Excellent start to summer schools in Applied Molecular Microbiology

In the summer of 2003 the first FEMS Congress of European Microbiologists was held in Ljubljana.  On the way there, David and Joyce Hopwood reminisced with a good friend, Dušica Vujaklija of the Rudjer Bošković Institute in Zagreb, about lapsed plans for holding courses in applied microbiology on the Dalmatian coast.  A course on “Microbial Breeding” had been organised by Marija Alačević, a pioneer in the study of the genetics of antibiotic production in Zagreb, for late 1991 in Dubrovnik but it was cancelled on the outbreak of war in the former Yugoslavia that year.  In discussions at the FEMS congress with Julian Davies of the University of British Columbia, a veteran of summer schools in the Mediterranean, we vowed to revive the idea of such courses under the general title of John Innes-Rudjer Bošković Summer Schools in Applied Molecular Microbiology, with different sub-topics each year.  For the first course we decided to link back to the original concept of microbial breeding for the improvement of secondary metabolite-producing strains.

Possibilities for microbial breeding have morphed almost beyond recognition over the past 15 years, with the sequencing of hundreds of microbial genomes, the development of powerful tools for genome annotation and comparison, and functional genomics in all it forms.  There have been notable efforts by biotech companies to harness genomics for the heterologous expression of proteins and secondary metabolic gene clusters, and to use combinatorial biosynthesis to engineer novel metabolites.  Meanwhile big pharma largely abandoned the isolation and screening of secondary metabolite-producing microbes in favour of combinatorial chemistry to make synthetic drug candidates, only to be largely disappointed when the novel compounds were found not to be “drugable”.  With the discovery that the genomes of actinomycetes encode the capacity to make many more interesting small molecules than are found during typical screening campaigns the time is ripe for a new marriage of genetics and antibiotic discovery.

The title of the first summer school was therefore chosen as “Microbial Genomics and Secondary Metabolites”, with David Hopwood and Julian Davies as co-directors and Duška Vujaklija as local organiser. Funding came from the sale of the John Innes Centre book “Practical Streptomyces Genetics” and from monies remaining from meetings on antibiotics organised by Julian Davies and colleagues in Spain in the 1980s.

In November 2003 Marija Alačević and Miroslav Radman founded the Mediterranean Institute for Life Sciences in a building on the grounds of the Villa Dalmatia in Split, one of Tito’s former summer retreats, with the vision of setting up several permanent research groups as well as making the Institute an attractive venue for summer schools.  We were fortunate in being able to hold our first lecture course there, from 23 June to 1 July 2007.

 

Thirty-eight participants, mostly at PhD or early post-doctoral level, were chosen from nearly twice that number of well-qualified applicants.  They came from 16 countries and represented 20 nationalities.  All brought posters, and lively poster sessions were one of the highlights of the summer school.  The lecturers (Haike Antelmann, Greifswald, Germany; Stephen Bentley, Cambridge, UK; Mervyn Bibb, Norwich, UK; Greg Challis, Warwick, UK; Lutz Heide, Tübingen, Germany; Flavia Marinelli, Varese, Italy; as well as Julian Davies and David Hopwood) covered subjects including the isolation and screening of secondary metabolite-producing microorganisms; the regulation of secondary metabolism; metagenomics and heterologous pathway expression; genome sequencing and comparative genomics; characterisation of secondary metabolic pathways; and pathway engineering and combinatorial biosynthesis of novel molecules.  Guest speakers were Boris Maček from Martinsried, Germany and Maja Pavela-Vrančić from the University of Split, as well as the Director of MedILS, Miroslav Radman, who told the group about the amazing system of DNA repair that Deinococcus radiodurans has evolved to survive enormous doses of ionising radiation.  A hands-on computer tutorial on genome annotation and comparison was a much-appreciated component of the course.

A feature of the summer school, highly commended by the participants, was provided by constant opportunities for interactions with teachers and fellow-participants at meal times and breaks, reinforced by small-group afternoon discussions at attractive spots around the MedILS campus.  But it was not all work: there was also enough time for swimming in the blue waters of the Adriatic, either from the rocky shore of the MedILS campus or from nearby public beaches.  And the middle day of the course was devoted to an all-day excursion to the island of Brač, a 45-minute ferry ride from the harbour of Split, providing further opportunities for social and scientific interactions.

 

What were some of the scientific highlights revealed during our week at MedILS?

