Reduction of the P5A-ATPase Spf1p phosphoenzyme by a Ca2+-dependent phosphatase

Abstract

P5 ATPases are eukaryotic pumps important for cellular metal ion, lipid and protein homeostasis; however, their transported substrate, if any, remains to be identified. Ca 2+ was proposed to act as a ligand of P5 ATPases because it decreases the level of phosphoenzyme of the Spf1p P5A ATPase from Saccharomyces cerevisiae. Repeating previous purification protocols, we obtained a purified preparation of Spf1p that was close to homogeneity and exhibited ATP hydrolytic activity that was stimulated by the addition of CaCl 2. Strikingly, a preparation of a catalytically dead mutant Spf1p (D 487 N) also exhibited Ca 2+-dependent ATP hydrolytic activity. These results indicated that the Spf1p preparation contained a copurifying protein capable of hydrolyzing ATP at a high rate. The activity was likely due to a phosphatase, since the protein i) was highly active when pNPP was used as substrate, ii) required Ca 2+ or Zn 2+ for activity, and iii) was strongly inhibited by molybdate, beryllium and other phosphatase substrates. Mass spectrometry identified the phosphatase Pho8p as a contaminant of the Spf1p preparation. Modification of the purification procedure led to a contaminant-free Spf1p preparation that was neither stimulated by Ca 2+ nor inhibited by EGTA or molybdate. The phosphoenzyme levels of a contaminant-free Spf1p preparation were not affected by Ca 2+. These results indicate that the reported effects of Ca 2+ on Spf1p do not reflect the intrinsic properties of Spf1p but are mediated by the activity of the accompanying phosphatase.

Highly exposed segment of the Spf1p P5A-ATPase near transmembrane M5 detected by limited proteolysis

“…Total membranes from eight liters of yeasts expressing the recombinant protein were obtained as previously described [ 24 , 44 ]. Briefly, the microsomal membranes were solubilized at 4°C for 30 min by adding 2 g of Triton X-100/g of total membrane proteins and the supernatant was loaded onto a column with 2-ml nickel-nitrilotriacetic acid—agarose resin (Qiagen).…”

“…Briefly, the microsomal membranes were solubilized at 4°C for 30 min by adding 2 g of Triton X-100/g of total membrane proteins and the supernatant was loaded onto a column with 2-ml nickel-nitrilotriacetic acid—agarose resin (Qiagen). The purification was carried out following the protocol for the production of “pure Spf1p” previously described [ 44 ]. Finally, the protein was eluted in purification buffer containing 0.005% C 12 E 10 and 150 mM imidazole.…”

“…The ATPase activity was estimated at 28°C by measuring the release of Pi from ATP using the colorimetric method of Baginski [ 45 ] as previously described [ 44 ]. Reaction medium was identical to that used for proteolysis and contained 1 μg of Spf1p supplemented with PC plus 3 mM ATP.…”

ATP hydrolytic activity of purified Spf1p correlate with micellar lipid fluidity and is dependent on conserved residues in transmembrane helix M1

“…Spf1p was expressed and purified using a protocol that was previously optimized to remove contaminants that can interfere with Spf1p activity [ 12 , 23 ]. The purified preparations were confirmed to contain mainly Spf1p protein as determined by amino acid analysis and mass spectrometry, although in some of the produced fractions we could identify glyderaldehyde-3-phosphate dehydrogenase as a minor contaminant ( S1 Fig ).…”

“…Spf1p was expressed and purified using a protocol that was previously optimized to remove contaminants that can interfere with Spf1p activity [12,23]. The purified preparations were confirmed to contain mainly Spf1p protein as determined by amino acid analysis and mass…”