Am J Physiol Gastrointest Liver Physiol 285: G1225–G1234, 2003.
First published July 31, 2003; 10.1152/ajpgi.00236.2003.
Identification of an apical Cl⫺/HCO3⫺ exchanger in
gastric surface mucous and duodenal villus cells
Jie Xu,1 Sharon Barone,1 Snezana Petrovic,1,2 Zhaohui Wang,1
Ursula Seidler,3 Brigitte Riederer,3 Krishnamurthy Ramaswamy,4,5
Pradeep K. Dudeja,4,5 Gary E. Shull,6 and Manoocher Soleimani1,2
Departments of 1Medicine and 6Molecular Genetics, Biochemistry, and Microbiology,
University of Cincinnati, Cincinnati 45267; 2Veterans Affairs Medical Center, Cincinnati, Ohio 45220;
3
University of Tubingen, D-72076 Tubingen, Germany; and 4Department of Medicine, University of
Illinois at Chicago, and 5Veterans Affairs Medical Center at West Side, Chicago, Illinois 60612
Submitted 22 May 2003; accepted in final form 28 July 2003
Address for reprint requests and other correspondence: M. Soleimani, Division of Nephrology and Hypertension, Dept. of Medicine, Univ. of Cincinnati, 231 Albert Sabin Way, MSB G259, Cincinnati, OH 45267-0585 (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
anion exchanger isoform 4; chloride/bicarbonate exchange;
parietal cells; mucous cells; acid injury
ONE OF THE MAJOR MECHANISMS
http://www.ajpgi.org
0193-1857/03 $5.00 Copyright © 2003 the American Physiological Society
G1225
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
for protection of the surface epithelium of the stomach from injury by gastric
acid is the presence of a layer of mucus that contains
HCO3⫺ (5, 8, 28), both of which are secreted by mucous
cells. As acid diffuses into the mucus gel, it is neutralized by the secreted HCO3⫺, resulting in the formation
of a gradient in which the pH at the luminal surface of
the epithelium is relatively neutral (5). It has been
suggested that the HCO3⫺ secreted from surface cells
during gastric acid secretion may originate in part
from parietal cells (10, 11). Intracellular HCO3⫺, generated during acid secretion, exits the basolateral
membrane of the parietal cell via electroneutral Cl⫺/
HCO3⫺ exchange (21, 23) and alkalinizes the blood,
which perfuses the basal aspect of surface epithelial
cells (10, 11). The available evidence suggests that the
HCO3⫺ is taken up by mucous cells, most likely via
Na⫹-HCO3⫺ cotransport (4, 25), and is then secreted
across the luminal membrane of surface mucous cells,
thereby providing protection from gastric acid.
The mechanism of HCO3⫺ secretion by surface epithelial cells has been studied by several groups but
remains controversial. In frog fundic mucosa, HCO3⫺
secretion was reported to have no effect on the transepithelial electrical potential and to be dependent on
luminal Cl⫺, suggesting that electroneutral Cl⫺/HCO3⫺
exchange was involved (9). These results have been
disputed (31), however, and there is evidence for a
HCO3⫺ conductance pathway (4). Nevertheless, electroneutral Cl⫺/HCO3⫺ exchange remains an attractive
possibility because the diffusion of H⫹ and Cl⫺ into the
mucus layer would both neutralize outwardly transported HCO3⫺ and provide Cl⫺ for inward transport,
thereby maintaining a strong driving force for Cl⫺/
HCO3⫺ exchange. This hypothesis has received support
by the recent finding of a HCO3⫺ secretory mechanism
in rat gastric mucosa that is sensitive to DIDS (23), an
inhibitor of some of the electroneutral Cl⫺/HCO3⫺ exchangers and related transporters.
The identity of the apical anion transporter that
mediates HCO3⫺ secretion by surface mucous cells is
unclear, but members of either the SLC4 (Cl⫺/HCO3⫺
exchangers and Na⫹-HCO3⫺ cotransporters) or SLC26
(anion exchangers, including Cl⫺/HCO3⫺ exchangers)
families are potential candidates. Recently, a Cl⫺/
HCO3⫺ exchanger of the SLC4 family, termed anion
exchanger (AE) isoform 4, was cloned from rabbit kid-
Xu, Jie, Sharon Barone, Snezana Petrovic, Zhaohui
Wang, Ursula Seidler, Brigitte Riederer, Krishnamurthy Ramaswamy, Pradeep K. Dudeja, Gary E. Shull,
and Manoocher Soleimani. Identification of an apical Cl⫺/
HCO3⫺ exchanger in gastric surface mucous and duodenal
villus cells. Am J Physiol Gastrointest Liver Physiol 285:
G1225–G1234, 2003. First published July 31, 2003; 10.1152/
ajpgi.00236.2003.—The molecular identity of the apical
HCO3⫺-secreting transporter in gastric mucous cells remains
unknown despite its essential role in preventing injury and
ulcer by gastric acid. Here we report the identification of a
Cl⫺/HCO3⫺ exchanger that is located on apical membranes of
gastric surface epithelial cells. RT-PCR studies of mouse
gastrointestinal tract mRNAs demonstrated that this transporter, known as anion exchanger isoform 4 (AE4), is expressed in both stomach and duodenum. Northern blot analysis of RNA from purified stomach epithelial cells indicated
that AE4 is expressed at higher levels in mucous cells than in
parietal cells. Immunoblotting experiments identified AE4 as
a ⬃110- to 120-kDa protein in membranes from stomach
epithelium and apical membranes from duodenum. Immunocytochemical staining demonstrated that AE4 is expressed in
apical membranes of surface cells in both mouse and rabbit
stomach and duodenum. Functional studies in oocytes indicated that AE4 functions as a Cl⫺/HCO3⫺ exchanger. These
data show that AE4 is an apical Cl⫺/HCO3⫺ exchanger in
gastric mucous cells and duodenal villus cells. On the basis of
its function and location, we propose that AE4 may play an
important role in mucosal protection.
