Toxicology in Vitro 24 (2010) 1441–1449

Contents lists available at ScienceDirect

Toxicology in Vitro
journal homepage: www.elsevier.com/locate/toxinvit

Inflammatory parameters in Caco-2 cells: Effect of stimuli nature,

concentration, combination and cell differentiation
Jacqueline Van De Walle, Aurélie Hendrickx, Béatrice Romier, Yvan Larondelle, Yves-Jacques Schneider *
Institut des Sciences de la Vie and UCLouvain, B 1348 Louvain-la-Neuve, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: Enterocytes regulate gut maintenance and defence by secreting and responding to inflammatory media-
Received 24 November 2009 tors and by modulating the intestinal epithelial permeability. In order to develop an in vitro model of the
Accepted 12 April 2010 acute phase of intestinal inflammation, Caco-2 cells were exposed to the inflammatory mediators IL-1b,
Available online 18 April 2010
TNF-a, IFN-c and LPS, and the importance of several experimental parameters, i.e. cell differentiation,
stimulus nature, concentration and combination on the inflammatory response was assessed by measur-
Keywords: ing the production of IL-6, IL-8, PGE-2 and NO and by evaluating the monolayer permeability. A maximal
Inflammation
increase in IL-8, IL-6 and PGE-2 production and monolayer permeability was observed when using the
Caco-2
IBD modeling
cytokines simultaneously at their highest level, but this relied mainly on IL-1b. The effects of TNF-a on
Cytokine IL-8 and IL-6 or NO production were stronger upon combination with IL-1b or IFN-c, respectively,
Lipopolysaccharide whereas cells were unaffected by the presence of LPS. Although NO production, induced by IFN-c-con-
Costimulation taining combinations, was observed only in differentiated cells, general inflammatory response was
Differentiation higher in proliferating cells. The use of a mixture of IL-1b, TNF-a and IFN-c thus accurately mimics intes-
tinal inflammatory processes, but cell differentiation and stimuli combination are important parameters
to take into account for in vitro studies on intestinal inflammation.
Ó 2010 Published by Elsevier Ltd.

1. Introduction
Intestinal inflammation is a natural and protective process,
which is crucial to maintain gut integrity and functioning (Martin
and Wallace, 2006). It requires a continuous crosstalk between the
different cell types present in the gut, and results from and in the
secretion of soluble mediators such as cytokines, eicosanoids, nitric
oxide (NO) and growth factors (Böcker et al., 2003; Martin and
Wallace, 2006; Danese et al., 2008). In turn, these mediators stim-
ulate or attenuate the secretion of other mediators by the target
cells, and their expression thus has to be finely regulated to obtain
a coordinated host’s response.
Intestinal epithelial cells (IECs) have a strategic position at the
interface between the antigenic luminal environment and the inter-
nal milieu. They actively contribute to the gut immune system, medi-
ating processes as mucosal defense, tolerance to resident flora and
barrier repair (Cario et al., 2000). IECs establish bidirectional interac-
Abbreviations: COX, cyclo-oxygenase; IBDs, inflammatory bowel diseases; IECs,
intestinal epithelial cells; iNOS, inducible nitric oxide synthase; LDH, lactate
dehydrogenase; TEER, transepithelial electrical resistance.
* Corresponding author. Address: Institut des Sciences de la Vie and UCLouvain,
Croix du Sud, 5, boîte 3, B 1348 Louvain-La-Neuve, Belgium. Tel.: +32 0 10 47 27 91;
fax: +32 0 10 47 48 95.
E-mail addresses: yjs@uclouvain.be, jacqueline.vandewalle@uclouvain.be (Y.-J.
Schneider).
tions with the underlying immune cells, and participate to the muco-
sal inflammatory response in two ways (Böcker et al., 2003; Danese
et al., 2008). Their property of selectively modulating the permeabil-
ity of the epithelial monolayer – and thus immune cell exposure to
antigens – as well as their ability to synthesize and secrete inflamma-
tory mediators themselves, allow them to trigger immune cells and
initiate the inflammatory process. Conversely, IECs also respond to
various inflammatory mediators secreted by the immune cells, by
modulating the epithelial monolayer permeability and secretion,
thus further amplifying or attenuating the inflammatory process
(Jung et al., 1995; Panja et al., 1998; Bruewer et al., 2005).
Although intestinal inflammation is a continuous and protective
process, a dysregulation of one of its components can lead to se-
vere intestinal disorders. Inflammatory bowel diseases (IBDs), the
collective name for Crohn’s disease and ulcerative colitis, are char-
acterized by chronic and unpredictable attacks of inflammation of
the intestine, causing weight loss, diarrhoea, rectal bleeding,
abdominal pain, fever and anemia (Neuman, 2007). IBDs affect
about 0.5–1% of the population in Western countries and incidence
levels are increasing worldwide (Russel, 2000). The pathogenesis of
IBDs is not fully elucidated, but two central features associated
with IBDs, i.e. defective epithelial cell barrier functioning and an
exaggerated immune activity, are thought to cooperate in a
self-amplifying loop, where barrier dysfunction causes increased
paracellular permeation of harmful luminal antigens and thus

0887-2333/$ - see front matter Ó 2010 Published by Elsevier Ltd.

