PhD Summer School: Methods for Mathematical and Empirical Analysis of Microbial Communities
Monday 27th October 2014 to Wednesday 29th October 2014
|12:00 to 12:30||Registration|
|12:30 to 13:20||Lunch at Wolfson Court|
|13:20 to 13:30||Welcome from John Toland (INI Director)|
|13:30 to 14:30||
S Brown (University of Edinburgh)
Plenary Lecture 1: Understanding bacterial communication and cooperation: combinatorial quorum-sensing
Quorum sensing (QS) is a cell–cell communication system that controls gene expression in many bacterial species, mediated by diffusible signal molecules. Although the intracellular regulatory mechanisms of QS are often well-understood, the functional roles of QS remain controversial. In particular, the use of multiple signals by many bacterial species poses a serious challenge to current functional theories. Here, we address this challenge by showing that bacteria can use multiple QS signals to infer both their social (density) and physical (mass-transfer) environment. Analytical and evolutionary simulation models show that the detection of, and response to, complex social/physical contrasts requires multiple signals with distinct half-lives and combinatorial (nonadditive) responses to signal concentrations. We test these predictions using the opportunistic pathogen Pseudomonas aeruginosa and demonstrate significant differences in signal decay between its two primary si gnal molecules, as well as diverse combinatorial responses to dual-signal inputs. QS is associated with the control of secreted factors, and we show that secretome genes are preferentially controlled by synergistic “AND-gate” responses to multiple signal inputs, ensuring the effective expression of secreted factors in high-density and low mass-transfer environments. Our results show that combinatorial communication is not restricted solely to primates and is computationally achievable in single-celled organisms.
|14:30 to 14:45||
E Cunnington (Massey University)
Contributed Talk 1: The Crabtree effect and its influences on fitness of yeast populations from natural isolates
Co-author: Thomas Pfeiffer (Massey University)
Yeasts degrade sugars in order to produce ATP. Two metabolic pathways are distinguished in ATP production: respiration and fermentation. While the respiration pathway occurs in presence of oxygen and produces up to 38 ATP to the cell, fermentation does not require oxygen but is also much less efficient (2 ATP produced by sugar converted into ethanol). Despite the low efficiency of fermentation, a certain number of yeasts species (including the brewer’s yeast Saccharomyces cerevisiae) have the ability to ferment sugar in aerobic conditions, this in addition to the respiration pathway when sugar concentration is sufficiently high. This is known as the Crabtree effect. It remains unclear why certain yeasts exhibit an aerobic alcoholic fermentation, and one explanation to this phenomenon relies on the increase in ATP production rate, which come at the cost of the production yield. This explanation is supported by the yield/rate trade-off theory. However this theory has not yet been conclusively supported by experiments. In my talk, I will introduce novel experimental approaches that might be used to investigate the yield/rate trade-off theory under the Crabtree effect in yeast from natural isolates.
|15:00 to 15:15||
G D'Souza (Max-Planck-Institut für Chemische Ökologie)
Contributed Talk 2: Less is more: Selective advantages can explain the loss of biosynthetic functions in bacteria
Co-authors: Silvio Waschina (Experimental Ecology and Evolution Research Group, Department of Bioorganic Chemistry, Max Planck Institute for Chemical Ecology, 07745 Jena, Germany; Research Group Theoretical Systems Biology, Friedrich Schiller University of Jena, 07743 Jena, Germany), Christoph Kaleta (Research Group Theoretical Systems Biology, Friedrich Schiller University of Jena, 07743 Jena, Germany), Christian Kost (Experimental Ecology and Evolution Research Group, Department of Bioorganic Chemistry, Max Planck Institute for Chemical Ecology, 07745 Jena, Germany)
Bacteria that have adapted to nutrient-rich, stable environments are typically characterized by reduced genomes. The loss of biosynthetic genes frequently renders these lineages auxotroph, hinging their survival on an environmental uptake of certain metabolites. However, the factors that govern this ‘genome streamlining’ remain poorly understood. Our analysis of 1532 metabolic networks revealed that auxotrophies are likely to be highly prevalent in both symbiotic and free-living bacteria. To unravel whether selective advantages can account for the rampant loss of anabolic genes, we systematically determined the fitness consequences that result from deleting conditionally essential biosynthetic genes from the genome of Escherichia coli in the presence of the focal nutrient. Pairwise competition experiments with each of 16 mutants auxotrophic for different amino acids, vitamins, and nucleobases against the prototrophic wild type unveiled a pronounced, concentration-dependent growth advantage of around 13% for virtually all mutants tested. Our in silico analysis also suggests that bacteria are frequently auxotrophic for multiple metabolites. Also bacteria are frequently subjected to changes in resource environments. Hence we also determined the effect of different carbon environments and epistasis on the fitness of Escherichia coli genotypes from whose genome one, two, or three different amino acid biosynthesis genes have been deleted. Competition experiments between auxotrophic mutants and prototrophic wild type cells in one of two carbon environments revealed that plasticity and epistasis strongly affected the mutants’ fitness individually and interactively. Taken together, our findings suggest adaptive benefits could drive the loss of conditionally essential biosynthetic genes and that both the genetic background and environmental conditions determine the adaptive value of the loss of these biosynthetic functions.
