Am J Physiol Cell Physiol Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Cell Physiol 274: C1017-C1029, 1998;
0363-6143/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Okamoto, C. T.
Right arrow Articles by Goldenring, J. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Okamoto, C. T.
Right arrow Articles by Goldenring, J. R.
Vol. 274, Issue 4, C1017-C1029, April 1998

Identification of clathrin and clathrin adaptors on tubulovesicles of gastric acid secretory (oxyntic) cells

Curtis T. Okamoto1, Sherif M. Karam2, Young Y. Jeng1, John G. Forte2, and James R. Goldenring3

1 Department of Pharmaceutical Sciences, School of Pharmacy, University of Southern California, Los Angeles 90033; 2 Department of Molecular and Cell Biology, University of California, Berkeley, California 94720; and 3 Institute for Molecular Medicine and Genetics, Medical College of Georgia, Augusta, Georgia 30912-3175

    ABSTRACT
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References

gamma -Adaptin and clathrin heavy chain were identified on tubulovesicles of gastric oxyntic cells with the anti-gamma -adaptin monoclonal antibody (MAb) 100/3 and an anti-clathrin heavy chain MAb (MAb 23), respectively. In Western blots, crude gastric microsomes from rabbit and rat and density gradient-purified, H-K-ATPase-rich microsomes from these same species were immunoreactive for gamma -adaptin and clathrin. In immunofluorescent labeling of isolated rabbit gastric glands, anti-gamma -adaptin and anti-clathrin heavy chain immunoreactivity appeared to be concentrated in oxyntic cells. In primary cultures of rabbit oxyntic cells, the immunocytochemical distribution of gamma -adaptin immunoreactivity was similar to that of the tubulovesicular membrane marker in oxyntic cells, the H-K-ATPase. Further biochemical characterization of the tubulovesicular gamma -adaptin-containing complex suggested that it has a subunit composition that is typical of that for a clathrin adaptor: in addition to the gamma -adaptin subunit, it contains a beta -adaptin subunit and other subunits of apparent molecular masses of 50 kDa and 19 kDa. From solubilized gastric microsomes from rabbit, gamma -adaptin could be copurified with the major cargo protein of tubulovesicles, the H-K-ATPase. Thus this tubulovesicular coat may bind directly to the H-K-ATPase and may thereby mediate the regulated trafficking of the H-K-ATPase at the apical membrane of the oxyntic cell during the gastric acid secretory cycle. Given the similarities of the regulated trafficking of the H-K-ATPase with recycling of cargo through the apical recycling endosome of many epithelial cells, we propose that tubulovesicular clathrin and adaptors may regulate some part of an apical recycling pathway in other epithelial cells.

hydrogen-potassium-adenosine 5'-triphosphatase trafficking; apical membrane recycling; internalization motif

    INTRODUCTION
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References

VESICULAR TRAFFICKING AND protein sorting by transport vesicles are mediated by coat proteins (7, 38, 48, 50). The regulation of the trafficking of solute and water transporters in epithelial cells represents a growing area of investigation (6). HCl secretion by the gastric oxyntic cell is a model system to study the regulation of the trafficking of ion transporters (21). Two major protein sorting steps are required for gastric HCl secretion by the oxyntic cell. In the nonsecreting, resting oxyntic cell, the gastric proton pump, the H-K-ATPase, resides in an intracellular tubulovesicular compartment lying below the apical membrane. On stimulation of acid secretion, these tubulovesicles fuse with the apical membrane (or invaginations thereof known as intracellular canaliculi). The H-K-ATPase is thus inserted into the apical membrane where it can function to acidify the lumen of the stomach. On the cessation of gastric acid secretion, the H-K-ATPase is retrieved from the apical membrane, and the tubulovesicular compartment is reestablished. Thus the regulation of the gastric HCl secretory cycle involves the regulated recycling of the H-K-ATPase to and from the apical membrane of the oxyntic cell. As has been characterized for many other vesicular trafficking and processes, the trafficking of tubulovesicles and the sorting of the H-K-ATPase are likely to be mediated by coat proteins.

Given that the H-K-ATPase is the major membrane protein in tubulovesicles, the H-K-ATPase should represent the major cargo protein for putative tubulovesicular coat proteins. The gastric H-K-ATPase belongs to the growing family of heterodimeric P-type ATPases, to which the ubiquitous Na-K-ATPase also belongs. The minimal functional unit of these ATPases is a 100-kDa catalytic alpha -subunit and a noncovalently associated, glycosylated beta -subunit. The alpha -subunit of the P-type ATPases is polytopic (has multiple transmembrane domains), and it contains a site for the binding and hydrolysis of ATP and the binding sites for cation transport. The beta -subunit is a type II transmembrane protein (NH2 terminus is cytoplasmic), and it can apparently modulate the ion transport capabilities of the associated alpha -subunit (16, 25). Although the ion transport functions of the gastric H-K-ATPase have been extensively studied (46), specific sorting signals responsible for the regulated recycling of the H-K-ATPase at the apical membrane remain obscure. However, when expressed in a heterologous epithelial cell system, each subunit of the H-K-ATPase is apically targeted (28). In the case of the alpha -subunit of the gastric H-K-ATPase (HKalpha ), the putative apical targeting signal apparently resides in the NH2-terminal half of the protein. In addition, intriguingly, all gastric H-K-ATPase beta -subunits (HKbeta ) cloned thus far contain a tetrapeptide motif in their cytoplasmic domain, FR(or Q)XY (where F = Phe, R = Arg, Q = Gln, X = any amino acid, and Y = Tyr); this motif is highly reminiscent of the internalization signal found in the transferrin receptor (28) and conforms to the consensus motif for binding to the µ-subunits of the AP-1, AP-2, and AP-3 clathrin adaptors (5, 13, 40, 41). Thus both subunits of the H-K-ATPase may have the potential to interact with tubulovesicular coat proteins via these putative sorting signals. In fact, supportive evidence for a functional role for the motif in HKbeta has been recently provided in which the Tyr in this motif appears to be involved in the targeting of the H-K-ATPase to a regulated compartment and also appears to be required for the cessation of gastric acid secretion, presumably by forming part of an internalization motif (11).

Other than two members of the Rab family of small GTPases, Rab11 and Rab25, (8, 26, 27), the proteins involved in the regulated apical recycling of the H-K-ATPase have not been characterized. Despite the potential for the H-K-ATPase to interact with clathrin adaptors, a classical clathrin coat on tubulovesicles has not been morphologically identified at the electron microscopic level (4, 20, 23, 33, 51). However, tubulovesicles are apparently derived from an elaboration of the Golgi apparatus during the development of the acid secretory machinery in oxyntic cells (19). Thus our hypothesis is that the putative tubulovesicular coat may be related to other previously characterized Golgi-associated coats, the clathrin adaptors AP-1 (47) or AP-3 (13, 55) or the nonclathrin coat COPI (15, 53).

In this study, we have identified components of a tubulovesicular coat complex. Despite the lack of morphological evidence by electron microscopy for clathrin on tubulovesicles, we find that the tubulovesicular coat contains immunoreactive clathrin and the Golgi-associated AP-1 clathrin adaptor subunits, gamma - and beta -adaptin. The localization of clathrin and the AP-1-related adaptor to tubulovesicular membranes may represent a novel localization for clathrin-coated membranes in epithelial cells. In addition, the H-K-ATPase, the major cargo protein of tubulovesicles, may interact directly with the tubulovesicular adaptor. Thus clathrin and the AP-1-related clathrin adaptor may be involved in regulating the trafficking of the H-K-ATPase during the HCl secretory cycle.