doi:10.1016/j.tiv.2010.04.002
1442 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449
increases activation of mucosal immune cells, leading in turn to in-
creased release of inflammatory stimuli, IEC response and further
barrier dysfunction (Bruewer et al., 2003; Wang et al., 2006;
Al-Sadi and Ma, 2007).
Several inflammatory mediators are thought to be implicated in
the development of IBDs. However, in vitro research on IECs has
mainly focused on the involvement of three major cytokines, i.e.
Interleukin (IL)-1b, tumor necrosis factor (TNF)-a and interferon
(IFN)-c. IL-1b is a multifunctional cytokine playing a major role
both in the initiation and the amplification of many inflammatory
conditions (Böcker et al., 1998). It is released by various cell types
including monocytes–macrophages, neutrophils and endothelial
cells (Martin and Wallace, 2006) and has been found in increased
concentrations in the intestinal tissue of IBD patients (Ligumsky
et al., 1990; Reinecker et al., 1993; Reimund et al., 1996). IL-1b
mediates important features of IBD, such as the generation of fever,
the reduction of appetite, the release of mediators and the recruit-
ment of leukocytes (Martin and Wallace, 2006). TNF-a and IFN-c
are cytokines that play a key role in the acute phase of inflamma-
tion and diverse immunological processes, modulating e.g. the
recruitment and adhesion of leukocytes and the antigen presenta-
tion (Rosenstiel et al., 2003; Schlottmann et al., 2004). Their
expression levels have been found elevated in the mucosa of IBD
patients (Fais et al., 1991; Reimund et al., 1996), notably due to
an increased secretion by Th-1 lymphocytes of the lamina propria
and, in case of TNF-a, by the IECs themselves.
The presence of bacterial components, such as lipopolysaccha-
rides (LPS), is also considered as an important factor both in the
initiation and in the reactivation of IBDs: a dysregulation of IEC
sensitivity to the common local microflora is believed to cause
an initial inappropriate inflammatory stimulus, leading to an exag-
gerated cytokine presence and IBD development (Caradonna et al.,
2000; Cario et al., 2000; Böcker et al., 2003). Moreover, the leaky
IEC barrier observed during IBDs leads to increased permeation
to bacteria during inflammation, which further enhances immune
activity (Aoki, 1978; Bruewer et al., 2005).
The in vitro effects of IL-1b, TNF-a, IFN-c and LPS on IECs consist
in activation of intracellular cascades, leading to an increased
transcriptional activity and the secretion of interleukin (IL)-8
(Schuerer-Maly et al., 1994), IL-6 (Parikh et al., 1997; Ogle et al.,
1997; Vitkus et al., 1998), prostaglandin (PG)-E2 (Grishin et al.,
2004; Wright et al., 2004; Duque et al., 2006) and/or NO (Chavez
et al., 1999; Forsythe et al., 2002), as well as increase of the para-
cellular permeability through defects in tight junction functioning
or assembly (Forsythe et al., 2002; Ma et al., 2005; Bruewer et al.,
2005; Al-Sadi and Ma, 2007).
These studies however show a lack of homogeneity in experi-
mental conditions such as stimulus concentration, exposure dura-
tion, and nature and differentiation stage of cells. Moreover,
although IL-1b, TNF-a, IFN-c and LPS are individually known to
contribute to the relapse of inflammation, reports on their con-
certed action as occurring during the acute phase of IBDs are scarce
(Ou et al., 2009). In view to establish a valid and consistent in vitro
model of the physio-pathological behaviour of enterocytes during
the acute phase of IBDs, we exposed Caco-2 cells, a well-estab-
lished and widely used model of the human intestinal barrier, to
IL-1b, TNF-a, IFN-c and LPS exposure, and evaluated inflammatory
parameters with respect to cell differentiation, stimulus concentra-
tion and stimuli combination.
2. Materials and methods
2.1. Chemicals
Culture reagents were from Lonza (Verviers, BE) unless
mentioned otherwise. IL-1b, TNF-a, and LPS were purchased from
Sigma–Aldrich (St. Louis, MO) and IFN-c was from Calbiochem
(Darmstadt, DE). Arachidonic acid, taurocholate and Triton-X-100
were from Sigma.
2.2. Cell culture
The Caco-2 cell line, derived from a human colon adenocarci-
noma (ATCC, Rockville, MD), was used between passages 30 and
50 and cultured in DMEM containing 4.5 g/l glucose, 25 mM hepes,
10% (v/v) heat-inactivated FBS (Hyclone Perbio-Sciences, Erem-
bodegem, BE), 2% (v/v) L-glutamine 200 mM and 1% (v/v) nones-
sential amino acids (NEAA) (Invitrogen, Carlsbad, CA). Cells were
grown on 175 cm2 flasks (Greiner Bio-One, Strickenhausen, DE)
in an atmosphere of 5% CO2/95% air (v:v) at 37 °C. For experiments,
cells were seeded at a density of 40  103 cells/cm2 on type I colla-
gen (Sigma–Aldrich) precoated 24-well plates (Nunc, Roskilde, DK)
and cultured in standard medium supplemented with 1% pen
(10  103 U/ml)-Strep (10 mg/ml) during 21 days to obtain fully
differentiated cells. The cells were then washed with phosphate
buffered saline (PBS, 137 mM NaCl, 2.68 mM KCl, 1.14 mM KH2PO4,
8 mM Na2HPO4, pH 7.2) and each treatment was applied for 24 h in
culture medium containing 1% (v/v) FBS. In an additional experi-
ence, a selection of the treatments was applied on differentiated
as well as proliferating cells; the latter consisting in Caco-2 cells
treated 24 h after seeding.
2.3. Determination of cytotoxicity
Cytotoxicity of the treatments was evaluated by measuring the
activity of lactate dehydrogenase (LDH) in the culture media, since
this enzyme is released by damaged or necrotic cells (Cytotoxicity
detection kit, Roche Diagnostics GmbH, Mannheim, DE). Results
were expressed as a percentage of the positive control, consisting
of cells exposed to 1% (v/v) Triton-X-100.
2.4. Determination of IL-6 and IL-8 secretion
The extracellular media were collected and centrifuged at
16  103 g for 10 min. IL-6 and IL-8 secretion was evaluated using
a sandwich Elisa method (BD Biosciences Pharmingen, San Diego,
CA) as in Van De Walle et al. (2008) and quantified in pg/ml using
the standard provided with the kit, with sensitivity limits being of
2.2 and 0.8 pg/ml, respectively for the IL-6 and IL-8 assay. Results
were subsequently expressed in relative terms to the negative con-
trol (untreated cells) to facilitate comparison between groups.
2.5. Determination of the arachidonic acid cascade activity through
PGE-2 production
The activation of the arachidonic acid cascade was estimated
through the production of PGE-2 after addition of arachidonic acid
(Romier-Crouzet et al., 2009). Briefly, 24 h after treatment, cells
were washed and incubated with 10 mM arachidonic acid in PBS
during 10 min. PGE-2 secretion in extracellular medium was quan-
tified in pg/ml using a Prostaglandin E2 Enzyme Immunoassay Kit
(Cayman Chemical Corporation, Inc. Ann Arbor, MI) in accordance
with the manufacturer’s protocol, with a LOD of 15 pg/ml. The
amount of PGE-2 produced by untreated cells was subtracted from
the other values to obtain the amount of PGE-2 produced by induc-
tion of the arachidonic acid cascade (further referred to as ‘‘in-
duced PGE-2 production”).
2.6. Determination of NO production
Nitric oxide, present in the culture medium as nitrite and
nitrate, was assayed using the colorimetric Nitric Oxide Assay
 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449 1443