|15:15 to 15:30||Afternoon Tea|
|15:30 to 16:30||
J Weitz (Georgia Institute of Technology)
Plenary Lecture 2: Theoretical principles of virus-host population dynamics
In this talk I will introduce theoretical principles underlying the study of virus-host interactions from an ecological perspective. In doing so, I will show how viruses can affect population dynamics, evolutionary dynamics and ecosystem functioning.
|16:30 to 17:30||
J Kreft (University of Birmingham)
Plenary Lecture 3: Are simple models more general?
Co-author: Robert J Clegg (University of Birmingham) Using some examples I will show that simpler models can be less general and that complex models can be less realistic. It is therefore important to vary the complexity of a model to test the structural robustness of models, not just checking off the parameter sensitivity box. For example, one ought to test which processes (e.g. growth, diffusion, migration, predation, …) need to be included in a model, but having to spend a lot of time implementing further processes that may turn out not to matter means that this is often not done, especially towards the end of a project. Open-source, individual-based modelling platforms can help here if many groups contribute by implementing further processes enabling the user to quickly try out a bunch of processes. In the end, models are more useful if they are less wrong.
|17:30 to 18:30||Wine Reception and Poster Session|
|09:30 to 10:30||
I Klapper (Temple University)
Plenary Lecture 4: Biofilms, particularly Biofilm Models
|10:30 to 11:00||Morning Coffee|
|11:00 to 12:00||
Q Jin (University of Oregon)
Plenary Lecture 5: Biogeochemical Reaction Modeling: Theory and Application
Biogeochemical reaction modeling (BGRM) is a computational framework that couples the simulation of microbial metabolism with the simulation of chemical reactions. BGRM simulates chemical reactions using the classical approaches of geochemical reaction modeling. Specifically, it builds on the equilibrium simulation of chemical speciation, and applies kinetic or equilibrium modeling approach to redox reaction, mineral precipitation and dissolution, and other chemical reactions. BGRM simulates the metabolism of microbial groups in terms of the rates of microbial respiration/fermentation, growth, and maintenance. It describes microbial rates using rate laws that consider the availability of electron donors, acceptors, and growth nutrients, the thermodynamics of the environment, and the energetics of microbial metabolism. By capturing the thermodynamics and kinetics of chemical and microbial reactions, BGRM can be applied to assess the habitability and potential metabolic activit ies of natural environments, and to predict the dynamics of microbial populations and environmental chemistry. By simulating simultaneously chemical and microbial reactions, BGRM can also be applied to investigate the interactions among different microbial groups, and between microbes and their environment.
|12:00 to 12:15||
ER Watkins (University of Oxford)
Contributed Talk 3: The role of metabolic and immunological competition in structuring pneumococcal populations and the effects of vaccination
Streptococcus pneumoniae (the pneumococcus) is a major cause of bacterial pneumonia, septicemia and meningitis worldwide. Traditional serotyping methods have demonstrated that pneumococcal populations are highly diverse, with over 90 capsular serotypes. More recently, whole genome sequencing has revealed extensive diversity among the metabolic genes, and that metabolic alleles tend to segregate in non-overlapping associations with the antigenic serotype. Here, we use a multilocus model in which strains are composed of metabolic, antigenic and virulence components to explain these patterns of structuring through ecological and immunological competition in the host.
Currently, pneumococcal protein conjugate vaccines target only a small subset of its >90 known capsular serotypes. We use our model to show that a strain-targeted vaccination strategy could alter the genomic profile of non-vaccine strains, potentially leading to an increase in their transmissibility and virulence. The results are consistent with data on the changes in the population structure of the pneumococcus following vaccination, including support at the whole genome level in a collection of 600 pneumococcal genomes.
