Perspectives on lantibiotic discovery – where have we failed and what improvements are required?

Abstract

The increasing resistance of bacteria to conventional antimicrobial therapy within both the nosocomial and community environment has enforced the urgent requirement for the discovery of novel agents. This has stimulated increased research efforts within the field of lantibiotic discovery. Lantibiotics are ribosomally synthesised, post-translationally modified antimicrobial peptides that exhibit antimicrobial activity against a range of multi-drug-resistant (MDR) bacteria. The success of these agents as a novel treatment of MDR infections is exemplified by: the clinical development of MU1140 (mutacin 1140) and NAI-107 (microbisporicin), which are in late pre-clinical trials against gram-positive bacteria; NVB302 that has completed Phase I clinical trials for the treatment of Clostridium difficile infections and; duramycin that has completed Phase II clinical trials in the treatment of cystic fibrosis. Despite these potential successes, the traditional method of lantibiotic discovery involving the induction, production and identification is often an inefficient, time-consuming process creating a barrier to the efficient discovery of novel lantibiotics. The introduction of novel and innovative identification methods, including the application of probes and the ability to improve the stability and activity of agents via mutagenesis offer encouraging new areas to explore. The rapid expansion of available genome sequences of a wide variety of bacteria has revealed multiple interesting lantibiotic clusters that have the potential to be effective antimicrobials. However, due to the inefficient expression, screening and production methods currently employed, they are being assessed inefficiently and not rapidly enough to keep up with the ever-increasing demand for new agents.

Keywords:

• anti-microbial
• biosynthesis
• genome mining
• heterologous expression
• infectious disease
• microbial resistance

1. Introduction

The emergence of new agents onto the market exhibiting novel mechanisms of action has been extremely limited over the past few decades, compounding the need to discover novel agents, particularly those with a novel mechanism of action as this is thought to decrease the potential of resistance development.

A report by the European centre for disease prevention states that of the 167 novel antimicrobial agents currently being assessed only 2 have a novel mechanism of action and only 2 new classes of agents have been marketed to treat multi-drug-resistant (MDR) gram-positive bacteria, the oxazolidinones and lipopeptides, since the 1970s.

This is alarming considering each year 25,000 patients die in the EU alone from MDR infections. The economic impact of these infections is also high with associated increased health-care costs and reduction in productivity costing the economy approximately EUR 1.5 billion per year [1].

Lantibiotics are highly modified peptides containing unusual amino acids lanthionine and methyl-lanthionine that are divided into subclasses based on the type of structure they adopt and the biosynthetic enzymes involved in their production. The main classes include type I lantibiotics, which tend to be elongated, flexible structures that are modified by two enzymes encoding a dehydratase (LanB) and a cyclase (LanC). Type II lantibiotics form more globular, rigid structures and are modified by a single biosynthetic enzyme LanM performing both dehydratase and cyclase functions. A typical lantibiotic operon also contains genes for transport/processing (LanT), immunity (LanI), proteolytic processing (LanP) and structural gene (LanA) as shown in Figure 1. Despite similarities between lantibiotic types, over 15 post-translational modifications have been described thus leading to a wide variety of structures [2].

Figure 1. Schematic representation of a strategy that could be applied for a more streamlined protocol for the discovery of novel lantibiotics. The operon depicts the general organisation of genes involved in lantibiotic production.

The biosynthesis and lantibiotic ring structure were first described for lantibiotics epidermin and gallidermin [3-5], leading to the term lantibiotics being adopted to describe these agents. One of the most attractive features of these agents is their multiple modes of action [6,7], which often includes interaction with complex intermediates such as lipid II, a highly conserved molecule essential for cell wall synthesis, which is less prone to structural changes [8,9]. Nisin, for example, targets the pyrophosphate in lipid II as opposed to glycopeptides that interact with the variable pentapeptide [10]. The range of activity lantibiotics exhibit against MDR bacteria also makes them ideal candidates as novel agents against MDR infections [11-14]. Although traditional methods have aided the isolation of some promising agents, one of the real failures regarding lantibiotic discovery is the current inability to fully investigate and exploit the numerous potentially interesting lantibiotic encoding clusters identified in a wide range of bacterial genomes.

2. Lantibiotics – a long road to discovery

Despite nisin being successfully used as a food additive to control food-borne pathogens for over the past 50 years [15,16] and the recognition of lantibiotics potential against a range of pathogenic bacteria with the advancement of peptides MU1140 [17,18], NAI-107 [19], NVB302 [20] and duramycin [21] into pre-clinical and clinical trials, no lantibiotic is currently used within the clinical environment.

