Our Platform
How to use the NRPieceS Library
Abstract
Antibiotic resistance is spreading faster than we can discover new drugs. One problem is the low probability for a promising compound to be developed into an antibiotic. Therefore, it is essential to have a platform that reliably generates lead compounds (interview with Kim Lewis) . High-throughput screening against targets and rational drug design failed to deliver enough lead compounds[1] Our solution is the NRPieceS platform. At its core, the platform builds on NRPS engineering leveraged by synthetic biology tools. The concept is simple but powerful: every position within a natural product scaffold can, in principle, be accessed through the respective module in the NRPS assembly line. By exchanging a module in the NRPS enzyme, we can directly exchange the corresponding amino acid in its product, resulting in new-to-nature peptides.
Created in BioRender. Klute, S. (2025) https://BioRender.com/iu66by9
This page provides an overview of the different steps in our NRPieceS platform. It highlights key design considerations and serves as a central hub, linking to the dedicated subpages that explain each component of the platform in detail.
NRPieceS platform – Overview
Modular: Our engineering platform allows the easy exchange of modular NRPS building blocks within a cluster, using our Golden Gate technology. The platform can leverage the power of combinatorial biosynthesis and offers a plug-and-play approach to natural product engineering.
Standardized: Standardized overhangs allow the flexible use of different engineering sites and native clusters.
Module Selection: Our software mATChmaker, enables data-driven design of hybrid NRPS, accelerating rational NRPS engineering.
Peptide Functionalization: An azide-handle enables functional modifications, such as improved drug delivery for Gram-negative bacteria.
Rapid validation: Accessible phenotypic activity screening against the ESKAPE pathogens, using filter disk assay.
1. Heterologous Expression of NRPS BGCs
Natural products (NPs) have historically been the most productive source of antibiotics, providing the structural scaffolds for many clinically used drugs. These compounds represent an unparalleled reservoir of chemical diversity, having been refined by evolution to interact effectively with biological targets. In contrast to synthetic libraries, which often occupy limited chemical space, NP scaffolds exhibit complex, stereochemically rich architectures that enable potent and selective biological activity. Because they originate from biosynthetic pathways that evolved in microbial competition, these molecules inherently possess features such as target specificity, membrane permeability, and metabolic stability, key properties for successful antibiotics.[2] The NRPieceS platform builds on these natural blueprints.
In the first step of our platform (Fig. 1), a NRPS BGC is split across three separate, orthogonal plasmids and subsequently co-expressed in E. coli. The separately expressed NRPS fragments can reassemble into a functional protein complex using split inteins (NRPS and engineering, Cluster Selection). Split inteins provide post-translational covalent linkage of two independently expressed proteins (Library characterization results., Cluster Selection). We successfully expressed three clusters in E. coli using this strategy (Cluster Selection Results.)
The design of our platform allows future users to plug in their own NRPS of interest, exploring new peptide scaffolds. The cluster selection should be guided by key considerations such as GC content, codon usage, and the availability of metabolite precursors required for efficient peptide biosynthesis. Based on Rossiter et al. (2017) and related work, the choice of source organisms should focus on microbes that naturally engage in strong ecological competition, as these have evolved potent antibiotic scaffolds.[2]
2. NRPieceS toolbox: Donor Selection
In order to discover and develop novel antibiotics we need a reliable discovery platform with a meaningful output of lead compounds[1] Establishing an NRPS-based platform, we first needed to overcome the major bottlenecks limiting throughput and reliability in NRPS engineering. The first challenge was to develop an efficient method for the high-throughput generation of engineered NRPS libraries. The second was to address the issue of module incompatibility, which often prevents engineered NRPSs from functioning as intended.
The foundation of our high-throughput NRPS engineering approach was a recently published Golden Gate–based NRPS engineering method. This strategy employs acceptor vectors that encode native NRPS clusters and donor vectors carrying individual NRPS modules derived from different clusters. In a Golden Gate reaction, the donor module is inserted into the native cluster, replacing the corresponding native module. With a collection of various donors, different positions in the NRPS can be exchanged,resulting in a wide variety of peptide derivatives.[2]
This technology lacks the modularity and flexibility needed for a truly high-throughput engineering platform (for a more detailed insight DBTL).
We established a new golden gate standard for NRPS engineering. By introducing one set of standardized overhangs for all donors and vectors, we streamlined cloning towards a high throughput approach. It allows the insertion of one donor at any desired position within the NRPS, using one pair of overhangs for all positions (for a more detailed insight DBTL) .
To address the issue of module incompatibility, we developed mATChmaker, a computational pipeline. To improve engineering success rates, we combined domain annotations, substrate specificity prediction, high-throughput 3D modeling, and phylogenetic analysis into one software tool (link software). This approach represents a substantial step toward reliable NRPS engineering.
