Biosynthesis and exploitation of marine-derived post-translationally modified ribosomal peptides

Lead Research Organisation: University of Aberdeen
Department Name: Chemistry


Compounds from Nature are still a mainstay for drug discovery and in certain therapeutic areas such as cancer, 70% of drugs used in the clinic are of natural origin. Many of these compounds are highly complex and are difficult to re-create in the laboratory. A recent trend has been to try to understand the mechanisms the organisms use to biologically synthesise these molecules. This work is already in a very advanced stage for certain classes of these molecules, which are constructed from small subunits, and are known as polyketides and non-ribosomal peptides. Candidates from each category are already in the clinic, or are being tested on patients to assess their suitability as pharmaceuticals. In some cases the supply from nature is limited, and the organism's processes have been transplanted into easy-to-culture bacteria, which then produces the compound of interest. This is done by taking the DNA which encodes the instructions (the biosynthetic genes) to make the compounds and introducing these into a bacterium. This method has been successfully used to produce prospective drug candidates and to engineer new compounds by changing the instructions. We have recently discovered that potential anti-lymphoma compounds, originally isolated from a marine invertebrate (seasquirt) are in fact produced by its bacterial symbiont, Prochloron. Its instructions for biological synthesis of the compounds are encoded in a much more straightforward way than the classes of compounds mentioned above, and should make it possible to modify them more readily. This class of compounds is relatively unexplored, but evidence from the scientific literature suggests that there may be many more examples of this class of compound in certain types of marine invertebrate. Using the known chemical structure we can predict the way in which it is encoded in the DNA of the producing organism, and thus locate the relevant biosynthetic genes. We will chemically screen a number of target marine invertebrates (seasquirts, sponges) and extract their DNA. After this we will then screen the DNA for the presence of the relevant biosynthetic genes and determine the sequence of their DNA. Doing this on a number of species producing this unusual group of compounds will allow us to understand the rules by which these compounds are synthesised within the organisms. The combined information can then be used to screen marine invertebrates, which are suspected, but not known to, produce similar compounds. Combining this approach with ecological information will enable us to identify organisms, which are likely to produce other compounds of this type with potent biological activity. The outcomes of this work will be the discovery of new biologically active compounds from marine invertebrates together with methods to produce them in a sustainable fashion and modify them to modulate their activity. The method necessitates only a small specimen to be collected for DNA extraction, rather than large scale harvesting. In addition we will gain an understanding of how these unique compounds are biologically synthesised in these primitive organisms.

Technical Summary

Modified peptides display a wealth of biological activity and most are constructed via NRPS pathways. An emerging group of metabolites is the post-translationally modified ribosomal peptides. These metabolites are formed from a ribosomally synthesised prepeptide by a complex of monofunctional peptides. Knowledge of the metabolite's chemical structure will therefore enable the prediction of the gene sequence from which it is derived. The biosynthetic rules towards these metabolites are not yet clearly understood. The most complex examples of this class are the patellamides, originally isolated from the seasquirt Lissoclinum patella, but now confirmed to be synthesised by its symbiont, Prochloron didemni. Many seasquirts and sponges produce related peptides that are often attributed to the symbiont. The biosynthesis of many of these bioactive compounds may also be via a ribosomal pathway. The main aim is to use a combination of chemistry, chemoinformatics, molecular biology and bioinformatics to explore these natural products. The discovery of novel biosynthetic pathways towards these bioactive modified peptides would be of interest academically, and will make excellent candidates for biosynthetic engineering. There are two strands to this work; (a) understanding the biosynthetic rules and (b) appreciation of the diversity of these type of compounds in the natural environment and gaining an understanding of how the pathways vary. A large number of molecules have been described from marine sources that are likely to be synthesised by this pathway. We will reisolate these compounds, and isolate the symbiont and its genomic DNA. Following this, the design of degenerate primers for conserved domains in the gene cluster will allow the systematic discovery of novel compounds in this class. Detailed cataloguing of the types of sea squirt and the nature of their symbionts (16S rRNA determination) will enable relationships between these organisms to be understood.


10 25 50
Description Database Searching:

The initial database search was completed early on and a list of target species compiled including ascidians, porifera, cyanobacteria and actinobacteria (Objective 1). Features of post-translationally modified ribosomal cyclic peptides were searched for and a number of key species identified. These species were targeted during collections in Australia and Fiji with the assistance of local taxonomists.