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
G1226
EXPERIMENTAL PROCEDURES
RT-PCR of AE4 in Mouse and Rabbit
Gastrointestinal Tract
A BLAST search of the mouse expressed sequence tag
(EST) database against the rabbit AE4 sequence (GenBank
accession no. AB038263) identified a mouse EST (GenBank
accession no. AW018362) with a high degree of sequence
similarity. On the basis of the cDNA sequence of the mouse
EST, primers were designed (sense, CAT GCC TGG TGC
TCA AGA AAG CTA G; antisense, CAC TCA TGT TAC TGG
GCC TGG TGG) and used for RT-PCR experiments. RT-PCR
was performed on RNA isolated from stomach, duodenum,
and kidney by using the above primers, which amplified a
412-bp fragment corresponding to nucleotides 66–482 of the
EST.
For rabbit AE4, primers (sense, GAA ATG GGC CAC TTG
CAC C; antisense, AAT TGA CAG AGG CAG CCA TAG G)
were designed based on the rabbit AE4 cDNA sequence
(GenBank accession no. AB038263). The primers were used
for RT-PCR in various tissues, and a PCR fragment corresponding to nucleotides 992–1595 of the rabbit AE4 cDNA
was isolated. The reason for using different sets of primers
for mouse and rabbit was because of DNA sequence differences in these two species.
(using a mouse AE4 EST with Gene Bank accession no.
AW018362) was used to generate polyclonal antibodies in
two rabbits. This sequence is identical in mouse, rabbit, and
human AE4. Antibodies were purified by using a cysteine
affinity column. A human AE4 polyclonal antibody was purchased from Alpha Diagnostics (San Antonio, TX). Gastric
H-K-ATPase ␤-subunit antibody was a generous gift from Dr.
John Forte (University of California at Berkeley).
Immunoblotting of AE4 in stomach and duodenum. Mouse
microsomal membranes from the scrapings of gastric epithelium and apical membrane vesicles from the duodenum were
isolated according to established methods and as described
(22, 34). For mouse, immunoblotting experiments were carried out as previously described (34). Briefly, the solubilized
membrane proteins were size fractionated on 8% SDS polyacrylamide minigels (Novex, San Diego, CA) under denaturing conditions, electrophoretically transferred to nitrocellulose membranes, blocked with 5% milk proteins, and then
probed with the affinity-purified anti-AE4 immune serum at
an IgG concentration of 0.6 ␮g/ml. The secondary antibody
was donkey anti-rabbit IgG conjugated to horseradish peroxidase (Pierce). The antigen-antibody complexes on the nitrocellulose membranes were visualized by chemiluminescence
(SuperSignal substrate; Pierce), and the image was captured
on light-sensitive imaging film (Kodak).
For human duodenum, apical and basolateral membrane
vesicles were isolated from two organ donors by using established methods (24) and in accordance with the institutional
protocols approved by the University of Illinois at Chicago.
Seventy-five micrograms each of purified brush-border and
basolateral membranes were solubilized in Laemmli sample
buffer, separated on 7% SDS-PAGE, transferred to Hybond
nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ), and visualized with Ponceau stain to ensure the
efficiency of the transfer. After being blocked with 5% nonfat
dry milk in 1⫻ PBS-0.1% Tween 20, blots were incubated
with affinity-purified rabbit anti-human AE4 antibody (Alpha Diagnostics) diluted (1:150) in 1⫻ PBS-0.1% Tween
20–1% milk overnight at 4°C. After four washes with 1⫻
PBS-0.1% Tween 20, blots were then incubated with horseradish peroxidase goat anti-rabbit IgG secondary antibody
(Santa Cruz Biotechnology, Santa Cruz, CA) diluted in 1⫻
PBS-0.1% Tween 20–1% milk for 1 h at room temperature.
After five washes with 1⫻ PBS-0.1% Tween 20, the bands
were visualized by using an enhanced chemiluminescence
system (Amersham Biosciences). To determine antigen specificity, in a separate experiment AE4 control peptide (30
␮g/ml, sequence near the cytoplasmic COOH terminal of
human AE4; Alpha Diagnostics) was preincubated with the
antibody solution (5 ␮g/ml) at 37°C for 2 h before being added
to the blot.
RNA Isolation and Northern Blot Hybridization
Immunofluorescence Labeling Studies
Total cellular RNA was extracted from mouse duodenum,
stomach, and kidney and rabbit gastric parietal and mucous
cells using Tri reagent. Hybridization was performed according to the method of Church and Gilbert (3). The membranes
were washed, blotted dry, and exposed to a PhosphorImager
screen (Molecular Dynamics, Sunnyvale, CA). 32P-labeled
mouse or rabbit AE4 PCR fragments (see above) were used as
hybridization probes.
Mice were euthanized with an overdose of sodium pentobarbital and perfused through the left ventricle with 0.9%
saline followed by cold 4% paraformaldehyde in 0.1 M sodium
phosphate buffer (pH 7.4). Rabbits were euthanized with an
overdose of sodium pentobarbital via earlobe vein and perfused through the left ventricle similar to mice. Stomachs
and duodena were removed, cut in tissue blocks, and left in
the fixative solution overnight at 4°C. For cryosections, tissue
blocks were removed from the fixative solution and soaked in
30% sucrose overnight. The tissue was frozen on dry ice, and
5-␮m sections were cut with a cryostat and stored at ⫺80°C
until use.
Immunocytochemistry of AE4 in Stomach and Kidney
Antibodies. For AE4, a synthetic peptide corresponding to
amino acids residues CLMYQPKAPEINISVN of mouse AE4
AJP-Gastrointest Liver Physiol • VOL
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
ney cortical collecting duct (32). It exhibits 48% amino
acid identity to the NBC1 Na⫹-HCO3⫺ cotransporter
and 32% identity to the AE2 Cl⫺/HCO3⫺ exchanger (32).