(Calbiochem) according to the manufacturer’s protocol. Briefly, the 3. Results

nitrate present in the supernatant was enzymatically transformed
into nitrite by nitrate reductase and the total nitrite concentration 3.1. Determination of treatments and cytotoxicity
was measured photometrically using the Griess reagent (p-amino-
benzenesulfonamide in 3 N HCl and N-(1-napthyl)ethylenediamine Stimuli of various nature, i.e. the cytokines IL-1b, TNF-a and
dihydrochloride in water). The amount of NO was determined in IFN-c, and LPS from Escherichia coli, were chosen to induce inflam-
lM from a nitrite standard curve provided with the kit, with the mation of Caco-2 cells. Concentration ranges were determined
sensitivity limit being of 2.5 lM. Results were expressed in relative according to the literature. IL-1b was used at concentrations be-
terms to the negative control (untreated cells). tween 1 and 25 ng/ml (Panja et al., 1998; Al-Sadi and Ma, 2007).
TNF-a and IFN-c were applied at 1 or 50 ng/ml (Schuerer-Maly
et al., 1994; Jung et al., 1995). LPS was applied in a range between
2.7. Determination of barrier integrity
0.05 and 10 lg/ml (Cario et al., 2000; Böcker et al., 2003). The sub-
stances were applied either individually at their maximal concen-
Cells were seeded on type I collagen precoated Transwell inserts
tration, or simultaneously at their maximal or minimal
with a pore diameter of 0.4 lm (Corning Costar Corp., Cambridge,
concentration, or simultaneously at their maximal concentration
MA) at a density of 90  103 cells/cm2 and cultured for 21 days.
except for one substance whose concentration diminished gradu-
The use of cell culture inserts allows access to both the basolateral
ally to evaluate its individual contribution to the inflammatory
and the apical sides of the cells, respectively representative of the
parameters (see Table 1). Since it has been repeatedly reported in
circulatory and luminal poles of the intestinal epithelium. A selec-
the literature that TNF-a and IFN-c operate together to induce
tion of 9 out of the 16 treatments was made based on the observa-
inflammatory parameters (Kolios et al., 1999; Abreu et al., 2002;
tions with the experiments described above. Stimuli of endogenous
Rosenstiel et al., 2003; Wright et al., 2004; Wang et al., 2006),
nature (IL-1b, TNF-a and IFN-c were applied at the basolateral side
we performed an additional treatment in which the concentration
of the cells. LPS was applied at both sides, representing on the one
of both substances varied simultaneously and similarly, either or
hand the common luminal microflora and, on the other hand, the
not in the presence of the other stimuli. All treatments were ap-
increased presence of bacterial components in serum during IBDs.
plied for 24 h to allow inflammatory responses to fully develop.
Cells were exposed at the basolateral side to the appropriate dose
No significant cytotoxicity of vehicle and treatments was observed
of IL-1b, TNF-a and IFN-c and half of the total dose of LPS in 2.6 ml
as evidenced by the LDH assay in the culture media (data not
of culture medium containing 1% FBS. The apical side received
shown).
1.5 ml of the same culture medium, containing half of the dose of
LPS. The transepithelial electrical resistance (TEER) was measured
at the beginning and end of the experiment using an Endohm-24 3.2. Effects of treatments on IL-8 secretion by differentiated Caco-2
chamber (World Precision Instruments, Sarasota, FL). cells

Enterocytes secrete the chemokine IL-8 to attract and activate

2.8. Comparison of proliferating and differentiated Caco-2 cells
neutrophils during the acute phase of inflammation (Hoffmann
et al., 2002). IL-8 secretion by fully differentiated Caco-2 cells in
In order to evaluate possible differences in inflammatory sensi-
our experimental conditions, expressed in relative terms to the
tivity between immature and mature enterocytes, we performed
negative control, is shown in Fig. 1. Basal IL-8 secretion in culture
the dosage of IL-6, IL-8, PGE-2 and NO on both proliferating, non-
confluent Caco-2 cells (24 h after seeding), as well as fully differen-
tiated, 21-day post-confluent Caco-2 cells. Cells were either or not
treated with IL-1b, TNF-a, IFN-c and LPS at their maximal concen- Table 1
tration. After the experiment, cells were washed with PBS, col- Treatments used on differentiated Caco-2 cells. Caco-2 cells were cultured until
lected and centrifuged (1 min at 12,000g). The pellet was then 21 days after confluence. Treatments were applied during 24 h and consisted in
exposure to IL-1b (1–25 ng/ml), TNF-a (1–50 ng/ml), IFN-c (1–50 ng/ml) or LPS (0–
suspended in Reporter Lysis Buffer (Promega Corp., Madison, WI)
10 lg/ml). Treatments were applied as follows: (i) IL-1b, TNF-a, IFN-c or LPS
and frozen at 20 °C for 30 min., followed by thawing, centrifuga- individually; (ii) TNF-a and IFN-c together at their highest level; (iii) all the stimuli
tion (2 min at 16,000g) and collection of the supernatant. Total simultaneously (C:) at their maximal (C:MAX) or minimal (C:MIN) concentration; (iv)
protein content was evaluated using the bicinchoninic assay (Sig- all the stimuli simultaneously at their maximal concentration except for one whose
ma), which is a colorimetric method measuring the reduction of concentration was diminished; (v) all the stimuli at their maximal concentration
except for TNF-a and IFN-c whose concentration was diminished simultaneously
Cu2+ after the formation of a Cu2+–protein complex in alkaline
(C:TNF-a–IFN-c 1).
conditions.
Values of IL-6, IL-8, NO and PGE-2 secretion were reported to IL-1b LPS TNF-a IFN-c
(ng/ml) (lg/ml) (ng/ml) (ng/ml)
the total protein content. Statistical analysis was performed to de-
tect differences between the two cell maturity stages either with- i IL-1b 25 25
LPS 10 10
out or with inflammatory stimulation.
TNF-a 50 50
IFN-c 50 50
ii TNF-a–IFN-c 50 50 50
2.9. Statistical analysis iii C:MAX 25 10 50 50
C:MIN 0,1 0.05 1 1
All data are expressed as mean ± SEM (standard error of the iv C:IL-1b 5 5 10 50 50
C:IL-1b 1 1 10 50 50
means). Values were compared to the negative control (untreated
C:IL-1b 0.1 0,1 10 50 50
cells) or positive control (cells exposed to all stimuli at their high- C:LPS 1 25 1 50 50
est concentration) using the Dunnett’s test after performing ANO- C:LPS 0.1 25 0,1 50 50
VA. All statistical analyses were performed using Kaleidagraph C:LPS 0.05 25 0.05 50 50
version 4.0 (Synergy Software, Reading, PA). Differences were con- C:TNF-a 1 25 10 1 50
C:IFN-c 1 25 10 50 1
sidered statistically significant when the probability value p was
v C:TNF-a–IFN-c 1 25 10 1 1
lower than 0.05.
1444 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449