The discovery of lantibiotics has often involved a relatively long process relying on plate-based methods, including simultaneous and deferred antagonism and a range of heat treatment and enzyme characterisation studies [22,23]. Their discovery is also hindered by the fact that the producing strains often produce the peptides in very low amounts, if at all. Production is often reliant upon finding the specific trigger for production, which can include testing a range of nutrient-rich and poor media at a variety pH ranges, including the addition or exclusion of specific elicitor agents [24,25]. Quorum sensing systems have also been linked to lantibiotic production, for example the accessory gene regulator (agr) system in Staphylococci controls the extracellular processing of the leader peptide of epidermin [26]. Lantibiotics nisin and mersacidin can also auto-induce their own production [27,28]. To determine the specific factors that trigger lantibiotic production is therefore extremely time-consuming and does not guarantee success. When activity is observed further investigation is required for the exact identification of the agent responsible, which can potentially turn out to be a previously described agent. Various methods previously employed in lantibiotic discovery are highlighted in Table 1.

Table 1. Examples of the various methods utilised in the discovery of lantibiotics.

Lantibiotic	Discovery method	Ref.
Nisin A	Observed inhibition	[50]
Subtilin	Observed inhibition	[51]
Entianin	Microtitre auto-induction method	[52]
Microbisporicin (NAI-107)	High-throughput screening	[39]
Planosporicin	High-throughput screening	[40]
Mutacin 1140 (MU1140)	Deferred antagonism	[53]
Macedocin	Well diffusion assay	[54]
Salivaricin A	Deferred antagonism	[55]
Haloduracin (Bha A/A2)	In silico	[56]
Lichenicidin (BliAI/A2)	In silico	[33]

The development of techniques that could facilitate the rapid analysis and identification of novel lantibiotics is being developed; however, a cohesive combination of these methods is urgently required to make lantibiotic discovery a more efficient process.

When considering the detection of lantibiotics from bacterial cultures, the application of more rapid techniques including liquid chromatography coupled to ESI-mass spectrometry would be much more sensitive and productive when compared to traditional methods [29,30]. This would be even more effective if an extensive database of mass spectrometry and NMR data existed for lantibiotics and would drastically reduce dereplication time.

Lantibiotic probes are an interesting and potentially useful method of lantibiotic discovery, these works by exploiting the presence of thioether rings that can be reversibly opened; however, they have only been applied to pure cultures of known lantibiotics and are yet to be fully investigated when considering their specificity and sensitivity in mixed microbial cultures. The effect the probes may have on bioactivity also requires investigation [31,32].

3. Utilising genome data for the discovery of novel lantibiotics

One of the most promising areas of research regarding lantibiotic discovery is the application of knowledge gained through genome sequencing data. Specific web-based genome mining tools, including Bagel and antiSMASH, have been developed to aid the discovery of novel antimicrobial peptide gene clusters using bacteriocin and motif databases incorporating the analyses of surrounding genes. The biosynthetic enzymes involved in lantibiotic modification contain highly conserved domains, which can be utilised as driver sequences to screen genomic data for new lantibiotics. This method was applied in the discovery of novel lantibiotic linchenicidin, using genes encoding LanM proteins as a driver sequence to screen genomes of bacteria not previously recognised as lantibiotic producers [33]. It is thought, however, that this reliance on homologous sequences will potentially only uncover those with similarity to known lantibiotics and greater interrogation of genomes is required. The in silico bioinformatic analysis of available genomes could also be applied to allow a more directed approach to the selection of specific environments to target for the discovery of novel lantibiotics [34-36].

Despite advancements in genome sequencing and bioinformatics programs leading to the identification of unknown lantibiotic producing bacteria and interesting operons encoding lantibiotics, a robust heterologous expression system to enable their production has yet to be described. Traditional methods of lantibiotic production can result in manufacturing difficulties, including problems with up-scaling fermentations and batch-to-batch variations of agents, which can hinder the possibility of gaining FDA approval. The development of a robust, reproducible system for lantibiotic production could circumvent these issues and is essential for the discovery of novel lantibiotics as it allows the increased production of lantibiotics and also the efficient production of ‘silent’ or ‘sleeping’ lantibiotic clusters that may not be induced in the parent strain. The strain used for production also has to be carefully selected to ensure it is compatible with the expression of the various types of lantibiotic-producing operons.

Once an efficient heterologous system has been developed another major obstacle is the availability of a robust high-throughput screening system enabling interesting agents to be efficiently screened against a panel of medically relevant pathogens. Methods that could be applied include using a turbidometric assay [37] or using a pH indicator to determine bacterial metabolism [38]. High-throughput screening methods have proved effective in the discovery of lantibiotics NAI-107 [39] and planosporicin [40] and therefore warrant application in further studies.

In-vitro synthesis of lantibiotics is also an underexplored area of lantibiotic discovery, which includes semi-synthesis, using chemical modifications to increase diversity [41,42].