    MATERIALS AND METHODS
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References

Materials. Anti-HKalpha monoclonal antibody (MAb) (56) was a kind gift from Dr. Adam Smolka (Medical University of South Carolina, Charleston, SC). Anti-gamma -adaptin MAb 100/3 and anti-beta -adaptin MAb 100/1 (1), fish skin gelatin, poly-D-lysine hydrobromide and BSA were purchased from Sigma (St. Louis, MO). Anti-gamma -adaptin MAb 88 (mouse gamma -adaptin fragment corresponding to COOH-terminal amino acids 642-821 used as immunogen), anti-beta -adaptin MAb 74 (human beta -adaptin fragment corresponding to NH2-terminal amino acids 75-245 used as immunogen), and anti-clathrin heavy chain MAb 23 (rat clathrin heavy chain fragment corresponding to NH2-terminal amino acids 4-171 used as immunogen) were purchased from Transduction Laboratories (Lexington, KY). 6-[(N-[7-nitrobenz-2-oxa-1,3-diazol-4-yl]amino)hexanoyl]sphingosine (NBD-C6-ceramide) was purchased from Molecular Probes (Eugene, OR). Fast flow protein G-Sepharose was purchased from Pharmacia. Goat anti-mouse IgG coupled to horseradish peroxidase (HRP) and prestained molecular mass standards for SDS-PAGE were purchased from Bio-Rad (Hercules, CA). Goat anti-mouse IgG conjugated to rhodamine was purchased from Jackson Immunological Laboratories (Bar Harbor, ME). Wheat germ agglutinin (WGA)- and Ricinus communis agglutinin I (RCA I)-Sepharose were purchased from E-Y Laboratories (San Mateo, CA). 3,3'-Dithiobis(sulfosuccinimidyl propionate) (DTSSP) was purchased from Pierce Chemical (Rockford, IL). Protease inhibitors phenylmethylsulfonyl fluoride and 4-(2-aminoethyl)benzenesulfonyl fluoride HCl were purchased from Calbiochem (San Diego, CA). Protease inhibitors antipain, leupeptin, and pepstatin A were purchased from Chemicon (Temecula, CA). Lumi-Glo enhanced chemiluminescence (ECL) detection reagent was purchased from Kirkegaard & Perry Laboratories (Gaithersburg, MD). All other biochemical reagents were reagent grade.

Purification of gastric microsomes. Gastric mucosal subcellular membrane fractions and H-K-ATPase-rich microsomes were prepared from rabbit, rat, or hog gastric mucosae by differential centrifugation and discontinuous sucrose density gradient centrifugation according to established protocols (63). Purified gastric microsomes were collected in the density gradient media in aliquots of 300 µl and stored at -80°C. Gastric microsomes from rat gastric mucosa were purified on 10-40% continuous sucrose gradients according to the protocol of Crothers et al. (12). Purified gastric microsomes are virtually all oriented with the cytoplasmic membrane leaflet facing outward.

SDS-PAGE and related procedures. Protein determinations were made using the bicinchoninic acid protein assay (Pierce Chemical). SDS-PAGE was performed according to the protocol of Laemmli (35). Due to the sensitivity of the H-K-ATPase to extended boiling, samples containing H-K-ATPase were boiled for 2 min in sample buffer before being loaded onto gel. In the absence of H-K-ATPase, samples were boiled for 5 min. For silver staining of SDS gels, the protocol of Heukeshoven and Dernick (32) was used. Urea SDS-PAGE gels were run according to the protocol of Ahle et al. (1).

For Western blotting, dilutions of primary antibodies in PBS-0.05% Tween 20 were MAb 100/3, 1:5,000; anti-clathrin MAb 23, 1:1,000; MAb 100/1, 1:500; anti-beta -adaptin MAb 74, 1:1,000; anti-gamma -adaptin MAb 88, 1:1,000; and MAb 2/2E6, 1:200 (cell culture supernatant). Goat anti-mouse-HRP secondary antibody was used at 1:20,000 dilution. Blocking of nitrocellulose was done in 5% nonfat milk in PBS-Tween 20. HRP was detected by ECL, and the signal was visualized on Kodak Bio-Max X-ray film.

Immunofluorescent labeling of isolated rabbit gastric glands. Isolated rabbit gastric glands (3) were either fixed in 3.7% formaldehyde in PBS and subsequently permeabilized in 0.1% Triton X-100 in PBS or fixed and permeabilized in cold (-20°C) methanol. After blocking in either 0.66% fish skin gelatin or 0.1% BSA in PBS, glands were stained in suspension. All primary antibodies were used at 1:100 dilution in PBS-0.05% Tween 20-0.66% fish skin gelatin or 0.1% BSA, and all secondary antibodies were used at 1:500 dilution in the same buffer. Glands were immobilized on polylysine-coated coverslips before viewing with a Zeiss Axioskop epifluorescence microscope.

Labeling of isolated rabbit gastric glands with NBD-C6-ceramide. Isolated glands were washed several times with sterile 10 mM HEPES-buffered minimal essential medium and incubated with NBD-C6-ceramide (5 µM) for 20 min at 4°C. Labeled glands were washed in HEPES-buffered minimal essential medium and then incubated in the same medium for 30 min at 37°C. Labeled glands were examined under the microscope and immediately photographed. The images were digitized and processed with Adobe Photoshop.

Immunofluorescent labeling and scanning confocal microscopy of cultured oxyntic cells. Primary cultures of rabbit oxyntic cells were prepared as previously described (9, 58). Cells maintained in culture for 48 h were fixed in 4% paraformaldehyde for 15 min at 4°C. Cells were permeabilized with 0.3% Triton X-100 in 15% donkey serum for 30 min and then incubated with either MAb 100/3 (1:1,000) or anti-HKalpha (1:2,000) for 2 h at 22°C. Specific labeling was localized with Cy5-donkey anti-mouse IgG. All cells were double labeled with BODIPY FL phallacidin (Molecular Probes). Cells were visualized using scanning confocal microscopy (Molecular Dynamics, Sunnyvale, CA).

Stripping of gastric microsomal coat proteins. Two hundred micrograms of purified gastric microsomes (27 or 32% layer) from rabbit were stripped of coat proteins by two washes in 0.5 M Tris · HCl, pH 7.0, 2 mM Na-EDTA, and 0.2 mM dithiothreitol, according to the protocol of Keen et al. (34). Samples were separated into high-speed supernatants and pellets. The proteins in the supernatants were concentrated by precipitation in ice-cold 10% trichloroacetic acid. Samples were analyzed by Coomassie blue staining for total protein and by Western blot for the distribution of gamma -adaptin and HKbeta between the supernatants and pellets.

Buffers for immunoprecipitation. Triton dilution buffer (TDB) consists of 2.5% (or 5%) Triton X-100, 100 mM triethanolamine HCl, pH 8.6, 100 mM NaCl, 5 mM Na-EDTA, 0.02% NaN3, and protease inhibitors [4-(2-aminoethyl)benzenesulfonyl fluoride HCl, phenylmethylsulfonyl fluoride, leupeptin, antipain, and pepstatin].

Mixed micelle buffer (MMB) consists of 1% Triton X-100, 0.2% SDS, 150 mM NaCl, 20 mM triethanolamine HCl, pH 8.6, 5 mM Na-EDTA, 5% sucrose, 0.2% NaN3, and protease inhibitors.

Final wash buffer (FWB) is the same as MMB, except that detergents and sucrose are omitted.

Triton X-100, triethanolamine, and glycerol (TTG) consists of 1% Triton X-100, 100 mM triethanolamine HCl, pH 8.6, 10% glycerol, 5 mM Na-EDTA, 1 mM Na3VO4, and protease inhibitors. In the absence of glycerol, this buffer is referred to as TT.

