BackgroundMembers of the thermophilic, anaerobic Gram-positive bacterial genus Caldicellulosiruptor grow optimally at 65 to 78°C and degrade lignocellulosic biomass without conventional pretreatment. Decomposition of complex cell wall polysaccharides is a major bottleneck in the conversion of plant biomass to biofuels and chemicals, and conventional biomass pretreatment includes exposure to high temperatures, acids, or bases as well as enzymatic digestion. Members of this genus contain a variety of glycosyl hydrolases, pectinases, and xylanases, but the contribution of these individual enzymes to biomass deconstruction is largely unknown. C. hydrothermalis is of special interest because it is the least cellulolytic of all the Caldicellulosiruptor species so far characterized, making it an ideal naïve system to study key cellulolytic enzymes from these bacteria.ResultsTo develop methods for genetic manipulation of C. hydrothermalis, we selected a spontaneous deletion of pyrF, a gene in the pyrimidine biosynthetic pathway, resulting in a strain that was a uracil auxotroph resistant to 5-fluoroorotic acid (5-FOA). This strain allowed the selection of prototrophic transformants with either replicating or non-replicating plasmids containing the wild-type pyrF gene. Counter-selection of the pyrF wild-type allele on non-replicating vectors allowed the construction of chromosomal deletions. To eliminate integration of the non-replicating plasmid at the pyrF locus in the C. hydrothermalis chromosome, we used the non-homologous Clostridium thermocellum wild-type pyrF allele to complement the C. hydrothermalis pyrF deletion. The autonomously replicating shuttle vector was maintained at 25 to 115 copies per chromosome. Deletion of the ChyI restriction enzyme in C. hydrothermalis increased the transformation efficiency by an order of magnitude and demonstrated the ability to construct deletions and insertions in the genome of this new host.ConclusionsThe use of C. hydrothermalis as a host for homologous and heterologous expression of enzymes important for biomass deconstruction will enable the identification of enzymes that contribute to the special ability of these bacteria to degrade complex lignocellulosic substrates as well as facilitate the construction of strains to improve and extend their substrate utilization capabilities.Electronic supplementary materialThe online version of this article (doi:10.1186/s13068-014-0132-8) contains supplementary material, which is available to authorized users.