Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/11826397
Proliferation,notapoptosis,altersepithelialcell
migrationinsmallintestineofCFTRnullmice
ArticleinAJPGastrointestinalandLiverPhysiology·October2001
ImpactFactor:3.8·Source:PubMed
CITATIONS
READS
27
23
2authors,including:
RobertaA.Gottlieb
Cedars-SinaiMedicalCenter
196PUBLICATIONS16,752CITATIONS
SEEPROFILE
Availablefrom:RobertaA.Gottlieb
Retrievedon:17May2016
Am J Physiol Gastrointest Liver Physiol
281: G681–G687, 2001.
Proliferation, not apoptosis, alters epithelial cell
migration in small intestine of CFTR null mice
ANN MARIE GALLAGHER AND ROBERTA A. GOTTLIEB
Department of Molecular and Experimental Medicine, The Scripps Research
Institute, La Jolla, California 92037
Received 5 January 2001; accepted in final form 4 May 2001
Gallagher, Ann Marie, and Roberta A. Gottlieb. Proliferation, not apoptosis, alters epithelial cell migration in
small intestine of CFTR null mice. Am J Physiol Gastrointest
Liver Physiol 281: G681–G687, 2001.—Expression of a mutated cystic fibrosis transmembrane conductance regulator
(CFTR) has been shown to enhance proliferation within CF
airways, and cells expressing a mutated CFTR have been
shown to be less susceptible to apoptosis. Because the CFTR
is expressed in the epithelial cells lining the gastrointestinal
tract and all CF mouse models are characterized by gastrointestinal obstruction, we hypothesized that CFTR null mice
would have increased epithelial cell proliferation and reduced apoptosis within the small intestine. The rate of intestinal epithelial cell migration from crypt to villus was increased in CFTR null mice relative to mice expressing the
wild-type CFTR. This difference in migration could be explained by an increase in epithelial cell proliferation but not
by a difference in apoptosis within the crypts of Lieberkühn.
In addition, using two independent sets of CF cell lines, we
found that epithelial cell susceptibility to apoptosis was unrelated to the presence of a functional CFTR. Thus increased
proliferation but not alterations in apoptosis within epithelial cells might contribute to the pathophysiology of CF.
cystic fibrosis transmembrane conductance regulator; animal
model; pathophysiology
conductance regulator (CFTR) is a chloride channel that is mutated in CF.
Mutation of the CFTR results in a decrease in epithelial cell chloride permeability. Although the defect affects all exocrine tissues, in humans it manifests itself
primarily in the airways and intestine. CF is characterized by reduced chloride and fluid secretion, which
results in accumulation of mucus within the bronchioles and in meconium ileus in CF newborns and intestinal obstructions in older CF patients. In all mouse
models of CF in which the CFTR has been deleted by
insertional or deletional mutagenesis, intestinal pathophysiology is characterized by mucus accumulation in
the crypts of Lieberkühn, goblet cell hyperplasia, and
intestinal obstruction. In addition, the CFTR is expressed in the epithelial cells lining the gastrointestinal tract with a reduction in expression along the
crypt-to-villus axis (7, 26).
THE CYSTIC FIBROSIS TRANSMEMBRANE
Address for reprint requests and other correspondence: R. A.
Gottlieb, The Scripps Research Institute, 10550 N. Torrey Pines Rd.,
MEM220, La Jolla, CA 92037 (E-mail: [email protected]).
http://www.ajpgi.org
Previously, Gottlieb and Dosanjh (9) showed that
cultured C127 mouse mammary epithelial cells stably
transfected with the ⌬F508-mutated CFTR remained
viable and were less susceptible to apoptosis induced
by cycloheximide or etoposide than cells expressing the
wild-type CFTR. Programmed cell death, or apoptosis,
is a physiologically relevant process that occurs in
regenerating tissues such as the lung and intestine.
Cell turnover is a balance between proliferation and
elimination, and cell proliferation has been shown to be
greater in airway epithelia and nasal polyps in CF
patients than in tissue from non-CF patients (12, 15).
Although this phenomenon has traditionally been attributed to inflammation, the possibility that the proliferative differences might be due to differential expression of the CFTR has not been addressed.
Because all mouse models of CF (CFTR ⫺/⫺) develop
intestinal obstructions (10) and the CFTR is expressed
in epithelial cells lining the gastrointestinal tract, we
hypothesized that the CFTR ⫺/⫺ mouse intestine
would show evidence of altered apoptosis and/or enhanced proliferation relative to mice expressing the
wild-type CFTR. The hypothesis was tested in vivo
using a knockout mouse model of CF. Mice expressing
a mutant CFTR were found to have a greater rate of
migration of epithelial cells up the crypt-villus axis
within the small intestine. This increased rate could be
attributed to an increase in the proliferative capacity of
cells within the crypts of Lieberkühn but not to differences in apoptosis. Because these current results contradict the previous findings (9) of differences in apoptosis between cells transfected with the ⌬F508 CFTR
and the wild-type CFTR, studies were extended to
further examine apoptosis in two independent cell culture models of CF.
METHODS
Animal Model
A mouse model of CF generated by Dr. Beverly Koller
(University of North Carolina) was used for all studies (3). In
this model (cftrtm1Unc), the CFTR gene has been disrupted by
gene targeting, and as with all mouse models of CF, the
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.