Triton X-100, glycerol, and HEPES (TGH) consists of 1% Triton X-100, 10% glycerol, 50 mM Na-HEPES, pH 7.3, 1 mM Na3VO4, and protease inhibitors. This buffer was adapted from the protocol of Sorkin and Carpenter (57).

Immunoprecipitation of gastric microsomal coat complex with MAb 100/3. Purified gastric microsomes from rabbit were solubilized in 2.5% TDB and incubated overnight with MAb 100/3 and protein G-Sepharose. Immune complexes were washed three times in MMB and once in FWB. Immunoprecipitates were analyzed by silver staining of SDS gels.

Copurification of gamma -adaptin with solubilized H-K-ATPase. Purified gastric microsomes (100 µg) from rabbit were solubilized in either TTG, TT, or TGH. The solubilized samples were incubated with 40 µl WGA- or RCA I-Sepharose bead suspension (equivalent to 100-160 µg of lectin) for 2 h at 4°C. Lectin precipitates were washed with MMB and FWB before SDS-PAGE. Samples were analyzed on Western blots for gamma -adaptin and HKbeta . There was no difference in the amount of gamma -adaptin coimmunoprecipitated with the H-K-ATPase when the detergent-treated samples were centrifuged at 100,000 g to remove unsolubilized material; thus the high-speed centrifugation step was normally omitted. Also, as a control, microsomes were solubilized in 1% SDS before dilution with TGH and lectin affinity chromatography.

Cross-linking of gastric microsomal coat proteins to H-K-ATPase. Gastric microsomal proteins from rabbit were cross-linked with DTSSP according to a protocol adapted from Simpson et al. (54). Purified gastric microsomes (200 µg) were diluted into an equal volume of 50 mM Na-HEPES-2 mM MgCl2, to which DTSSP was added to 2 mM from a freshly prepared stock solution of 20 mM in dimethylformamide. The samples were cross-linked for 30 min at room temperature, after which unreacted DTSSP was quenched with 150 mM glycine. SDS was added to 1%, and the samples were boiled for 2 min. The samples were then diluted with TDB and subjected to WGA affinity chromatography overnight at 4°C as described above. The isolated, cross-linked sample was analyzed by Western blots for gamma -adaptin. This experiment was performed twice.

    RESULTS
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References

Distribution of clathrin and gamma -adaptin in crude gastric mucosal membrane fractions and in density gradient-purified gastric microsomes. To test the hypothesis that oxyntic cell tubulovesicles interact with Golgi-related vesicular coat proteins, the distribution of clathrin heavy chain and the gamma -adaptin subunit of the Golgi-associated AP-1 clathrin adaptor was determined by Western blot analysis of gastric mucosal subcellular membrane fractions (Fig. 1). Figure 1A shows Coomassie blue-stained electrophorograms of rabbit gastric homogenates fractionated by differential centrifugation (lanes 1-4) and of subsequent purification of H-K-ATPase-rich membrane vesicles from the crude microsomal pellet on discontinuous sucrose density gradients (lanes 5-7). Microsomes sedimenting at the 27 and 32% sucrose interfaces are highly enriched in the H-K-ATPase; HKalpha is the most prominent protein band in Coomassie blue-stained gels (Fig. 1A, lanes 5 and 6). The amount of HKalpha band correlates well with the amount of H-K-ATPase enzymatic activity in these membrane preparations. The greatest amounts of HKalpha protein and highest specific ATPase activity (not shown) were found in the 27% layer (Fig. 1A, lane 5). By virtue of the enrichment of the H-K-ATPase in the 27 and 32% layers of density gradient-purified microsomes, these membrane fractions represent fractions enriched in oxyntic cell tubulovesicles.


View larger version (83K):
[in this window]
[in a new window]
  		Fig. 1.   A: Coomassie blue-stained SDS gels of crude gastric membrane fractions (lanes 1-4) and of sucrose density gradient fractions (lanes 5-7) from rabbit gastric mucosa (24 µg protein/lane). Lane 1, nuclear pellet; lane 2, mitochondrial pellet; lane 3, microsomal pellet; lane 4, high-speed supernatant; lane 5, 27% sucrose layer; lane 6, 32% sucrose layer; lane 7, pellet from density gradient (i.e., >32%). Apparent molecular masses are shown, and identity of molecular mass markers are (left, from top) myosin, beta -galactosidase, phosphorylase B, BSA, ovalbumin, and carbonic anhydrase; alpha -subunit of the gastric H-K-ATPase (HKalpha ) is indicated (right). B: distribution of clathrin in crude gastric mucosal membrane fractions (lanes 1-4) and in sucrose density gradient fractions (lanes 5-7). Proteins (24 µg/lane) were analyzed by Western blot with monoclonal antibody (MAb) 23 as described in MATERIALS AND METHODS, and immunoreactive bands were visualized by enhanced chemiluminescence (ECL). All Western blots were exposed for 2 min. Position of prestained molecular mass markers myosin (203 kDa) and beta -galactosidase (118 kDa) are shown. C: distribution of gamma -adaptin in crude gastric mucosal membrane fractions (lanes 1-4) and in sucrose density gradient fractions (lanes 5-7). Proteins (24 µg/lane) were analyzed by Western blot with MAb 100/3 and visualized by ECL. Signals were developed for 2 min. D: anti-beta -adaptin immunoreactivity in membrane fractions from sucrose density gradients (lanes 5-7). Proteins (24 µg/lane) were analyzed by Western blot with MAb 74. Immunoreactivity was detected by ECL with a 2-min exposure.

Clathrin (Fig. 1B) and gamma -adaptin (Fig. 1C) were found in all crude membrane fractions but tended to be most enriched in the microsomal pellet, where most of the H-K-ATPase activity fractionated. Microsomes sedimenting at the 32% sucrose barrier appear to contain the highest specific content of clathrin (Fig. 1B, lane 6) and gamma -adaptin (Fig. 1C, lane 6). Lower, but significant, amounts of clathrin and gamma -adaptin were found in the 27% density gradient fraction (Fig. 1, B and C, lane 5). The lower specific content of clathrin and gamma -adaptin in the H-K-ATPase-rich membranes of the 27% layer may represent vesicles with a lower amount of clathrin and gamma -adaptin per vesicle, relative to the vesicles sedimenting at the 32% interface. Alternatively, the 27% layer may contain "uncoated" as well as coated vesicles, resulting in the apparently lower specific content of the coat proteins. The microsomal membranes constituting the pellet of the sucrose density gradient (Fig. 1, A and C, lane 7), although extremely low in H-K-ATPase specific activity, usually contained significant amounts of gamma -adaptin and clathrin. The clathrin coat proteins on membranes associated with the pellet of the sucrose density gradient may represent coat proteins that are associated with other membranes from oxyntic and chief cells that are poor in H-K-ATPase content, such as Golgi membranes.

Clathrin adaptors and their homologues are typically composed of four subunits: two large subunits (e.g., gamma  and beta 1 or alpha  and beta 2, ~100 kDa and above), a medium-sized subunit (µ chain, ~50 kDa), and a small subunit (sigma  chain, ~20 kDa) (7, 13, 47, 54). Thus density gradient-purified gastric microsomal membranes were probed for the presumptive beta -subunit with the anti-beta -adaptin MAb 74. This MAb is clearly reactive with a beta -adaptin on rabbit gastric microsomes (Fig. 1D), and its distribution in membrane fractions closely parallels that of gamma -adaptin (Fig. 1C) and clathrin (Fig. 1B).