75 TNF-a and IFN-c, both at 50 ng/ml, did significantly upregulate

IL-8 secretion

+++

+++
+++

++

+++
(x control)

60 IL-6 secretion by 7.9-fold, suggesting additive effects between

+++
++

45 these cytokines. The simultaneous exposure to all the stimuli at

++
*

*
+++

*
**
30 their highest levels resulted in a maximal increase (16-fold) in

++
***

**
**
++

***

+
+++
***

15 IL-6 secretion. Although this value failed to differ significantly (p-

***

***

0 value of 0.3) from the induction by IL-1b alone, it nevertheless

C: LPS 0.05
TNFa-IFNg 50

C: LPS 0.1
C: IL1b 0.1

C: TNF-IFN 1
C: MAX

C: LPS 1

C: IFNg 1
C: TNFa 1
IFNg 50
TNFa 50

C: MIN
C: IL1b 5
C: IL1b 1
LPS 10
IL1b 25

suggests possible additive effects of these stimuli to induce IL-6

secretion. The apparent indifference to LPS exposure allows us to
conclude that IL-6 secretion primarily relies on the presence of
IL-1b, with possible additive effects of TNF-a and IFN-c.
treatment

Fig. 1. IL-8 secretion by differentiated Caco-2 cells exposed to inflammatory 3.4. Effects of treatments on PGE-2 production related to the induction
stimuli. Differentiated Caco-2 cells were submitted during 24 h to inflammatory of the arachidonic acid cascade
stimuli as exposed in Table 1. IL-8 secretion in culture medium was evaluated by
ELISA. Values were expressed in function of the negative control (untreated cells)
and represent means ± SEM, n = 3. Basal IL-8 secretion level was of 96.7 ± 1.9 pg/ml. Under inflammatory conditions, there is an increased activation
Differences with the negative control or with C:MAX are respectively marked with of the arachidonic acid cascade, resulting in enhanced production
+
or * when p-value < 0.05, ++ or ** when p < 0.01, +++ or *** when p < 0.001. of prostanoids (Raub et al., 1995). We estimated the effect of
24 h exposure of fully differentiated Caco-2 cells to the different
treatments on the activity of the arachidonic acid cascade by mea-
medium by untreated Caco-2 cells was evaluated at 96.7 ± 1.9 pg/ suring the amount of PGE-2 secreted after a further 10 min expo-
ml after 24 h. Stimulation with IL-1b at 25 ng/ml or TNF-a at sure of cells to arachidonic acid (see Fig. 3). Basal PGE-2
50 ng/ml increased the secretion by 32.3- and 4.6-fold, respec- production by untreated Caco-2 cells reached 535 ± 169 pg/ml of
tively. A maximal induction (54.6-fold) was observed when using culture medium after 10 min. PGE-2 production increased signifi-
these stimuli together at their highest dose. IL-8 secretion was cantly by 151 ± 50 pg/ml using IL-1b at 25 ng/ml, while cells re-
unaffected by LPS or IFN-c, used either individually or in combina- mained unresponsive to LPS, TNF-a or IFN-c. Simultaneous
tion, and was unchanged by doses of IL-1b or TNF-a below 5 or incubation with all these substances at their highest dose further
50 ng/ml, respectively. increased the induced production of PGE-2 by 235 ± 55 pg/ml,
although this failed to differ significantly (p-value of 0.31) from
3.3. Effects of treatments on IL-6 secretion by differentiated Caco-2 induction with IL-1b alone. Interestingly, co-incubation of TNF-a
cells and IFN-c caused a slight (69 ± 37 pg/ml) increase in the induction
of PGE-2 production which, although not significant (p-value of
IL-6 is a cytokine causing widespread and pleiotropic effects in 0.1), could indicate a small effect of the simultaneous presence of
chronic and acute inflammation (Mudter and Neurath, 2007). We TNF-a and IFN-c. Consistent with this, exposure of cells to all stim-
quantified IL-6 secretion by differentiated Caco-2 cells exposed to uli at their maximal dose while diminishing TNF-a, IFN-c or both,
the various inflammatory conditions under investigation in this led to a significant reduction in induction compared to the maxi-
study and expressed the results in function of the negative control mum, confirming the synergic action of these substances. Cells re-
(see Fig. 2). Basal IL-6 secretion in the culture medium of untreated mained totally unresponsive to the presence of LPS. The induction
Caco-2 cells was of 1.01 ± 0.01 pg/ml. The exposure to individual of the arachidonic acid cascade in our experimental conditions
cytokines resulted in a 9.2-fold increase in IL-6 secretion at seems thus to rely principally on stimulation by IL-1b, with TNF-
25 ng/ml of IL-1b, while the 7.6-fold upregulation using 50 ng/ml a and IFN-c playing an additional but minor role when used
of TNF-a just failed to be significant (p-value of 0.059). About together.
10 lg/ml of LPS and 50 ng/ml of IFN-c had no effect (IL-6 upregu-
lation of respectively 1.5- and 1.8-fold). However, co-exposure to
Induced PGE-2 production

++

350
++

++
++

300
250
++
++

25
(pg/ml)

200
+

+
IL-6 secretion

+
*
+
+

20
+

150
+
*
+
*
(x control)

100
++
*
+

15
+
+

**

50
**
**

**
+

10
+

0
C: TNFa-IFNg 1
TNFa-IFNg 50

C: LPS 0.05
C: IL1b 0.1

C: LPS 0.1

C: TNFa 1
C: IFNg 1

5
C: LPS 1
TNFa 50
IFNg 50

C: MAX

C: IL1b 5
C: IL1b 1
C: MIN
*

IL1b 25
LPS 10
*
*

0
TNFa-IFNg 50

C: TNF-IFN 1
C: LPS 0.05
C: IL1b 0.1

C: LPS 0.1

C: TNFa 1
C: IFNg 1
C: LPS 1
TNFa 50
IFNg 50

C: MAX

C: IL1b 5
C: IL1b 1
C: MIN
IL1b 25
LPS 10

treatment

Fig. 3. Induction of the arachidonic acid cascade in differentiated Caco-2 cells

treatment exposed to inflammatory stimuli. Differentiated Caco-2 cells were exposed during
24 h to inflammatory stimuli as in Table 1. The activity of the arachidonic acid
Fig. 2. IL-6 secretion by differentiated Caco-2 cells exposed to inflammatory cascade was determined as the amount of PGE-2 secreted after an additional 10 min
stimuli. Differentiated Caco-2 cells were submitted during 24 h to inflammatory incubation with arachidonic acid (referred to as ‘‘induced PGE-2 production”). The
stimuli as exposed in Table 1. IL-6 secretion in culture medium was evaluated by amount of PGE-2 secreted by untreated cells (535 ± 169 pg/ml) was subtracted from
ELISA. Values were expressed in function of the negative control (untreated cells) the other values to obtain the amount of PGE-2 produced through induction by the
and represent means ± SEM, n = 3. Basal IL-6 secretion level was of 1.01 ± 0.01 pg/ treatments. Values represent means ± SEM, n = 6. Differences with the negative
ml. Differences with the negative control or with C:MAX are respectively marked control or with C:MAX are respectively marked with + or * when p < 0.05, ++ or **
with + or * (p-value < 0.05). when p < 0.01.
 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449 1445

