ENABLE - ExploitiNg Antarctica BiotechnologicaL potEntial
A project funded by "Piano Nazionale Ricerca in Antartide" (PNRA B4.02)
Recently, bacterial strains from extreme environments have begun to capture the attention of scientists since microorganisms inhabiting ecological niches characterized by adverse conditions need to adopt peculiar survival strategies to achieve competitive advantages. Indeed, the study of microbes inhabiting extreme environments offers an irreplaceable resource for the elucidation of antagonistic relationships inside complex ecological niches and for the individuation of metabolic traits with previously valuable biotechnological potential. Our project originates from these observations and from previous experimental tests that showed the huge biotechnological potential of bioactive molecules obtained from Antarctic strains. Nevertheless, the biosynthesis of these compounds in vivo is often sub-optimal and largely dependent on external factors, such as growth temperature and medium composition. Accordingly, methods for bioactive compounds biosynthesis optimization are highly needed since they may bridge microbiological knowledge to large scale, industrial production of, for example, antibiotics. Integrating "-omics" data with methodologies for modeling biological systems, we aim at developing and implementing computational and statistical models to elucidate the biosynthetic circuits responsible for the production of bioactive molecules and to gain a complete knowledge on the metabolisms of Antarctic strains for their possible biotechnological use.
ENABLE is a multi-center project, involving the following groups/experties:
Research unit 2 - Transcriptomics - Unipd (Group Leader: Barbara Cardazzo)
Research unit 3 - Bioactive strains testing - IPB CNR (Group Leader: Concetta De Santi)
Research unit 5 - Preliminary phenotypic screening - Unime (Group Leader: Angelina Lo Giudice)
Research unit 4 - Proteomics - Unipa (Group Leader: Annamaria Puglia)
The study of microbial communities living in extreme environments offers an untapped resource for the elucidation of antagonistic relationships inside complex ecological niches with possible huge biotechnological, for example, in bio-medical research. Following an integrated experimental-bioinformatic pipeline (Figure 1) we aim to:
1- identify and characterize a panel of Antarctic bacteria shown to be able to synthesize bioactive compounds and identify the pathways (and the genes) involved in their biosynthesis
2- develop of computational methods (metabolic network reconstruction and in silico growth simulation) to be used in metabolic engineering of the most promising tool from experimental screening
3- maximize the biosynthesis of the bioactive compounds and identification of those parameters (e.g. nutrients supply, gene deletions) able to maximize the production of the metabolites of interest, e.g. antimicrobial substances)
In the first part of the project we have performed a genome-scale reconstruction of PhTAC125's metabolism was performed based on its genome annotation. The predictive capability of the model [named iMF721 according to the current naming convention]] was successfully validated, comparing constraint-based modelling outcomes with experimentally determined growth rates and large-scale growth phenotype data (phenotype microarray). The iMF721 model was then used to globally investigate possible metabolic adjustments of PhTAC125 during growth at low temperature by means of robustness analysis and functional integration of protein abundance data into the reconstructed network.
The iMF721 model predictive potential was initially tested. The PhTAC125 enzymatic capacity for these compounds was calculated as the ratio of the growth rate to the biomass yield in batch experiments and was set to 0.7, 3.6, 2.5 and 3.4 (mmol g-1 × h-1) for leucine, alanine, aspartate and glutamate respectively. The predicted growth rates were compared to those experimentally determined for PhTAC125 , revealing an overall agreement between experimentally determined growth rates and in silico predictions. Furthermore, we used Biolog Phenotype Microarray (PM) data (obtained at 15°C) to evaluate and iteratively refine the iMF721 model. This is typically achieved by (qualitatively) comparing the estimated flux value across biomass assembly reaction of the model with the activity directly measured during PM experiment. Of the 192 carbon sources tested with PM microplates, 64 (∼ 33%) were accounted for by the iMF721 model and thus could be used to directly test model predictions. In silico growth on these substrates was simulated by setting each of them as sole carbon source and the uptake rate to the arbitrary value of 1 mmol g-1 × h-1 (under aerobic conditions). Simulation results (either ‘growth' or ‘no growth') were compared with in vivo determined phenotypes. Inconsistencies between simulation results and PM data allowed the identification of metabolic gaps in the model and/or missing transport reactions. These included, for example, the gluconate:H + symporter (encoded by PSHAb0479), the pyruvate transporter (putatively) encoded by PSHAa0587 and the ATP:D-fructose 6-phosphotransferase (encoded by PSHAb0209); these genes were missing in the initial draft reconstruction and thus precluded the model from using some of the tested carbon sources.
After this iterative refinement procedure, iMF721 growth phenotypes predictions were compared again with PM results, revealing that in 84% of the cases (54 out of 64) the outcomes of in silico simulations correctly matched growth phenotypes assessed by in vivo experiments
We analysed changes in fluxes distribution across the PhTAC125's metabolism among the two temperature conditions (4°C and 18°C); in both cases iMF721 was optimized for biomass production using FBA. Overall, we found 209 reactions whose fluxes varied between the two conditions. In particular, 141 reactions displayed a reduced flux at 4°C in respect to 18°C, whereas 68 reactions increased their flux in the shift between 18°C and 4°C. This is in line with the overall number of induced versus repressed genes in the two conditions and with the observed decrease of the PhTAC125's growth rate at 4°C. Red and blue lines indicate a decrease or an increase (of at least a factor 2) in reaction fluxes when shifting between the two conditions respectively. Grey lines represent reactions for which a significant change in fluxes was not observed.
It can be anticipated that the iMF721 model presented here will be a valuable platform for a further understanding of PhTAC125 cellular physiology at the system level, including the design of more focused strategies for its possible biotechnological exploitation.
Results of this analysis have been published in the journal Environmental Microbiology. You can access the paper clicking here.
We are currently performing large scale cross streaking assays and transcriptomics to investigate the influence of medium composition in the inhibitory potential of these strains. Also, we are performing -omics integration with metabolic modeling to investigate the mechanisms implied in cold and heat shocks. Results will be made available here soon.