During mixed-acid fermentation, Escherichia coli initially translocates formate out of the cell, but re-imports it at lower pH. This is performed by FocA, the archetype of the formate-nitrite transporter (FNT) family of pentameric anion channels. Each protomer of FocA has a hydrophobic pore through which formate/formic acid is bidirectionally translocated. It is not understood how the direction of formate/formic acid passage through FocA is controlled by pH. A conserved histidine residue (H209) is located within the translocation pore, suggesting that protonation/deprotonation might be linked to the direction of formate translocation. Using a formate-responsive lacZ-based reporter system we monitored changes in formate levels in vivo when H209 in FocA was exchanged for either of the non-protonatable amino acids asparagine or glutamine, which occur naturally in some FNTs. These FocA variants (with N or Q) functioned as highly efficient formate efflux channels and the bacteria could neither accumulate formate nor produce hydrogen gas. Therefore, the data in this study suggest that this central histidine residue within the FocA pore is required for pH-dependent formate uptake into E. coli cells. We also address why H209 is evolutionarily conserved and provide a physiological rationale for the natural occurrence of N/Q variants of FNT channels.

The formate channel A (FocA) belongs to the formate-nitrite transporter (FNT) family, members of which permeate small monovalent anions. FocA from Escherichia coli translocates formate/formic acid bi-directionally across the cytoplasmic membrane during fermentative growth. Two residues are particularly well conserved within the translocation pores of FNTs: threonine-91 and histidine-209, based on E. coli FocA numbering. These residues are located at the tips of two broken transmembrane helices and control anion passage. H209 is the only charged residue within the pore and interacts with T91. Here, we addressed the role of the T91-H209 interaction network in the permeation of formate in vivo through FocA by performing an extensive amino acid-exchange study. Monitoring changes in intracellular formate using a formate-responsive fdhFP::lacZ reporter system revealed that T91 is essential for the ability of FocA to translocate formate bi-directionally. Only exchange for serine was partially tolerated, indicating that the hydroxyl group of T91 is mechanistically important. Substitution of H209 with N or Q was previously shown to convert FocA into a formate efflux channel. We show here that residue-exchanges A, I and T at this position resulted in a similar phenotype. Moreover, efflux function was confirmed for these FocA variants by measuring excreted formate in the culture medium. Substitution of bulky or charged residues for H209 prevented bi-directional formate passage. Studies using hypophosphite, a toxic analogue of formate taken up by FocA, and which causes impaired growth, confirmed that T91 and H209 substitutions essentially abolished, or drastically reduced, FocA´s translocation activity, as shown by reduction of growth rate. The exceptions were T91S- and T91Y-exchange variants that retained partial ability to take up inhibitory hypophosphite. Together, our findings indicate that T91 is essential for formate permeation in both directions; however, it is particularly important to allow anion efflux. Moreover, H209 is essential for formate uptake by FocA, strongly suggesting that protonation-deprotonation of this residue plays a role in formate uptake. Finally, our results substantiate the premise that efflux and influx of formate by FocA are mechanistically distinct processes that are controlled by the interplay between T91 and H209.