3.5. Effects of treatments on NO production by differentiated Caco-2 of 5.0 ± 1.8 nmoles/ml. Treatment with IL-1b, TNF-a or LPS individ-
cells ually had no effect at the concentrations used, but IFN-c tended to
cause an upregulation (2.1 ± 0.7-fold at 50 ng/ml, p-value of 0.15).
NO is an important actor in inflammatory processes, regulating The combination of IFN-c and TNF-a at 50 ng/ml induced a signif-
blood flow, pain transmission and activation of the immune system icant increase (1.7 ± 0.2-fold) in NO secretion, and the presence of
(Kolios et al., 2004). We assayed NO, present as nitrite and nitrate, all stimuli at their maximal dose further increased this to 5.8-fold.
in the culture medium of differentiated Caco-2 cells and expressed A progressive decrease of the level of IL-1b seemed to cause a grad-
the results in function of the negative control (see Fig. 4). Basal NO ual attenuation of NO production but this failed to be statistically
secretion in culture medium of unstimulated cells during 24 h was significant. By contrast, the decrease of IFN-c almost completely
abolished NO upregulation, suggesting that NO induction strongly
depends on the presence of IFN-c in combination with other
8.0 stimuli.
+
NO secretion

+++

+++
6.0
+++

+++

+++
(x control)

3.6. Effects of treatments on the cell barrier integrity of differentiated

4.0
++

Caco-2 cells
++
++
*

*
*

*
*
*

2.0
*

An increased paracellular permeability is a hallmark of intesti-

0.0 nal inflammation (Bruewer et al., 2003). We evaluated the effect

C: TNFa-IFNg 1
TNFa-IFNg 50

C: LPS 0.05

of our treatments on the Caco-2 cell barrier integrity through the

C: IL1b 0.1

C: LPS 0.1

C: TNFa 1
C: IFNg 1
C: LPS 1
TNFa 50
IFNg 50

C: MAX

C: IL1b 5
C: IL1b 1
C: MIN
IL1b 25
LPS 10

measurement of the transepithelial electrical resistance (TEER),

which quantifies the movement of ions across the paracellular
pathway. TEER in untreated cells was of 1500 X cm 2. A preli-
treatment minary experiment revealed no significant effect on TEER when
stimuli were all applied in the apical compartment (results not
Fig. 4. NO secretion by differentiated Caco-2 cells exposed to inflammatory stimuli. shown), which is consistent with the polarized localization of cyto-
Differentiated Caco-2 cells were exposed during 24 h to inflammatory stimuli as in
Table 1. Nitric oxide, present in the culture medium as nitrite and nitrate, was
kine receptors in IEC. When applying cytokines individually in the
assayed using a colorimetric assay. Values were expressed in function of the basolateral compartment, only IL-1b at 25 ng/ml was effective to
negative control (untreated cells) and represent means ± SEM, n = 4. Basal NO significantly decrease (20 ± 4%) the TEER. The use of additional
secretion in culture medium was of 5.0 ± 1.8 nmoles/ml. Differences with the stimuli further diminished the TEER up to 25% but this was not sta-
negative control or with C:MAX are respectively marked with + or * when p < 0.05, ++
tistically different from the values obtained when using IL-1b
or ** when p < 0.01, +++ or *** when p < 0.001.
alone. The TEER remained insensitive to LPS applied in equal pro-
portions at the apical and basolateral compartment, in the pres-
ence or absence of the additional stimuli. No significant LDH
leakage was observed in the culture media (results not shown)
(Fig. 5).

3.7. Comparison between proliferating and differentiated Caco-2 cells

We evaluated differences in sensitivity between immature and

mature enterocytes by performing the dosage of IL-8, NO and in-
duced PGE-2 production on proliferating, non-confluent as well
as on fully differentiated, 21-days post-confluent Caco-2 cells. As
can be seen in Table 2, basal secretion level of IL-8, IL-6 and PGE-
Fig. 5. TEER of differentiated Caco-2 cells exposed to inflammatory stimuli. Caco-2
2 failed to differ between the two maturity stages whereas NO
cells were cultured on inserts until 21 days after confluence and inflammatory secretion was detected only in differentiated cells. The exposure
treatments were applied during 24 h. IL-1b, TNF-a and IFN-c were applied at the to the inflammatory inducers IL-1b, TNF-a, IFN-c and LPS simulta-
basolateral side of the cells. LPS was applied at both sides. No significant LDH neously at their highest concentration, caused an increase in IL-8
leakage was observed in the culture media. The TEER was measured at the
secretion and PGE-2 production, which was significantly higher
beginning and at the end of the experiment. Values were expressed in function of
the negative control (untreated cells, TEER of 1500 X cm 2) and represent in proliferating (reaching 2000 pg/lg for IL-8 secretion and
means ± SEM, n = 3. Differences with the negative control or with C:MAX are 45 pg/lg for PGE-2 production) than differentiated cells (reach-
respectively marked with + or * when p < 0.05, ++ or ** when p < 0.01. ing respectively 100 and 14 pg/lg). Upregulation of IL-6

Table 2
Comparison of IL-8, IL-6, NO and PGE-2 production between proliferating and differentiated Caco-2 cells. Proliferating (24 h after seeding) and differentiated (21 days after
confluence) Caco-2 cells were untreated (control cells) or treated with inflammatory stimuli consisting in IL-1b (25 ng/ml), TNF-a (50 ng/ml), IFN-c (50 ng/ml) and or LPS (10 lg/
ml). The levels of IL-8 (pg/ml), IL-6 (pg/ml), PGE-2 (pg/ml) and NO (pM) were determined as described in Section 2. The results were reported to the total protein content and
relative differences between untreated and treated cells were calculated (D). Values represent means ± SEM, n = 3–5. Significant differences (p < 0.05) in the basal or induced
production of IL-8, PGE-2 and NO between proliferating and differentiated cells are respectively marked with + or *. n.d. = not detectable.