Specimen Collection:

Four collections were made, one ahead of the fellowship starting to Fiji and Australia, funded by the University of Aberdeen and three collections in Fiji during the term of the fellowship. A reprioritisation of research direction away from marine biodiscovery at the Australian Institute of Marine Science (AIMS) meant that after 2006, further expeditions on the AIMS vessels became impossible. During the first two expeditions it was found that organism growth was strongly seasonal with best growth and species diversity at the end of the antipodean summer. Depth at which organisms were found was from shallow intertidal to moderate depth (10 m). Despite lengthy collection times, some sample sizes were still very small, due to high current conditions and small organism size. Throughout the fellowship many rare ascidians and candidate sponges were collected which were used in this project and will also form a valuable resource for my research group for many years into the future (Objective 2). In all cases extracts of the organisms or pure compounds derived from these have been tested in a number of biological assays, including cytotoxicity, anti-inflammatory/anticancer and antiparasitic with some excellent results in against trypanosomes which are currently being followed up.

Developing gDNA Extraction Protocols:

During the first expedition all specimens were treated using two different on-site DNA extractions procedures, as well as sample preservation in RNALater and ethanol. It was found that the best protocol involved on-site preservation in RNALater for DNA extraction. Subsequent DNA extraction from preserved tissue gave in many cases high quality high molecular weight gDNA (Objective 4). This gDNA was used to further the other objectives of this fellowship and forms part of a valuable resource for future investigations by my research group.

Developing methods to detect these metabolites in extracts:

Initial efforts aimed at extracting the organisms with ethanol as soon as possible after collection, but this proved impractical due to the volume of samples collected. During the first expeditions laboratory extraction using organic solvents in the host laboratory was viable, but after this the burden on the host lab was too great and therefore sending ethanol preserved samples was the best solution to this problem. Previous studies in our lab show that this does not lead to extensive degradation of components. Extracts from these specimens were subjected to LC-MS. The most promising target organism was Lissoclinum bistratum giving the expected mass spectral fragmentation patterns for the bistratamides and high quality gDNA. However, as we were pursuing this we became aware of a publication by Schmidt's group in Utah (Nat Chem Bio 2006, 2, 729) which reported on the biosynthetic capabilities of a number of Prochloron containing Didemnid ascidians. This work meant we had to redirect the objectives of the fellowship, to focus on rarer Didemnid ascidians, the porifera, cyanobacteria and actinobacteria (see objectives 8-10).

Close collaboration with analytical chemists at Aberdeen led to the development of a novel parallel chromatographic method to identify metal complexed marine metabolites. The method relies on coupling of the HPLC with both ICP-MS, which can detect trace elements below 1 ppb, and ES-MS which ionises and allows the detection of metal-ligand complexes in the extract. We had long suspected that the patellamides complexed copper, and used these copper complexes for method development, although the actual proof of concept was carried out using a the ascidian Eudistoma gilboviride containing a different metal ligand. This represents a major step forward in our understanding of the role of small molecules as metal sequestering agents in ascidians. This work was published in the world's top chemistry journal, Angewandte Chemie, and we are currently writing the manuscript on the method development work on Lissoclinum patella metabolites. We have also contributed a book chapter on metal complexing metabolites from marine invertebrates.

For the first time we are able to provide concrete evidence that such small molecules have a role in the acquisition and sequestration of biologically important metals. The similarity of the metabolite in question to one produced by a Pseudomonas suggested that it was of symbiotic origin. DNA extraction on the ascidian tissue preserved in RNAlater was carried out with a view to determining if any Pseudomonas DNA was present, but this was one of the few samples in which extensive shearing of gDNA had occurred. We have asked the researchers at AIMS to recollect this species, but they have not been able to do so to date. This has meant that objective 10 has not been realised.

Detecting Post-Translationally Modified Ribosomal Peptide Biosynthetic Pathways:

Trials were conducted to detect the presence of the pathways encoding the biosynthesis of these metabolites using PCR and sets of degenerate primers designed according to the putative metabolite coding sequences and the highly conserved region within the patD gene (Objective 3). Initial work focused on the development of highly degenerate primers for these regions to localise the metabolite coding sequences for the heptapeptide lissoclinamides using gDNA obtained from Prochloron contained within the ascidian Lissoclinum patella. However, the level of degeneracy necessary meant that the method could not be applied successfully.

