Cellular vacuoles induced by Helicobacter pylori originate from late endosomal compartments

Proc. Natl. Acad. Sci. USA

Vol. 91, pp. 9720-9724, October 1994 Medical Sciences

Cellular vacuoles induced by Helicobacter pylon originate from late endosomal compartments (pists/ulcers/vwacuolar degeneratlon/vacuolatlng toin/rab) E. PAPINI*, M. DE BERNARD*, E. MILIA*, M. BUGNOLIt, M. ZERIALt, R. RAPPUOLIt, AND C. MONTECUCCO* *Dipartimento di Scienze Biomediche, Universit& di Padova, Via Trieste 75, 35121 Padova, Italy; tIstituto di Ricerche Immunobiologiche Siena, Via Fiorentina 1, 53100 Siena, Italy; and *European Molecular Biology Laboratory, Postfach 10.2209, D-6900 Heidelberg, Germany

Communicated by R. John Collier, May 31, 1994

Pathogenic strains of Helicobacter pylon ABSTRACT cause progressive vacuolatlon and death of epithelial cells. To identify the nature of vacuoles, the distribution of markers of various membrane traffic compartments was studied. Vacuoles derive from the endocytic pathway since they include the fluid-phase marker Lucifer yellow. Early endosome markers such as rab5, trerrin, and trnsferrin receptor, as well as the lysosomal hydrolase cathepsin D, are excluded from these structures. In contrast, the vacuolar membrane is specifically stained by ity-purified antibodies against rab7, a small GTPase, localized to late endosomal compartments. The labeling of rab7 on vacuolar membranes increases as vacuolation progresses, without a concomitant increase of cellular rab7. Cell vacuolation is inhibited by the microtubule-depolymerizing agents nocodazole and colchicine. Taken together, these findins indicate that the vacuoles specifically originate from late endosomal compartments.

(V-ATPase) (15), indicates that vacuoles originate from intracellular compartments endowed with a V-ATPase, such as endosomes, lysosomes, or the trans-Golgi network (TGN) (16-18). To obtain information on the origin and composition of the vacuoles, we have followed a panel of markers of different intracellular compartments involved in membrane trafficking: early endosomes (rab5, transferrin, and transferrin receptor), late endosomes (rab7), lysosomes (cathepsin D), and TGN (rab6), as well as Lucifer yellow, a fluid-phase marker (19-28). rab9 is involved in the control of the route from late endosomes to TGN (29). Among all markers tested, H. pylori-induced vacuoles are stained only by anti-rab7 antibodies, thus suggesting that they originate from late endosomal compartments. This conclusion is also supported by the effects of nocodazole and colchicine.

Strong evidence indicates that toxigenic strains of Helicobacter pylori play a major role in the development of type B gastritis, peptic ulcers, and gastric adenocarcinoma (1-3). Biopsies of H. pylori-colonized stomach epithelium show cellular swelling, expansion of endosomal compartments, and extensive vacuolation (4). A vacuolar degeneration of epithelial cells, followed by cell death, is induced in vitro by H. pylori bacterial extracts (5-7). The luminal pH of these vacuoles is acidic, as deduced from the uptake of neutral red, a membrane-permeant amine that becomes protonated in the vacuolar lumen, and from the potentiating effect of ammonia (8). Two virulence factors, produced by pathogenic strains of H. pylori are thought to be mainly responsible for this cell degeneration and death: a urease and a cytotoxin (3, 4, 7, 8). The urease hydrolyzes urea and the ammonia produced permeates membranes, as neutral red does, and accumulates as ammonium ions inside intracellular acidic compartments, thus causing their osmotic swelling (9-11). However important, this process alone cannot account for the manyfold increase in vacuole volume, because, without membrane addition, it would lead to vacuole rupture. Recently a cytotoxin has been isolated from culture supernatants of a virulent strain of H. pylori (12) and shown to cause vacuolation in vitro (12) and in vivo (13). The genes encoding this protein (vacA; H. pylori vacuolating toxin A), and a bacterial antigen (cagA) characteristic of H. pylori pathogenic strains, have been recently cloned and sequenced, and they share no similarity with any other known protein (13, 14). Thus, the molecular mechanism of action of vacA and the sequence of events leading to vacuolar cell degeneration and death remain largely obscure. The inhibition of vacuole formation by bafilomycins, specific inhibitors of the vacuolar-type ATPase proton pump