Solid-phase peptide synthesis is one of the most promising techniques in the synthetic production of lantibiotics and has proved effective for the production of certain agents [43,44]. However, for this method to be truly efficient the optimisation of reactions is required to allow increased yields and cost-effectiveness of this method. A synthetic analogue of the lantibiotic MU1140 (mutacin 1140), termed MU1140-S is currently in late pre-clinical trials. The company producing this agent, Oragenics, claim to have overcome the manufacturing difficulties associated with lantibiotics by application of its Differentially Protected Orthogonal Lanthionine Technology. It is believed cost-effective, large-scale and consistent production can be achieved using this method of synthesis, if so, this offers a real breakthrough in the ability to produce highly effective lantibiotics with consistent yields and purity.

4. The application of methods to improve known and novel lantibiotics

Various technologies are now available for the improvement of lantibiotics, including point mutations that have produced improved agents exhibiting increased stability, solubility and potency [45,46]. This method has mainly been applied to reference agents and requires a robust high-throughput screening system to be in place to screen all the variants produced. Site-directed mutagenesis has also revealed the importance of specific amino acids and provided essential information to allow the rational design of lantibiotics with greater potency, stability and lower toxicity. The mode of action of lantibiotics often includes pore formation and specific interactions with cell wall precursors, including lipid II and lipid I [47]. Increasing knowledge of specific structure–function relationships can highlight specific areas of interest. For example, site-directed mutagenesis of the hinge region of nisin Z which originally possessed activity exclusively against gram-positive bacteria resulted in gram-negative activity being developed [45]. Extensive characterisation of lantibiotics is essential to facilitate their application within the medical environment.

Combinatorial biosynthesis is also a promising method for the rational design of novel lantibiotics [48]; however, for this to be truly efficient a more in-depth knowledge of the actual impact the specific modifications can have on lantibiotics stability, solubility and activity is required. The combination of lantibiotics and known antimicrobial agents could also be investigated. This method has been successfully applied in the combination of nisin and vancomycin that when combined were active against vancomycin resistant enterococcus [49] and therefore warrants further investigation.

Greater understanding is also required into the effect specific modifications have on the lantibiotics ability to interact with target pathogens and exert their antimicrobial effect. This could aide the development of tools to predict ‘in silico’ interactions of specific agents enabling only the most promising agents to be taken forward into the lab. Figure 1 highlights methods that could be applied for the streamlined discovery and engineering of novel lantibiotics.

In conclusion, the development of technologies that can be applied to the discovery of lantibiotics offers great potential in streamlining the process. However, these need to be refined before their benefit will be realised.

5. Expert opinion

Due to the constantly increasing rates of resistance developing against currently used antimicrobials, the search for novel agents with favourable activity against these resistant bacterial strains is essential.

The lantibiotics appear to be ideal candidates for the treatment of MDR infections owing to their activity against a range of MDR strains and multiple modes of action, which is thought to restrict the rate of resistance development towards these agents.

Despite the wealth of novel lantibiotic clusters in a wide range of bacterial strains that have been highlighted through genome sequencing, the actual discovery and identification of novel lantibiotics often relies on time-consuming traditional isolation and screening techniques. The success of this approach is determined by the ability to find the precise inducing conditions to stimulate lantibiotic production. Even when strains contain multiple lantibiotic clusters within their genome, there is no guarantee their production will be observed in the laboratory as they can often be present but are ‘silent’ gene clusters. Despite novel technologies that can be applied to the discovery of lantibiotics, they are not currently being applied at a rate that enables the efficient analysis of these agents and this must be addressed. There are two areas that novel technologies can be applied, one in the identification of agents from bacterial cultures and possibly the most promising area, from the analysis of bacterial genomes. Although technologies exist that can aide lantibiotic identification, there is currently no standard method of efficiently processing these samples and many of the technologies are in the early stage of development and require application to a greater number of samples and studies to identify and highlight there true potential and efficiency.

The ideal solution to this problem of processing data arising from genome analysis is for the development of a robust heterologous system that enables interesting gene clusters to be expressed under a strong constitutive promoter with the host strain optimised for production in specified culture conditions. This would eliminate the need to spend excessive time screening multiple different medium or elicitors to observe production and also make up-scaling production easier and increase batch-to-batch reproducibility. The success of this approach would be highly dependent upon the application of a high-throughput screening system that could be applied using panels of medically relevant pathogenic bacteria. Once a promising agent is identified site mutagenesis and combinatorial biosynthesis could then be applied to increase the stability, solubility and potency of the agent. However, for this approach to be efficient and create meaningful results further work is necessary to understand the impact of specific modifications on the agent and structure–activity relationships. Although the interesting methods of discovery and modifications including the use of probes, mutagenesis, combinatorial biosynthesis and semi and full synthesis, none of these methods have been extensively applied to a wide range of lantibiotics and are fully optimised for the high-throughput analysis of agents. Despite the clinical promise of lantibiotics and novel methods of discovery further work is required to streamline these methods enabling the accurate and efficient assessment of the vast array of lantibiotic clusters available.

Declaration of interest

The author has no relevant affiliation or financial involvement with any organisation or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.