Proliferating cells Differentiated cells

Control Stimulus D Control Stimulus D
IL-8 (pg/lg) 11.3 ± 6.4 2026.6 ± 750.4 179.8 * 1.5 ± 0.1 95.5 ± 10.6 61.5
IL-6 (pg/mg) 13.0 ± 2.9 116.2 ± 20.1 8.9 16.8 ± 0.2 154.9 ± 30.2 9.2
PGE-2 (pg/lg) 11.5 ± 2.9 46.5 ± 12.6 4.1 * 8.6 ± 2.7 13.5 ± 3.6 1.6
NO (pM/lg) n.d. n.d. + 8.0 ± 2.9 35.9 ± 9.6 4.5
*
1446 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449
secretion (9-fold) was not different between the cell maturity
stages, whereas NO upregulation (4.5-fold) was only observed in
differentiated Caco-2 cells.
4. Discussion
This study aimed at determining how the experimental param-
eters of stimuli nature, concentration and combination, as well as
the degree of cell differentiation affect the inflammatory response
of an IEC model consisting of the human Caco-2 line, in view to
establish a coherent and valid in vitro cell culture system of the
acute phase of IBDs.
Although different cell types in the gut contribute to the inflam-
matory response in vivo, we focused our research on the mediation
of intestinal epithelial cells, since they play a crucial role in IBD
pathogenesis (Panja et al., 1998). The Caco-2 line was selected
since differentiated Caco-2 cells express functional tight junctions,
brush border characteristics and biotransformation enzymes (Pinto
et al., 1983), and their differentiation process is thought to reflect
the maturation occurring in vivo along the crypt-villus axis
(Cavicchi and Whittle, 1999; Vignoli et al., 2001), which makes
Caco-2 cells probably the best in vitro model of human enterocytes
currently available.
The intestinal inflammatory response requires the correct inte-
gration of the various signals that are released or received by the
different cell types present in the gut. Each of these signals has a
specific role in the development of the response, and the coordi-
nated regulation of their expression is crucial to obtain a net
inflammatory response (Jung et al., 1995). We analyzed how the
endogenous mediators IL-1b, TNF-a, IFN-c whose mucosal levels
are increased during IBDs as well as exogenous LPS (Fais et al.,
1991; Reimund et al., 1996; Caradonna et al., 2000), individually
or concomitantly affect the inflammatory response in IECs. The
concentrations of stimuli were determined according to the litera-
ture (Jung et al., 1995; Cario et al., 2000; Rosenstiel et al., 2003;
Al-Sadi and Ma, 2007) and were verified for the lack of cytotoxicity.
Although the in vivo physiological plasma levels of the cytokines
rather lay in the pg/ml range (Fais et al., 1991; Reimund et al.,
1996), the concentrations used in vitro could be considered as phy-
sio-pathologically relevant, reflecting the much localized increase
in cytokine production in the gut mucosa during active IBD. Phys-
iological intestinal inflammation occurs through the synchronized
and orchestrated release and interaction of inflammatory media-
tors. IBDs development however relies on an increased inflamma-
tory stimulation by exogenous (e.g. bacterial components) as well
as endogenous (e.g. pro-inflammatory cytokines) mediators, and is
characterized by ongoing latency and relapse of the disease,
involving subsequent initialization, amplification and attenuation
of the inflammatory process. We used experimental conditions
integrating simultaneously the well-known IBD causative agents
IL-1b, TNF-a, IFN-c and LPS in order to represent the acute inflam-
matory reaction as it might occur during the active phase of IBDs.
We focused our analysis on several parameters that aim at acti-
vating leukocytes and IEC and thus further promote inflammation
in an autocrine or paracrine fashion, either in a direct way by
secreting more soluble mediators, or in an indirect way by increas-
ing cell permeability and thus gut antigen permeation. IL-8 is a
chemokine attracting neutrophils during the early stage of intesti-
nal inflammation, and its expression level is increased during IBD
(McCormack et al., 2001). IL-6 is a cytokine mediating pleiotropic
effects during both acute and chronic inflammation, and its expres-
sion level has been found to coincide with the disease severity and
activity (Mudter and Neurath, 2007), which makes it particularly
interesting to target for this study. Prostaglandins are also impor-
tant components in the intestinal immune processes. They stimu-
late mucus and bicarbonate secretion, elevate mucosal blood flow,
regulate the recruitment of leukocytes and mediate fever and pain
(Tanabe and Tohnai, 2002; Martin and Wallace, 2006). Increased
production of prostaglandins has been observed during IBD and
this can rely both on the induction of the COX-2 enzyme (Singer
et al., 1998), as well as the overexpression of microsomal PGE-syn-
thase (Subbaramaiah et al., 2004). Nitric oxide is a mediator pro-
duced through the conversion of L-arginine to L-citrulline by the
nitric oxide synthases (NOS) (Aktan, 2004) and its role in IBD
seems complex. In the gut, small amounts of NO are constitutively
synthesized to regulate gastro-intestinal blood flow, motility and
secretion. In inflammatory conditions, there is a large increase in
NO production through upregulation of the expression of the
inducible-NOS (iNOS) isoform, and this can lead to altered perme-
ability and motility, cytokine production, oxidative damage and
tissue injury (Vignoli et al., 2001; Kolios et al., 2004).
Our results identify the cytokine IL-1b as the major determinant
of the production of IL-6, IL-8 and PGE-2 in Caco-2 cells. IL-1b
receptors are constitutively expressed on IECs (Panja et al., 1998),
and mediate the activation of the two major inflammatory cas-
cades, depending respectively on the transcription factor nuclear
factor (NF)-jB and the Mitogen Activated Protein Kinases (MAPKs)
(Saklatvala et al., 2003). Response elements to NF-jB and down-
stream targets of the MAPKs such as the transcription factors Acti-
vating-Protein (AP)-1 and CCAAT/Enhancer Binding Protein (C/EBP)
are found in the promotors of IL-6 (Hershko et al., 2004), IL-8
(Hoffmann et al., 2002) and COX-2 (Tanabe and Tohnai, 2002),
which is consistent with our results and confirms the central role
of this mediator in the induction of inflammation.
Most interestingly, we found that exposure to TNF-a alone was
only efficient in upregulating IL-8, while IL-6, NO and PGE-2 pro-
duction were significantly upregulated by the use of TNF-a in com-
bination with IFN-c. According to Wang et al. (2006), the presence
of IFN-c induces an upregulation of the expression of the TNF-