0193-1857/01 $5.00 Copyright © 2001 the American Physiological Society
G681
G682
PROLIFERATION, NOT APOPTOSIS, IN CF EPITHELIAL MIGRATION
animals develop intestinal obstructions and typically die
within weeks of birth. However, by replacing the drinking
water with Colyte (Schwarz Pharma, Milwaukee, WI), a
commercially available electrolyte solution containing 6%
polyethylene glycol, the mice remain alive past 18 mo of age
(10). Animals were maintained within the custom breeding
colony at The Scripps Research Institute with all animals
receiving Colyte in place of drinking water. At the time of
weaning, the genotypes of all mice were determined by isolation of tail DNA and PCR amplification of the CFTR alleles
as described previously by Koller et al. (14). All animal
studies adhered to approved institutional animal protocols
and American Association for Accreditation of Laboratory
Animal Care guidelines.
Epithelial Cell Migration
Bromodeoxyuridine (BrdU; Sigma Chemical, St. Louis,
MO) was suspended in PBS at 20 mg/ml and administered
via intraperitoneal injection at 150 ␮g/g body wt. The mice
were killed by halothane overdose 2–48 h after BrdU injection. The small intestine was isolated and excised, rinsed in
PBS, splayed open along its length, and rinsed twice to
remove contents. The splayed intestine was fixed in Z-Fix
(10% buffered formalin containing zinc; Anatech, Battle
Creek, MI), coiled from proximal to distal end to form a
“Swiss roll,” dehydrated, and embedded in paraffin for tissue
sectioning. Tissue sections of Swiss rolls allow for examination of the entire length of the intestine in a single section.
BrdU incorporation on Swiss rolls was analyzed by immunohistochemistry. Briefly, 3-␮m-thick tissue sections were
cleared of paraffin and rehydrated through graded alcohols.
Endogenous peroxidase activity was blocked by 0.3% hydrogen peroxide in methanol for 15 min at room temperature.
After pepsin digestion (0.4 mg/ml in 0.1 N HCl, 15 min at
37°C), tissue was treated with 2 N HCl for 30 min at 37°C
and rinsed in 0.1 M sodium borate (pH 8.5). Tissue sections
were incubated with anti-BrdU (Sigma Chemical) at 1:500 in
20% normal goat serum (90 min at 37°C). Incubation with
biotinylated goat anti-mouse IgG (1:2,000 in PBS containing
1 mg/ml IgG-free BSA; Caltag, Burlingame, CA) for 60 min at
room temperature was followed by enzymatic labeling with
horseradish peroxidase-conjugated streptavidin (1:1,000 in
PBS/BSA, 30 min at room temperature; Caltag). Diaminobenzidine (0.04%, Sigma Chemical) in 0.05 M Tris, pH 7.4,
and 0.01% H2O2 were used for color development. Tissues
were counterstained with Gill’s hematoxylin (Fisher Scientific, Pittsburgh, PA), dehydrated through graded alcohols,
and placed on a coverslip with Permount (Fisher Scientific).
Epithelial cell migration was evaluated in CFTR ⫹/⫺ and
CFTR ⫺/⫺ mice by measuring the distance between the base
of the intestinal crypts at the most distal Paneth cell along
the epithelial cells to each BrdU-positive nuclei within the
crypt-villus structure. The investigator was blinded to animal identification and time after BrdU injection. Through the
use of a ⫻20 objective on a Nikon TE300 microscope and a
Spot Diagnostics Instruments digital camera (Sterling
Heights, MI), 5 to 24 crypt-villus structures were captured
for each animal and analyzed using MetaMorph imaging
software (Universal Imaging, West Chester, PA). Three individual time courses for BrdU migration were performed for a
total of 20 animals (3 animals/genotype at 2 and 24 h, 4
mice/genotype at 48 h). ANOVA was used to assess whether
there was a genotypic difference in migration rate, and a
two-tailed t-test was utilized to determine if there was a
difference in distance of migration at 48 h after BrdU injection.
AJP-Gastrointest Liver Physiol • VOL
Spontaneous and Induced Epithelial Cell Apoptosis
In situ terminal deoxynucleotidyltransferase dUTP nickend labeling (TUNEL) has been found to be unreliable as a
method of detecting apoptosis in the gastrointestinal tract
(21). Therefore, a method previously described by Potten (21)
in which apoptosis is quantified by calculating the percentage of cells with condensed and/or fragmented nuclei within
the first 10 cells following the Paneth cells in the crypts of
Lieberkühn was used to quantify apoptosis in the small
intestine of CFTR mice. Crypts were scored by an investigator blinded to the identity of animals that met the following
criteria: 1) Paneth cells could be viewed, 2) the crypt lumen
could be seen, and 3) a minimum of 10 nonoverlapping
epithelial cells in the crypt column could be distinguished.
