We develop new technologies to study how microbes interact with their environment to influence global carbon cycling. Our systems biology methods (genomics, proteomics, and computer modeling) characterize the behavior of cellular regulatory and metabolic networks. We, in turn, use molecular genetics to modify microbial genomes to test our systems biology predictions and to develop strains with novel properties. Our long term goal is to reveal the core principles allowing us to redesign and improve microbes for the sustainable production of biochemicals and for balanced environmental carbon management.
Currently, I am collaborating with the Church lab and the Gygi lab lab at Harvard Medical School, the Leschine and Blanchard Labs at UMass Amherst, the Kasif lab at BU, and Qteros Inc to study fermentation of plant biomass.
How do microbes efficiently deconstruct and ferment plant biomass? This question is critical to understand how microbes affect environmental carbon cycling and to develop lignocellulosic biomass as a cost-effective feedstock for renewable energy. We have recently developed and applied two new technologies that work towards a systems-level understanding of cellulosic fermentation in the model system Clostridium phytofermentans. This forest soil anaerobe secretes dozens of enzymes to cleave cellulose and hemicellulose and then ferments the resulting hexose and pentose sugars to ethanol and hydrogen. First, we performed a global analysis of the enzymes for cellulosic fermentation by integrating analysis of growth, fermentation, electron microscopy, and mass spectrometry-based quantification more than 2500 extra- and intracellular proteins (Project 1 below). Second, we developed tools for gene inactivation and overexpression in C. phytofermentans to determine the functions of key enzymes (Project 2 below). We are currently working to develop additional molecular tools to understand and ultimately optimize the fermentation of plant biomass.
While cellulosic biomass is the world's most abundant biological energy source (Leschine, 1995), it is primarily composed of a matrix of high molecular weight polysaccharides that can only be degraded by certain cellulolytic microbes such as C. phytofermentans. The C. phytofermentans genome encodes 161 carbohydrate-active enzymes (CAZy), highlighting the elaborate set of enzymes needed to breakdown different biomass types. We are working towards a systems-level understanding of how C. phytofermentans ferments different cellulosic substrates by combining quantitative mass spectrometry-based proteomics (ReDi proteomics) of over 2500 proteins with analyses of growth, fermentation, and electron microscopy (Fig 1). We identified the different combinations of carbohydratases used to degrade cellulose and hemicellulose as well as the repertoires of glycolytic enzymes and alcohol dehydrogenases enabling ethanol production at near maximal yields. Growth on cellulose also altered processes such as tryptophan synthesis, motility, and fatty acid metabolism. These data were distilled into a model of cellulosic fermentation that gives a blueprint to engineer microbes for more efficient conversion of biomass.
Fig 1 Integrated systems biology strategy to study cellulosic bioconversion. Cultures metabolizing different biomass substrates were examined for A growth and biomass consumption rates, B fermentation production rates and yields, and C ability of the microbe to adhere to cellulosic substrates. D Supernatant and cellular protein samples were taken for ReDi proteomics and analyzed for enzyme secretion, abundances of cellulolytic enzymes, and proteome-wide changes. E These data were integrated to identify key enzymes for each step in biomass deconstruction and fermentation. Image from Tolonen et al, 2010.
We developed methods to experimentally manipulate gene expression in C. phytofermentans. Specifically, we enabled targeted gene inactivation and over-expression using interspecific conjugation with E. coli to transfer a plasmid into C. phytofermentans that has a designed group II intron (Fig 2A) that can be targeted to insert anywhere in the chromosome (Fig 2B). In an initial study (Tolonen et al, 2009), we inactivated the cphy3367 gene, encoding the second most highly up-regulated carbohydratase on cellulose. The cphy3367 mutant strain (AT02-1) grew normally on some carbon sources such as glucose, but surprisingly had completely lost the ability to degrade cellulose (Fig 2C). Although C. phytofermentans up-regulates the expression of numerous enzymes to breakdown cellulose, a single cellulase is thus essential for cellulose degradation. This finding was the subject of a recent commentary (Wilson DB, 2009). We are applying our proteomics-based systems analyses (Project 1) to prioritize additional genes to study by targeted inactivation and to over-expression as well as developing additional tools for the genetic manipulation of C. phytofermentans.
Fig 2 A Plasmid for targeted gene inactivation in C. phytofermentans with a group II intron. B The group II intron can be re-targeted by two-step PCR to insert anywhere in the C. phytofermentans genome.C Inactivation of the cphy3367 gene in strain AT02-1 resulted in an inability to degrade cellulose.