In the second step of our platform (Fig. 1), mATChmaker identifies donor modules that have a high likelihood of being compatible with the target cluster. This software enables the systematic selection of suitable donor modules for new NRPS clusters, eliminating the need for time-consuming and labor-intensive trial-and-error approaches in the laboratory.
We demonstrated this by generating the NRPieceS toolbox, a collection of 35 donor modules designed to derivatize the chaiyaphumine cluster (see Part Collection, DBTL Cycle) With corresponding 3 x 35 = 105 expression plasmids in our NRPieceS plasmid collection, it is possible to express 42,875 engineered NRPS.
3. Target cluster: Expanding peptide diversity
In the third step, acceptor plasmids of the native cluster are generated. Modules at and specific positions were replaced with an mCherry dropout cassette. For each module intended for exchange, a dedicated acceptor vector must be constructed. (DBTL Cycle). We created a total of 9 acceptor vectors, for the three clusters Chaiyaphumine, Szentiamid, and Xentrivalpeptide (see Part Collection), to serve as versatile chassis for module exchange.
Using Golden Gate assembly, the donor cassette is
inserted into the acceptor plasmid, replacing the mCherry. This design
allows for a simple visual verification of successful cloning,
pink colonies indicate the presence of mCherry (no
insert), while white colonies confirm successful
replacement with a heterologous donor module (see Results Page)
By inserting all 35 donors in the three different positions of
Chaiyaphumine, we generated 105 expression plasmids, from which 63
constructs yielded detectable peptide production, which was confirmed by
HPLC-MS analysis (Fig. 1)(Library characterization results.)
4. Bioactivity Screening
After generating potential bioactive compounds, their activity must be evaluated using a bioactivity assay. However, bioactivity screening is inherently selective toward the chosen target or organism, meaning that compounds active against other biological pathways or species might remain undetected. Therefore, careful consideration of the assay design and screening target is essential from the very beginning to ensure that the testing strategy aligns with the intended biological or therapeutic application[3] .
As our goal was to develop novel antibiotics effective against multidrug-resistant bacteria, we evaluated the bioactivity of our derivatized peptides against the ESKAPE pathogens, a group of clinically highly relevant organisms responsible for the majority of hospital-acquired infections worldwide[4] .
As suggested by Dr. Haldimann, (interview with Dr. Haldimann) we decided to do phenotypic screening instead of the often used target based screening. Phenotypic screening outperforms target-based approaches because it identifies compounds that are active against whole bacterial cells, inherently accounting for factors like cell permeability and efflux. In contrast, many target-based hits fail to translate into effective antibiotics in vivo (see Testing library.)[5] We evaluated both Filter disk assay and Drop-spot assay (Library characterization results).
After identifying a promising hit compound, we performed an
additional derivatization cycle to further enhance its bioactivity. For
this, we employed intein shuffling to recombine our
existing NRPS expression plasmids, enabling rapid generation of new
hybrid variants based on our initial lead (Fig. 1)
(
5. Peptide Functionalization
Using NRPS engineering to introduce a module capable of incorporating a clickable amino acid offers a powerful strategy for site-specific functionalization of peptides. This approach expands the accessible chemical diversity by providing a reactive handle for post-synthetic modification, enabling the conjugation of diverse functional groups. Such modifications can be leveraged to enhance drug-like properties, including pharmacokinetics, stability, and target specificity of the identified peptides.
This approach allows precise functionalization of peptides through Cu(I)-catalyzed azide–alkyne cycloaddition “click-click” in which a azide group reacts with an alkyne to form stable triazole linkages. When combined with advanced delivery platforms such as trojan-horse strategy (Drug delivery.).
Outlook & Future Applications
The NRPieceS platform provides a foundation for ongoing exploration and expansion:
Derivatization potential: Generate up to 42,000 possible peptide variants using our current NRPS modules.
Library expansion: Adding more peptide backbones and donor units to increase chemical diversity.
High-throughput testing: Screening compound libraries in a high-throughput manner often represents the main bottleneck in their application.
Low threshold workflow
NRPieceS provides a streamlined, end-to-end platform for peptide antibiotic discovery and is designed to evolve alongside future iGEM teams, empowering continuous innovation in NRPS engineering and drug discovery. Recognizing that not all teams have access to advanced analytical tools such as LC-MS, we propose an alternative, low-threshold workflow. After generating the expression plasmids through standard cloning, teams can directly perform bioactivity screening on production cultures. Once a promising hit is identified, compound verification via LC-MS or other analytical methods can follow, drastically reducing the number of samples that require detailed analysis.