Levels of spontaneous apoptosis were evaluated in control
animals. Whole body cesium gamma irradiation was used to
induce apoptosis in the small intestine of CFTR mice in an
attempt to uncover genotypic differences in susceptibility to
apoptosis. An initial dose-response curve using CFTR ⫹/⫺
mice revealed that significant epithelial cell apoptosis could
be seen 3 h after whole body cesium gamma irradiation with
100 rads. Higher doses of gamma irradiation did not result in
detection of more apoptotic cells. Therefore, subsequent experiments were performed with a dose of 100 rads, and the
animals were killed 3 h after gamma irradiation. Following
fixation and sectioning as described above, intestinal tissue
was deparaffinized, stained with hematoxylin, and placed on
a coverslip with Permount. The average number of apoptotic
nuclei was determined in a minimum of 50 crypts/animal in
18 CFTR ⫺/⫺ mice (8 control, 10 irradiated) and 14 each of
⫹/⫺ and ⫹/⫹ mice (5 control, 9 irradiated) and expressed as
a percentage of the first 10 crypt cells. ANOVA was used to
test whether irradiation increased intestinal cell apoptosis,
and two-tailed t-tests were used to determine the statistical
significance of irradiation within each genotype.
Intestinal Epithelial Cell Proliferation
Immunohistochemical detection of proliferating cell nuclear antigen (PCNA) was evaluated within the crypts of
control (nonirradiated, no BrdU) CFTR mice by an investigator blinded to animal identity. Tissues were acquired and
processed as described above, and paraffin-embedded Swiss
rolls were dewaxed and rehydrated through graded alcohols.
Tissues were incubated with Target retrieval solution
(DAKO, Carpinteria, CA) at 95°C for 20 min and at room
temperature for an additional 20 min. Endogenous peroxidase activity was blocked by 0.3% hydrogen peroxide in
methanol before incubation with anti-PCNA (clone PC10,
DAKO) for 90 min at 37°C in 20% normal goat serum. After
rinsing with PBS, tissue was incubated with biotinylated
goat anti-mouse IgG (Caltag) suspended 1:2,000 in PBS containing 1 mg/ml IgG-free BSA. After 30 min incubation with
horseradish peroxidase-conjugated streptavidin (Caltag),
color was developed with SigmaFast diaminobenzidine (0.7
mg/ml diaminobenzidine in 0.06 M Tris buffer containing 0.2
mg/ml urea-H2O2), counterstained with hematoxylin, dehydrated through alcohols, and placed on a coverslip with
Permount. Proliferation was evaluated by determining the
average number of PCNA-positive cells present within the
first 10 cells following the Paneth cells in a minimum of 25
crypts/animal. Twelve animals (17) were used to determine if
there was a genotypic difference in epithelial cell proliferation. A two-tailed t-test was used to determine statistical
significance.
281 • SEPTEMBER 2001 •
www.ajpgi.org
PROLIFERATION, NOT APOPTOSIS, IN CF EPITHELIAL MIGRATION
Cell Culture Models
C127 cell lines. C127 mouse mammary epithelial cells
stably transfected with the wild-type CFTR or the CFTR with
the deletion of phenylalanine at position 508 (C127 ⌬F508)
were obtained from Dr. G. M. Denning (Howard Hughes
Medical Institute, University of Iowa College of Medicine)
(4). These cells were grown in DMEM (GIBCO BRL, Rockville, MD) supplemented with 10% FBS (HyClone, Logan,
UT), penicillin, streptomycin, and Fungizone (Sigma Chemical) and passaged 1:5 two times per week. The cell lines were
grown in the presence of Geneticin (500 ␮g/ml; GIBCO BRL).
CFT1 cell lines. CFT1 cell lines, which were derived from
tracheal epithelial cells isolated from a 24-year-old CF patient homozygous for the ⌬F508 mutation, were obtained
from J. Yankaskas (University of North Carolina). The parental cell line (CFT1-C2) was used to establish the other
three cell lines and has been immortalized with human
papilloma virus. CFT1-LCFSN cells overexpress the wildtype CFTR introduced by retroviral vector (20). CFT1-⌬F508
cells express an additional copy of the CFTR with the ⌬F508
mutation, and the CFT1-LC3 is a retroviral vector control
cell line overexpressing the ␤-galactosidase gene. These cells
were grown in Ham’s F-12 (Fisher Scientific) supplemented
with insulin (10 ␮g/ml), hydrocortisone (1 ␮M), endothelial
cell growth supplement (ECGS; 3.75 ␮g/ml), epidermal
growth factor (25 ng/ml), T3 (30 nM), transferrin (5 ␮g/ml),
and cholera toxin (10 ng/ml) (all hormones were obtained
from Sigma Chemical). All cells, except the CFT1-C2 cells,
were grown in the presence of Geneticin (500 ␮g/ml) and
were passaged 1:3 two times per week.
Analyses of death
Membrane destabilization detected by YOPRO-1 uptake.
Cells were plated at 10,000 cells/well into 96-well tissue
culture plates. After 24 h, the medium was aspirated and
replaced with 100 ␮l sterile PBS. Cells were ultraviolet (UV)
irradiated at 0–8,000 J/m2 using a Fisher Scientific UV
crosslinker, the PBS was replaced with fresh medium, and
the cells were incubated at 37°C with 5% CO2. After 16 h, 25
␮M YOPRO-1 (10 ␮l) was added to each well and incubated
at 37°C for 10 min. YOPRO-1 is taken up by cells with an
unstable membrane and binds to double-strand DNA; therefore, apoptotic or necrotic cells will be labeled with YOPRO-1
(13). With the use of a fluorescence plate reader (Bio-Tek
Instruments, Winooski, VT), YOPRO-1 was found to be excited at 530 nm and emissions were detected at 480 nm. Cells
were lysed with 10 ␮l of lysis buffer (125 mM EDTA and 2.5%
Nonidet P-40) and incubated at 37°C with 5% CO2 for 30 min
after which fluorescence was evaluated to determine the total
number of cells per well. The ratio of the first fluorescence
(apoptotic and necrotic cells) to the second fluorescence (total
cells) was taken to evaluate percent death and plotted
against UV dose.
