Prof. Dr. Nir Ben-Tal / Tel Aviv University George S. Wise Faculty of Life Sciences, Tel Aviv-Yafo, Israel
"Transport determinants in cation/proton antiporters: phylogenetic analysis and simulations guide experiments"
Cation/proton antiporters (CPAs) critically contribute to cell homeostasis. With over 100,000 entries in UniProt, they comprise a large superfamily, including 13 paralogues in humans. Thus, the CPAs phylogeny, while complex and challenging to reconstruct, can be very informative, for example for deciphering the molecular determinants of specificity among these antiporters. CPAs are divided into two functional groups, electroneutral and electrogenic antiporters, exchanging one or two protons per one cation (Na+ or K+), respectively. They are also divided into two phylogenetic groups, CPA1 and CPA2, which have been suggested to respectively correlate with the two electrogenicity phenotypes. However, such a correlation has not been established, and the mechanistic underpinnings of the electrogenicity and cation selectivity are still unclear. The first part of the talk will present a phylogenetic analysis of a representative set of 6,597 antiporters [Masrati et al Nature Communs 2018, PMID: 30310075]. A well-defined, eight-residue sequence motif, was defined, which distinguishes CPA1s from CPA2s and seems to determine both their electrogenicity and cation selectivity. The analysis provides a new way to classify CPAs, and shows that contrary to previous suggestions, the CPA1/CPA2 division only partially correlates with the electrogenicity of the antiporters. We were able to identify specificity determining residues in this seemingly redundant family, which enabled us, based on predictions, to modify CPAs function. As a strong support of our findings, we successfully used mutagenesis to recover transport activity in an inactive mutant of the electrogenic CPA2 antiporter EcNhaA. This variant was originally designed to make EcNhaA electroneutral, but was inactive. We rationally designed two compensatory mutations that recovered activity, and showed that the resulting EcNhaA triple mutant became electroneutral, as it was meant to be. In the second part of the talk we will explore the transport mechanism [Masrati et al. PNAS, Accepted for publication]. We now know the structure of a handful of CPAs, including both an inward-facing (IF) conformation and an outward-facing (OF) conformation of NapA from Thermus thermophilus (TtNapA). However, we still do not have a clear understanding of their transport mechanism in atomic detail. In particular, how do they alternate between the IF and OF conformations? Two different mechanisms have been suggested: the rocking bundle vs. the elevator. Structurally, CPAs comprise a core domain and a dimerization domain, which differ in their mutual interactions between the IF vs OF conformations. According to the rocking bundle mechanism, the transition between the IF and OF conformations mostly involves a rotation motion of the core domain with respect to the dimerization domain. According to the elevator mechanism, in contrast, the transition mostly involves vertical translation of the core domain relative to the interface domain. Our all-atom molecular dynamics simulations of TtNapA, within the metadynamics framework, show that TtNapA’s conformational changes are driven predominantly by the rotation, and that translation just follows, in support of the rocking bundle mechanism. We then turn to another CPA of known structure: Escherichia coli NhaA (EcNhaA). The EcNhaA structure, in the IF conformation, was the first CPA structure to be determined (in 2005), and yet, in spite of years of effort, we still do not know the structure of an OF conformation. Thus, we conduct metadynamics simulations of ECNhaA starting from the known IF conformation, where we bias the rotation angle, and observe what the OF conformation could be. Next we designed cross-linking mutations to trap this putative conformation. Reassuringly, our experiments, using growth assays and pH assays in everted membrane vesicles, showed that in oxidizing conditions the cross-linked transporter became inactive, and that it regained activity upon reducing, when the cross-link was detached, as it should.