Functional studies in oocytes and cultured mammalian
cells indicated that AE4 mediates electroneutral Cl⫺/
HCO3⫺ exchange and does not transport Na⫹ (32). Because of this functional property and despite its greater
similarity to the NBC subfamily, it was named AE4.
On the basis of its localization on the apical membrane
of ␤-intercalated cells in the cortical collecting duct, it
was postulated that AE4 mediates HCO3⫺ secretion
(32).
In the current studies, we investigated the gastrointestinal distribution of AE4, because little is known
about the expression of this transporter outside the
kidney. Our results indicate that AE4 mRNA is expressed in both stomach and duodenum and that it
functions as a Cl⫺/HCO3⫺ exchanger when expressed in
oocytes. Immunocytochemical studies demonstrate
that AE4 is localized on the apical membranes of gastric surface epithelial cells and villi of the duodenum.
We propose that AE4 mediates apical HCO3⫺ secretion
in gastric surface epithelium and duodenum, thereby
protecting these tissues against acid injury.
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
Single and double immunofluorescence labeling were performed as described recently (22). For double labeling with
H-K-ATPase, gastric H-K-ATPase ␤-subunit antibody was
diluted at 1:250. Goat anti-rabbit IgG conjugated with Oregon green 488 and goat anti-mouse IgG conjugated with
Alexa Fluor 568 dye (Molecular Probes, Eugene, OR) were
used at 1:150 dilution for AE4 and H-K-ATPase, respectively.
Sections were examined and images were acquired on a
Nikon PCM 2000 laser confocal scanning microscope as 0.5
␮m optical sections of the stained cells.
G1227
medium (63 mM Na-gluconate and 33 mM NaHCO3). All
other Cl⫺-containing chemicals (KCl, etc.) were replaced
with gluconate salts. This maneuver results in cell alkalinization due to reversal of Cl⫺/HCO3⫺ exchange (16). On pHi
stabilization in Cl⫺-free medium, oocytes were switched back
to the Cl⫺-containing solution. This should result in cell
acidification back to baseline due to activation of Cl⫺/HCO3⫺
exchange. The initial rate of cell pHi recovery was used as the
rate of Cl⫺/HCO3⫺ exchange activity (34). In additional studies, experiments were repeated with Na⫹-free solutions, with
Cloning of AE4 cDNA
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Full-length human AE4 cDNA was cloned from a human
kidney cDNA library by PCR using the following primers:
5⬘-GAC TGT ACT GGT TCT GAG ATT CTG TGC AAG and
5⬘-CTG CGG GCT TAC TGG AGA TAT TAT TTA GGG,
which were designed based on sequences from the human
genome database. PCR amplification of the human AE4
cDNA was performed by using the Clontech Advantage 2
PCR kit protocol. Each PCR reaction contained 5 ␮l cDNA, 5
␮l 10⫻ PCR buffer, 1 ␮l 10 mM dNTPs, 10 pmol of each
primer, and 1 ␮l Advantage 2 polymerase mix in a final
volume of 50 ␮l. Cycling parameters were 95°C, 1 min; 95°C,
30 s; and 68°C, 4 min. An ⬃3.2-kb PCR fragment was obtained that contained the full-length coding region of the
exchanger. The product was gel purified and subcloned into
the pGEM-T Easy vector for expression studies.
Expression of AE4 cRNA in Xenopus Oocytes
The capped AE4 cRNA was generated by using the
mMESSAGE mMACHINE kit (Ambion) according to the
manufacturer’s instructions. Xenopus oocytes were injected
with 50 nl of the human AE4 cRNA (0.2–1.3 ␮g/␮l) by using
a Drummond 510 microdispenser via a sterile glass pipette.
Intracellular pH Studies
Intracellular pH (pHi) in oocytes was measured with the
pH-sensitive fluorescent probe BCECF (Molecular Probes)
as described (34). Briefly, oocytes were loaded with 10 ␮M
BCECF-AM for 20 min at room temperature in the same
medium in which they were kept after cRNA injection.
Oocytes were than transferred to the 1-ml perfusion chamber, placed on the nylon mesh with vegetable pole facing
the fluorescent beam, and perfused at rate of 3 ml/min with
the following solution (in mM): 63 NaCl, 33 NaHCO3, 2
KCl, 1.8 CaCl2, 1 MgCl2, and 5 HEPES. Another 20–30
min were allowed for the equilibration and dye ester hydrolysis, which yields the fluorescent probe. Ratiometric
fluorescence measurements were performed by using an
Attofluor digital imaging system (Atto, Rockville, MD).
Excitation wavelengths were 450 and 490 nm, and fluorescence emission intensity was recorded at 520 nm. Ratios
were calculated and stored. Data analyses were performed
by using the Attograph and Attoview software packages
provided with the imaging system. Background signal (always ⬍1%) was recorded and automatically subtracted
from the subsequent fluorescence measurements. The ratios were obtained from the submembrane region of the
oocytes that were visualized with a ⫻10 objective. Measured excitation ratios were converted to pHi by using a
calibration curve that was constructed with a high K⫹/
nigericin method at the end of each experiment (34).
To test for Cl⫺/HCO3⫺ exchange activity, the chamber was
first perfused with solution containing 63 mM NaCl and 33
mM NaHCO3. The oocytes were then switched to Cl⫺-free
AJP-Gastrointest Liver Physiol • VOL
Fig. 1. Expression of anion exchanger (AE) isoform 4 mRNA in the
gastrointestinal tract. A: mouse kidney and stomach RNA were
analyzed by RT-PCR in the presence or absence of reverse transcriptase (RT) by using mouse AE4-specific primers, and the
products were fractionated by agarose gel electrophoresis.
