Sir, A recent MicroGenomics review article presented the ATP-binding cassette (ABC) proteins encoded within the Escherichia coli genome (Linton and Higgins, 1998, Mol Microbiol28: 5–13). We have also analysed the ABC transporters found universally in living organisms (see Saier, 1998, In Advances in Microbiological Physiology. Poole, R.K. (ed.) 40: 81–136). Our most recent analyses led us to update the descriptions in the article of Linton and Higgins. These descriptions are completed or corrected below, and we provide further interesting information regarding some of the relevant proteins (see Table 1 in the article by Linton and Higgins). The following systems are probably exporters instead of importers: (i) f583 is most closely related to the DrrA protein of Streptomyces peucetius, a constituent of a known drug efflux pump. Homologues of DrrA are associated with one or two inner membrane proteins (f377 and f368 in the case of f583) and a protein homologous to membrane fusion proteins (MFPs, Dinh et al., 1994, J Bacteriol176: 3825–3831) present in many export systems (f332 in the case of f583). (ii) YhiG and YhiH exhibit striking sequence identity to f583, f377 and f368. YhiI would be the associated MFP. (iii) o648 is homologous to o228 and YbbA, and homologues are found in many bacterial genomes. They are accompanied by one or two hydrophobic membrane proteins, which, in the case of o648, is fused to the ATP-binding protein. These are sometimes found encoded in operons with genes encoding proteins having strong similarity to members of the MFP family. YcbE could be part of an operon encoding f278 (a conserved inner membrane protein) and f333 (a putative substrate-binding protein). This system is therefore probably an importer and not an exporter. The functions of the following two systems are known: (i) YebM is similar to proteins involved in the import of Mn2+, Zn2+ and Fe2+. YebM is associated with YebI (a conserved inner membrane protein) and YebL (a putative substrate-binding protein). This system has been shown to catalyse high-affinity zinc uptake (Patzer and Hantke, 1998, Mol Microbiol28: 1199–1210). (ii) The YjcVWX operon is known to function in the utilization of D-allose (Kim et al., 1997, J Bacteriol179: 7631–7637). Some known or putative constituents of ABC systems were not described in the Linton and Higgins review as follows: (i) the CysATW (sulphate–thiosulphate) system functions with two distinct periplasmic binding receptors, CysP (thiosulphate uptake) and SbpA (sulphate uptake). (ii) The LivFGHM (hydrophobic amino acid) system functions with two distinct periplasmic binding receptors, LivJ (leucine/isoleucine/valine receptor) and LivK (leucine-specific receptor). (iii) Homologues of f248 (cited as f284 in Table 1 of Linton and Higgins) are found in several bacterial and chloroplast genomes, and they are always associated with a conserved protein. In E. coli two such proteins (f508 and f423) flank the f248 gene. (iv) o322 is an ATP-binding protein most similar in sequence to those in the oligosaccharide ion family (family 5). It is probably part of an operon where o430 encodes the substrate-binding protein and o293 and o280 encode the cytoplasmic membrane proteins. This system is therefore most likely to be an oligosaccharide uptake system. (v) YecC belongs to the polar amino acid transporter family (family 4). It is associated with YecS (a conserved cytoplasmic membrane protein) and FliY (a putative substrate-binding protein). FliY has been identified as the cystine-binding protein, and consequently this permease probably transports this amino acid as well as diaminopimelate (B. Schneider, C. Furlong and M. H. Saier, Jr, unpublished observations). (vi) f535 is an orphan periplasmic-binding receptor that clusters in the peptide-binding receptor family. It is possible to predict the functions of the following systems. (i) YagC could be part of an operon that has been disrupted by an insertion sequence (IS) element. Partial open reading frames (ORFs) around this IS element encode protein fragments that display strong similarity to cytoplasmic membrane proteins of the oligosaccharide ion family (family 5). (ii) o530 is an ATP-binding protein homologous to the yeast GCN-20 protein, a protein involved in the regulation of translational initiation. Based on the sequence similarity observed, a comparable function in E. coli can be proposed. With respect to the phylogenetic characterization of ABC transporter protein constituents, we have provided evidence for the co-evolution of the constituents of ABC-type uptake systems from a common evolutionary origin with minimal shuffling (Tam and Saier, 1993, Microbiol Rev57: 320–346; Saurin and Dassa, 1994, Prot Sci3: 325–344; Kuan et al., 1995, Res Microbiol146: 271–278). One of us (E. Dassa) devised a tree (Fig. 1) of E. coli ATP-binding proteins using the same method as that used by Linton and Higgins (1998, ibid.). This tree proved to be similar to that reported in this paper, but since it is unrooted, we have chosen a representation that highlights the fact that exporters and importers segregated early, as documented (Saurin et al., 1999, J Mol Evol48: 22–41). This representation suggests that import and export systems separated very early during the evolution of the ABC superfamily, before divergence of the proteins that comprise either of these two groups. . Phylogenetic tree of Escherichia coli ATP-binding proteins. Groups of proteins from systems with known or predicted specificity are represented within coloured rectangles. The upper part of the figure contains all known import systems and the lower part putative export systems. A complete list of the ABC and other transporters in E. coli and in various other organisms (Paulsen et al., 1998a, J Mol Biol277: 573–592; Paulsen et al., 1998b, FEBS Lett430: 116–125) is available on the World Wide Web (http://www-biology.ucsd.edu/∼ipaulsen/transport/titlepage.html), and the classification of the ABC superfamily into 44 families as well as the classification of 200 other transporter families is described in a separate web site (http://www-biology.ucsd.edu/∼msaier/transport/titlepage.html).
No takes yet. Share an insight, caveat, or question.
Dassa et al. (1999) studied this question.