The association of clathrin coat proteins with purified gastric microsomes is also observed in species other than rabbit. Western blot analysis of density gradient-purified microsomes from hog gastric mucosa revealed that these membranes are enriched in gamma -adaptin (not shown). In addition, rat gastric microsomes were assayed for clathrin and gamma -adaptin by fractionating microsomal membranes on a continuous linear sucrose density gradient and assaying for the distribution of H-K-ATPase (Fig. 2A), clathrin (Fig. 2B), gamma -adaptin (Fig. 2C), and beta -adaptin (Fig. 2D) by Western blots. Although the distributions of clathrin, gamma -adaptin, and beta -adaptin are more widespread than that of the H-K-ATPase, the fractions most enriched in clathrin and adaptin immunoreactivity correspond to the peak of H-K-ATPase immunoreactivity. A strict interpretation of these data would be that, by this approach, H-K-ATPase-rich microsomes are not separable from those containing clathrin and gamma -adaptin. On the other hand, a liberal interpretation of these data suggest that immunoreactivity of clathrin and gamma -adaptin appears to cofractionate with H-K-ATPase immunoreactivity. On closer inspection, the distribution of clathrin and adaptins in the continuous sucrose density gradient fractions appears to be made up of two peaks (fractions 11-15 and fractions 18-23). This result suggests that two populations of H-K-ATPase-rich, clathrin-coated membranes of differing densities may exist. A similar bimodal distribution has been reported for vesicle-associated membrane protein (VAMP) and syntaxin 3 on gastric microsomes fractionated on continuous linear sucrose gradients (43). These two populations of clathrin-coated vesicles detected in continuous linear sucrose gradients may represent the two distinct membrane populations fractionated on discontinuous sucrose gradients as described above (Fig. 1A, lanes 5 and 6).


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 2.   Fractionation of H-K-ATPase and clathrin coat proteins on continuous linear 10-40% sucrose gradients. Crude gastric microsomal pellet from rat gastric mucosa was further fractionated on a continuous linear 10-40% sucrose gradient; 31 fractions were collected (fraction 1 is top of gradient) and analyzed by SDS-PAGE and Western blotting. For HKalpha , detection was performed directly on blot with alkaline phosphatase-conjugated secondary antibody; for coat proteins, visualization of signal was done by ECL with exposure times of 2 min. Molecular mass markers are same as in Fig. 1, B-D. A: negative image of immunoblot of HKalpha with anti-HKalpha MAb. Corresponding fraction numbers are shown above image. B: immunoblot of clathrin heavy chain with MAb 23. C: immunoblot of gamma -adaptin with MAb 88. D: immunoblot of beta -adaptin with MAb 74. Gels are representative of results from 3 gradients.

In summary, the copurification of gamma -adaptin-containing membranes with density gradient-purified gastric microsomes from three different species (rabbit, hog, and rat) is consistent with the presence of high-affinity binding sites for a gamma -adaptin-containing complex on gastric microsomes that, in turn, may serve as docking sites for clathrin. Given the overlapping distributions of the H-K-ATPase, clathrin, and gamma -adaptin in these membrane fractions, oxyntic cell tubulovesicles may represent a novel class of clathrin-coated vesicles in epithelial cells.

Immunofluorescent labeling of clathrin and gamma -adaptin in oxyntic cells. The functional secretory unit of the gastric mucosa is the gastric gland (21). In vivo, the gland is mainly composed of four types of epithelial cells: surface mucous cells, mucous neck cells, zymogen (pepsinogen)-secreting chief cells, and HCl-secreting oxyntic (parietal) cells. Gastric glands can be isolated by collagenase digestion of gastric mucosa. These isolated rabbit gastric glands are primarily composed of larger, bulging HCl-secreting oxyntic cells interspersed with the smaller mucous neck cells and chief cells, as shown in the phase-contrast micrograph in Fig. 3A. In isolated glands, the apical membranes of the oxyntic and nonoxyntic cells form a central lumen that can be delineated by staining apical membranes with the lectin Helix pomatia agglutinin conjugated with FITC (Fig. 3B). A typical distribution of oxyntic cells within an isolated gland is shown in Fig. 3C, in which oxyntic cells have been stained with a MAb against HKbeta (10).


View larger version (140K):
[in this window]
[in a new window]
  		Fig. 3.   Distribution of clathrin and gamma -adaptin in gastric glands. A: phase-contrast micrograph of isolated rabbit gastric glands. Larger, bulging oxyntic cells (O) and smaller chief cells (C) are indicated. B: staining of central glandular lumen (lu) with FITC-conjugated HPA in fixed, unpermeabilized gastric glands. C: immunofluorescent staining of beta -subunit of the gastric H-K-ATPase (HKbeta ) with MAb 2/2E6 in fixed gastric glands. Immunostaining protocols are outlined in MATERIALS AND METHODS. Anti-HKbeta staining reflects a typical distribution of oxyntic cells in an isolated gland. Within body of oxyntic cell, distribution of HKbeta largely reflects distribution of tubulovesicular compartment. D-F: immunofluorescent staining of clathrin heavy chain with MAb 23 in fixed, permeabilized gastric glands. Clathrin immunoreactivity in oxyntic cells is indicated by arrowheads; in chief cells, by arrows. G-I: immunofluorescent staining of gamma -adaptin in isolated rabbit gastric glands. Fixed and permeabilized isolated gastric glands were stained with anti-gamma -adaptin MAb 100/3. gamma -Adaptin immunoreactivity in oxyntic cells is indicated by arrowheads; in chief cells, by arrows. Bars, 10 µm for all micrographs.

To provide supportive evidence that clathrin resides on oxyntic cell tubulovesicles, isolated rabbit gastric glands were immunostained for clathrin heavy chain. Most of the clathrin immunoreactivity in glands appeared to be concentrated around the lumen of the gland. This distribution suggests that clathrin is concentrated at the apical pole of the oxyntic and nonoxyntic cells (Fig. 3, D-F). Closer inspection of the distribution of anti-clathrin staining in oxyntic cells confirmed that clathrin immunoreactivity appeared to be concentrated at the apical pole (Fig. 3D). Diffuse, less intense anti-clathrin staining was also evident in the supranuclear and perinuclear regions of oxyntic cells (Fig. 3E). Although they were not apparently precisely coincident, the subcellular distributions of anti-clathrin and anti-HKbeta immunoreactivity appeared to overlap somewhat (Fig. 3C).

Isolated gastric glands were also immunostained for gamma -adaptin (Fig. 3, G-I). Oxyntic cells reacted strongly to the well-characterized anti-gamma -adaptin MAb 100/3 (1), suggesting that gamma -adaptin, or an immunoreactive homologue thereof, is relatively abundant in oxyntic cells. The subapical (Fig. 3G), supranuclear (Fig. 3I), and perinuclear distribution of immunostaining within oxyntic cells overlaps with that of anti-HKbeta immunostaining (Fig. 3C). Relative to the distribution of clathrin (Fig. 3, D-F), gamma -adaptin appeared to be more abundant in the supranuclear and perinuclear regions. However, there also appears to be some overlap in staining for gamma -adaptin and clathrin, particularly in the subapical region of oxyntic cells (Fig. 3, D, G, and I). Although the limit of resolution of the immunofluorescent signal does not allow for the precise assignment of gamma -adaptin and clathrin to specific organelles, the distribution of immunoreactivity is consistent with a tubulovesicular localization for gamma -adaptin and clathrin. Moreover, these immunocytochemical data are consistent with the results from the Western blot analysis of gastric microsomes. Together, these data represent the first demonstration of clathrin coat proteins on oxyntic cell membranes.

A consistent, but less intense, punctate staining by MAb 100/3 was also observed near the basolateral membrane in oxyntic cells (Fig. 3H). This staining may represent the Golgi apparatus, since this staining pattern correlates with the staining pattern in oxyntic cells of the vital dye for Golgi membranes, NBD-C6-ceramide (Fig. 4). These data suggest that most of the gamma -adaptin and clathrin in oxyntic cells appears to be associated with a membrane compartment that is distinct from the Golgi apparatus, presumably the tubulovesicular compartment or the apical membrane.