receptor 2, thus potentiating the action of TNF-a. TNF-a is cur-
rently the principal target of IBD therapy, and our findings open
the perspective to synergize the efficacy of this strategy by simul-
taneously inhibiting IFN-c.
Surprisingly, we found that IFN-c used alone failed to upregu-
late IL-8 secretion in our experimental conditions, which seems
to conflict with the characteristic, high inducibility of the IL-8 gene
(Hoffmann et al., 2002). Moreover, we also failed to observe any
significant NO release upon exposure to IFN-c, while the iNOS pro-
moter is known to contain Interferon Regulatory Factor (IRF)-cis
elements that are activated through the Jak (IFN-c activated ki-
nase)-STAT signaling pathway upon IFN-c exposure (Aktan,
2004). However, important variations in the sensitivity and re-
sponse to IFN-c seem to exist among different IEC lines (Eckmann
et al., 1993; Schuerer-Maly et al., 1994; Jung et al., 1995;
Schlottmann et al., 2004) and, although iNOS is known to be an
early-response gene, it could be that longer incubation periods
(48 h or more) are recommended to fully induce NO production
(Chavez et al., 1999). IFN-c appeared nevertheless very efficient
to induce NO production when applied in combination with IL-1b
or TNF-a, providing evidence for their additive effects on cis-acting
elements located in the iNOS promoter (Cavicchi and Whittle,
1999; Chavez et al., 1999; Aktan, 2004).
An increased mucosal permeability is considered as a hallmark
of IBDs (Bruewer et al., 2005; Al-Sadi and Ma, 2007), contributing
to both the initiation and the amplification phases of the inflam-
matory condition through increased passage of luminal antigens.
Our results suggest a major role for the cytokine IL-1b in the
reduction of cell barrier integrity occurring during inflammatory
conditions. It was surprising that neither TNF-a, IFN-c nor LPS
seemed to affect monolayer permeability nor we hypothesize that
sustained incubation periods seem more appropriate to evaluate
 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449
the contribution of inflammatory stimuli to barrier hyperperme-
ability. In this respect, Ma et al. (2005) showed that an 48 h incu-
bation period with TNF-a caused tight junction disruption both
through increased Myosin Light Chain Kinase (MLCK) protein
expression and subsequent contraction of actin filaments, and de-
creased ZO-1 expression (Ma et al., 2004). Also sustained dura-
tions of incubation with IFN-c induced the endocytosis of tight
junction proteins claudin-1, occludin and JAM-1 (Bruewer et al.,
2006). Moreover, Chavez et al. (1999) indicated an important con-
tribution of iNOS to Caco-2 cell monolayers hyperpermeability
using IL-1b, TNF-a and IFN-c concomitantly during an 48 h incu-
bation. These data additionally demonstrate that cytokines can
modulate cell barrier integrity at various levels, which makes it
uneasy to discriminate between the direct effects of an individual
cytokine and those that rely on the prior induction of additional
inflammatory mediators (Chavez et al., 1999; Forsythe et al.,
2002; Han et al., 2003; Clark et al., 2005; Wang et al., 2006).
The modulation of the duration of exposure should therefore be
a useful tool to discriminate between rapid/direct and time-
consuming/indirect effects of the treatment, with prolonged
incubation periods allowing the development of supplementary
features of inflammation.
The intestinal epithelium is constantly renewing and its main-
tenance relies on a balance between three crucial processes, i.e.
(i) the division of stem cells at the bottom of the crypts, (ii) the
migration and the differentiation of cells along the crypt-villus
axis, and (iii) finally the apoptosis of the cells after reaching the
top of the villus (Chell et al., 2006). The cells that constitute
the intestinal epithelium in vivo thus differ in maturity, which
may alter their sensitivity or response to inflammatory stimuli
through differential expression of receptors or modified intracel-
lular signaling (Panja et al., 1998; Vignoli et al., 2001) which
could result from a downregulation of the NF-jB and MAPKs cas-
cade (Böcker et al., 2000) and their downstream transcriptional
and post-transcriptional events, a decreased arachidonic acid cas-
cade activity (Martin-Venegas et al., 2006), as well as from a de-
crease of receptor expression occurring with cell differentiation
(Schling et al., 2006), as well as a decrease of receptor expression
occurring with cell differentiation. These results confirm the
general opinion that IECs become less sensitive to inflammatory
stimulation upon differentiation. However, we and others, Cavicchi
and Whittle (1999) and Vignoli et al. (2001), detected NO secre-
tion only in differentiated Caco-2 cells. These findings indicate
that mature adult as young proliferating cells could contribute
differentially to the inflammatory response in the gut and this
is an important aspect to take into account when studying intes-
tinal inflammation.
The study of inflammatory response as a function of cell differ-
entiation was particularly interesting regarding to stimuli of bac-
terial nature: IECs are constantly exposed to a variety of products
derived from the commensal microflora, but in normal physiolog-
ical conditions, these components do neither cross the epithe-
lium (Clark et al., 2005) nor trigger IEC inflammatory signals
(Rosenstiel et al., 2003). IECs seem to progressively acquire this
state of hyporesponsiveness during their maturation, where the
gradually increased contact with bacterial components would re-
sult in diminished expression of the coreceptors TLR-4, CD-14 and
MD2 (Abreu et al., 2001; Böcker et al., 2003; Suzuki et al., 2003).
Moreover, the cytokine upregulation observed during IBDs has
been found to lead both to defective IEC barrier functioning and
thus increased permeability to bacteria, as well as to a higher
expression level of receptors to bacterial components, which
hence modulates the general responsiveness of IECs to bacterial
components (Abreu et al., 2002; Suzuki et al., 2003). However,
important differences in the expression of LPS receptors seems
to exist between the in vitro models with variations due to the
1447
type of clone used, the culture conditions and the coordinated
expression of specific inflammatory stimuli (Cario et al., 2000;
Böcker et al., 2003; Abreu et al., 2002), making it difficult to dis-
criminate which cell line is the best representative of intestinal
sensitivity to LPS In the present study using Caco-2 cells, we