Ethidium bromide staining shows amplification of the expected
PCR fragment (⬃420 bp) in both tissues. B: RNA from rabbit
mucous and parietal cells was analyzed by RT-PCR in the presence (⫹) or absence (⫺) of RT by using rabbit AE4-specific primers, and the products were fractionated by agarose gel electrophoresis. Ethidium bromide staining shows amplification of the
expected PCR fragment (⬃600 bp) in both cell types. C: Northern
blot analysis of AE4 mRNA from rabbit mucous and parietal cells
reveals AE4 expression in both cell types.
285 • DECEMBER 2003 •
www.ajpgi.org
G1228
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
AJP-Gastrointest Liver Physiol • VOL
285 • DECEMBER 2003 •
www.ajpgi.org
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
G1229
tetramethylammonium chloride replacing NaCl and choline
bicarbonate replacing NaHCO3.
Rabbit Gastric Cell Purification
Rabbit gastric mucous and parietal cells were isolated as
described (26) and were processed for total RNA isolation.
The use of rabbit in these studies was for the purpose of
determination of the gastric epithelial cell origin of AE4,
because reasonably purified mucous and parietal cells have
been successfully isolated from rabbit stomach.
Materials
32
P-dCTP was purchased from New England Nuclear (Boston, MA). Nitrocellulose filters were purchased from Sigma
(St. Louis, MO). RadPrime DNA labeling kit was purchased
from GIBCO-BRL. All other chemicals were purchased from
Sigma unless otherwise stated.
Statistical Analyses
Values are expressed as means ⫾ SE. The significance of
differences between mean values were examined by using
ANOVA. P ⬍ 0.05 was considered statistically significant.
RESULTS
AE4 mRNA Expression in the Upper
Gastrointestinal Tract
To examine the distribution of AE4 mRNA in stomach, RT-PCR was performed by using RNA isolated
from mouse stomach and was compared with that of
kidney by using mouse-specific primers (see EXPERIMENTAL PROCEDURES). Figure 1A shows ethidium bromide
staining of an agarose gel, which revealed that the
expected PCR fragment of 412 bp is amplified from
mouse stomach and kidney. The sequence of the gel-
Fig. 2. Immunoblotting and immunocytochemical staining of AE4 in mouse stomach. A: immunoblot analysis of
AE4 in microsomal membranes from stomach. Left: AE4 immune serum identifies an ⬃120-kDa protein. Right:
preadsorbed immune serum. B: immunocytochemical staining of AE4 in stomach indicates that AE4 is localized on
the apical membrane of surface epithelial cells (left). Staining with the preadsorbed immune serum failed to detect
any labeling in the stomach (right). C: immunofluorescence double labeling with AE4 (green staining, left) and
gastric H-K-ATPase (red staining, right). The distribution of gastric H-K-ATPase ␤-subunit is distinct from AE4
when dual images were acquired (middle).
AJP-Gastrointest Liver Physiol • VOL
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 3. Immunocytochemical staining
of AE4 in rabbit stomach and kidney.
A: immunofluorescence labeling with
anti-AE4 antibody in rabbit stomach
reveals expression of AE4 on the apical
membrane of surface epithelial cells. B:
immunofluorescence double labeling of
AE4 (left) and peanut lectin-binding
protein (right) in rabbit kidney demonstrates colocalization of both transporters to the same membrane domain
(merged images in middle). Arrows
show labeling of apical membrane surface cells.
G1230
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
purified PCR product from kidney and stomach was
determined, verifying the product as AE4. The above
results demonstrate that in addition to the kidney,
AE4 is expressed in the stomach.
Expression of AE4 mRNA in Gastric Mucous and
Parietal Cells
Immunoblotting and Immunofluorescence Labeling of
AE4 in Stomach and Duodenum
To determine the protein expression of AE4, microsomal membrane proteins from scrapings of the gastric
epithelium and apical membrane vesicles from the
duodenum were isolated from mouse and were subjected to immunoblot analysis. Figure 2A shows an
immunoblot analysis of microsomal membranes from
mouse stomach, which demonstrates that AE4 appears
as a ⬃120-kDa protein. The reaction is specific because
Fig. 4. Immunoblot localization of AE4 in the duodenum. A: immunoblot analysis of AE4 in human duodenum from 2 donors reveals an ⬃110-kDa protein in
brush-border membrane (BBM) vesicles (left), but no
bands were detected in basolateral membrane (BLM)
vesicles. Preadsorbed immune serum did not detect any
bands (right). B: immunoblot analysis of AE4 in mouse
duodenum using immune serum reveals an ⬃120-kDa
protein in apical membrane vesicles (left). The labeling
of the ⬃120-kDa band was prevented by preadsorbed
immune serum (right).
AJP-Gastrointest Liver Physiol • VOL
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
To determine the cellular distribution of AE4 mRNA
in the stomach, RNA was isolated from parietal and
mucous cells of rabbit stomach and was analyzed by
RT-PCR using rabbit-specific primers. Figure 1B
shows an ethidium bromide-stained agarose gel, demonstrating that the expected PCR fragment (⬃600 bp),
which was verified by sequence analysis, was amplified
from both mucous and parietal cells. The results of
these RT-PCR experiments were verified by Northern
blot analysis, which demonstrated that AE4 mRNA is
expressed in both gastric mucous and parietal cells
(Fig. 1C). The use of rabbit in these series of studies
was for the purpose of determination of the epithelial
cell origin of AE4. As is known, rabbit is the only
mammalian species that has successfully been used for
the isolation of reasonably purified mucous and parietal cells.
the preadsorbed immune serum failed to detect the
band (Fig. 2A). To determine the cellular distribution
and subcellular localization of AE4, immunocytochemical staining was performed on sections of mouse stomach. As shown in Fig. 2B, AE4 is expressed on the
apical membrane of surface epithelial cells (left). The
preadsorbed immune serum did not detect any labeling
(right). Figure 2C shows double immunofluorescence
labeling with AE4 and gastric H-K-ATPase ␤-subunit
in mouse stomach. It demonstrates that the H-KATPase is restricted to parietal cells, as expected,
whereas AE4 is expressed primarily in surface epithelial cells, with lower levels of expression in gastric
parietal cells. The staining in parietal cells was variable, with intracellular labeling being more frequent
than the apical labeling. No basolateral labeling in
parietal cells was observed.