View larger version (92K):
[in this window]
[in a new window]
  		Fig. 4.   Golgi apparatus in gastric glandular cells. Negative images of staining of Golgi apparatus in viable isolated gastric glands with vital dye NBD-C6-ceramide. Freshly isolated rabbit gastric glands were stained with vital dye NBD-C6-ceramide as described in MATERIALS AND METHODS. Oxyntic and chief cells are indicated. Staining in oxyntic cells is indicated by arrowheads; in chief cells, by arrows. Bars, 10 µm.

Staining of gamma -adaptin in chief cells and mucous neck cells was significantly less than in oxyntic cells (Fig. 3, H and I). Anti-gamma -adaptin immunostaining that was observed within chief cells and mucous neck cells appeared to be restricted to the apical pole. This staining pattern in nonoxyntic cells may be more typical of many epithelial cells, in which the Golgi apparatus is located toward the apical pole of epithelial cells. The apical staining pattern of gamma -adaptin in chief cells (Fig. 3H) correlates well with that of NBD-C6-ceramide (Fig. 4), supporting the conclusion that the structures in nonoxyntic cells stained by MAb 100/3 are Golgi membranes.

Immunofluorescent labeling of gamma -adaptin in primary cultures of oxyntic cells. Isolated oxyntic cells maintained in primary culture assume a morphological arrangement distinct from their tissue and glandular form (9, 58). During the isolation of oxyntic cells, the intracellular canaliculi are pinched off at the lumen, resulting in the formation of intracellular canalicular vacuoles (representing the apical membrane). Thus cultured oxyntic cells tend to acquire a more simplified morphology, thereby facilitating the subcellular localization of proteins. The apical membrane and subapical tubulovesicular compartment may be more easily identified by staining for F-actin (Fig. 5, b and e) and H-K-ATPase (Fig. 5d), respectively.

The distribution of gamma -adaptin immunoreactivity in cultured oxyntic cells was revealed by scanning confocal fluorescence microscopy (Fig. 5a). The subcanalicular distribution of gamma -adaptin immunostaining correlates well with that of anti-H-K-ATPase staining (Fig. 5d). The observed subapical distributions of both gamma -adaptin and H-K-ATPase provide further immunocytochemical support for the localization of gamma -adaptin on tubulovesicles of oxyntic cells.

Soluble pools of clathrin coat proteins and stripping of gamma -adaptin from gastric microsomal membranes. The gastric microsomal (tubulovesicular) clathrin coat proteins were further characterized biochemically. Coat proteins exist in membrane-bound and soluble pools (50). As expected, besides the membrane-bound pool of clathrin and gamma -adaptin on gastric microsomes, the high-speed supernatant of gastric mucosal homogenates also contains clathrin and gamma -adaptin (Fig. 1, B and C, lane 4). However, this soluble pool of clathrin and gamma -adaptin may also be derived from sources other than tubulovesicles, such as from Golgi membranes of oxyntic and chief cells.

An established protocol to strip gamma -adaptin from gastric microsomal membranes was used to characterize the clathrin and adaptor coat proteins on gastric microsomes and to rule out the possibility that the gamma -adaptin antibodies were spuriously cross-reacting with other gastric microsomal proteins, particularly integral membrane proteins. The membrane-bound complex should be stripped from membranes by washing in 0.5 M Tris · HCl, pH 7.0 (34). Density gradient-purified microsomes were washed with this buffer, and the distribution of total protein, gamma -adaptin, and HKbeta between the membrane pool and the stripped, soluble pool was qualitatively analyzed on Coomassie blue-stained gels and Western blots (Fig. 6).


View larger version (115K):
[in this window]
[in a new window]
  		Fig. 5.   Distribution of gamma -adaptin in isolated, cultured oxyntic cells analyzed by confocal scanning microscopy. a-c: Fixed, permeabilized oxyntic cell double labeled for gamma -adaptin and F-actin with MAb 100/3 and BODIPY FL phallacidin, respectively. a: Immunostaining of gamma -adaptin. b: Staining of F-actin at canalicular membrane. c: Merged image. d-f: Fixed, permeabilized oxyntic cell double labeled for H-K-ATPase and F-actin. d: Immunostaining of H-K-ATPase with an anti-HKalpha MAb. e: Staining of F-actin at canalicular membrane. f: Merged image. Bars, 2 µm.


View larger version (43K):
[in this window]
[in a new window]
  		Fig. 6.   Stripping of gamma -adaptin from purified gastric microsomes. Putative coat proteins on gastric microsomes were stripped from 27 or 32% microsomes by washing in 0.5 M Tris · HCl, pH 7.0, as described in MATERIALS AND METHODS. Samples of starting membranes (SM, 24 µg protein, 1/8 of total starting membranes), stripped membranes (P, entire sample), and stripped soluble proteins (Sup, entire sample) were analyzed by SDS-PAGE and Western blot. A: Coomassie blue-stained gel. Migration of molecular mass markers is shown. B: Western blot probed for gamma -adaptin with MAb 100/3 (top) and for HKbeta with MAb 2/2E6 (bottom). Signal from ECL detection was developed after 2 min. Migration of prestained molecular mass markers is shown. Identities of prestained molecular mass markers are beta -galactosidase (118 kDa) and BSA (86 kDa).

The Coomassie blue-stained gel of the sample of supernatant proteins shows that several 100-kDa proteins were stripped from gastric microsomes (Fig. 6A); these proteins could represent the ~100-kDa subunits found in all clathrin adaptor complexes. In addition, protein bands of ~160-180 kDa were found in the panel of stripped proteins; one of these could represent the clathrin heavy chain.

Western blots showed that, of the total recovered gamma -adaptin, over one-half of the gamma -adaptin was found in the panel of proteins stripped from gastric microsomes with 0.5 M Tris (Fig. 6B, top). As a control, the distribution of the integral membrane subunit of the H-K-ATPase, HKbeta , between these pools was also determined (Fig. 6B, bottom). As expected, all of the recovered HKbeta were found exclusively in the membrane pellet. These results suggest that gamma -adaptin behaves as a vesicular coat complex and that the gamma -adaptin immunoreactivity in gastric microsomes is not a result of artefactual cross-reactivity.

Characterization of the other subunits of the clathrin adaptor complex on gastric microsomes. Although immunoreactivity with the anti-beta -adaptin MAb 74 was robust in density gradient-purified gastric microsomes (Figs. 1D and 2D), we were surprised to find that little, if any, reactivity to the well-characterized anti-beta -adaptin MAb 100/1 (1) was observed in gastric microsomes, although such reactivity was robust in control membranes from rabbit adrenal gland (not shown). This dichotomy in reactivity to the anti-beta -adaptin MAbs was also observed in gastric microsomes from rat and hog (not shown), suggesting that the beta -adaptin of gastric microsomal membranes may be distinct from that of Golgi-associated AP-1 and/or may possess some unique features. The basis for the difference in immunoreactivity of the gastric microsomal beta -adaptin with one MAb (MAb 74) and not another (MAb 100/1) is currently being investigated.

In another approach to characterize the other subunits of the tubulovesicular adaptor, gastric microsomes were solubilized under nondenaturing conditions and gamma -adaptin-containing complexes were immunoprecipitated with MAb 100/3. The immunoprecipitate was analyzed by SDS-PAGE (Fig. 7). In addition to the expected ~100-kDa band for gamma -adaptin (confirmed by Western blot of immunoprecipitates; not shown), other bands were also clearly coimmunoprecipitated; they migrated with apparent molecular masses of ~100 kDa (clearly observed in SDS-urea gels, Fig. 7C), 50 kDa, and 20 kDa. The apparent molecular masses of the other proteins coimmunoprecipitating with gamma -adaptin correlate well with those of the subunits comprising Golgi-associated AP-1 adaptors. Thus, based on the profile of the apparent molecular masses of the other subunits coimmunoprecipitating with gamma -adaptin and on the immunologic reactivity of the beta -adaptin, the tubulovesicular adaptor complex appears to be closely related to, but perhaps not identical to, the well-characterized Golgi-associated AP-1 adaptor.