Julian Davies was on characteristically provocative form for his first lecture on the wonders of small molecules, providing a useful antidote to the idea that they exist only as weapons of mass destruction, like the antibiotics we use in the clinic.  In nature, some of them are involved in signalling within and between organisms, and many have global effects on transcription at sub-inhibitory levels.  If we are to exploit them effectively – notably to meet the diverse threats of antibiotic resistance as described in Julian’s second lecture, in which he especially highlighted the wonders of integrons as a mechanism for the dissemination of multiple drug resistance between bacteria – we need to gain a much more complete understanding of their roles in natural habitats.

Flavia Marinelli’s account of two case-histories to illustrate the power of directed screening for novel secondary metabolites, taken from the work of her former colleagues at Vicuron Pharmaceuticals, was an eye-opener for most molecular biologists unfamiliar with modern screening techniques.  In one study they reasoned that nearly all successful antibiotics targeting translation inhibit the elongation step. They therefore designed a high-throughput screen to identify inhibitors of initiation and were rewarded by the discovery of a novel tetrapeptide.  In the other they used a cell wall-biosynthesis inhibition assay that would reveal antibiotics with modes of action different from those of beta-lactam and glycopeptide antibiotics to find a new lantibiotic.

Lutz Heide, in a model of audience-participation teaching, took us through the cloning and analysis of secondary metabolic pathway gene clusters, and the possibilities for combinatorial biosynthesis via heterologous gene expression and genetic engineering.  The work of his group and collaborators on the coumarin antibiotics, novobiocin, clorobiocin and coumermycin, provided a microcosm of recent developments in a field which, for many people, has been dominated by the polyketides and non-ribosomal peptides.

In discussing pathway prediction from in silico analysis of DNA sequences, Greg Challis highlighted the generally linear, and very helpful, logic of the NRPS and modular PKS systems, but also pointed out the pitfalls of non-linear exceptions.  His description of experimental methods for verifying in silico predictions was illustrated by the examples of coelichelin, germicidin and a sesquiterpene, all encoded by hitherto ‘cryptic’ gene clusters revealed in the complete genome sequence of Streptomyces coelicolor, providing a powerful reminder of the wealth of untapped biosynthetic potential in the genomes of soil microorganisms.  Maja Pavela-Vrančić in her seminar provided further examples of the wonders of the NRPS logic and its exploitation. David Hopwood added a new approach devised by scientists at Kosan Biosciences called gene morphing in which a combination of heterologous expression of polyketides in E. coli, error-free synthesis of long stretches of DNA, and rapid combinatorial expression of synthetic modules, domains and linkers allows the production of polyketides to order, including in principle the products of uncultured microorganisms.

 

Reviewing metagenomics, Mervyn Bibb illustrated the concepts of DNA isolation, heterologous expression and product detection with some of the experiences of Diversa Corporation in searching for novel antibiotics.  There is no doubt that there is a huge untapped reservoir of interesting compounds to be found in “intractable” microbes from hitherto unexplored sources, but success will only come by a happy marriage of science and business to overcome the many technical and financial barriers to discovering and licensing new antibacterial agents. The regulation of antibiotic biosynthesis at its many levels needs to be understood, an endeavour that will depend heavily on the techniques of functional genomics, including gene inactivation, transcriptomics and proteomics.

Proteome analysis was described using two different approaches.  Haike Antelmann described recently-developed sophistications in 2-D gel technology, including the Decodon software, primarily illustrated by work on stress responses and the exported proteome of Bacillus subtilis.  Complementary to the use of 2-D gels, Boris Maček used an analysis of the bacterial Ser/Thr/Tyr phosphoproteome to illustrate gel-free methodology, including the powerful Stable Isotope Labelling by Amino Acids in Cell Culture (SILAC) technique.  Both techniques rely on mass spectrometry for protein identification.

The fundamental breakthrough in microbial genomics was provided by the invention of whole-genome sequencing and of software for handling and understanding the deluge of Gs, Cs, As and Ts that results from it.  Stephen Bentley described established and more recently introduced technologies – 454 and Solexa – for sequencing, and focussed on two powerful sets of software for genome presentation, analysis and comparison.  These programs – Artemis for annotation and the Artemis Comparison Tool (ACT) – were presented in a hands-on tutorial which many participants, both novices and established users, found particularly useful.  In light of this experience, future summer schools will incorporate a wider range of bioinformatics tutorials.

Next year’s summer school is already at the planning stage and will be announced in due course on the web site (http://www.jic.ac.uk/science/molmicro/summerschool2007/index.htm).  Please draw it to the attention of anyone who might be interested in applying.