In the next series of studies, we examined the localization of AE4 in rabbit stomach by immunofluorescence labeling. As shown in Fig. 3A, AE4 is located on
the apical membrane of surface epithelial cells in rabbit stomach. As a control and to verify that the AE4
antibody indeed recognizes the AE4 protein, double
immunofluorescence labeling with AE4 and peanut
lectin-binding protein was performed in rabbit kidney.
As indicated in Fig. 3B, AE4 and peanut lectin-binding
protein, which labels only the apical membrane of
␤-intercalated cells, colocalize to the same membrane domain. These results indicate that the AE4
immune serum recognizes only the AE4 protein and
does not cross-react with other proteins. It further
verifies the results of molecular studies in which AE4
was cloned from ␤-intercalated cells of rabbit kidney
and was shown to be expressed only in those same
cells (32).
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
The experiments shown in Fig. 4 examine the expression and subcellular localization of AE4 in mouse,
rabbit, and human duodenum. In human, AE4 appears
as an ⬃110-kDa band in apical membranes of the
duodenum but not in basolateral membranes (Fig. 4A,
left). Preadsorbed immune serum did not react with
any proteins (Fig. 4A, right). In the mouse, AE4 appears as an ⬃120-kDa protein in apical membrane
vesicles from the duodenum (Fig. 4B, left). No labeling
G1231
was observed with the preadsorbed immune serum
(Fig. 4B, right).
The results of the above experiments indicate that
AE4 is localized to the apical membrane proteins of the
duodenum in both mouse and human. To determine
the cellular distribution and subcellular localization of
AE4 in the duodenum in more detail, immunofluorescence labeling was performed in mouse and rabbit
duodenum. In mouse duodenum, AE4 is expressed
AJP-Gastrointest Liver Physiol • VOL
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Fig. 5. Immunocytochemical staining of
AE4 in mouse and rabbit duodenum. A:
immunocytochemical staining of AE4 in
mouse duodenum indicates that AE4 is
expressed on the apical membrane of
the villi with lower expression levels on
the apical membrane of crypt cells (left).
This reaction was specific, because the
labeling was prevented by immune
preadsorption (right). B: immunocytochemical staining of AE4 in rabbit duodenum. The preadsorbed immune serum shows nonspecific labeling on the
basolateral membrane of villi in the duodenum (right). The AE4 immune serum (left) specifically labels the apical
membrane of the villi of the duodenum
(arrows).
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
G1232
exclusively on apical membranes of epithelial cells in
the villi (Fig. 5A, left). No labeling was detected with
the preadsorbed immune serum (Fig. 5A, right). In
rabbit, the preadsorbed immune serum showed nonspecific labeling in the basolateral membrane of villi in
the duodenum (Fig. 5B, right). Compared with the
preadsorbed serum, the AE4 immune serum specifically labeled the apical membranes of villi in the duodenum (Fig. 5B, left).
Functional Expression of AE4
DISCUSSION
The lumen of the stomach is exposed to an acidic
solution secreted from gastric parietal cells, which can
achieve a pH as low as 1.5 (1, 7, 12). To protect itself
against the corrosive effects of luminal acid, the gastric
surface epithelium secretes a HCO3⫺-rich fluid into the
mucus gel layer (5). Despite the essential role of HCO3⫺
secretion in mucosal protection, the molecular identity
and functional characteristics of the transporter(s) responsible for this process have remained unclear. The
results of the present study show that in addition to its
expression in kidney, AE4 mRNA is expressed in the
stomach and in the duodenum, which is also exposed
AJP-Gastrointest Liver Physiol • VOL
Fig. 6. Functional expression of AE4. A: representative tracings
demonstrating Cl⫺/HCO3⫺ exchange activity in oocytes expressing
human AE4 cRNA. Oocytes were loaded with BCECF and perfused
with solutions as indicated (see EXPERIMENTAL PROCEDURES for details). Solutions were gassed with 95% O2-5% CO2. As indicated,
switching to a Cl⫺-free solution resulted in intracellular alkalinization in oocytes injected with AE4 cRNA. The intracellular pH (pHi)
returned to baseline on switching back to the Cl⫺-containing solution. The pHi in oocytes that were injected with water (control group)
remained unchanged in response to the removal or addition of Cl⫺.
Solutions containing HCO3⫺ were continuously bubbled with 95% O2
and 5% CO2. Solutions had a pH of ⬃7.4 at 37°C and an osmolality
of 290 ⫾ 3 mosmol/kgH2O. Solutions without HCO3⫺ were gassed
with 100% O2. Solutions containing gluconate salts had 4 mM Caacetate to account for complexing of Ca2⫹ with gluconate. B: summary of pHi experiments. AE4 mediates Cl⫺/HCO3⫺ exchange at the
rate of 0.14 pH units/min (P ⬍ 0.001 vs. control).
to gastric acid. Furthermore, immunohistochemical
staining demonstrated that expression of AE4 protein
in the stomach occurs predominantly in the apical
membrane of surface epithelial cells and that its expression in the duodenum is restricted to the apical
membranes of villus cells. Expression of AE4 in parietal cells was mostly limited to the cytoplasmic region,
with faint and occasional apical labeling. Functional
expression of AE4 in oocytes confirmed previous studies showing that this transporter mediates Cl⫺/HCO3⫺
exchange (32). These observations suggest that AE4 is
responsible, at least in part, for HCO3⫺ secretion into
the mucus layer of the gastric epithelium and across
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
Two recent studies indicated that AE4 is a Cl⫺/
HCO3⫺ exchanger (16, 32). The purpose of the next
series of experiments was to examine and verify the
functional identity of AE4 by using the oocyte expression system. In the first series of experiments, oocytes
were injected with human AE4 cRNA and were loaded
with BCECF in the presence of a Cl⫺/HCO3⫺-containing
solution (see EXPERIMENTAL PROCEDURES), and intracellular pH was monitored. The representative pHi tracings
in Fig. 6A demonstrate that switching to a Cl⫺-free
solution resulted in a rapid intracellular alkalinization
in oocytes expressing AE4. Switching back to the Cl⫺containing solution caused a return of pHi to baseline.
