Saliva is the principal protective agent for the mouth and thus is of primary importance to oral health maintenance. Perturbations of salivary secretory mechanisms can consequently lead to serious oral health problems. The objective of this project is to study the membrane and cellular processes that underlie the phenomenon of salivary fluid secretion and thus to contribute to our understanding of the fluid secretory process. Because similar secretory mechanisms are thought to be common to a number of other tissues, this information should be of rather broad applicability and interest. During the present reporting period we have continued our in-depth studies of the salivary Na-K-2Cl cotransporter (NKCC1). We have also extended some of our methodology to examine the structural properties of presenilin 1, the putative proteolytic component of gamma-secretase. NKCC1 is thought to be the major Cl entry pathway into salivary acinar cells and thus to be primarily responsible for driving Cl secretion, and thereby fluid secretion, in salivary glands. Obtaining a better understanding of the structure/function relationships of this protein and its behavior in acinar cells will improve our knowledge of salivary function and dysfunction, as well as possibly providing indications of how to treat the latter. Over the past year we have concentrated on two projects involving NKCC1: (i) We have investigated the transmembrane topology and biogenesis of NKCC1 using a new method recently developed in our laboratory for this purpose. Briefly stated this method involves the construct a fusion protein consisting of the green fluorescent protein followed by a portion of the membrane spanning region of the protein under study and a C-terminal glycosylation tag. By assaying for glycosylation of the tag we can establish whether it is located in the lumen of the endoplasmic reticulum or the cytosol. Thus by systematically varying the length of the membrane spanning region of the protein included in the construct we can trace its biogenesis and topology as it integrates into the membrane. NKCC1 contains 12 hydrophobic regions that previous studies from our laboratory had indicated were membrane spanning segments. Applying our new method to NKCC1 largely confirmed these earlier results and provided several new insights regarding its biogenesis. Specifically, (1) the highly conserved intracellular loop between hydrophobic regions 2 and 3 of NKCC1 appears to be membrane inserted. ?Re-entrant loops? of this type have now been identified in several membrane transport proteins and are thought to form a part of the selectivity filter for substrate translocation. (2) The sixth hydrophobic region of NKCC1 requires the presence of the seventh and eighth hydrophobic regions in order to integrate into the membrane. This result suggests a specific interaction of these hydrophobic regions during the intramembrane folding of NKCC1. (3) Glycosylation of the endogenous glycosylation sites in NKCC1 located between the seventh and eighth hydrophobic regions does not occur until after all 12 hydrophobic regions of the protein are integrated into the membrane. This latter result indicates that a maturation event requiring the entire membrane spanning region of the protein is necessary before the endogenous glycosylation sites are made available to the glycosylation machinery within the endoplasmic reticulum. (ii) We have investigated the mechanism of integration of hydrophobic regions 11 and 12 of NKCC1 into the membrane. Together these hydrophobic regions make up a sequence only ~36 amino acids long indicating that they insert into the membrane in a hairpin-like helix-loop-helix structure. Previous studies on model proteins suggested that such helical hairpins formed as a result of specific amino acids in the short loop between the alpha-helical transmembrane regions that had a high ?turn propensity? and thus produced the tight turn required for the helix-loop-helix structure. However, our initial experiments on this natural hairpin showed that mutating amino acids in the loop to those with low turn propensity had little effect on hairpin formation. Instead our results indicate that specific interactions between hydrophobic regions 11 and 12 are responsible for this effect. Our present efforts are directed toward trying to incorporate our mutational data into a molecular model of this helical hairpin structure. Mutations in presenilin 1 have been linked to cases of familial early-onset Alzheimer's disease. This protein is thought to be the proteolytic component of gamma-secretase, the protease that is responsible for the intramembrane cleavage of a number of substrates including the beta-amyloid precursor protein, the protein that is primarily responsible for the senile plaques characteristic of Alzheimer's disease. Presenilin 1 contains ten hydrophobic regions sufficiently long to be alpha-helical membrane spanning segments. Previous topology studies agree that the N-terminus of presenilin 1 is cytosolic and hydrophobic regions 1-6 span the membrane but HR 7 does not. However, the conformation of hydrophobic regions 8-10 remains controversial. We have examined the biogenesis and transmembrane topology of this region of human presenilin 1 using the fusion protein approach described above. In contrast to previous reports, our results indicate that hydrophobic regions 8, 9 and 10 all span the membrane. Although our conclusions regarding the topology of presenilin 1 differ from previous proposals our data actually reconcile rather than contradict most earlier observations.
Showing the most recent 10 out of 22 publications