View larger version (59K):
[in this window]
[in a new window]
  		Fig. 7.   Characterization of other subunits of putative tubulovesicular adaptor complex by immunoprecipitation of native gamma -adaptin-containing complexes. Purified gastric microsomes, 27 and 32% layers, were each solubilized under nondenaturing conditions as described in MATERIALS AND METHODS. gamma -Adaptin-containing complexes were immunoprecipitated with MAb 100/3 and protein G-Sepharose. Immunoprecipitations were also performed with protein G-Sepharose alone as a negative control. Immunoprecipitates were run on SDS-PAGE, and proteins in immunoprecipitates were stained with silver. Migration of molecular mass markers is shown. Identities of markers are same as in Fig. 1A. A: 10% SDS-PAGE. Immunoprecipitates from gastric microsomes from 27% layer (lane 1) and 32% layer (lane 2). Putative adaptor subunits are marked with asterisks. Other bands present in immunoprecipitate are derived from either Ig subunits or proteins leaching from protein G-Sepharose beads. B: 7.5% SDS-PAGE. Immunoprecipitate with MAb 100/3 from gastric microsomes from 32% layer. Putative adaptor subunits (large and medium chains) are indicated by asterisks. C: 7.5% SDS-urea PAGE. Lane 1: negative control; immunoprecipitate with protein G-Sepharose alone from gastric microsomes from 27% layer. Lane 2: immunoprecipitate with MAb 100/3 from 27% layer. Lane 3: immunoprecipitate with MAb 100/3 from gastric microsomes from 32% layer. Both ~100-kDa adaptor subunits are indicated with asterisks.

Copurification of gamma -adaptin with solubilized H-K-ATPase. The H-K-ATPase is the major membrane protein of purified gastric microsomes. Thus this enzyme would represent the major cargo protein for a tubulovesicular coat complex.

To determine whether the H-K-ATPase represents the cargo molecules for the tubulovesicular clathrin coat complex, the ability of gamma -adaptin to copurify with solubilized H-K-ATPase was assessed. The H-K-ATPase was solubilized from gastric microsomes under nondenaturing conditions with Triton X-100 diluted into three different buffers, TTG, TT, and TGH. Under these nondenaturing solubilizing conditions, HKalpha and HKbeta remain associated with each other, and they can be recovered together by lectin affinity chromatography by the binding of glycosylated HKbeta (comprising the overwhelming majority of tubulovesicular glycoproteins) to the lectins WGA or RCA I conjugated to Sepharose (42). The solubilized, lectin-purified H-K-ATPase was then assayed on Western blots for the presence of gamma -adaptin. As shown in Fig. 8A, gamma -adaptin is coprecipitated with the solubilized H-K-ATPase when WGA-Sepharose (lane 1) or RCA I-Sepharose (lane 2), but not Sepharose CL-2B (lane 3), is used. In addition, there appears to be a qualitative correlation in the amount of H-K-ATPase recovered by lectin affinity chromatography (as determined by HKbeta immunoreactivity) and that of gamma -adaptin (TTG and TGH, lane 1 vs. lane 2; TTG vs. TT, lane 2). Finally, the copurification of gamma -adaptin with the H-K-ATPase was not observed when gastric microsomes were solubilized under denaturing conditions before lectin affinity chromatography (Fig. 8B).


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 8.   Copurification of gamma -adaptin with solubilized H-K-ATPase. A: coisolation of gamma -adaptin with solubilized H-K-ATPase. Purified gastric microsomes were solubilized under nondenaturing conditions with TTG, TT, or TGH buffers as described in MATERIALS AND METHODS. Solubilized H-K-ATPase was purified by lectin affinity chromatography on wheat germ agglutinin (WGA)-Sepharose (lane 1) or Ricinus communis agglutinin I-Sepharose (lane 2), or, as a negative control, on Sepharose CL-2B (lane 3). Recovered H-K-ATPase was assayed on Western blots for HKbeta with MAb 2/2E6 and for coprecipitating gamma -adaptin with MAb 100/3. Blots incubated with ECL were exposed to film for 2 min for HKbeta and for 20 min for gamma -adaptin. Gels are representative of results from 3 experiments. Migration of prestained molecular mass markers is shown. Identities of prestained markers are myosin (203 kDa), beta -galactosidase (118 kDa), BSA (86 kDa), and ovalbumin (52 kDa). B: sensitivity of copurification of gamma -adaptin with H-K-ATPase to SDS. Purified gastric microsomes were solubilized in presence (+) or absence (-) of 1% SDS before isolation of H-K-ATPase by lectin affinity chromatography on WGA-Sepharose. Recovered H-K-ATPase was assayed on Western blots for HKbeta with MAb 2/2E6 and coprecipitating gamma -adaptin with MAb 100/3. Blots incubated with ECL were exposed to film for 2 min for HKbeta and for 20 min for gamma -adaptin. C: cross-linking of gamma -adaptin with H-K-ATPase. Purified gastric microsomes were cross-linked in 2 mM 3,3'-dithiobis(sulfosuccinimidyl propionate) (DTSSP) as described in MATERIALS AND METHODS. Cross-linked microsomal proteins were solubilized by boiling in 1% SDS. H-K-ATPase was recovered by lectin affinity chromatography as described in MATERIALS AND METHODS and assayed on Western blots for gamma -adaptin with MAb 100/3.

In an alternate approach to test whether gamma -adaptin can be copurified with the H-K-ATPase, gastric microsomes were cross-linked with the thiol-cleavable cross-linker DTSSP. After cross-linking, gastric microsomes were boiled in SDS. HKbeta (and anything cross-linked to it) was recovered by affinity chromatography on WGA. As shown in Fig. 8C, gamma -adaptin was specifically associated with the cross-linked sample. Thus, with two different approaches, gamma -adaptin can be copurified with the H-K-ATPase. The most straightforward explanation is that the gamma -adaptin-containing adaptor complex binds to the H-K-ATPase. Alternatively, gamma -adaptin may also be interacting with other uncharacterized tubulovesicular proteins that, in turn, are noncovalently associated with the H-K-ATPase. Experiments are in progress to determine which specific tubulovesicular protein serves as the receptor for the gamma -adaptin-containing coat.

    DISCUSSION
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References

On secretagogue-induced stimulation of gastric acid secretion in the oxyntic cell, the gastric proton pump, the H-K-ATPase, is recruited to the apical membrane from subapical H-K-ATPase-rich tubulovesicles. With cessation of gastric acid secretion, the H-K-ATPase is retrieved from the apical membrane as the tubulovesicular compartment is reestablished. Given the extensive, relatively synchronous vesicular trafficking steps associated with this process, the oxyntic cell may represent a good model system in which to study the regulation of membrane trafficking and protein sorting by vesicular coat proteins in epithelial cells.

