Structural aspects of the gastric H,K ATPase: the M5/M6 domain and alpha beta association

This review summarizes some of the structural information that has been obtained on the gastric H,K ATPase. Methods such as tryptic digestion, site specific labeling and in vitro translation combine to provide a ten membrane segment model with however reservations as to the full transmembrane nature of M5 or M6. Labeling this region with the thiophilic luminal face reagent omeprazole provided cogent evidence that cys 813 but not cys 822 was labeled. On the other hand, cysteine mutagenesis provided evidence that removal of cys 813 did not affect inhibition of Rb transport by omeprazole whereas removal of cys 822 although not affecting ATPase activity abolished omeprazole inhibition of transport. A model to reconcile these data is presented where M5 and M6 although intramembranal are not transmembrane hairpin structures. Analysis of the region of alpha beta interaction by tryptic digestion and WGA chromatography to define those fragments of alpha that remain beta associated shows that leu 853 to arg 922 in the TM7-loop are a major region of association with the beta subunit. Yeast two hybrid analysis, when combined with these data and those from a chimeric construct, indicates that the sequence Q 907 to R 922 is the important element of interaction in the alpha subunit and no other extracytoplasmic domain was found to interact. Two regions of the beta subunit interact with this region of the alpha subunit between Q64 and N130 as well as A156 and R188. Apparently the beta subunit is folded around a small region of the large extracytoplasmic loop between TM7 and TM8, closer to TM8.