Similar non-overlapping associations among metabolic and antigenic alleles have been observed in a number of other bacterial pathogens, suggesting that metabolic and immunological competition may play important roles in the maintenance of pathogen population structure more generally. Our vaccination findings also have implications for strain-targeted vaccination in a range of bacterial and viral systems.
|12:15 to 12:30||
C Mills (University of Edinburgh)
Contributed Talk 4: How the coexistence of specialist and generalist species is influenced by the size of environmental graining
Co-authors: Rosalind Allen (University of Edinburgh), Richard Blythe (University of Edinburgh)
Consider an ecosystem with limited space. For a specialist species to survive, there must be enough of its favoured habitat to support itself. If there is a lot of a particular habitat, then necessarily there must be less of another type of habitat.
A central question in ecology is why so many types of species can coexist in the same place. A popular explanation is niche partitioning, in which different species adopt different strategies and thus avoid competition. A possible difference in strategies is to either become a specialist, which uses one resource very well but cannot use anything else, or a generalist, which can use many resources, but less well. Specialist and generalist species coexist in many environments on earth, but why?
One possible factor is the size of similar patches that occur in the environment, which can also be thought of as the coarseness of environmental grain. Using an agent-based model, I investigate the effects of grain size on the coexistence of specialists and generalists and show why intermediately sized graining encourages coexistence and a higher overall species diversity.
|12:30 to 13:30||Lunch at Wolfson Court|
|14:00 to 15:00||
C Quince (University of Glasgow)
Plenary Lecture 6: tba
|15:00 to 15:15||
F Goldschmidt (ETH Zürich)
Contributed Talk 5: Successive range expansions of interacting microbial populations promote spatial diversity
Succession of species is a process that is widespread in nature. Succession occurs when a pioneering species that expands into new territory is followed by a secondary species that depends on the change in the formerly pristine environment caused by the primary species. It is known that when species expand their territory, genetic drift leads to a reduction of diversity in this species. On the other hand was recently shown that interactions between species, like mutualistic dependencies, can oppose genetic drift during simultaneous expansion. However, it remains unclear how genetic drift affects diversity during successive expansions of interacting species. Our main questions are two-fold; first, can a temporally segregated interaction can oppose genetic drift during the succession and if this affects mainly the secondary or also the primary pioneer?
To address our questions, we constructed a system of two syntrophic bacteria that undergo successive range expansions. The producing strain degrades a parent substrate into an intermediate, which is then excreted. The consuming strain then consumes the secreted intermediate. We manipulate the interaction between the two strains from commensal to a non-obligate mutualism by changing the reactivity of the intermediate using the pH (i.e., the toxicity of the intermediate increases as the pH decreases).
We found that the producing strain forms a regular wave front, while the consuming strain, which has to penetrate the biofilm of the producing strain, forms dendritic structures with fractal properties. There are two processes that oppose the diversity reducing effect of genetic drift during the successive expansion. First the number of expanding dendrites is higher at non-obligate mutualistic conditions. This process maintains the initial diversity. Second the degree of dendritic branching is higher at commensal conditions. Branching generates new spatial diversity by splitting up the populations into small sub-populations. In short we find that successive expansion promotes diversity by maintenance of pre-existing diversity and generation of new diversity during the expansion.
|15:15 to 15:30||
S Waschina (Max-Planck-Institut für Chemische Ökologie)
Contributed Talk 6: Carbon source-dependent metabolic costs of amino acid biosynthesis in Escherichia coli
Co-authors: Glen D'Souza (Experimental Ecology and Evolution Research Group, Max Planck Institute for Chemical Ecology), Christian Kost (Experimental Ecology and Evolution Research Group, Max Planck Institute for Chemical Ecology), Christoph Kaleta (Theoretical Systems Biology Research Group, FSU Jena)
Bacteria invest a significant proportion of their available energy and resources into the biosynthesis of amino acids, which are required for cell growth and maintenance. As a consequence, the costs of amino acid biosynthesis profoundly limit the growth rate and, hence, the fitness of a bacterial species. Despite the substantial role for the metabolic economics of a cell, little is known about how these costs may shape the dynamics of cooperative cross-feeding interactions in bacterial communities. Here we show that the growth rates of Escherichia coli amino acid auxotrophic strains relative to the growth rate of the E. coli wild type are strongly carbon source-dependent, when amino acid availability is limiting. To understand these differences we developed a computational framework to quantitatively estimate biosynthetic costs of amino acid anabolism. This approach is based on a genome-scale metabolic network of E. coli and essentially estimates the amount of a given carbon source which is needed to synthesize a specific amino acid. The observed carbon source-dependent increase of the auxotroph's maximum growth rate µmax with increasing amino acid concentration correlated positively with the predicted biosynthetic costs. We conclude that the differences in the increase of µmax are due to the metabolic costs, which the auxotrophs save by taking up the focal amino acid from the environment. These findings imply that there is a high potential for mutualistic amino acid cross-feeding interactions to evolve among sympatric populations of the same bacterial species that specialize in the utilization of different substrates when multiple carbon sources are available in the environment.
