Hydrogen profiles and localization of methanogenic activities in the highly compartmentalized hindgut of soil-feeding higher termites (Cubitermes spp.).

It has been shown that the coexistence of methanogenesis and reductive acetogenesis in the hindgut of the wood-feeding termite Reticulitermes flavipes is based largely on the radial distribution of the respective microbial populations and relatively high hydrogen partial pressures in the gut lumen. Using Clark-type microelectrodes, we showed that the situation in Cubitermes orthognathus and other soil-feeding members of the subfamily Termitinae is different and much more complex. All major compartments of agarose-embedded hindguts were anoxic at the gut center, and high H(2) partial pressures (1 to 10 kPa) in the alkaline anterior region rendered the mixed segment and the third proctodeal segment (P3) significant sources of H(2). Posterior to the P3 segment, however, H(2) concentrations were generally below the detection limit (<100 Pa). All hindgut compartments turned into efficient hydrogen sinks when external H(2) was supplied, but methane was formed mainly in the P3/4a and P4b compartments, and in the latter only when H(2) or formate was added. Addition of H(2) to the gas headspace stimulated CH(4) emission of living termites, indicating that endogenous H(2) production limits methanogenesis also in vivo. At the low H(2) partial pressures in the posterior hindgut, methanogens would most likely outcompete homoacetogens for this electron donor. This might explain the apparent predominance of methanogenesis over reductive acetogenesis in the hindgut of soil-feeding termites, although the presence of homoacetogens in the anterior, highly alkaline region cannot yet be excluded. In addition, the direct contact of anterior and posterior hindgut compartments in situ permits a cross-epithelial transfer of H(2) or formate, which would not only fuel methanogenesis in these compartments, but would also create favorable microniches for reductive acetogenesis. In situ rates and spatial distribution of H(2)-dependent acetogenic activities are addressed in a companion paper (A. Tholen and A. Brune, Appl. Environ. Microbiol. 65:4497-4505, 1999).