No pHi alteration in response to exposure to the Cl⫺free medium was observed in control (water-injected)
oocytes (Fig. 6A). In the absence of HCO3⫺ in the medium, the rate of cell alkalinization in response to Cl⫺
removal was minimal. These results are consistent
with AE4 functioning as a Cl⫺/HCO3⫺ exchanger. A
summary of multiple experiments is shown in Fig. 6B
and demonstrates that the rate of Cl⫺/HCO3⫺ exchange
activity was 0.14 ⫾ 0.01 in oocytes expressing AE4 (n ⫽
5). The baseline pHi in CO2/HCO3⫺-containing solutions
was 7.14 ⫾ 0.03 in water-injected oocytes and 7.17 ⫾
0.03 in AE4-injected oocytes (P ⬎ 0.05; n ⫽ 5). The
Cl⫺/OH⫺ exchange activity in control oocytes was not
significantly different from zero (Fig. 6, A and B). The
absence of Na⫹ in the medium did not prevent cell
alkalinization in response to Cl⫺ removal, indicating
that AE4-mediated Cl⫺/HCO3⫺ exchange is independent of Na⫹. In the absence of HCO3⫺, the rate of
Cl⫺/OH⫺ exchange activity (mediated by Cl⫺/OH⫺ exchange) was minimal, indicating the low affinity of
AE4 for OH⫺. AE4 did not function in a Na⫹-HCO3⫺
cotransport mode (data not shown).
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
AJP-Gastrointest Liver Physiol • VOL
small intestine with very low expression levels in the
large intestine (34). Pendrin is located on the apical
membrane of thyroid follicular cells and kidney collecting ducts (27, 30). None of these exchangers are expressed on the apical membrane of surface epithelial
cells in the stomach. In mouse stomach, immunocytochemical staining localized PAT1 to the tubulovesicle
membranes of gastric parietal cells (22), whereas it
failed to detect the expression of pendrin despite detectable mRNA levels (data not shown).
In addition to the stomach, AE4 is expressed in the
duodenum. Immunoblotting or immunocytochemical
staining demonstrates that AE4 is expressed on the
apical membrane of mouse, rabbit, and human duodenum. This observation is intriguing because it confirms
that more than one apical Cl⫺/HCO3⫺ exchanger is
expressed in the duodenum. Recently, PAT1
(SLC26A6) was identified as a major Cl⫺/HCO3⫺ exchanger in the apical membrane of the duodenum (34),
along with the DRA (SLC26A3) Cl⫺/HCO3⫺ exchanger
(20). Our studies in the duodenum indicate that the
expression levels of PAT1 are the highest, followed by
DRA and AE4 (Ref. 34 and unpublished data). It would
be difficult to conclude that, based on the expression
level studies in the duodenum, one anion exchanger
(i.e., PAT1) might be the major transporter mediating
HCO3⫺ secretion (or Cl⫺ absorption) and the others are
not. The presence of more than one apical HCO3⫺ transporter in the duodenum suggests possible differential
regulation of these transporters in physiological or
pathophysiological states.
In conclusion, AE4 is expressed in the upper gastrointestinal tract, with expression in the small intestine
and stomach. Immunohistochemical staining or immunoblotting studies localized AE4 to the apical membranes of surface epithelial cells in the stomach and to
the apical domain of villus cells in the duodenum.
Functional studies demonstrated that AE4 operates in
an electroneutral Cl⫺/HCO3⫺ exchange mode. We propose that AE4 is an apical HCO3⫺ transporter in surface epithelial cells of the stomach and villus cells of
the duodenum and that it may play an important role
in protecting the gastric and duodenal epithelium
against injury by the acid secreted from gastric parietal cells.
DISCLOSURES
These studies were supported by National Institute of Diabetes
and Digestive and Kidney Diseases Grants DK-54430 (to M. Soleimani), DK-50594 (to G. E. Shull), DK-54016 (to P. K. Dudeja), and
DK-33349 (to K. Ramaswamy), Merit Review Grants from the Department of Veterans Affairs (to M. Soleimani, P. K. Dudeja, and K.
Ramaswamy), a Cystic Fibrosis Foundation Grant, and grants from
Dialysis Clinic Incorporated (to M. Soleimani).
REFERENCES
1. Allen A, Flemstrom G, Garner A, and Kivilaakso E. Gastroduodenal mucosal protection. Physiol Rev 73: 823–857, 1993.
2. Bissig M, Hagenbuch B, Stieger B, Koller T, and Meier PJ.
Functional expression cloning of the canalicular sulfate transport system of rat hepatocytes. J Biol Chem 269: 3017–3021,
1994.
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
the apical membrane of villus cells in the duodenum,
consistent with the hypothesis that it plays an important role in mucosal protection in both organs.
There are conflicting studies on the nature of HCO3⫺secreting transporter(s) in gastric surface epithelial
cells. Although a Cl⫺/HCO3⫺ exchanger has been proposed to be the primary mechanism for HCO3⫺ secretion in frog gastric mucosa (9), other investigators have
proposed that a HCO3⫺-conductive pathway is responsible for this activity (32). AE4 functions as a Cl⫺/
HCO3⫺ exchanger in heterologous expression systems
(Refs. 16 and 32 and the current study), indicating that
at least a portion of the HCO3⫺ secretion in surface
epithelial cells is likely to be mediated by this transporter operating in a Cl⫺/HCO3⫺ exchange mode. A
recent preliminary study (21a) suggested that AE4
may also function as an electroneutral Na-HCO3⫺ cotransporter. Our functional expression studies in oocytes did not reveal Na-HCO3⫺ cotransport activity by
AE4, confirming recent results indicating that AE4
only functions in a Cl⫺/HCO3⫺ exchange mode (32).