Nuclear condensation and fragmentation. Cells were
plated at 30,000 cells/chamber into four-well chamber slides.
One cell type was plated per chamber slide with cells plated
in the first and fourth chambers only. After allowing the cells
to adhere and reach ⬃80% confluence, the medium was
aspirated and replaced with 500 ␮l sterile PBS. One chamber
(containing control cells) was covered with an opaque material to prevent irradiation while the chamber slide was irradiated at 2,000 J/m2. The position of the control and irradiated cells was randomized to ensure that the analysis was
done under blinded conditions. The PBS was aspirated and
replaced with fresh medium. Cells were incubated at 37°C
with 5% CO2 for 6 h after which 37% formaldehyde was
AJP-Gastrointest Liver Physiol • VOL
G683
added to each chamber (final concentration: 3.75% formaldehyde). During fixation, the slides were centrifuged at 1,200
rpm for 15 min to ensure minimal loss of cells during the
fixation procedure. The cells were fixed for an additional 15
min, rinsed with PBS, and allowed to air dry. At the time of
analysis, 30 ␮M Hoechst no. 33342 in PBS were added to
each chamber, slides were coverslipped, and a minimum of
300 cells/chamber were analyzed for nuclear condensation
and fragmentation by fluorescence microscopy. The percentage of fragmented and condensed nuclei from three separate
experiments was obtained with the experimenter blinded to
cell type and irradiation dose.
CFTR Adenovirus Infection
To confirm that the mutated CFTR was responsible for the
resistance to apoptosis, C127 ⌬F508 cells were infected with
a CFTR adenovirus in an attempt to restore susceptibility to
death in these cells. C127 ⌬F508 cells were plated at 5,000
cells/well in a 96-well tissue culture plate and allowed to
adhere overnight. Preliminary studies with the ␤-galactosidase adenovirus (Ad2/CMV␤gal-4, gift from Genzyme, Framingham, MA) showed that these cells were fairly resistant
to infection by this adenovirus and required 20,000 infective
units/cell for 95% infectivity. Cells were infected with the
20,000 infective units of CFTR adenovirus (Ad2/CFTR-16,
gift from Genzyme) suspended in DMEM containing 2%
heat-inactivated FBS for 60 min at room temperature, followed by 48 h at 37°C with 5% CO2. After this infection, cells
were UV irradiated and evaluated with YOPRO-1 as described above to determine susceptibility to death. ␤-Galactosidase adenovirus-infected and uninfected cells were used
as negative controls.
RESULTS
Mouse Model of CF
Mouse models of CF are consistent in their susceptibility to gastrointestinal obstruction; therefore, studies were designed to determine whether a nonfunctional CFTR alters the kinetics of epithelial cell
migration within the gastrointestinal tract and ultimately might underlie the pathophysiology within the
intestine.
The rate of migration of BrdU-labeled cells up the
crypt-villus axis is faster in CFTR ⫺/⫺ mice than in
their CFTR ⫹/⫺ littermates (P ⫽ 0.048). By 48 h, the
distance that the epithelial cells from the CFTR ⫺/⫺
mice have migrated is significantly greater than the
distance of migration in the CFTR ⫹/⫺ mouse intestine
(P ⫽ 0.018) (Fig. 1). Because epithelial cell migration
from the crypts of Lieberkühn up the intestinal villi is
the net result of proliferation and apoptosis within the
crypts, both components were evaluated separately to
see if either or both are altered as a result of the
expression of the mutant CFTR. An increase in epithelial cell migration might occur if apoptosis is reduced
and/or proliferation within the crypts of Lieberkühn is
increased.
Levels of spontaneous apoptosis are not different
between CFTR genotypes, with CFTR ⫺/⫺ mice exhibiting 5.4 ⫾ 4.6% apoptosis within the crypts and CFTR
⫹/⫺ and CFTR ⫹/⫹ mice presenting 6 ⫾ 2.5% and
2.1 ⫾ 2.1% apoptosis, respectively (Fig. 2). To deter281 • SEPTEMBER 2001 •
www.ajpgi.org
G684
PROLIFERATION, NOT APOPTOSIS, IN CF EPITHELIAL MIGRATION
mine whether the low levels of spontaneous apoptosis
seen within the crypts masks a difference in susceptibility to apoptosis, intestinal apoptosis was induced
with whole body cesium gamma irradiation. Irradiation significantly increased apoptosis in all genotypes
(P ⬍ 0.001) but did not uncover any differences in
susceptibility to apoptosis between genotypes (Fig. 2).
Epithelial cells at the villus tips do not show nuclear
fragmentation and condensation and, additionally, detection of apoptosis in the gastrointestinal tract by
TUNEL is highly variable (19). Therefore, apoptosis at
the villus tips could not be examined in this study.
To determine whether the difference in epithelial cell
migration is the result of differences in crypt cell proliferation, epithelial cells within the crypts of
Lieberkühn were evaluated for expression of PCNA. A
significantly greater percentage of epithelial cells from
CFTR ⫺/⫺ mice (49.4 ⫾ 6.6%) express PCNA than
those from animals expressing at least one copy of the
functional gene (CFTR ⫹/⫺, 36.8 ⫾ 2.7%; P ⫽ 0.003)
(Fig. 3).
