| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Division of Gastroenterology,*
Washington University School of Medicine, St. Louis, Missouri; the
Division of Digestive Disease,
University of
North Carolina School of Medicine, Chapel Hill, North Carolina; the
Department of Integrative Biology and
Pharmacology,
University of Texas Medical
School, Houston, Texas; and the Technical University of
Munich,§
Munich, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, can induce protein production in somatic
cells.9,10
Moreover, IL-1 receptors are found on the
surface of parietal cells.11
Therefore, models of chronic
inflammation of the zymogenic mucosa were selected to test whether
increased elaboration of cytokines could enlarge the population of
parietal cells that produce IF. HLA-B27, a serologically defined class 1 major histocompatibility complex allele, is associated with several human inflammatory disorders. An attempt was made to produce an animal model in transgenic mice, but despite physiologically normal function of B27 in hybrid mice, no features of human disease appeared.12 Because rats are susceptible to experimental arthritis that cannot be produced in mice, transgenic technology was developed in Fisher 344 and Lewis rats, and animals expressing HLA-B27 and human ß2-microglobulin genes developed spontaneous multiorgan inflammatory disease.13 Two of the seven transgenic lines developed inflammation, including gastrointestinal disease and arthritis, increasing in severity from 5 to 15 weeks of age. F344 transgenic rats raised in a specific pathogen-free environment develop histological evidence of colitis and gastroduodenal inflammation by 10 to 12 weeks of age. No disease occurs in the absence of bacterial stimulation14 and disease correlates with cecal luminal anaerobic bacterial concentrations.15 The incidence of disease varied from 30 to 100% in litters, and was affected by the genetic strain and density of animals/cage. The gastric body and antrum were involved with widely scattered mononuclear foci in the lamina propria and submucosa. The infiltrate occurred in some, but not all, crypt units, and was associated with a marked reduction in the number of parietal cells in some glands.13 Subsequently the model was reproduced in transgenic mice (mostly males), where it was associated with the presence of free heavy chains on a subpopulation of B27-expressing lymphocytes.16 In this model inflammation did not develop in germ-free animals, although no pathogen was detected in the gastrointestinal lumen.
Helicobacter pylori infection causes clinical gastritis and peptic ulcers and is associated with the development of gastric cancer. Helicobacter felis is a spiral bacteria isolated from feline stomach that colonizes gastric mucosa in mice, rats, and dogs.17 H. felis-associated gastritis in mice reproduces many of the features of human disease, including a marked polymorphonuclear and mononuclear cell infiltrate.18 Production of H. felis infection in C57BL/6 mice was achieved by inoculating mice orally, and observing a similar infiltrate with a marked reduction in parietal cells noted by 6 months of infection.19 The organism colonizes the gastric mucous layer and causes gastritis in the antrum and body of the stomach by 4 weeks, with inflammation increasing further by 8 weeks after inoculation.20
These animal models were used to test the hypothesis that chronic inflammation could increase ectopic IF production in other gastric cell lineages, particularly the parietal cell that expresses IF in human mucosa. Despite the decrease in parietal cell populations that occurred during development of inflammation, both models demonstrated more than a twofold increase in the percent of parietal cells that expressed IF, but only when an inflammatory infiltrate was present in the tissue. Other gastric mucosal lineages did not express IF. These results are consistent with a cytokine-induced expression of IF in parietal cells, and suggest that IF expression in cells is not entirely predetermined, at least in certain subpopulations that might be potential precursors for zymogen cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A colony of transgenic HLA-B27 rats and nontransgenic littermates were derived in a sterile environment at the University of North Carolina School of Medicine, populated with specific pathogen-free bacteria (documented Helicobacter sp. free), and maintained in that environment.15 The colony, derived from specific pathogen-free inbred Lewis/CrlBR (LEW) rats, was originally obtained from Dr. Joel D. Taurog (Southwestern Medical School, Dallas, TX). Female C57BL/6 mice, purchased from Taconic Laboratories (Germantown, NY), were inoculated intragastrically three times on alternate days at 5 weeks of age with 1 x 109 cfu of H. felis (ATCC 49179).