found an effect of LPS on none of the parameters tested, and this
was independent on both the cell differentiation stage and the
presence of additional stimuli. Although this might conflict with
the general ideas described above, it might be consistent with
the hypothesis that the different kind of cells present in the gut
also diverge in their individual sensitivity and response to bacte-
rial exposure (Suzuki et al., 2003), with absorptive IECs as mod-
eled by the Caco-2 line showing a total indifference to LPS.
Therefore, the choice of the cell line seems a crucial parameter
for research on bacterial interactions with the intestinal epithe-
lium, and should therefore be taken into account when modeling
intestinal inflammation in vitro.
It has become increasingly evident that IECs could be an inter-
esting target for IBD prevention and treatment (Böcker et al., 2003),
and in vitro models thus constitute a valuable tool for experimental
therapeutic research. Our experimental conditions, i.e. the use of a
cocktail of inflammatory stimuli at plausible concentrations on dif-
ferentiated Caco-2 cells, aimed at better reflecting the in vivo situ-
ation during ongoing IBDs. Our results show the induction of
additional features associated with IBDs, such as the increased re-
lease of the mediators IL-8, IL-6, NO and PGE-2, and a dysregulated
barrier functioning. Therefore, our conditions allow a more accu-
rate, consistent and integral approach of the behaviour of IECs dur-
ing active IBD, and encourage their use in in vitro preclinical
research. In this respect, an interesting perspective consists in
the screening of dietary substances that could be used for either
the prevention or the follow up of the IBDs, as performed in Romier
et al. (2008).
Future fine-tuning of these experimental conditions including
the in vitro modeling of (i) the sequential and time-dependent re-
lease and interaction of inflammatory mediators as it occurs in a
well-orchestrated, physiological inflammatory reaction; (ii) the
different cytokine profiles that are observed in the gut of Crohn’s
disease and ulcerative colitis patients; (iii) the anti-inflammatory
mechanisms that aim at counteracting the inflammatory reaction;
and (iv) the involvement of additional cellular partners such as
macrophages, dendritic cells or lymphocytes, will provide supple-
mentary applications for our model of the IBD pathology at the
intestinal epithelial level.
Acknowledgements
JVDW is a fellow from the Belgian FRIA. This work was sup-
ported by a grant from the Fondation Louvain. The authors are
grateful to Thomas Raas for technical assistance.
References
Abreu, M.T., Vora, P., Faure, E., Thomas, L.S., Arnold, E.T., Arditi, M., 2001. Decreased
expression of Toll-like receptor-4 and MD-2 correlates with intestinal epithelial
cell protection against dysregulated proinflammatory gene expression in
response to bacterial lipopolysaccharide. J. Immunol. 167, 1609–1616.
Abreu, M.T., Arnold, E.T., Thomas, L.S., Gonsky, R., Zhou, Y.H., Hu, B., Arditi, M., 2002.
TLR4 and MD-2 expression is regulated by immune-mediated signals in human
intestinal epithelial cells. J. Biol. Chem. 277, 20431–20437.
Aktan, F., 2004. iNOS-mediated nitric oxide production and its regulation. Life Sci.
75, 639–653.
Al-Sadi, R.M., Ma, T.Y., 2007. IL-1 beta causes an increase in intestinal epithelial
tight junction permeability. J. Immunol. 178, 4641–4649.
Aoki, K., 1978. A study of endotoxemia in ulcerative colitis and Crohn’s disease. I.
Clinical study. Acta Med. Okayama 32, 147–158.
Böcker, U., Damião, A., Holt, L., Han, D.S., Jobin, C., Panja, A., Mayer, L., Sartor, R.B.,
1998. Differential expression of interleukin 1 receptor antagonist isoforms in
human intestinal epithelial cells. Gastroenterology 115, 1426–1438.
1448 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449
Böcker, U., Schottelius, A., Watson, J.M., Holt, L., Licato, L.L., Brenner, D.A., Sartor,
R.B., Jobin, C., 2000. Cellular differentiation causes a selective down-regulation
of interleukin (IL)-1 beta-mediated NF-kappa B activation and IL-8 gene
expression in intestinal epithelial cells. J. Biol. Chem. 275, 12207–12213.
Böcker, U., Yezerskyy, O., Feick, P., Manigold, T.P., Panja, A., Kalina, U., Herweck, F.,
Rossol, S., Singer, M.V., 2003. Responsiveness of intestinal epithelial cell lines to
lipopolysaccharide is correlated with Toll-like receptor 4 but not Toll-like
receptor 2 or CD14 expression. Int. J. Colorectal. Dis. 18, 25–32.
Bruewer, M., Luegering, A., Kucharzik, T., Parkos, C.A., Madara, J.L., Hopkins, A.M.,
Nusrat, A., 2003. Proinflammatory cytokines disrupt epithelial barrier function
by apoptosis-independent mechanisms. J. Immunol. 171, 6164–6172.
Bruewer, M., Samarin, S., Nusrat, A., 2006. Inflammatory bowel disease and the
apical junctional complex. Ann. N Y Acad. Sci. 1072, 242–252.
Bruewer, M., Utech, M., Ivanov, A.I., Hopkins, A.M., Parkos, C.A., Nusrat, A., 2005.
Interferon-gamma induces internalization of epithelial tight junction proteins
via a macropinocytosis-like process. Faseb J. 19, 923–933.
Caradonna, L., Amati, L., Magrone, T., Pellegrino, M., Jirillo, E., Caccavo, D., 2000.
Enteric bacteria, lipopolysaccharides and related cytokines in inflammatory
bowel disease: biological and clinical significance. J. Endotoxin Res. 6, 205–214.
Cario, E., Rosenberg, I.M., Brandwein, S.L., Beck, P.L., Reinecker, H.C., Podolsky, D.K.,
2000. Lipopolysaccharide activates distinct signalling pathways in intestinal
epithelial cell lines expressing Toll-like receptors. J. Immunol. 164, 966–972.
Cavicchi, M., Whittle, B.J.R., 1999. Regulation of induction of nitric oxide synthase
and the inhibitory actions of dexamethasone in the human intestinal epithelial
cell line, Caco-2: influence of cell differentiation. Br. J. Pharmacol. 128, 705–715.
Chavez, A.M., Morin, M.J., Unno, M., Fink, M.P., Hodin, R.A., 1999. Acquired
interferon c responsiveness during Caco-2 cell differentiation: effects on iNOS
gene expression. Gut 44, 659–665.
Chell, S., Kadi, A., Williams, A.C., Paraskeva, C., 2006. Mediators of PGE(2) synthesis
and signalling downstream of COX-2 represent potential targets for the
prevention/treatment of colorectal cancer. Biochim. Biophys. Acta – Rev.
Cancer 1766, 104–119.
Clark, E., Hoare, C., Tanianis-Hughes, J., Carlson, G.L., Warhurst, G., 2005. Interferon
gamma induces translocation of commensal Escherichia coli across gut epithelial
cells via a lipid raft-mediated process. Gastroenterology 128, 1258–1267.
Danese, S., