by non-canonical amino acids (ncAAs). By reprogramming protein translation, the ncAAs can be incorporated into recombinant proteins using in vivo and in vitro methods. These insertions offer useful modifications that can be made to the protein main chain (backbone) and the amino acid side chains (aliphatic or aromatic). Modified backbones and side chains often have unique physicochemical properties with the potential to dramatically expand the chemical and functional space of ribosomally synthesized peptides and proteins.

FIGURE 5

Expansion of scope of the nisin A biosynthesis by ncAAs. The biosynthetic production of active nisin involves the expression of peptide precursors with 20 canonical amino acids as standard building blocks, followed by posttranslational modifications (PTMs). The active form of natural nisin is characterized by the presence of unusual posttranslationally introduced bridge structures known as lanthionines (depicted in orange). It is generally expected that the incorporation of ncAAs (i.e., co-translational modifications) along with post-synthetic modification options (e.g., click chemistry) will create additional functional and structural levels of RiPPs diversification. For example, new variants with improved properties such as stability, specificity, bioavailability, action spectrum and half-life could be created. This is particularly important in view of the fact that nisin is highly effective against a number of human clinical pathogens, including many multi-drug resistant strains (Piper et al., 2009). In recent years it has been clearly demonstrated that several variants of nisin increase activity or stability in different experimental setups (reviewed in Field et al., 2015).

FIGURE 6

Incorporation of Met analogs into the one subunit of the two-component lantibiotic lichenicidin using the SPI method. Top left: Structures of methionine and its analogs. Met, L-methionine; Aha, L-azidohomoalanine; Hpg, L-homopropargylglycine; Nle, L-norleucine and Eth, L-ethionine. These analogs were used to individually replace Met residues in the precursor peptide of lichenicidin (bottom right), which was also analytically confirmed (bottom left). The substituted α-peptide of lichenicidin underwent PTM-activity, whereby the linear precursor peptide was converted into a polycyclic form that exhibits antimicrobial activity when mixed with the β-peptide (top right). For more details see Oldach et al. (2012). Most recently, Kuipers and co-workers have demonstrated the use of click chemistry in the engineering of nisin by using SPI with the methionine analogs Hpg and Aha (Deng et al., 2020).

FIGURE 7

Flow chart of the SPI and SCS methods. Both methods can be performed separately or combined in a single expression experiment (Hoesl and Budisa, 2011). Further details, advantages and disadvantages of both methods are described in the text.