MATERIALS AND METHODS Materials. Neutral red, Lucifer yellow, nocodazole, colchicine, and 3-methyladenine were from Sigma. Rhodaminelabeled pig antibodies, anti-rabbit and anti-mouse IgG, were from Dako. Cell culture media and fetal calf serum (FCS) were from Flow Laboratories. Affinity-purified rabbit polyclonal antibodies specific for rab5, rab7, rab9, and a monoclonal anti-transferrin receptor antibody were prepared as described (24-26). Samples of affinity-purified rabbit polyclonal antibodies, anti-rab5, anti-rab6, cathepsin D, and p53, were kind gifts of Philippe Chavrier (Marseille-Luminy), Bruno Goud (Institute Pierre et Marie Curie, Paris), Bernard Hoflack (European Molecular Biology Laboratory, Heidelberg), and J. Paul Luzio (University of Cambridge, U.K.), respectively. vacA Preparation. H. pylori cytotoxic strain CCUG 17874 was grown in Brucella broth containing 0.2% cyclodextrin and Skirrow selective supplement in a micro-aerobic environment at 370C for 3-4 days. Bacterial cultures were centrifuged (16,000 x g, 30 min), and the pellet was resuspended in phosphate-buffered saline (PBS) and sonicated five times for 30 sec at 40C with a Soniprep model 150 ultrasonic generator. The material was centrifuged (40,000 x g, 20 min), and the supernatant was sterilized by filtration through 0.22-,um cellulose filters, frozen in liquid nitrogen, and stored at -800C. The procedure outlined by Cover and Blaser (12) was then followed. Cells. HeLa cells were cultured as monolayers in plastic flasks in Earle-modified minimal essential medium (MEM) containing 10% FCS in a 5% CO2 atmosphere at 370C. Twenty-four hours before experiments, cells were released with trypsin/EDTA and seeded in 96-, 48-, or 24-well titration plates in MEM/10%6 FCS at a density of 15 x 103 per cm2.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Abbreviations: vacA, Helicobacterpylori vacuolating toxin A; TGN, trans-Golgi network. 9720

e'sr*.:;t4,6XS}^

Medical Sciences:

Papini et al.

Cell Vacuolation. HeLa cells were incubated with H. pylon cytotoxin vacA in MEM/2% FCS (15 pg/ml). At different time periods, cells were inspected for vacuolation under the optical microscope. To quantify the extent of vacuolation, cells in 96-well titer plates were incubated for 8 min at 250C with 0.1 ml of freshly prepared 0.05% neutral red in PBS containing 0.3% bovine serum albumin (BSA) and washed three times with 0.2 ml of the same buffer. After addition of 0.1 ml of 70%o ethanol in water containing 0.37% HCl, absorbance was measured with a Packard Argus 400 misubtraction of absorbance at croplate reader at 534 nm with ? 405 nm. When needed, the cells were preincubated with 1 ug of nocodazole per ml or 20 A&M colchicine for 90 min or 5 mM 3-methyladenine for 30 min before the addition of H. pylon vacA, and the effect of these drugs on vacuolation was determined as above. Heat shock was performed by incubating the cells in MEM without carbonate plus 20 mM NaHepes (pH 7.4) at 43TC for 1 hr. vacA was added after cooling the cells to 37°C. Cells were incubated with serumfree MEM for 3 hr, and the toxin was then added in the same medium. Immunofluorescence of Vacuolated HeLa Cells. HeLa cells, grown on glass coverslips, were treated with vacA as above, washed twice with PBS, and then fixed with 3% (wt/vol) paraformaldehyde/0.025% (vol/vol) glutaraldehyde in PBS (pH 7.4) for 15 min. They were permeabilized with precooled

A

tbs ss,

'K ) :.;... ,@

Proc. Nati. Acad. Sci. USA 91 (1994)

9721

methanol (-20°C) for 10 sec and incubated with 0.1% NaBH4 (wt/vol) in PBS (pH 7.4) for 10 min. After washing the cells three times with PBS, primary antibody binding was visualized with pig anti-rabbit or pig anti-mouse rhodamine-labeled antibodies diluted in PBS/3% BSA. After one wash in PBS/3% BSA and three washes in PBS (20 min total) the coverslips were mounted on glass slides in elvanol. Lucifer yellow (500 pg/ml) was added to medium of already vacuolated cells or just after the addition of vacA and cells were washed with ice-cold PBS. Samples were observed and photographed with a Zeiss Axioplan fluorescence microscope. Immunoblotting with Anti-rab7 Antibodies. HeLa cells were plated in Petri dishes (8 cm in diameter) at a density of 106 cells per Petri dish and incubated with vacA (15 ,g/ml final concentration) for 160, 250, and 540 min. After three washes in PBS, the cells were scraped, centrifuged, and resuspended in 0.5 ml of homogenation buffer (250 mM sucrose/3 mM imidazole chloride/0.5 mM EDTA/0.05% SDS). Cells were then passed through a 22-gauge needle with a 1-ml syringe and the homogenate was centrifuged at 2500 rpm for 10 min. Postnuclear supernatants were collected, electrophoresed on 10%o polyacrylamide gels with a 4% polyacrylamide step gel (30), and transferred onto nitrocellulose membranes for 3 hr at 400 mA, as described (31). Membranes were treated with affinity-purified rabbit antirab7 antibodies (1:10 dilution) for 2 hr. After washing, the

-s

%UB

,'

.i n

>'

C.

S

:.'i

':'

i;X.,:: ,,,: ..S. c: ... t:f,...

*>iL'ffi';

: ?

9

}

.'

.,

..

'%" :''. ..

;,

q

..

XSciS s 1