The results of the animal studies contradict the
previous cell culture findings presented by Gottlieb
and Dosanjh (9), in which cells expressing the mutated
CFTR were more resistant to apoptosis than cells expressing the wild-type CFTR. Therefore experiments
were designed to determine whether the cell culture
results were misleading.
Mouse mammary epithelial cells expressing a mutated CFTR (C127 ⌬F508) are more resistant to death
induced by 1,000 or 2,000 J/m2 UV irradiation than
cells expressing the wild-type CFTR (Fig. 4). With UV
irradiation doses at or above 4,000 J/m2, a similar
percentage of cell death was seen in both cell types.
This is consistent with the previous findings (9) that
C127 ⌬F508 cells are more resistant than C127 wildtype cells to induction of apoptosis by etoposide and
cycloheximide.
To determine whether the difference in susceptibility to UV irradiation seen in the C127 cell lines is a
result of differences in expression of a functional
CFTR, C127 ⌬F508 cells were infected by an adenovirus expressing the wild-type CFTR before induction of
apoptosis. Although cells overexpressing the mutant
CFTR may have a low level of CFTR activity (2),
expression of the wild-type CFTR in the C127 ⌬F508
cells did not increase the susceptibility of these cells to
death relative to cells infected with the ␤-galactosidase
adenovirus or uninfected cells (Fig. 5). These data
suggest that expression of the functional CFTR does
Fig. 2. Irradiation induced crypt cell apoptosis in CFTR mice. CFTR
mice were subjected to whole body cesium gamma irradiation and
killed 3 h after irradiation. Apoptosis was evaluated in the first 10
cells following the Paneth cells in 50 crypts of Lieberkühn per
animal. The no. of animals per group is noted above each bar. Values
are means ⫾ SD. Gamma irradiation (100 rad) induced 10-fold
increases in apoptosis within the epithelial cells in the crypts of
Lieberkühn in CFTR ⫺/⫺, CFTR ⫹/⫺, and CFTR ⫹/⫹ mice (P ⬍
0.001). There were no genotypic differences in susceptibility to apoptosis.
Fig. 3. Intestinal epithelial cell proliferation in CFTR mice. Proliferating cell nuclear antigen (PCNA) was detected by immunohistochemistry in the first 10 cells following the Paneth cells in 25 crypts
of Lieberkühn in 7 CFTR ⫺/⫺ mice and 5 CFTR ⫹/⫺ mice. The
%PCNA-positive cells per crypt revealed significantly more proliferative capacity in the crypts of Lieberkühn of CFTR ⫺/⫺ mice (49.4 ⫾
6.6%) than in CFTR ⫹/⫺ mice (36.8 ⫾ 2.7%) (P ⫽ 0.003). *Significant
difference (P ⫽ 0.003).
Fig. 1. Intestinal epithelial cell migration in cystic fibrosis transmembrane conductance regulator (CFTR) ⫺/⫺ and CFTR ⫹/⫺ mice.
CFTR mice given an intraperitoneal injection of bromodeoxyuridine
(BrdU) were killed 2–48 h after BrdU injection. Cells that incorporated BrdU were localized immunohistochemically, and the distance
from the base of the crypt of Lieberkühn to each BrdU-positive cell
was determined in 20 animals (3 animals/genotype were evaluated
at 2 and 24 h and 4 animals/genotype were examined at 48 h).
BrdU-labeled epithelial cells from CFTR ⫺/⫺ mice migrate up the
intestinal villi at a faster rate than do cells from CFTR ⫹/⫺ mice.
Each point represents the mean ⫾ SE of the most distal BrdUpositive cell for the animals in the group. At 48 h after BrdU
migration, the distance in migration is significantly different between the 2 genotypes (P ⫽ 0.018).
AJP-Gastrointest Liver Physiol • VOL
281 • SEPTEMBER 2001 •
www.ajpgi.org
PROLIFERATION, NOT APOPTOSIS, IN CF EPITHELIAL MIGRATION
Fig. 4. Dose-response curve for ultraviolet (UV)-induced cell death in
C127 cells expressing the wild-type CFTR (C127 wt) and C127
⌬F508 cells. C127 cells UV irradiated at doses of 0–8,000 J/m2 were
evaluated for cell death 16 h after irradiation by uptake of the DNA
binding fluorochrome YOPRO-1. Only cells with an unstable plasma
membrane will take up YOPRO-1; therefore, the ratio of fluorescence
emissions before and after cell lysis allows for the determination of
%cell death. C127 wt are more susceptible to UV irradiation-induced
death than are cells expressing the ⌬F508-mutated CFTR. Irradiation with 1,000 (P ⫽ 0.04) and 2,000 J/m2 (P ⫽ 0.001) produces
significant genotypic differences in death. Irradiation with doses at
or above 4,000 J/m2 results in similar %cell death in the two cell
lines. Values are means ⫾ SD of 2–6 experiments/dose with a
minimum of triplicates for each dose in each experiment.
not make C127 cells more susceptible to apoptosis than
cells expressing a mutated CFTR. Instead, the data
suggest that the observed difference in apoptosis is
unrelated to the CFTR but instead may have been a
property acquired during the clonal selection process.