We have identified clathrin and an AP-1 adaptor complex on oxyntic cell tubulovesicles. This report is the first report of clathrin on oxyntic cell tubulovesicles; these tubulovesicles may represent a novel compartment to which clathrin and AP-1 adaptors are localized. The presence of clathrin on tubulovesicles is intriguing, given that morphologically distinct clathrin coats have not been reported in any of the numerous electron microscopic analyses of the exocytic and endocytic trafficking of the H-K-ATPase associated with HCl secretion by the oxyntic cell (4, 20, 23, 33, 51). This incongruence suggests that tubulovesicular clathrin may be functionally and/or structurally different from previously characterized conventional clathrin. Relevant to this speculation, a novel clathrin heavy chain gene was recently cloned and characterized; this gene is selectively expressed in skeletal muscle in adults but ubiquitously in all of the limited number of fetal tissues that were assayed (36). Alternatively, the clathrin light chains (7) or beta -adaptin (2, 24), the other proteins thought to influence the polymerization of clathrin and the structure of clathrin coats, may regulate an alternative mode of polymerization of clathrin in oxyntic cells such that morphologically distinct clathrin baskets are not visible by electron microscopy. Molecular characterization of this clathrin heavy chain and its accompanying light chains will be required to determine the relationship of the tubulovesicular clathrin to the other previously cloned isoforms and may account for its novel morphology in oxyntic cells. Indeed, with respect to the structure of clathrin coats, a similar situation may exist for tubulovesicular elements in early endosomal compartments; the presence of conventional clathrin on endosomes had gone unnoticed until recent critical immuno-electron-microscopic analyses were performed (59).

The tubulovesicular compartment of the oxyntic cell may represent a gross elaboration of apical recycling endosomes found in other epithelial cells. Several similarities exist between membrane trafficking during gastric acid secretion and apical recycling in other epithelial cells. First, as with regulation of trafficking associated with the apical recycling endosome (17, 30, 45), gastric acid secretion is stimulated by increasing intracellular cAMP (62). Second, as with apical trafficking (19, 29, 37), gastric acid secretion is highly dependent on intact microfilaments and associated proteins such as ezrin (31, 39, 64, 65). Finally, Rab11, a marker of the tubulovesicular compartment (8, 27), also regulates trafficking in another recycling endosomal compartment, the pericentriolar (perinuclear) recycling endosome (61). Thus this tubulovesicular coat complex may represent an adaptor complex that is common to many epithelial cells and is involved in the regulation of membrane traffic to and/or from an apical recycling endosome.

For this tubulovesicular adaptor coat, the cargo protein appears to be the H-K-ATPase. Although direct binding of the tubulovesicular coat to the H-K-ATPase has not been definitively demonstrated, the H-K-ATPase appears to reside in a complex with the tubulovesicular coat. The simplest explanation is that the tubulovesicular coat binds to the alpha - and/or beta -subunit of the H-K-ATPase, similar to the manner in which clathrin adaptors bind to motifs present in the cytoplasmic domains of other membrane proteins (5, 38, 40, 41, 60). In this regard, the cytoplasmic domain of HKbeta contains a tetrapeptide sequence [FR(or Q)XY] highly reminiscent of the internalization signal of the transferrin receptor (28); binding of adaptors to the H-K-ATPase may be mediated by this putative sorting signal. Indeed, recent work has shown that the motif YXXphi (where Y = Tyr, X = any amino acid, and phi  = bulky aromatic amino acid) is an optimal one for interaction with the medium chains of AP-1 and AP-2 adaptors (5, 40). However, as tempting as this speculation may be, identification of the binding site on HKalpha and/or HKbeta for the tubulovesicular adaptor will rely on in vitro binding assays. Recently, evidence has been presented suggesting that the Tyr in the putative motif in HKbeta serves as a signal to target HKbeta to a regulated compartment and is required for the cessation of acid secretion (11), implying that this Tyr also serves as an internalization motif. However, because of the uncertainty of the physiological evidence presented in Ref. 11, we must be circumspect in our interpretation. Alternatively, binding of coat complexes to tubulovesicles may involve another non-ATPase tubulovesicular membrane protein, such as a receptor for soluble N-ethylmaleimide-sensitive factor (SNARE) (8, 38, 43). However, because the coat complex is coimmunoprecipitated with the H-K-ATPase, any other putative docking receptor for the adaptor would also have to be noncovalently associated with the H-K-ATPase. The final alternative is that the coat complex recognizes both the H-K-ATPase and another docking protein.

The relative abundance of clathrin and adaptors on oxyntic cell tubulovesicles, together with the previous identification of SNARE proteins on tubulovesicles (8, 43), supports the hypothesis that the mechanism of HCl secretion involves membrane translocation and fusion events (22) rather than an osmotically regulated expansion (secreting state) and collapse (nonsecreting state) of preexisting tubules that are contiguous with the apical membrane (44). Thus coincident on tubulovesicles are key components of the cellular machinery necessary for both sorting of the H-K-ATPase (clathrin coat) and vesicular fusion (syntaxins and VAMPs).

In conclusion, a clathrin and an AP-1 adaptor coat complex are associated with the tubulovesicular compartment of the gastric oxyntic cell. This finding appears to represent another distinct non-Golgi localization of gamma -adaptin (14). The step at which the coat proteins regulate the recycling of its cargo, the H-K-ATPase, is unknown. In a study by Schofield et al. (52), coated vesicles, although not distinctly clathrin-coated vesicles, were observed during the return of oxyntic cells from the stimulated (secreting) state to the resting (nonsecreting) state; thus clathrin and adaptors may regulate the reuptake of the H-K-ATPase from the apical membrane with the cessation of HCl secretion. The function of clathrin and AP-1 adaptors in the gastric acid secretory cycle may be tested by determining the sensitivity of particular steps of the cycle to reagents [e.g., brefeldin A, guanosine 5'-O-(3-thiotriphosphate), and aluminum fluoride] that have been used to modify AP-1 adaptor function in other cell types (49, 64). Further biochemical and molecular characterization of the tubulovesicular coat and its interaction with its cargo should help provide important details with respect to the role of clathrin and adaptors in gastric acid secretion. Moreover, characterization of the role of adaptor coat proteins in the regulation of gastric acid secretion may consequently provide important clues regarding the regulation of apical recycling in many other epithelial cells.

    ACKNOWLEDGEMENTS

We thank Dr. Catherine Chew for providing cultured oxyntic cells and Dr. Sarah Hamm-Alvarez for a critical reading of this manuscript.

    FOOTNOTES

This work was supported by a Zumberge Award from the University of Southern California, American Heart Association National Grant-in-Aid 96-1205, a New Investigator Award from the American Association of Colleges of Pharmacy, a Pilot Project Grant from the University of Southern California Gastrointestinal and Liver Diseases Center (to C. T. Okamoto), National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-10141 (to J. G. Forte), DK-38063 (to J. R. Goldenring), and DK-4370 (to J. R. Goldenring), and the Kuwait Foundation for the Advancement of Sciences KFAS-95-07-02 (to S. M. Karam).

Address for reprint requests: C. T. Okamoto, Dept. of Pharmaceutical Sciences, School of Pharmacy, University of Southern California, 1985 Zonal Ave., Los Angeles, CA 90033.

Received 3 October 1997; accepted in final form 6 January 1998.

    REFERENCES
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References

1.   Ahle, S., A. Mann, U. Eichelsbacher, and E. Ungewickell. Structural relationships between clathrin assembly proteins from the Golgi and the plasma membrane. EMBO J. 7: 919-929, 1988[Medline].

2.   Ahle, S., and E. Ungewickell. Identification of a clathrin binding site in the HA2 adaptor protein complex. J. Biol. Chem. 264: 20089-20093, 1989[Abstract/Free Full Text].

3.   Berglindh, T., and K. J. Öbrink. A method for preparing isolated glands from the rabbit gastric mucosa. Acta Physiol. Scand. 96: 150-159, 1976[Medline].

4.   Black, J. A., T. M. Forte, and J. G. Forte. The effects of microfilament disrupting agents on HCl secretion and ultrastructure of piglet gastric oxyntic cells. Gastroenterology 83: 595-604, 1982[Medline].

