This laboratory has developed a procedure for using beta- lactamase (beta-1a) for analysis of the topology of transmembrane proteins expressed in yeast. A C-terminal beta-1a reporter is fused to N-terminal fragments of target proteins via the P fragment of Killer preprotoxin. In fusions in which this P-beta-1a reporter is lumenal, processing of P in a late Golgi compartment by the Kex2 protease results in secretion of beta-1a. Analysis of Kex2- dependent beta-1a secretion, with and without an intervening signal sequence S, provides positive identification of both cytoplasmic (internal) and "periplasmic" (external) fusion sites. A model for the Ste2 receptor was produced that agreed with predictions. Corroborative data should result from analysis of the size, N- glycosylation pattern and protease sensitivity of the intact fusion proteins expressed in a kex2 mutant, or expressed in vitro. The technique is being used to analyze the topology of Pma1, the yeast plasma membrane proton ATPase, a model for the family of P-type ATPase ion pumps found in all cell types. Any ambiguities in deduced topology will be resolved using a sandwich technique, using either beta-1a or P-beta-1a as the inserted reporter. Analyses of Pma1 function by site-directed mutagenesis have identified many residues important to the conserved cytoplasmic kinase and phosphatase domains, but have had very limited success in identifying residues critical to proton channel function. Our preliminary Pma1 topology identifies new candidate residues; if the topology is confirmed, the phenotypes of mutants in these residues should clarify the ion pump mechanism. %%% A thorough understanding of protein function (and the ability to use such an understanding to engineer proteins with desired properties) depends on knowledge of three-dimensional structure. Unfortunately, the very properties of integral membrane proteins that allow their transmembrane segments to interact with the lipid biolayer make it impossible to apply the usual approaches for purification and crystallization that are normally employed for determining 3-D protein structures. Thus, structural information on membrane-spanning proteins is particularly difficult to obtain. This is frustrating, because some of the most interesting proteins from a functional point of view are membrane-spanning proteins (receptors, pumps, channels, etc.). This project offers a clever approach to the dilemma, which involves topological analysis of membrane proteins through genetic engineering. The idea is to construct chimeric mutants of the membrane protein of interest in which the introduced piece has a known proteolytic cleavage site that only gets cleaved if it is facing the lumen of the Golgi. By determining which constructs undergo cleavage, one can deduce the locations of the transmembrane regions of the protein, and from that deduce 3-D structure. Although the project focuses on a specific transmembrane proton pump, the approach can be generalized to a wide variety of biologically important proteins. Many downstream biotechnological applications of both the approach and the structural information obtained can easily be envisioned.