Tissues
Transgenic and control rats were killed at 1, 2, or 3 months of age. H. felis inoculated mice were maintained for up to 4 weeks after infection.18 After rats or mice were euthanized with CO2 inhalation, the glandular rat stomachs were fixed in Bouin’s solution for 2 hours at room temperature, and transferred to 70% ethanol overnight at 4°C before embedding in paraffin. The body and antrum of the murine stomachs were fixed in 10% buffered formalin for a few hours, stored in 80% ethanol, and then embedded in paraffin. Sections were cut 4-µm thick and stained for hematoxylin and eosin (H&E), or used for immunocytochemistry. Inflammatory infiltrate was estimated as absent, moderate, or severe (0, +, or ++) by one observer (J-SS) in a blinded manner. These samples were obtained from the same animals reported previously.20
Immunocytochemistry
Immunostaining was performed using the standard avidin-biotin-peroxidase complex method as reported previously,6 and counterstained with hematoxylin. Endogenous peroxidase was quenched by pretreatment with 1% H2O2 in methanol for 20 minutes. The primary antiserum used was rabbit polyclonal anti-human IF (1:200), raised against recombinant human IF produced in baculovirus-infected Sf9 cells.5 In some experiments recombinant purified human IF made in Pichia pastoris was added to the antiserum before adding to the section.21 Rabbit antibody against the ß-subunit of rat H/K ATPase (amino acids 2 to 23) was obtained from Dr. Jeffrey I. Gordon (Washington University School of Medicine, St. Louis, MO). Second antibody was goat anti-rabbit IgG obtained from Vector Laboratories (Burlingame, CA). Parietal cells were identified by their large size, triangular shape, and different staining pattern compared with chief cells. Quantitation of parietal cells staining positively for IF was performed manually. All parietal cells in sections of rat and mouse stomachs were counted in eight x200 magnified fields per section, prepared from four rats and three mice per group/time point. The mean ± SD values were compared by Student’s t-test of paired samples.
Parietal Cell Incubation
Highly enriched preparations of parietal cells were isolated from
rat gastric mucosa as described previously,6
and cultured
for 24 to 48 hours.11
Enriched preparations of parietal
cells were washed 3 times in serum-free culture medium, and cells were
cultured for 48 hours in a 1:1 mixture of Ham’s F-12
medium/Dulbecco’s modified Eagle’s medium with HEPES plus
L-glutamine without bicarbonate, and supplemented with
insulin (5 µg/ml), transferrin (5 µg/ml), sodium selenite (5
ng/ml), hydrocortisone (4 mg/ml), epidermal growth factor (25 ng/ml),
gentamicin (10 mg/100 ml), and bovine serum albumin (2
mg/ml).11
Thereafter, recombinant human IL-1ß (2.5
pg/ml) or recombinant human interferon-
(1.5 ng/ml) were added to
fresh culture media without added growth factors for 4 hours at 37°C.
The medium was aspirated, centrifuged, and stored at -20°C. Adherent
cells were removed by trypsinization, washed in phosphate-buffered
saline (PBS), centrifuged, and fixed in Bouin’s solution overnight
before transfer to graded ethanol concentrations from 70 to 100%.
The cells were embedded in paraffin and 5-µm sections of the cell
pellet were prepared on slides, which were probed with antiserum
against IF as described above. The cells on each slide were counted
completely, each high-power field contained 108 to 384 parietal cells
(total of 1773 to 4384 in each slide were counted), and the IF-positive
cells recorded. The results were compared using a two-tailed Student’s
t-test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Transgenic animals were studied at 1, 2, or 3 months of age.
Congenic control animals were examined only at 3 months of age. In the
gastric body of the control rats, 8.9% of the parietal cells were
positive for IF (Table 1
and Figure 1
). Figure 1
demonstrates the specificity
of the antiserum used against IF. Neither normal rabbit serum (Figure 1B)
nor anti-IF antiserum incubated with recombinant IF (Figure 1D)
showed any reactivity in either chief cells or parietal cells.
|
|
. There was no
inflammation found in the body of these animals, although mild
inflammation had developed in the antrum by 2 months of age. No
parietal cells are found in the antrum of adult rats, however, and
stains for IF were also negative. After 2 months the infiltrate in the
gastric body was not uniform, involving some gastric glands, but not
others (Figure 2A)
. By 3
months submucosal abscesses were found occasionally in the antrum, and
in the body more extensive inflammation with dilation of the gastric
pits (Figure 2B)
. This inflammation was mononuclear (Figure 2C)
. In
control rats the IF-positive parietal cells were mostly located in the
lower part of the crypt (Figure 2D)
.
