Systems biology of the butanol-producing Clostridium acetobutylicum: new source of biofuel and chemicals/COSMIC2

Lead Research Organisation: University of Nottingham
Department Name: Sch of Mathematical Sciences


By 2030, worldwide energy consumption is projected to grow by 57%. Fossil fuels cannot meet this demand, as their continued use is causing global warming and they are in any case a finite reserve that will be exhausted before the end of this century. Liquid fuels derived from natural plant material (biofuels), are the most promising alternative energy source for use in the transportation sector. In this regard, an alcohol called butanol is currently receiving considerable interest due to its superior properties compared to ethanol and biodiesel. Butanol has a higher energy content than ethanol, can make use of existing petrol supply and distribution channels, can be blended with petrol at higher concentrations without engine modification, offers better fuel economy and has, unlike ethanol, potential as aviation fuel. Traditionally, butanol is produced by a bacterium called Clostridium acetobutylicum when growing on growth media containing sugars. The fermentation process is known as the Acetone-Butanol-Ethanol (ABE) process. In the first half of the last century, the ABE process was second in importance/scale only to ethanol production from yeast. However, with the advent of the petrochemical industry, the process became uneconomic and was abandoned in the western world from the 1960s onwards. The process has recently been re-established in China and Brazil, where lower operating costs make a process based on inefficient 20th century technology economic. For adoption in the western world, a more efficient process is required. Its derivation will be reliant on improvements to the bacterial strains employed. However, the rational creation of such strains will require a more detailed understanding of the complex processes within the cell that catalyse and control the formation of fermentation products. Deriving a better understanding of the process is the overall objective of this proposal. For this, a new holistic approach termed systems biology will be used, which involves iterative cycles of mathematical modelling and experimental testing. It is made possible by developments made in an earlier funding round in which crucial gene tools and methodologies were developed that will allow precise changes to be made to selected genes and the consequences assessed.

Technical Summary

The enzymes catalyzing and controlling conversion of glucose to solvents and acids are encoded by up to 40 genes. Starting with model-driven hypotheses, specific mutants will be generated by knock-in and knock-out strategies and analyzed. Selected mutants will be grown in continuous culture, allowing the imposition of reproducible, controlled perturbations. Fermentation analysis will include substrate/product concentrations, determination of key intracellular metabolites, and transcriptome time series. These data will lead to further iterative experimentation and ultimately to a fine-tuned quantitative description of the process of solventogenesis. Workpackages:- 1] Construction of artificially controlled genes required for solvent production/regulation: ACE technology will place specific chromosomal genes under inducible control. These changes, in combination with gene knock-outs, will allow rational perturbation of the system. 2] Analysis of mutants in continuous culture under standardized conditions: strains will be grown and analyzed in continuous culture. Quantitative analysis will include substrate and product determination, identification and quantification of key intracellular metabolites, and transcriptome time series. 3] Modelling, in silico generation of hypotheses and experimental design: an iterative process of model-based hypothesis generation and experimental testing by variations of the transcriptome and the environome will be adopted for refining the model. Controlled stimuli from the environome and rapid sampling experiments will be included. 4] Data management: DaMaSys will serve as an access-controlled repository for multi-'omics' datasets, as well as a platform for communication and joint model development. Data pre-treatment, data consistency checks, data curation for modelling purposes and, in part, the cyclic interaction between model-based hypothesis generation and experimental testing will be organized into automatable workflows.

Planned Impact

Beneficaries: The ultimate goal of this project is to generate strains of Clostridium acetobutylicum that produce biobutanol with greater productivity and which can form the basis of a commercial process for biobutanol production. The major direct beneficary is, therefore, the industrial private sector concerned with chemical commodity production. A boost for the agricultural sector is also expected, as farmers will be able to profit from the demand for cellulosic waste products which form the substrates for biobutanol production. The results will also benefit national and international government green policies in helping to replace fossil fuels with biofuel. Achieving government targets in terms of green issues will also indirectly benefit the wider general public. How they benefit from this research: At the basic level, industrial producers will benefit through the availability of strains which may be employed as the basis of an economic process for biobutanol production. These beneficaries may be expected to gain a commercial advantage over competitors. The ultimate development of a biobutanol process will reduce national, and international, reliance on fossil fuels in the transportation sector, providing a cleaner environment and therefore indirectly impacting on human health. The technological developments will also provide an opportunity for export to third countries providing revenue for UK Plc. The project will also provide the opportunity for staff working directly on the project, together with postgraduate students indirectly affiliated to the project, to become trained in the arena of the strategically important areas of 'Systems Biology' and 'Bioenergy'. These skills should prove applicable to many different projects outside of butanol metabolism. To ensure that they benefit: The PI of the experimental partner programme at the University of Nottingham (Professor Nigel Minton) already has strong links with a major UK Biofuel company, who part fund other BBSRC projects at the University and with whom a commercial agreement is already in place. A high degree of collaboration will be maintained with this company, while at the same time other collaborative ventures will be explored. Through liaison with the University of Nottingham's Research and Innovation Services, Professor Minton will continue to monitor the IP and commercial potential of the research to be undertaken here, and will additionally chair a consortium wide committee concerned with IP and commercialisation of the consortia outputs. Professor King will draw upon his wide-ranging experience in industrial mathematics and multidisciplinary research in assisting in such developments and in ensuring that the mathematical modelling work encompasses the key biological mechanisms that control the productivity of biobutanol production, ensuring its impact in the above contexts. Over and above these activities, the project members will endeavour to communicate their work widely, both through the scientific and non-scientific press and through various media outlets, including websites.


