Book of Abstracts: Albany 2009
June 16-20 2009
© Adenine Press (2008)
Role of Transmembrane Segment M6 in the Biogenesis and Function of the Yeast Plasma-Membrane Pma1 H+-ATPase
P-type ATPases, which are found throughout prokaryotic and eukaryotic cells, use the energy from ATP hydrolysis to pump cations across biological membranes. Recently, crystal structures of the mammalian Ca2+- and K+, Na+-ATPases and fungal and plant H+-ATPase, have appeared, providing a valuable framework to study the molecular mechanism of P-type ATPases. The structure includes a cytoplasmic headpiece that is folded into three discrete domains connected by a thick stalk to the membrane domain, consisting of 10 α-helices with varying lengths and inclinations. Site-directed mutagenesis of Ca2+-ATPase had located residues essential for Ca2+ transport in four of them: M4 (E309), M5 (N768 and E771), M6 (N796, T799, and D800), and M8 (E908). The crystal structure showed that side-chain oxygen atoms from these residues contribute to two Ca2+-binding sites (I and II), situated in a pocket near the middle of the membrane; three additional M4 residues (V304, A305, and I307) also furnish main-chain carbonyl oxygens to site II.
P-ATPases are noteworthy for their ability to pump a wide range of cations, including H+, Na+, K+, Mg2+, Ca2+, Cu2+, Cd2+, Mn2+. There are also major differences in cation stoichiometry, ranging from 1 H+/ATP in the plasma-membrane H+-ATPase of yeast and other fungi to 3 Na+/2 K+/ATP in the Na+,K+-ATPase of animal cells. Based on sequence alignments and cryoelectron microscopic images of the Neurospora and Arabidopsis plasma-membrane H+-ATPases and the mammalian Na+, K+-ATPase, it seems likely that a common folding pattern has been conserved throughout the P-ATPases. Thus, a reasonable guess is that the determinants of cation specificity and cation stoichiometry lie in a core of membrane segments M4, M5, M6, and M8 of these enzymes. Indeed, mutagenesis of the Na+, K+-ATPase has identified at least 8 residues in M4, M5, and M6 that are essential for cation occlusion and/or transport, and mutagenesis of the yeast Pma1 H+-ATPase has located positions in M5 and M8 at which amino acid substitutions alter the coupling between ATP hydrolysis and H+ transport (1, 2).
M6 forms part of the Ca2+-binding pocket in the sarcoplasmic reticulum ATPase, contributing T799 to site I, N796 to site II, and D800 to both. In Pma1, these residues correspond to A729, A726, and D730, respectively. It therefore seemed worthwhile to carry out Ala/Ser-scanning mutagenesis along M6 of the yeast H+-ATPase, searching for residues that may play a role in the enzyme functioning. Each mutant allele was cloned into the expression vector Ycp-2HSE and expressed in secretory vesicles (SV), as described (1, 2). The Saccharomyces cerevisiae strain SY4 used in this study carries the temperature-sensitive sec6-4 mutation which, upon incubation at 37 °C, blocks the last step in plasma membrane biogenesis and leads to the SV accumulation in the cell; SV could readily be isolated and used to assay the ATPase activity and expression.
Of the 19 mutations studied, only two (D730A and D739A) led to complete blocks in membrane trafficking that prevented the ATPase from reaching SV. Other mutations of the same residues (D730N, D730V, D739N, and D739V) gave similar results (1). This kind of behavior can be traced to a severe defect in protein folding, causing the abnormal ATPase to be retained by quality control mechanisms in the endoplasmic reticulum; consistent with misfolding, direct assays of metabolically labeled D730N, D730V, and D739V ATPases have shown that they are highly sensitive to trypsin (1). Mutations L721A, I722A, I725A, and I727A were expressed not very well (18 to 35% of the wild-type control) in SV and, accordingly, displayed ATPase activities that were very low. The remaining mutations were expressed at 46 to 100% of the wild-type level and had ATPase activities, ranging from 7 to 71% of the control. Worth noting is the stretch of 7 almost successive positions (L721, I722, F724, I725, I727, F728, and D730), starting from the extracytoplasmic end of M6, at which Ala substitutions interfered markedly with ATPase activity, biogenesis, or both; by contrast, only three (L734, Y738, and D739) of the nine Ala replacements towards the cytoplasmic end of M6 led to pronounced effects on biogenesis and/or activity. Two mutants, V723A and I736A, also had altered kinetics. Both were strikingly resistant to orthovanadate, with Ki values 20-fold higher compared to the wild-type control, also displaying 2.5- to 15-fold decreases in Km for MgATP and, in the case of V723A, an alkaline shift in pH optimum. Such changes can be accounted for by a shift in equilibrium from the vanadate-sensitive E2 conformation towards E1, which has a much lower affinity for orthovanadate but a higher affinity for MgATP.
Given the known contribution of M6 to the transport pathway of Ca2+-ATPase, it was of particular interest to ask whether any of the mutations affected H+ pumping by the Pma1 ATPase. For most of the mutants, including V723A and A729S towards the extracytoplasmic end of M6 and V731A, A735S, I736A, and A737S towards the cytoplasmic end, the pumping slope was close to that seen in the wild type. The mutants A726S (A726 corresponded to N796 of Ca2+-ATPase), A732S, and T733A, however, gave slopes significantly lower than the wild-type value, consistent with a partial uncoupling between ATP hydrolysis and H+ transport.
Thus, mutagenesis of the M6 residues gave different results compared to M8 of the Pma1 H+-ATPase (2). In M8, four of 21 substitutions were not expressed; two others were poorly expressed and were non-active. By contrast, only two M6 mutants (D730A and D739A) were not expressed (2-6%); another (I725A) was expressed poorly (18%); and the rest was well (100%) to reasonably (29%) expressed. However, among 16 expressed mutants only half was active enough to measure the ATPase activity; two of them have kinetics significantly altered. Three M6 mutants showed undercoupling, but the differences were not as dramatic as in M8 where substitutions at 5 positions led to strong or even severe uncoupling while two others caused significant overcoupling (2). Therefore, one can suggest that M6 plays an important role in H+-ATPase functioning, being probably responsible for cation selectivity similar to pmr1 ATPase (3), while M8 is mostly responsible for stoichiometry (2). Based on crystallographic structures of Ca2+-ATPase, we built a homology model showing H+ site(s) in the Pma1 H+-ATPase (Fig. 1): like Ca2+-ATPase the yeast H+-ATPase may also have two binding sites for H+ (hydroniums). This model can explain change in stoichiometry reported earlier (1, 2).
Figure 1: H+ binding sites of the yeast Pma1 H+-ATPase. Residues I331, I332, V334 in M4 correspond to V304, A305, and I307 of the site II in Ca2+-ATPase; D730 (M6) corresponds to D800 and E803 (M8) corresponds to E908 of the sites I and II in Ca2+-ATPase. Coordinated hydrated H+ are represented by circles in spheres. The homology model was built based on crystallographic structures of Ca2+-ATPase as described in (2).
The authors are grateful to Prof. C. W. Slayman (Yale University) who was a scientific adviser of this project. This study was supported in part by the Grant Number 5G12RR008124 (to the Border Biomedical Research Center (BBRC)/University of Texas at El Paso) from the National Center for Research Resources (NCRR, NIH) (MMA) and by the RFFI grant 07-04-00419, and Grant of the President of Russian Federation for the Leading Scientific Schools SS-1004.2008.4 (VVP).
References and Footnotes
1Department of Biological Sciences and Border Biomedical Research Center