These results not withstanding, either Cl⫺/HCO3⫺ exchange or Na-HCO3⫺ cotransport activity via AE4
would function in a HCO3⫺-secreting mode, because
luminal acidity precludes the functioning of AE4 (or
any HCO3⫺ transporter) in a net HCO3⫺-absorbing
mode. As an apical Cl⫺/HCO3⫺ exchanger, AE4 can
neutralize the gastric acidity according to the following
scheme. Following secretion into the gastric lumen via
coordinated action of gastric H-K-ATPase and apical
Cl⫺ channel in parietal cells, the resulting HCl dissociates into H⫹ and Cl⫺. The secreted Cl⫺ will be exchanged with intracellular HCO3⫺ via the Cl⫺/HCO3⫺
exchanger AE4 located on the apical membrane of
mucous cells. The net effect of these processes is the
secretion of HCl by parietal cells and the secretion of
HCO3⫺ from mucous cells for protection against injury
by the acid. In the case of proton pump inhibitors,
inhibition of acid secretion is associated with decreased
generation of HCO3⫺, which in turn leads to reduced
HCO3⫺ exit across the basolateral membrane of parietal
cells, eventually reducing the uptake of HCO3⫺ by mucous cells (presumably via basolateral NBC). The reason that gastric luminal pH does not climb very high in
response to proton pump inhibiter (PPI) might be due
to decreased intracellular HCO3⫺ concentration in mucous cells secondary to decreased HCO3⫺ generation in
parietal cells.
Recent studies (2, 6, 13, 14, 15, 17, 18, 19, 20, 24, 27,
29, 30, 33, 35) have identified a family of anion exchangers, referred to as the SLC26A family, that include at least 10 distinct genes. Three well-known
members of this family are SLC26A3 [or downregulated in adenoma (DRA)], SLC26A4 (or pendrin) and
SLC26A6 [putative anion transporter 1 (PAT1) or chloride format exchange (CFEX)] (6, 18, 29), which are
located apically in a limited and distinct number of
epithelia and function as Cl⫺/OH⫺/HCO3⫺ exchangers
(20, 30, 34). DRA is predominantly expressed in the
large intestine with lower levels in the small intestine
(20), whereas PAT1 is predominantly expressed in the
G1233
G1234
A HCO3⫺ TRANSPORTER IN SURFACE EPITHELIAL CELLS
AJP-Gastrointest Liver Physiol • VOL
terization of three novel tissue-specific anion exchangers:
SLC26A7, A8 and A9. J Biol Chem 277: 14246–14254, 2002.
20. Melvin JE, Park K, Richardson L, Schultheis PJ, and
Shull GE. Mouse down-regulated in adenoma (DRA) is an intestinal Cl⫺/HCO3⫺ exchanger and is upregulated in colon of mice
lacking the NHE-3 Na⫹/H⫹ exchanger. J Biol Chem 274: 22855–
22861, 1999.
21. Muallem S, Burnham C, Blissard D, Berglindh T, and
Sachs G. Electrolyte transport across the basolateral membrane
of parietal cells. J Biol Chem 260: 6641–6653, 1985.
21a.Parker MD, Boron WF, and Tanner MJA. Characterization
of human AE4 as an electroneutral, sodium-dependent bicarbonate transporter (Abstract). FASEB J 16: A796, 2002.
22. Petrovic S, Wang Z, Ma L, Seidler U, Forte JG, Shull GE,
and Soleimani M. Colocalization of the apical Cl⫺/HCO3⫺ exchanger PAT1 and gastric H-K-ATPase in stomach parietal cells.
Am J Physiol Gastrointest Liver Physiol 283: G1207–G1216,
2002.
23. Phillipson M, Atuma C, Henriksnas J, and Holm L. The
importance of mucus layers and bicarbonate transport in preservation of gastric juxtamucosal pH. Am J Physiol Gastrointest
Liver Physiol 282: G211–G219, 2002.
24. Rajendran VM, Harig JM, Adams MB, and Ramaswamy K.
Transport of acidic amino acids by human jejunal brush border
membrane vesicles. Am J Physiol Gastrointest Liver Physiol 252:
G33–G39, 1987.
25. Rossmann H, Bachmann O, Vieillard-Baron D, Gregor M,
and Seidler U. Na⫹/HCO3⫺ cotransport and expression of NBC1
and NBC2 in rabbit gastric parietal and mucous cells. Gastroenterology 116: 1389–1398, 1999.
26. Rossmann H, Bachmann O, Wang Z, Shull GE, Obermaier
B, Stuart-Tilley A, Alper SL, and Seidler U. Differential
expression and regulation of AE2 anion exchanger subtypes in
rabbit parietal and mucous cells. J Physiol 534: 837–848, 2001.
27. Royaux IE, Wall SM, Karniski LP, Everett LA, Suzuki K,
Knepper MA, and Green ED. Pendrin, encoded by the Pendred syndrome gene, resides in the apical region of renal intercalated cells and mediates bicarbonate secretion. Proc Natl Acad
Sci USA 98: 4221–4226, 2001.
28. Schade C, Flemstrom G, and Holm L. Hydrogen ion concentration in the mucus layer on top of acid-stimulated and -inhibited rat gastric mucosa. Gastroenterology 107: 180–188, 1994.
29. Schweinfest CW, Henderson KW, Suster S, Kondoh N, and
Papas TS. Identification of a colon mucosa gene that is downregulated in colon adenomas and adenocarcinomas. Proc Natl
Acad Sci USA 90: 4166–4170, 1993.
30. Soleimani M. Molecular physiology of the renal chloride-formate exchanger. Curr Opin Nephrol Hypertens 10: 677–683,
2001.
31. Takeuchi K, Merhav A, and Silen W. Mechanism of luminal
alkalinization by bullfrog fundic mucosa. Am J Physiol Gastrointest Liver Physiol 243: G377–G388, 1982.
