UC San Diego

Energy-coupling factor (ECF) porters catalyze uptake of vitamins and trace minerals with high affinity. They consist of two membrane constituents called S (substrate recognition) and T (energy transducing) subunits as well as one or two energizing ATPase(s), the A subunit(s). The S subunit is thought to recognize the substrate while the T subunit interacts with the ATPase. We here show that each of these three subunits is monophyletic, that the S and T subunits are homologous but distantly related to each other, and that these subunits are homologous to the integral membrane subunits of conventional ATP-binding cassette (ABC) porters. ECF porters therefore comprise an offshoot (a "sub-superfamily") of the ABC superfamily in spite of some distinctive features. We propose a pathway by which all of these transport systems may have evolved. An intragenic duplication of a genetic element encoding a 3 transmembrane segment (TMS) peptide gave rise to 6 TMS proteins, and these sometimes lost the C-terminal TMS to give 5 TMS proteins. The transmembrane subunits of the ECF porters are also homologous to certain secondary carriers, and we provide preliminary evidence that the S subunit of a thiamine porter can function by itself (secondary active transport) or when complexed with both T and A subunits (primary active transport). Phylogenetic analyses of the three ECF subunits revealed that extensive shuffling of these constituents occurred over evolutionary time although the T and A subunits, when encoded separately from the S subunits, frequently coevolved. Additionally, our genomic analyses using riboswitch regulatory sequences regulating expression of ECF transporter genes promises to reveal potential substrates of these diverse transporters. In the second part of this dissertation, I analyzed the distribution of riboswitches and used the results for comparative genome analysis. Riboswitch analysis offers a unique advantage in that the substrate binding (aptamer) domain of a riboswitch are highly conserved, allowing more accurate functional predictions. With the rapid accumulation of complete prokaryotic genomes and experimental validation of riboswitch sequences, such phylogenetic/functional analyses allow more extensive annotation of previously uncharacterized genes and comparison of metabolic pathways between different organisms. Our study is focused on the discovery of candidate riboswitches in fully sequenced bacterial genomes and regulon reconstruction of riboswitch-regulated genes, which are collected and manually curated using a RegPredict web-based tool. From these annotations we can thoroughly analyze the conservation of orthologous metabolic pathway and the distribution of different types of riboswitches and establish their probable evolutionary pathways in the Domain Bacteria. My results indicated that some riboswitches, especially those that correspond to coenzymes TPP and cobalamin, are widespread and possibly originated very early. However, many more riboswitches in our analyses are restricted to just a single Order/Family of bacteria and likely represent more recent evolutionary innovations