5.   Boll, W., H. Ohno, Z. Songyang, I. Rapoport, L. S. Cantley, J. S. Bonifacino, and T. Kirchhausen. Sequence requirements for the recognition of tyrosine-based endocytic signals by clathrin AP-2 complexes. EMBO J. 15: 5789-5795, 1996[Medline].

6.   Bradbury, N. A., and R. J. Bridges. Role of membrane trafficking in plasma membrane solute transport. Am. J. Physiol. 267 (Cell Physiol. 36): C1-C24, 1994[Abstract/Free Full Text].

7.   Brodsky, F. M. New fashions in vesicle coats. Trends Cell Biol. 7: 175-179, 1997.[Medline]

8.   Calhoun, B. C., and J. R. Goldenring. Two Rab proteins, VAMP-2, and SCAMPs are present on immunoisolated parietal cell tubulovesicles. Biochem. J. 325: 559-564, 1997.

9.   Chew, C. S., M. Ljungstrom, A. Smolka, and M. R. Brown. Primary culture of secretagogue-responsive parietal cells from rabbit gastric mucosa. Am. J. Physiol. 256 (Gastrointest. Liver Physiol. 19): G254-G263, 1989[Abstract/Free Full Text].

10.   Chow, D. C., and J. G. Forte. Characterization of the beta -subunit of the H,K-ATPase using an inhibitory monoclonal antibody. Am. J. Physiol. 265 (Cell Physiol. 34): C1562-C1570, 1993[Abstract/Free Full Text].

11.   Courtois-Coutry, N., D. Rousch, V. Rajendran, J. B. McCarthy, J. Geibel, M. Kashgarian, and M. J. Caplan. A tyrosine based signal targets H/K-ATPase to a regulated compartment and is required for the cessation of gastric acid secretion. Cell 90: 501-510, 1997[Medline].

12.   Crothers, J. M., Jr., D. C. Chow, and J. G. Forte. Omeprazole decreases H+-K+-ATPase protein and increases permeability of oxyntic secretory membranes in rabbits. Am. J. Physiol. 265 (Gastrointest. Liver Physiol. 28): G231-G241, 1993[Abstract/Free Full Text].

13.   Dell'Angelica, E. C., H. Ohno, C. E. Ooi, E. Rabinovich, K. W. Roche, and J. S. Bonifacino. AP-3: an adaptor-like protein complex with ubiquitous expression. EMBO J. 16: 917-928, 1997[Medline].

14.   Dittié, A. S., N. Hajibagheri, and S. A. Tooze. The AP-1 adaptor complex binds to immature secretory granules from PC12 cells, and is regulated by ADP-ribosylation factor. J. Cell Biol. 132: 523-536, 1996[Abstract/Free Full Text].

15.   Duden, R., G. Griffiths, R. Frank, P. Argos, and T. E. Kreis. beta -COP, a 110 kd protein associated with non-clathrin-coated vesicles and the Golgi complex shows homology to beta -adaptin. Cell 64: 649-665, 1991[Medline].

16.   Eakle, K. A., M. A. Kabalin, S.-G. Wang, and R. A. Farley. The influence of beta -subunit structure on the stability of Na,K-ATPase complexes and interaction with K+. J. Biol. Chem. 269: 6550-6557, 1994[Abstract/Free Full Text].

17.   Eker, P., P. Kaae Holm, B. van Deurs, and K. Sandvig. Selective regulation of apical endocytosis in polarized Madin-Darby canine kidney cells by mastoparan and cAMP. J. Biol. Chem. 269: 18607-18615, 1994[Abstract/Free Full Text].

18.   Fath, K. R., and D. R. Burgess. Golgi-derived vesicles from developing epithelial cells bind actin filaments and possess myosin-I as a cytoplasmically oriented peripheral membrane protein. J. Cell Biol. 120: 117-127, 1993[Abstract/Free Full Text].

19.   Forte, G. M., L. Limlomwongse, and J. G. Forte. The development of intracellular membranes concomitant with the appearance of HCl secretion in oxyntic cells of the metamorphosing bullfrog tadpole. J. Cell Sci. 4: 709-727, 1969[Abstract/Free Full Text].

20.   Forte, J. G., J. A. Black, T. M. Forte, T. E. Machen, and J. M. Wolosin. Ultrastructural changes related to functional activity in gastric oxyntic cells. Am. J. Physiol. 241 (Gastrointest. Liver Physiol. 4): G349-G358, 1981[Abstract/Free Full Text].

21.   Forte, J. G., and A. Soll. Cell biology of hydrochloric acid secretion. In: Handbook of Physiology. The Gastrointestinal System. Bethesda, MD: Am. Physiol. Soc., 1989, sect. 6, vol. III, chapt. 11, p. 207-228.

22.   Forte, J. G., and X. Yao. The membrane-recruitment-and-recycling hypothesis of gastric HCl secretion. Trends Cell Biol. 6: 45-48, 1996.[Medline]

23.   Forte, T. M., T. E. Machen, and J. G. Forte. Ultrastructural changes in oxyntic cells associated with secretory function: a membrane recycling hypothesis. Gastroenterology 73: 941-955, 1977[Medline].

24.   Gallusser, A., and T. Kirchhausen. The beta-1-subunit and beta-2-subunit of the AP complexes are the clathrin coat assembly components. EMBO J. 12: 5237-5244, 1993[Medline].

25.   Geering, K., A. Beggah, P. Good, S. Girardet, S. Roy, D. Schaer, and P. Jaunin. Oligomerization and maturation of Na,K-ATPase: functional interaction of the cytoplasmic NH2 terminus of the beta  subunit with the alpha  subunit. J. Cell Biol. 133: 1193-1204, 1996[Abstract/Free Full Text].

26.   Goldenring, J. R., K. R. Shen, H. D. Vaughn, and I. M. Modlin. Identification of a small GTP-binding protein, rab25, expressed in the gastrointestinal mucosa, kidney, and lung. J. Biol. Chem. 268: 18419-18422, 1993[Abstract/Free Full Text].

27.   Goldenring, J. R., C. J. Soroka, K. R. Shen, L. H. Tang, W. Rodriguez, H. D. Vaughn, S. A. Stoch, and I. M. Modlin. Enrichment of rab11, a small GTP-binding protein, in gastric parietal cells. Am. J. Physiol. 267 (Gastrointest. Liver Physiol. 30): G187-G194, 1994[Abstract/Free Full Text].

28.   Gottardi, C. J., and M. J. Caplan. An ion-transporting ATPase encodes multiple apical localization signals. J. Cell Biol. 121: 283-293, 1993[Abstract/Free Full Text].

29.   Gottlieb, T. A., I. E. Ivanov, M. Adnesnik, and D. D. Sabatini. Actin microfilaments play a critical role in endocytosis at the apical but not the basolateral surface of polarized epithelial cells. J. Cell Biol. 120: 695-710, 1993[Abstract/Free Full Text].

30.   Hansen, S., and J. E. Casanova. Gsalpha stimulates transcytosis and apical secretion in MDCK cells through cAMP and protein kinase A. J. Cell Biol. 126: 677-687, 1994[Abstract/Free Full Text].

31.   Hanzel, D., H. Reggio, A. Bretscher, J. G. Forte, and P. Mangeat. The secretion-stimulated 80K phosphoprotein of parietal cells is ezrin, and has properties of a membrane cytoskeletal linker in the induced apical microvilli. EMBO J. 10: 2363-2373, 1991[Medline].

32.   Heukeshoven, J., and R. Dernick. Simplified method for silver staining of proteins in polyacrylamide gels and the mechanism of silver staining. Electrophoresis 6: 103-112, 1985.

33.   Ito, S., and G. C. Schofield. Studies on the depletion and accumulation of microvilli and changes in the tubulovesicular compartment of mice parietal cells in relation to gastric acid secretion. J. Cell Biol. 63: 364-382, 1974