To further assess whether the original data were
misleading, a second set of cell lines was tested to
determine susceptibility to apoptosis based on expres-
Fig. 5. Dose effect of UV irradiation on cell death in adenovirusinfected C127 ⌬F508 cells. C127 ⌬F508 cells were infected with an
adenovirus construct containing the wild-type CFTR (wt CFTR) or
the ␤-galactosidase gene (b-gal) or were uninfected. After 48 h of
infection, apoptosis was induced by 2,000 J/m2 UV irradiation. Sixteen hours after irradiation, cells were evaluated for cell death by
uptake of YOPRO-1. Expression of the wild-type CFTR did not alter
susceptibility to death by UV irradiation relative to the uninfected
control cells. ␤-galactosidase expression made the cells more sensitive to UV irradiation. The data reflect 2 experiments performed in
quadruplicate with the range of data reflected by the error bars.
AJP-Gastrointest Liver Physiol • VOL
G685
sion of a functional or mutated CFTR. In the CFT1 cell
lines, at 16 h after UV irradiation, cell death increased
in a dose-dependent manner to a similar extent across
cell lines (Fig. 6). Similar data were obtained 6 h after
UV irradiation using nuclear condensation and fragmentation as markers of apoptosis (Fig. 7). In addition,
UV irradiation of CFT1 cell lines resulted in a similar
time course of DNA degradation (data not shown) with
DNA laddering present at 3 h after UV irradiation in
all four cell lines. Furthermore, cleavage of caspase 3
(PharMingen, La Jolla, CA) and caspase 9 (New England Biolabs, Beverly, MA) after UV irradiation was
not different across cell lines (data not shown). Thus
the current data from the C127 and CFT1 cell lines
suggest that the CFTR does not play a role in apoptosis.
DISCUSSION
Many CF patients and all mouse models of CF have
a common pathology in their susceptibility to gastrointestinal obstruction. In the normal intestine, epithelial
cells line the villi of the absorptive surfaces. These cells
are continuously regenerated by proliferation in the
crypts of Lieberkühn, differentiation, and migration up
the villi. At the tips of the villi, the cells undergo
apoptosis and are shed into the lumen (11, 24). In
addition, Martin et al. (18) have shown that both proliferation and apoptosis occur in the stem cells within
the crypts and that apoptosis within the crypts is likely
to be an important part of the homeostatic process for
tissue maintenance. Under normal conditions, stem
cell proliferation and epithelial sloughing are in equilibrium. As we have shown in the current study, epithelial cells in CFTR ⫺/⫺ mice migrate up the cryptvillus axis at a greater rate than in animals expressing
a functional CFTR. The epithelial cells from CFTR ⫺/⫺
Fig. 6. Dose-response curve for UV-induced cell death in CFT1 cell
lines. CFT1 cells were UV irradiated at doses of 0–8,000 J/m2 and
evaluated for cell death as described in the Fig. 4 legend. No differences were seen across cell lines. Cells expressing the wild-type
CFTR (LCFSN) were equally susceptible to death by UV irradiation
as were cells from the parental cell line (C2), cells expressing additional copies of the ⌬F508 mutation (dF508), or cells expressing the
␤-galactosidase vector control (LC3). The data reflect 2 experiments
performed in quadruplicate with error bars reflecting the range of
the data.
281 • SEPTEMBER 2001 •
www.ajpgi.org
G686
PROLIFERATION, NOT APOPTOSIS, IN CF EPITHELIAL MIGRATION
Fig. 7. Effect of UV irradiation on nuclear condensation and fragmentation in CFT1 cells. CFT1 cells were UV irradiated at 0 and
2,000 J/m2, fixed after 6 h, and evaluated for nuclear condensation
and fragmentation by fluorescence microscopy after staining the
nuclei with 30 ␮M Hoechst no. 33342. A: examples of typical nuclear
condensation and fragmentation for each cell line. B: means ⫾ SD of
%apoptosis seen in 3 separate experiments in which a minimum of
300 cells was counted per cell type per experiment. The amount of
apoptosis seen for each cell line was similar and independent of the
expression of the wild-type or mutated CFTR.
mice migrated 27% farther in 48 h than those from the
CFTR ⫹/⫺ mice. This difference in rate is due to
differences in proliferation and not apoptosis within
the crypts of Lieberkühn with ⬃12% more cells with
proliferative capacity in the CFTR ⫺/⫺ mice than the
CFTR ⫹/⫺ mice. Because villus length is not different
between the genotypes (data not shown), the increased
number of cells proliferating within the crypt and migrating up the villi would lead to an increase in the
mass of cells that are sloughed into the gastrointestinal tract.
The increased proliferative capacity within the
CFTR-null mice may be attributed to a cellular defect
within the cells as a result of a nonfunctional CFTR,
although the mechanism by which this might happen
is unknown. Potentially, differences in intracellular
pH might account for the enhanced proliferation in
AJP-Gastrointest Liver Physiol • VOL
CFTR ⫺/⫺ cells. Cytosolic pH is thought to be one of
the factors that control the rate of proliferation (8) with
cytoplasmic alkalinization constituting a signal for mitogenesis (1, 22, 27). In addition, cells with higher
intracellular pH are more likely to be in the S, G2, or M
phases of the cell cycle, whereas cells with lower pH
are more likely to be in G1 (19). Cells expressing a
mutated CFTR have been reported (5, 6) to have an
elevated cytosolic pH, and overexpression of the wildtype CFTR but not a mutant CFTR results in growth
arrest (23). Thus a mutation in the CFTR may result in
an elevated intracellular pH, which stimulates the cell
to undergo proliferation more readily and thus increase the rate at which epithelial cells of the crypts of
Lieberkühn proliferate and migrate up the intestinal
villi.