Other work has concentrated on culturable species obtainable from culture collections and those for which full or partial genome sequences are available. Genome mining is a powerful tool for the discovery of silent (cryptic) biosynthetic pathways of novel secondary metabolites. We have used this approach to search for patellamide-like biosynthetic cluster in the genomes of different cyanobacterial species. We have identified one species that contains a cluster of nine different patE-homologues among other genes that are homologous to patA, patB, patD, patF and patG. After obtaining the strain from a culture collection and growing it in bulk we confirmed the presence of the relevant genes in this species of cyanobacterium using the relevant PCR primers. A solvent extract of the organism was subjected to LCMS and showed an interesting metabolic profile. We have analysed the nine possible metabolite coding sequences, but the number of possible post-translational modifications such as heterocyclisation, oxidation and macrocyclisation make it difficult to predict the exact metabolites that will be produced. To efficiently search for possible metabolites in the complex LC-MS chromatogram we are currently using MetWorks® software. This package was provided with our Thermo Orbitrap LC-MS instrument which is originally designed to search for drug metabolites but can be modified for our purpose. We are exploring looking for MS evidence of the leader sequence which is released during the processing of the PatE peptide.

Other options are to culture this strain under different conditions to switch on this biosynthetic pathway. We have been discussing this possibility with Prof. Phil Wright at Sheffield, an expert in the design and optimal use of photobioreactors for cyanobacterial culture. This part of the work goes a long way towards meeting the original objectives 4 and 6, revised objective 8 as well as the proposed outcomes that stated we would obtain "a collection of novel biosynthetic pathways for ribosomally synthesised natural products that relate to their structure, the organism and their symbiotic hosts" as well as gaining "an understanding of the diverse range of molecules that can be made by this route."

Constructing gDNA libraries:

As this work was using precious gDNA from field samples we investigated the possibility of obtaining culturable organisms which could be used to develop and validate the methodology which could then be applied to the field collected samples (Objectives 5, 8). The initial culture identified was that of the cyanobacterium Nostoc sp. 31 obtained from ATCC (ATCC 43529), which is reported to produce bistratamide-like compounds, and this was confirmed by LC-MS analysis. The initial culture was contaminated with Xanthomonas and obtaining a pure culture took a significant amount of time. We started work with Dr Jem Stach at Newcastle to produce a library and conduct a full genome scan. This work was discontinued after the group of Dittmann published a full analysis of the gene architecture for bistratamide-type compounds (App. Env. Microbio 2008, 74, 1791).

Protein Overexpression and Planned Crystallisation Trials:

The original objectives 4, 5 and 6 were superseded to a large extent by the published work of Schmidt and Dittmann referred to previously. For this reason we concentrated on objective 7 and the revised objective 9. We therefore planned to carry out overexpression studies to lead to identification of the structures and functions of proteins involved in the biosynthesis of patellamides and other related cyclic peptides. This work is being done in collaboration with Prof Jim Naismith at St Andrews, a world authority on protein crystallography. As the genes for these pathways show low or no homology to known sequences, there is currently no structural information in the literature and this will be highly important for a full understanding of the biosynthetic pathway. Initially we cloned each of the genes in the biosynthetic cluster in pEhistev vector which tags the expressed proteins with 6 histidine residues to facilitate purification on affinity nickel column. However, we managed to obtain only one protein, PatC, which was expressed in the insoluble form. We also cloned all biosynthetic genes into several different expression vectors, but this did not improve our success rate. Based on the knowledge that both PatB and PatC are not essential for the biosynthesis, we optimised the sequence of PatA, PatD, PatE and PatF and synthesised sequences optimised for expression in E coli. We have used a robust algorithm developed by the company DNA2.0 to optimise the sequence of each gene to increase the possibility of its expression in E. coli. Codon usage, codon adaptation index, the secondary structure of the mRNA, and the presence deleterious motifs are among the factors that have been considered during sequence optimisation. So far we have successfully expressed PatA and PatF from the synthetic constructs. The only problem is the high tendency of the expressed proteins to precipitate from solution after purification. Trials to overcome this problem as well as to express the other proteins are underway. We will continue work with Prof Jim Naismith who has appointed a post-doctoral research assistant on this project to assist with protein expression and crystallisation trials. Continued work on this (funding for two post-docs, 1 in Aberdeen and 1 in St Andrews until Mid 2011) will lead to a successful realisation of the originally proposed outcome to develop "a platform from which to undertake combinatorial biosynthesis of these products to generate novel compounds."

