Purification of the lactose:H+ carrier of Escherichia coli and characterization of galactoside binding and transport.

The lactose carrier, a galactoside:H+ symporter in Escherichia coli, has been purified from cytoplasmic membranes by pre-extraction of the membranes with 5-sulfosalicylate, solubilization in dodecyl-O-beta-D-maltoside, Ecteola-column chromatography, and removal of residual impurities by anti-impurity antibodies. Subsequently, the purified carrier was reincorporated into E. coli phospholipid vesicles. Purification was monitored by tracer N-[3H]ethylmaleimide-labeled carrier and by binding of the substrate p-nitrophenyl-alpha-D-galactopyranoside. All purified carrier molecules were active in substrate binding and the purified protein was at least 95% pure by several criteria. Substrate binding to the purified carrier in detergent micelles and in reconstituted proteoliposomes yielded a stoichiometry close to one molecule substrate bound per polypeptide chain. Large unilamellar proteoliposomes (1-5-micron diameter) were prepared from initially small reconstituted vesicles by freeze-thaw cycles and low-speed centrifugation. These proteoliposomes catalyzed facilitated diffusion and active transport in response to artificially imposed electrochemical proton gradients (delta mu H+) or one of its components (delta psi or delta pH). Comparison of the steady-state level of galactoside accumulation and the nominal value of the driving gradients yielded cotransport stoichiometries up to 0.7 proton/galactoside, suggesting that the carrier protein is the only component required for active galactoside transport. The half-saturation constants for active uptake of lactose (KT = 200 microM) or beta-D-galactosyl-1-thio-beta-D-galactoside (KT = 50-80 microM) by the purified carrier were found to be similar to be similar to those measured in cells or cytoplasmic membrane vesicles. The maximum rate for active transport expressed as a turnover number was similar in proteoliposomes and cytoplasmic membrane vesicles (kcat = 3-4 s-1 for lactose) but considerably smaller than in cells (kcat = 40-60 s-1). Possible reasons for this discrepancy are discussed.