32. Tsuganezawa H, Kobayashi K, Iyori M, Araki T, Koizumi
A, Watanabe S, Kaneko A, Fukao T, Monkawa T, Yoshida
T, Kim DK, Kanai Y, Endou H, Hayashi M, and Saruta T. A
new member of the HCO3⫺ transporter superfamily is an apical
anion exchanger of ␤-intercalated cells in the kidney. J Biol
Chem 276: 8180–8189, 2001.
33. Vincourt JB, Jullien D, Kossida S, Amalric F, and Girard
JP. Molecular cloning of SLC26A7, a novel member of the
SLC26 sulfate/anion transporter family, from high endothelial
venules and kidney. Genomics 79:249–256, 2002.
34. Wang Z, Petrovic S, Mann E, and Soleimani M. Identification of an apical Cl-/HCO3⫺ exchanger in the small intestine.
Am J Physiol Gastrointest Liver Physiol 282: G573–G579, 2002.
35. Zheng J, Shen W, He DZ, Long KB, Madison LD, and
Dallos P. Prestin is the motor protein of cochlear outer hair
cells. Nature 405: 149–55, 2000.
285 • DECEMBER 2003 •
www.ajpgi.org
Downloaded from http://ajpgi.physiology.org/ by 10.220.32.246 on June 18, 2017
3. Church GM and Gilbert W. Genomic sequencing. Proc Natl
Acad Sci USA 81: 1991–1995, 1984.
4. Curci S, Debellis L, Caroppo R, and Fromter E. Model of
bicarbonate secretion by resting frog stomach fundus mucosa. I.
Transepithelial measurements. Pflügers Arch 428: 648–654,
1994.
5. Engel E, Guth PH, Nishizaki Y, and Kaunitz JD. Barrier
function of the gastric mucus gel. Am J Physiol Gastrointest
Liver Physiol 269: G994–G999, 1995.
6. Everett LA, Glaser B, Beck JC, Idol JR, Buchs A, Heyman
M, Adawi F, Hazani E, Nassir E, Baxevanis AD, Sheffield
VC, and Green ED. Pendred syndrome is caused by mutations
in a putative sulphate transporter gene (PDS). Nat Genet 17:
411–422, 1997.
7. Feldman M. Gastric secretion: normal and abnormal. In: Gastrointestinal and Liver Disease: Pathophysiology/Diagnosis/
Management (2nd ed.), edited by Feldman MD, Scharschmidt
BF, Sleisenger MH, and Klein SWB. Philadelphia, PA: Saunders, 1998, p. 587–603.
8. Flemstrom G. Gastric and duodenal mucosal bicarbonate secretion. In: Physiology of the Gastrointestinal Tract (3rd ed.), edited
by Johnson LR, Jacobson ED, Christensen J, Alpers D, and
Walsh JH. New York: Raven, 1994, p.1285–1309.
9. Flemstrom G. Cl⫺ dependence of HCO3⫺ transport in frog gastric mucosa. Ups J Med Sci 85: 303–309, 1980.
10. Gannon B, Browning J, and O’Brien P. The microvascular
architecture of the glandular mucosa of rat stomach. J Anat 135:
667–683, 1982.
11. Gannon B, Browning J, O’Brien P, and Rogers P. Mucosal
microvascular architecture of the fundus and body of human
stomach. Gastroenterology 86: 866–875, 1984.
12. Garner A, Flemstrom G, Allen A, Heylings JR, and McQueen S. Gastric mucosal protective mechanisms: roles of epithelial bicarbonate and mucus secretions. Scand J Gastroenterol
Suppl 101: 79–86, 1984.
13. Hastabacka J, de la Chapelle A, Mahtani MM, Clines G,
Reeve-Daly MP, Daly M, Hamilton BA, Kusumi K, Trivedi
B, Weaver A, Coloma A, Lovett M, Buckler A, Kaitila I, and
Lander ES. The diastrophic dysplasia gene encodes a novel
sulfate transporter: positional cloning by fine-structure linkage
disequilibrium mapping. Cell 78: 1073–1087, 1994.
14. Hoglund P, Haila S, Socha J, Tomaszewski L, SaarialhoKere U, Karjalainen-Lindsberg ML, Airola K, Holmberg C,
Chapelle A, and Kere J. Mutations of the down-regulated in
adenoma (DRA) gene cause congenital chloride diarrhea. Nat
Genet 14: 316–319, 1996.
15. Jacob P, Rossmann H, Lamprecht G, Kretz A, Neff C,
Lin-Wu E, Gregor M, Groneberg DA, Kere J, and Seidler U.
Down-regulated in adenoma mediates apical Cl⫺/HCO3⫺ exchange in rabbit, rat, and human duodenum. Gastroenterology
122: 709–724, 2002.
16. Ko SB, Luo X, Hager H, Rojek A, Choi JY, Licht C, Suzuki
M, Muallem S, Nielsen S, and Ishibashi K. AE4 is a DIDSsensitive Cl-/HCO3⫺ exchanger in the basolateral membrane of
the renal CCD and the SMG duct. Am J Physiol Cell Physiol 283:
C1206–C1218, 2002.
17. Knauf F, Yang CL, Thomson RB, Mentone SA, Giebisch G,
and Aronson PS. Identification of a chloride-formate exchanger
expressed on the brush border membrane of renal proximal
tubule cells. Proc Natl Acad Sci USA 98: 9425–9430, 2001.
18. Lohi H, Kujala M, Kerkela E, Saarialho-Kere U, Kestila M,
and Kere J. Mapping of five new putative anion transporter
genes in human and characterization of SLC26A6, a candidate
gene for pancreatic anion exchanger. Genomics 70: 102–112,
2000.
19. Lohi H, Kujala M, Makela S, Lehtonen E, Kestila M, Saarialho-Kere U, Markovich D, and Kere J. Functional charac-