10 25 50
Description Many bacteria are industrially useful for their ability to produce important chemicals from a variety of substrates. For
example, the bacterial species Clostridium acetobutylicum can consume glucose to produce the highly useful solvents
acetone, butanol and ethanol, via the so-called ABE fermentation process; optimising this process could thus make the production of these important chemicals cheaper, while reducing dependence on fossil fuels.

The metabolism of C. acetobutylicum comprises two main phases, acid-producing and solvent-producing. In batch culture, the initial supply of glucose is consumed as the culture grows, producing acids. As the pH reduces, the culture moves into stationary phase and the metabolism shifts towards solvent production, acetic and butyric acids being reabsorbed and converted into acetone and butanol. In continuous culture, however, the pH can be controlled and the above phases induced separately by altering the pH: increasing it above a critical value induces acid formation, while decreasing it induces solvent formation.

The present research (as part of a multi-centre EU consortium) sought to model a variety of aspects of C. acetobutylicum
metabolism mathematically using differential equations in order to provide understanding that could lead to improvements to the efficiency of solvent production. The modelling was iterated in the light of data from the experimental partners and in turn informed the experimental work.

The new models for the metabolic switch in continuous culture capture the chemical concentrations involved in the
metabolism alongside those of the enzymes catalysing the reactions. These models can be used to simulate how different
culture conditions or different enzyme production rates might alter the spectrum of solvents produced, guiding further
experimental work. A particular prediction is that mutations or over-expression of single genes will not be sufficient to
enhance butanol production significantly beyond wild-type levels: a more sophisticated approach targeting multiple genes is required.

A further aspect of the modelling has focussed on so called 'quorum-sensing' processes associated with communication
between bacteria. This work has clarified differences between the quorum-sensing circuitry of C. acetobutylicum and
virulent bacteria such as Staphylococcus aureus. Furthermore, the results of our modelling of the putative direct effects of
the quorum-sensing system on solventogenesis and sporulation initiation fit well with data from our experimental partners and indicate that the quorum-sensing system could be capable of inducing both solventogenic and sporulation responses (or of inducing sporulation only), thus identifying it as a potential regulator of these processes.

The models developed represent significant developments of earlier work, in particular treating the population of bacteria
as a heterogeneous mixture of a primarily acid-producing subpopulation (predominant during acid production) and a
primarily solvent-producing one (predominant during solvent production). This allowed a much improved fit to experimental
solvent data with a forward shift (from high to low pH), the results being characterised as insensitive to the values of
specific combinations of the model parameters - this has fed into experimental work to determine the concentrations of
chemicals intermediate in the reactions in order to break these indeterminancies. This modelling was also tested against a
reverse-shift experiment (low to high pH), the simulated concentrations again generally fitting well to the data; however, there is a short delay in the response of the model compared to experiments and the reverse-fitted parameters suggest (unexpectedly) that the concentration of a particular enzyme (involved in the reuptake of acids) does not change noticeably between the two phases. This could have important implications and will need to be tested in further experiments.
Exploitation Route The modelling studies of gene-regulatory and metabolic processes in clostridial bacterial could have bearing on both industrial-biotechnology and medical contexts.
Sectors Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

Description Most of the impact of the work thus far has been academic (including within the cross-national SysMO research team). In particular the first PDRA on the grant subsequently secured an MRC Fellowship on related medical applications, followed by a permanent academic post.
First Year Of Impact 2011
Sector Education
Description A systems biology approach to understanding and combating Clostridium difficile infection
Amount £385,072 (GBP)
Funding ID G1002093 
Organisation Medical Research Council (MRC) 
Sector Public
Country United Kingdom
Start 06/2011 
End 05/2014
Description Collaborations in clostridial systems biology 
Organisation University of Rostock
Country Germany 
Sector Academic/University 
PI Contribution We have worked on improving the parameter estimation in the metabolic model, helping to quantify the uncertainty in the parameters, and, in collaboration University of Rostock, we have been working on improving the metabolic model further to include more genomic regulation.
Collaborator Contribution See above.
Impact Joint research has been published: see publication listing. The collaboration is multidisciplinary, involving both dry and wet researchers.
Start Year 2010
Description COSMIC2 model dissemination 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other academic audiences (collaborators, peers etc.)
Results and Impact Models are being made publicly available via JWS Online and SysMO SEEK. Publicly accessible models

no actual impacts realised to date
Year(s) Of Engagement Activity 2010
Description Parameter estimation for modelling the metabolic shift in Clostridium acetobutylicum 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact Research seminar, University of Rostock.

no actual impacts realised to date
Year(s) Of Engagement Activity 2012