All the mouse models of CF show evidence of inflammation and intestinal obstructions (10); thus the possibility that the stimulus to enhanced proliferation
within the intestine of the CFTR ⫺/⫺ mice might be
due to differences in inflammation cannot be ruled out.
Proliferation in nasal polyps (12, 16) and epithelium of
bronchial airways (15) has been shown to be increased
in CF patients and has been attributed to airway
inflammation. Chronic inflammation has been shown
to be present in the CF intestine with increased luminal content of immune cell constituents, secretory immunoglobulins, and cytokines (25). Therefore, inflammation within the gastrointestinal tract potentially
can be a stimulus to enhanced proliferation.
Initial studies by Gottlieb and Dosanjh (9) showed
decreased susceptibility to apoptosis in cells expressing
the mutated CFTR relative to cells expressing the
wild-type CFTR. In this initial study, cycloheximide
and etoposide were used to induce apoptosis. The
CFTR is an anion channel that when mutated does not
allow for chloride to be transported across the plasma
membrane. Because this is a transport problem, one
concern with the use of any drug or chemical to induce
apoptosis is a difference in the ability of the cells to
transport the drug or chemical across the cell membrane. To avoid this potential problem in the current
studies, UV irradiation was utilized as the apoptotic
stimulus. Cells expressing the mutated CFTR were
more resistant to apoptosis induced by UV irradiation
than cells expressing the wild-type CFTR, which is in
accordance with the findings of Gottlieb and Dosanjh
(9). However, one control that was lacking from the
original set of experiments was the introduction of the
wild-type CFTR into these resistant cell lines to restore
susceptibility to apoptosis. In the current study, infection of C127 ⌬F08 cells with an adenovirus expressing
the wild-type CFTR failed to restore susceptibility to
apoptosis. ␤-galactosidase adenovirus-infected cells
were found to be more susceptible to death than both
the uninfected and infected CFTR cells. The reason
behind this increased propensity toward death is not
known. The ␤-galactosidase construct used in these
experiments was the nuclear-targeted protein, which,
because of its location of expression, may make the
cells more susceptible to the DNA-damaging effect of
281 • SEPTEMBER 2001 •
www.ajpgi.org
PROLIFERATION, NOT APOPTOSIS, IN CF EPITHELIAL MIGRATION
UV irradiation. To confirm the findings from the C127
cells, an independent set of cell lines was used to
determine whether expression of a mutated or wildtype CFTR results in differences in susceptibility to
apoptosis. Cells with a mutated CFTR were found to be
no different in their sensitivity to UV irradiation than
cells expressing the wild-type CFTR when evaluated
for membrane destabilization detected by the uptake of
YOPRO-1, nuclear condensation and fragmentation,
DNA degradation, or caspase cleavage. Thus taken
together the data suggest that the CFTR does not play
a role in a cell’s susceptibility to apoptosis.
In the CFTR-null mice, epithelial cell kinetics within
the gastrointestinal tract differ from mice expressing
at least one copy of the functional CFTR. The altered
proliferation within the crypts of Lieberkühn is not
accompanied by changes in apoptosis but results in an
increased rate of cell migration up the villus. The
increased number of cells reaching the intestinal lumen might contribute to the viscosity of the luminal
contents and increase the likelihood of intestinal blockage, one of the pathophysiological hallmarks of CF. If
epithelial cell proliferation and turnover is increased in
parallel in the CF lung and pancreas, this might contribute to the viscosity of the bronchial and pancreatic
secretions and exacerbate the symptoms of CF.
This research was supported by a postdoctoral fellowship from the
Cystic Fibrosis Foundation (A. M. Gallagher), a grant from the
Cystic Fibrosis Foundation (R. A. Gottlieb), and an award from
the Pew Foundation Scholars Program in Biomedical Sciences (R. A.
Gottlieb).
REFERENCES
1. Burns CP and Rozengurt E. Extracellular Na⫹ and initiation
of DNA synthesis: role of intracellular pH and K⫹. J Cell Biol 98:
1082–1089, 1984.
2. Cheng SH, Fang SL, Zabner J, Marshall J, Piraino S,
Schiavi SC, Jefferson DM, Welsh MJ, and Smith AE. Functional activation of the cystic fibrosis trafficking mutant ⌬F508CFTR by overexpression. Am J Physiol Lung Cell Mol Physiol
268: L615–L624, 1995.
3. Clarke LL, Grubb BR, Gabriel SE, Smithies O, Koller BH,
and Boucher RC. Defective epithelial chloride transport in a
gene-targeted mouse model of cystic fibrosis. Science 257: 1125–
1128, 1992.
4. Denning GM, Anderson MP, Amara JF, Marshall J, Smith
AE, and Welsh MJ. Processing of mutant cystic fibrosis transmembrane conductance regulator is temperature sensitive. Nature 358: 761–764, 1992.
5. Elgavish A. High intracellular pH is lowered by retrovirusmediated CFTR gene transfer. Biochem Biophys Res Commun
180: 342–348, 1991.
