Development of pyrF-Based Genetic System for Targeted Gene Deletion in Clostridium thermocellum and Creation of a pta Mutant

We report development of a genetic system for making targeted gene knockouts in Clostridium thermocellum, a thermophilic anaerobic bacterium that rapidly solubilizes cellulose. A toxic uracil analog, 5-fluoroorotic acid (5-FOA), was used to select for deletion of the pyrF gene. The ΔpyrF strain is a uracil auxotroph that could be restored to a prototroph via ectopic expression of pyrF from a plasmid, providing a positive genetic selection. Furthermore, 5-FOA was used to select against plasmid-expressed pyrF, creating a negative selection for plasmid loss. This technology was used to delete a gene involved in organic acid production, namely pta, which encodes the enzyme phosphotransacetylase. The C. thermocellum Δpta strain did not produce acetate. These results are the first examples of targeted homologous recombination and metabolic engineering in C. thermocellum, a microbe that holds an exciting and promising future in the biofuel industry and development of sustainable energy resources.Conversion of cellulosic biomass using saccharolytic fermentative microorganisms without the addition of purified cellulase and hemicellulase enzymes is a promising approach for low-cost production of renewable fuels and chemicals (22, 23). Thermophilic, cellulolytic bacteria are one departure point for development of microorganisms with the requisite capabilities for such consolidated bioprocessing (CBP), with Clostridium thermocellum being exemplary in this regard. As reviewed elsewhere (6, 22), C. thermocellum is a Gram-positive organism able to ferment cellulose and products of cellulose solubilization to ethanol, acetic acid, lactic acid, formic acid, hydrogen, and CO₂. C. thermocellum appears to be a cellulose-utilizing specialist (6, 8) and produces a multienzyme cellulose-solubilizing complex termed a cellulosome (2, 3, 9).Metabolic engineering is required in order to increase the yield of ethanol or other desired products from mixed-product fermentation, such as that carried out by Clostridium thermocellum. Comprehensive work directed to this end has been carried out with genetically tractable organisms, such as Escherichia coli, resulting in high or near-theoretical yields achieved for ethanol (35, 36), other native products (21, 25), and nonnative products (7, 12). In these organisms, genetic systems involving both positive and negative selection markers have been employed in order to facilitate reuse of the same marker and to develop marker-free strains. One prominent system in the category involves use of the gene encoding the enzyme orotidine 5-phosphate decarboxylase (PyrF) (4, 11, 20, 27-29, 39). PyrF participates in de novo pyrimidine biosynthesis but is also a target for the antimetabolite 5-fluoroorotic acid (5-FOA) (4). Thus, cells lacking pyrF are uracil auxotrophs and resistant to 5-FOA, creating an opportunity whereby ectopic expression of pyrF can be selected or counterselected (4).Reliable genetic tractability has been elusive for Clostridium species. Prior to this report, the only Clostridia species in which gene deletion via homologous recombination has been demonstrated are Clostridium acetobutylicum, Clostridium perfringens, and Clostridium septicum. In the first organism, the use of a replicating plasmid for transformation followed by selection and screening for plasmid segregation resulted in a single clone that when analyzed contained a disruption in the gene of interest but not by the expected recombination events (13). The last two organisms have either an unusually high transformation frequency or feasibility for acquiring DNA from E. coli via conjugation, allowing the use of suicide plasmids (1, 16, 19). By comparison, the recently reported method of C. thermocellum transformation consists of a complex and cumbersome electroporation protocol using a custom pulse delivery system (37, 38). In our hands, efficiency of the C. thermocellum electrotransformation system does not compare with that of typical model organisms and does not enable the use of nonreplicating plasmids as a means of gene manipulation. Alternatively, group II intron technology has been used to inactivate gene targets in clostridia that were previously characterized as genetically intractable, but systems described to date have a temperature restriction that make such approaches incompatible with thermophilic clostridia (14, 15, 34).The only C. thermocellum mutant characterized genetically was isolated following a random mutagenesis and enrichment for cells that did not adhere to cellulose (43). The random mutagenesis approach is limited, in the sense that it does not lend itself well to reverse genetics, as many desired mutations lack selectable or screenable phenotypes. For instance, attempts have been made, with little success, to isolate saccharolytic thermophiles containing lesions in the pta-ack operon responsible for acetate production by selective enrichment using antimetabolites (26). In contrast, the creation of a Thermoanaerobacterium saccharolyticum Δpta-ack strain has been achieved using selectable markers that serve as a proxy for the events leading to targeted gene deletion (32). Motivated by the potential of microbial cellulose processing and the attributes of C. thermocellum, we undertook to develop a gene deletion system based on the pyrF gene.