Structural Work:

We have studied the solution structure of the highly conserved leader sequence of the patellamide precursor peptide (PatE1-34) using CD and NOE restrained molecular dynamics calculations (Objective 6). Our results showed that the peptide adopts a helical structure spanning residues 13-28. The helix has a hydrophobic surface which may act as the first binding site with the post-translational enzyme machinery. This work has been submitted for publication and we are currently responding to the referee's comments. Mutational analysis studies must now be conducted on the leader sequence to corroborate the suggested hydrophobic interaction surface. Success in the overexpression work mentioned previously means this can now be realised. Based on observations from the studies carried out on microcin and lantibiotics, it was assumed that the leader sequence of the patellamide-precursor peptide similarly targets it toward the post-translational machinery. In the past three years, several patellamide-like pathways have been identified in different species. ClustalW alignment of the precursor peptides from these pathways indicated that the leader sequence is highly conserved. This observation strengthens the previous assumption and encourages us to study the 3D structure of the precursor peptide in solution in order to identify the main interaction sites with the enzymes of the post-translational machinery. This ties in with the overexpression and structural characterisation work referred to above.

Theoretical Work:

A theoretical study, together with Dr Sam de Visser in Manchester, probing the possible catalytic functions of patellamide-dicopper complexes has been published. This work showed extremely strong binding of carbonate to the patellamide dicopper complex consistent with experimental results and also suggests this bridged complex prepares the system for a potential catalytic function.

Review Articles:

During this fellowship, I co-authored two reviews on the approaches to access metagenomes from marine invertebrates for the expression of novel marine metabolites. These encapsulate much of the philosophy used in during the fellowship and will form the basis of my future research.
Exploitation Route Bioscience for Business Science Manager Role:

This role was taken on with the approval of the BBSRC from 01/06/2007 to 30/08/2009 with a number of additional objectives. In my original letter to the BBSRC Fellowships Committee I stated that acceptance of the post "is compatible with the exploitation aims of my fellowship and will encourage the translation of my own research and that of others into applied blue biotechnology." Indeed, I am now more efficiently translating my skills for the benefit of UK industry. I am a paid consultant to Aquapharm Biodiscovery and an unpaid consultant to Glycomar Ltd. I have had several funded projects with Aquapharm, and have recently secured translational funding from the Scottish Universities Life Science Alliance. I have also just received the announcement of a funded Knowledge Transfer Partnership co-funded by the BBSRC.

I have effectively carried out my role as science manager through face to face meetings with agencies, industry and academia, presentations at industry conferences and authoring of reports and published papers. I was delighted to be able to lead an industry mission to BIO Pacific Rim and to lead an academic/industry mission to Japan on the topic of marine bioresources with the assistance of NERC and the Foreign and Commonwealth Office. I provided advice to Alf Game at the BBSRC on the role of genomics in marine sciences as input to the creation of The Genome Analysis Centre. I organised a BBSRC sponsored meeting on anti-infectives from natural sources which brought together clinicians, chemists, pharmacologists, microbiologists and industrialists and has led to tangible benefits (funded link between Galapagos and Dr Mark Moloney at Oxford University). In terms of written output I have published two articles in trade journals and am now co-authoring the European Science Foundation's Marine Board White Paper on Marine Science. Although the Bioscience for Business Knowledge Transfer Network has ceased to exist I am now a member of the Industrial Biotechnology Sector Group of the newly formed Biosciences Knowledge Transfer Network. I believe that the work mentioned here and in the 'recognition' section meets all the stated objectives and proposed outcome of the science manager role.
Sectors Pharmaceuticals and Medical Biotechnology

Description Biosynthesis of five-membered heterocyclic rings
Amount £60,382 (GBP)
Funding ID BB/K015176/1 
Organisation Biotechnology and Biological Sciences Research Council (BBSRC) 
Sector Public
Country United Kingdom
Start 09/2013 
End 08/2017
Description Development of Pgp antagonists to target inhibition of multiple drug resistance in cancer
Amount £49,073 (GBP)
Funding ID MSDPD41 
Organisation Scottish Universities Life Sciences Alliance 
Sector Academic/University
Country United Kingdom
Start 02/2013 
End 01/2015
Description Enhanced discovery and scaleable synthesis of therapeutic cyclic peptides
Amount £142,653 (GBP)
Funding ID TS/K00476X/1 
Organisation Innovate UK 
Sector Public
Country United Kingdom
Start 10/2013 
End 03/2015
Description Natural Product Nanotechnology
Amount £218,274 (GBP)
Funding ID RPG-2012-504 
Organisation The Leverhulme Trust 
Sector Academic/University
Country United Kingdom
Start 05/2012 
End 09/2014
Description New chemical biology for tailoring novel therapeutics
Amount £452,154 (GBP)
Funding ID 339367 
Organisation European Commission 
Sector Public
Country European Union (EU)
Start 03/2014 
End 02/2019