|
. The
absolute number of parietal cells was decreased.17
Control
rats had 110 ± 27 (mean ± SD) parietal cells per magnified
field (n = 32), whereas inflamed areas in
transgenic rats showed 77 ± 19 parietal cells
(n = 32). In the inflamed glands, 21% of
parietal cells were IF-positive. Most of the positive cells were found
in the lower neck and basal regions of the glands, and none appeared to
be dying or apoptotic cells. In adjacent glands that were not inflamed
there were still 16% of parietal cells that were positive for IF
(Figure 2, E and F
, and Table 1
). These values were significantly
greater than that for the control rats. In addition, the inflamed
glands and those adjacent to them were also different from each other
(P < 0.0001).
The identity of IF-positive cells with morphology-like parietal cells
was confirmed using antiserum against H/K ATPase, a parietal cell
marker. Figure 2G
shows all parietal cells positive in a
3-month-infected transgenic animal. Figure 2H
shows an adjacent section
stained with antiserum against IF, demonstrating some cells positive
for both markers (isolated arrows), and some parietal cells positive
for H/K ATPase but negative for IF.
H. felis Model
After either 4 or 8 weeks of infection, the mucosa and the
submucosa both showed inflammatory infiltration, although the degree
and distribution of inflammation was somewhat variable. After 4 weeks
the parietal cell density in some gastric glands was decreased, and
scattered inflammatory cells were found in the mucosa of the gastric
body (Figure 3, B
compared with the control in A). The number of parietal cells seen in
H&E sections was diminished by 8 weeks (see Figure 3C
versus
Figure 1,A and B
).20
Control mice
(n = six animals, 70 fields) showed 70 ±
29 parietal cells/magnified field compared with the value in infected
mice (n = 6, 62 fields) of 50 ± 41, but no
mucosal hyperplasia was noted as in the HLA-B27 rat model. Moreover,
submucosal inflammation was present, and atrophic parietal cells were
noted by their smaller size and diminished eosinophilic stain. These
cells were seen mostly in the middle part of the gastric gland (Figure 3, C and E)
. Further analysis of these atrophic cells was made using
antiserum against H/K ATPase. In the normal control mice, all parietal
cells stained strongly positive (Figure 3D)
. In the 8-week-infected
mice, on the other hand, the stain for H/K ATPase is much diminished
compared with control tissues, and some cells that appear by morphology
to be atrophic parietal cells are negative (Figure 3E)
.
|
, and did not
change from 4 to 8 weeks (P = 0.49). However,
the degree of mucosal inflammation did not alter qualitatively either
(compare Figure 3, B and C
). Again, the increase in IF-positive cells
was located primarily in the lower neck and basal regions of the
gastric glands and these cells seemed normal morphologically (Figure 3, F–H)
.
|
The percent of IF-positive parietal cells incubated with the
cytokine interleukin-1ß increased from 3.14 ± 1.41% to
6.35 ± 4.05% (mean ± SD, P < 0.005,
n = 17). On the other hand, incubation with
interferon- produced a small decline in IF-reactive cells, from
1.83 ± 1.19% to 0.96 ± 0.85% (n =
10, P < 0.1). No difference in IF appearance in the
medium was detected between control and cytokine-incubated cells.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The finding of ectopic production of IF, especially located in the lower neck and basal regions of the gastric gland, suggests that the expression of chief cell products in parietal cells is either hard-wired (predetermined by lineage or by functional maturation), or is influenced by environmental factors. The increase in IF-positive parietal cells in two models of chronic inflammation of the gastric body suggests that local environmental factors may be important. The rat model demonstrates gastric mucosal hyperplasia, whereas in the mouse model patches of atrophic mucosa are seen without intervening hyperplasia. Both models cause a marked decrease in the total number of parietal cells, and some of the remaining parietal cells in the middle of the murine gastric glands were atrophic. It seems unlikely that IF expression is solely a measure reflecting cell damage, as none of the IF-positive cells in or near the base of the gastric glands appeared abnormal morphologically. In addition, H. felis infection of C57BL6 mice show a marked diminution in chief cell number with a decrease in IF immunostaining.23
It is possible that the proliferative zone containing relatively undifferentiated cells is expanded whether mucosal hyperplasia or atrophy is present. For example, in the presence of inflammation the parietal cell precursors capable of IF expression (eg, preneck precursor lineage in the mouse,1 might be preferentially preserved, creating a falsely elevated percent of IF-positive cells. However, in relatively normal gastric glands adjacent to the inflamed ones (in the rat model) the percent of IF-positive cells was also higher, although the number of parietal cells seemed normal. Thus, it seems unlikely that preservation of an IF-expressing lineage accounts for the observed changes.