6. Elgavish A and Meezan E. Altered sulfate transport via anion
exchange in CFPAC is corrected by retrovirus-mediated CFTR
gene transfer. Am J Physiol Cell Physiol 263: C176–C186, 1992.
7. French PJ, van Doorninck JH, Peters RH, Verbeek E,
Ameen NA, Marino CR, de Jonge HR, Bijman J, and
Scholte BJ. A delta F508 mutation in mouse cystic fibrosis
transmembrane conductance regulator results in a temperaturesensitive processing defect in vivo. J Clin Invest 98: 1304–1312,
1996.
AJP-Gastrointest Liver Physiol • VOL
G687
8. Gerson DF, Kiefer H, and Eufe W. Intracellular pH of mitogen-stimulated lymphocytes. Science 216: 1009–1010, 1982.
9. Gottlieb RA and Dosanjh A. Mutant cystic fibrosis transmembrane conductance regulator inhibits acidification and apoptosis
in C127 cells: possible relevance to cystic fibrosis. Proc Natl Acad
Sci USA 93: 3587–3591, 1996.
10. Grubb BR and Gabriel SE. Intestinal physiology and pathology in gene-targeted mouse models of cystic fibrosis. Am J
Physiol Gastrointest Liver Physiol 273: G258–G266, 1997.
11. Hall PA, Coates PJ, Ansari B, and Hopwood D. Regulation
of cell number in the mammalian gastrointestinal tract: the
importance of apoptosis. J Cell Sci 107: 3569–3577, 1994.
12. Hassid S, Degaute MP, Dawance S, Rombaut K, Nagy N,
Choufani G, Decaestecker C, Danguy A, Salmon I, and
Kiss R. Determination of proliferative activity in nasal polyps.
J Clin Pathol 50: 923–928, 1997.
13. Idziorek T, Estaquier J, De Bels F, and Ameisen JC. YOPRO-1 permits cytofluorometric analysis of programmed cell
death (apoptosis) without interfering with cell viability. J Immunol Methods 185: 249–258, 1995.
14. Koller BH, Kim HS, Latour AM, Brigman K, Boucher RCJ,
Scambler P, Wainwright B, and Smithies O. Toward an
animal model of cystic fibrosis: targeted interruption of exon 10
of the cystic fibrosis transmembrane regulator gene in embryonic
stem cells. Proc Natl Acad Sci USA 88: 10730–10734, 1991.
15. Leigh MW, Kylander JE, Yankaskas JR, and Boucher RC.
Cell proliferation in bronchial epithelium and submucosal
glands of cystic fibrosis patients. Am J Respir Cell Mol Biol 12:
605–612, 1995.
16. Lesprit E, Escudier E, Roger G, Pruliere V, Lenoir G,
Reinert P, and Coste A. Characterization of inflammatory
reaction in upper airways of cystic fibrosis patients. Histol Histopathol 15: 395–402, 2000.
17. Li J and Eastman A. Apoptosis in an interleukin-2-dependent
cytotoxic T lymphocyte cell line is associated with intracellular
acidification. J Biol Chem 270: 3203–3211, 1995.
18. Martin K, Kirkwood TB, and Potten CS. Age changes in
stem cells of murine small intestinal crypts. Exp Cell Res 241:
316–323, 1998.
19. Musgrove E, Seaman M, and Hedley D. Relationship between cytoplasmic pH and proliferation during exponential
growth and cellular quiescence. Exp Cell Res 172: 65–75, 1987.
20. Olsen JC, Johnson LG, Stutts MJ, Sarkadi B, Yankaskas
JR, Swanstrom R, and Boucher RC. Correction of the apical
membrane chloride permeability defect in polarized cystic fibrosis airway epithelia following retroviral-mediated gene transfer.
Hum Gene Ther 3: 253–266, 1992.
21. Potten CS. Epithelial cell growth and differentiation. II. Intestinal apoptosis. Am J Physiol Gastrointest Liver Physiol 273:
G253–G257, 1997.
22. Rozengurt E. Early signals in the mitogenic response. Science
234: 161–166, 1986.
23. Schiavi SC, Abdelkader N, Reber S, Pennington S, Narayana R, McPherson JM, Smith AE, Hoppe IV H, and Cheng
SH. Biosynthetic and growth abnormalities are associated with
high-level expression of CFTR in heterologous cells. Am J
Physiol Cell Physiol 270: C341–C351, 1996.
24. Shibahara T, Sato N, Waguri S, Iwanaga T, Nakahara A,
Fukutomi H, and Uchiyama Y. The fate of effete epithelial
cells at the villus tips of the human small intestine. Arch Histol
Cytol 58: 205–219, 1995.
25. Smyth RL, Croft NM, O’Hea U, Marshall TG, and Ferguson
A. Intestinal inflammation in cystic fibrosis. Arch Dis Child 82:
394–399, 2000.
26. Trezise AE and Buchwald M. In vivo cell-specific expression
of the cystic fibrosis transmembrane conductance regulator. Nature 353: 434–437, 1991.
27. Vicentini LM and Villereal ML. Inositol phosphates turnover,
cytosolic Ca2⫹ and pH: putative signals for the control of cell
growth. Life Sci 38: 2269–2276, 1986.
281 • SEPTEMBER 2001 •
www.ajpgi.org