Cysteine-scanning mutagenesis around transmembrane segments 1 and 11 and their flanking loop regions of Tn10-encoded metal-tetracycline/H+ antiporter

Putative transmembrane helices (TM) 1 and 11 in the metal-tetracycline/H+ a ntiporter are predicted to be close to each other on the basis of disulfide cross-linking experiments of the double-cysteine mutants in the periplasmi c loop regions (Kubo, Y., Konishi, S., Kawabe, T., Nada, S., and Yamaguchi, A, (2000) J. Biol. Chem. 275, 5270-5274). In this study, each amino acid f rom Asn-2 to Gly-44 in the putative TM1 and loop1-2 regions or that from Se r-328 to Gly-366 in TM11 and its flanking regions was individually replaced with cysteine. With respect to the TMI region, 10 mutants, from T5C to L14 C, were all not reactive with N-ethylmaleimide (NEM), and from D15C to I22C , NEM-reactive and non-reactive mutations periodically appeared every two r esidues. Three mutants, M23C to V25C, were all NEM-reactive, but the degree of the latter two mutants was very low. Seven mutants, from L26C to E32C, were all highly reactive with NEM. Therefore, the region of TM1 is composed of the 21 amino acid residues from Thr-5 to Val-25. It is a partially amph iphilic helix, that is, the N-terminal (cytoplasmic) half is embedded in th e hydrophobic interior, and the C-terminal (periplasmic) half faces a water -filled channel. With respect to TM11, nine mutants, from S328C to G336C, a nd six mutants, from L361C to G366C, were all reactive with NEM. On the oth er hand, out of the 24 mutants, from L337C to S360C, 17 were not reactive w ith NEM, and the 7 NEM-reactive mutants were scattered, indicating that thi s region is a transmembrane segment. The 7 residues from Val-347 to Phe-353 including Pro-350 formed a central hydrophobic core, and the 7 NEM-reactiv e mutations were periodically distributed in its flanking regions, indicati ng that both ends of TM11 face a water-filled channel. Ala-354 is located a t about 1/3 of the length from the periplasmic end of TM11. Disulfide cross -linking experiments on double-cysteine mutants having the combination of A 354C and a cysteine-scanning mutation in the loop1-2 region indicated that loop1-2 is very flexible and close to the pepiplasmic end of TM11. Tetracyc line prevented the cross-linking formation between the periplasmic ends of TM1 and TM11; however, it did not affect the crosslinking between loop1-2 a nd TM11, indicating that the substrate-induced conformational change involv es a shift in the relative locations of TM1 and TM11.