The most likely explanation for these findings induced by inflammation is that they are related to recruitment of parietal cells newly expressing IF. The IF-positivity was present only when the gastric body is inflamed and the percent of positive cells was higher in inflamed areas than in noninflamed glands, at least in the rat model that allows such comparisons. Moreover, the difference between inflamed and noninflamed areas suggests a gradient effect, consistent with the presence of a diffusible inducer, perhaps secreted by (or liberated from) infiltrating inflammatory cells. The inflammation might be related either to a primary effect on the mucosa or a secondary one mediated by microbial presence and stimulation.15
The cytokines produced by inflammatory cells that have been often
associated with increased protein expression are IL-1, TNF-, and
interferon-. The mRNAs encoding these cytokines (among others) are
abundantly expressed in the gastric mucosa of patients with chronic
gastritis,24
and expression is increased in the inflamed
colons of HLA-B27 transgenic rats.15
Gastric parietal
cells have receptors for IL-1 at least, and this cytokine has been
demonstrated to have antisecretory activity.11,25
IL-10
and TNF- production26
and IL-1ß, IL-6, IL-8, and
IL-12 as well,27
are increased in antral mucosal biopsies
from patients with chronic gastritis, and the cytokine secretion varies
with the degree of inflammation.26
IL-8 and TNF-
production were increased in patients with H. pylori
infection in one study,28
but not in
another.27
The studies with isolated parietal cells showed
that at least one cytokine that contributes to a Th1 response (IL-1ß)
reproduced in vitro the twofold increase in IF-positivity
seen in vivo. Atrophy of parietal and zymogenic cells is an
eventual outcome of chronic inflammation of the gastric body, but
dysplasia precedes the atrophy in most cases. The altered IF expression
might be an early marker of such dysplastic changes.
It is possible that the observed increase in IF immunoreactivity is related to a decrease in IF secretion, as IF and H+ secretion occur by the same canalicular pathway.22 For this explanation to be valid, it would be necessary for some parietal cells to be producing undetectable IF under basal conditions. Alternatively, the cytokines released during inflammation could up-regulate expression of IF in cells which constitutively express undetectable amounts, recruit new cells to produce IF, expand a population of parietal cells that normally produce IF, or have combined effects. The results with isolated parietal cells favor up-regulation of IF expression.
This study did not evaluate the quantitative secretion of IF from chief
cells, although there was no obvious change in their immunoreactivity
(Figures 2 and 3)
. IL-1ß, on the other hand, does not affect basal or
stimulated pepsinogen secretion from chief cells.29
Whatever the mechanism whereby cytokines may produce the effect noted
on parietal cells, it is possible that increased IF-positivity
represents a surrogate marker for the local effect of one or more
cytokines. If so, this would be very instructive in following the
tissue effects of cytokine stimulation and in evaluating this response
as a marker for eventual mucosal atrophy.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by National Institutes of Health Grants PO1 DK33487, RO1 DK14038 (to D. H. A.), RO1 DK40249 (to R. B. S.), and RO1 DK33239 (to L. M. L.).
Accepted for publication July 12, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. Vallon, F. Grahammer, H. Volkl, C. D. Sandu, K. Richter, R. Rexhepaj, U. Gerlach, Q. Rong, K. Pfeifer, and F. Lang KCNQ1-dependent transport in renal and gastrointestinal epithelia PNAS, December 6, 2005; 102(49): 17864 - 17869. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mahmood, J.-s. Shao, and D. H. Alpers Rat enterocytes secrete SLPs containing alkaline phosphatase and cubilin in response to corn oil feeding Am J Physiol Gastrointest Liver Physiol, July 7, 2003; 285(2): G433 - G441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. H Alpers How adaptable is the intestine in patients with short-bowel syndrome? Am. J. Clinical Nutrition, May 1, 2002; 75(5): 787 - 788. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
