Yeast mitochondrial phosphate transport activity has been reconstituted from the import receptor (MIR) expressed as inclusion bodies in Escherichia coli. This result undermines the suggestion [Murakami, H., et al. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 3358-3362] that the MIR has been misidentified as the phosphate transport protein (PTP). PTP was solubilized with N-lauroylsarcosinate and Triton X-100 and purified with a yield of about 2 mg/L of induced bacterial culture. This PTP, reconstituted in liposomes, catalyzes phosphate uptake with a V-max [24.5 degrees C, net (zero trans), pH(i) = 8.0, pH(e) = 6.8] of 0.61 mmol of phosphate min(-1) (mg of PTP)(-1) and a K-m of 1.30 mM. This V-max is higher and the K-m about the same as that obtained with PTP purified from mitochondria. Replacement of Thr43 and Ile141 by other amino acids results in three types of PTP: (a) 2.5-5.0% V-max of wild-type PTP (PTPwt) (Thr43Cys; Thr43Ser; Ile141Cys), (b) <0.1% V-max (detection limit of assay) of PTPwt (Thr43Ala; Thr43Asp), and (c) proton transport uncoupled from phosphate transport (Ile141Cys). K-m changes are not significant. Activity of Thr43Cys confirms results obtained with mitochondrially expressed protein. Thus, yeast PTP requires Thr43 and mammalian PTP the similarly located Cys42 for high transport activity. Thr43 and Ile141 are each situated between two basic residues (LysThrArg vs ArgIleArg). Cys substitutions in either of these positions confer the same high N-ethylmaleimide sensitivity to the yeast PTPwt as displayed by the mammalian PTP. On the basis of single substitutions and a bacteriorhodopsin-like pattern of Glu126, His32, and Glu137, helices I and III (second tandem repeat) have been suggested to facilitate proton transport. Thus it is interesting that replacing Ile141, which is located near the matrix end of transmembrane helix III (similar to Thr43 and helix I), with a polar Cys perturbs helix III and its local environment sufficiently to yield uncoupled proton transport.