|15:30 to 16:00||Afternoon Tea|
|16:00 to 17:00||
A McKane (University of Manchester)
Discussion Session: Different approaches to the modelling of microbial communities
|09:30 to 10:30||
C Tarnita (Princeton University)
Plenary Lecture 7: Mathematics of social behavior
I will begin with a discussion and mathematical description of the two different types of social construction: `staying together' and `coming together' (or aggregation). Staying together means that individuals form larger units (complexes, groups) by not separating after reproduction (eg. ant colonies, most multicellular organisms), while coming together means that independent individuals form aggregates (eg. most animal groups, including humans). For each of these operations I will discuss its strengths and vulnerabilities in promoting social behavior, which will lead naturally into a discussion of the various mechanisms (and the relationships between them) that have been proposed to explain the evolution and maintenance of social behavior and cooperation: direct and indirect reciprocity, kin selection, group/multilevel selection, spatial structure, punishment/ostracism, rewards. I will discuss the theoretical frameworks in which these mechanisms are generally studied and for each mechanism I will present a simple model that captures the essence of how it can be described mathematically. Examples will be given from multicellularity, eusociality, bacterial biofilms, animal and human behavior.
|10:30 to 10:45||
A Tunnicliffe (University of Cambridge)
Contributed Talk 7: Parameter Estimation for a Non-Neutral Evolutionary Process
Even with multiple time-point sampling, the estimation of the rate parameters which drive an evolutionary process can be difficult. Many existing methods work by fitting models with strong simplifying assumptions, such as fixed population size, discrete time steps between events, or a limited number of competing groups. In most cases the process is also modelled as neutral, wherein each group has equal fitness and survival or extinction are entirely down to luck. We aim to fit a more realistic model based on a birth-death process that is augmented to include both mutations which create new groups and relative fitness values for each group. We then develop a Monte Carlo method that separates simulation of the total population amount from that of the internal evolutionary dynamics, thereby increasing its overall sampling efficiency. The model and method are applied to a dataset taken from a long term E coli experiment where colony size and composition were sampled over ti me.
|10:45 to 11:00||
S O'Brien (University of Exeter)
Contributed Talk 8: Social evolution of toxic metal bioremediation in P.aeruginosa
Bacteria are often iron-limited, hence produce extracellular iron-scavenging siderophores. A crucial feature of siderophore production is that it can be an altruistic behaviour (individually costly but benefitting neighbouring cells), thus siderophore producers can be invaded by non-producing social “cheats”. Recent studies have shown that siderophores can also bind other heavy metals (such as Cu and Zn), but in this case siderophore chelation actually reduces metal uptake by bacteria. These complexes reduce heavy metal toxicity, hence siderophore production may contribute to toxic metal bioremediation. Here, we show that siderophore production in the context of bioremediation is also an altruistic trait and can be exploited by cheating phenotypes in the opportunistic pathogen Pseudomonas aeruginosa. Specifically, we show that in toxic copper concentrations: 1) siderophore non-producers evolve de novo and reach high frequencies; and 2) that producing stra in s are fitter than isogenic non-producing strains in monoculture, and vice versa in co-culture. Moreover, we show that the evolutionary effect copper has on reducing siderophore production is greater than the reduction observed under iron-limited conditions. We discuss the relevance of these results to the evolution of siderophore production in natural communities and heavy metal bioremediation.
|11:00 to 11:30||Morning Coffee|
|11:30 to 12:30||
T Pfeiffer (Massey University)
Plenary Lecture 8: Game theory for modelling microbial communities
|12:30 to 13:30||Lunch at Wolfson Court|
|19:30 to 22:00||Conference Dinner at Emmanuel College|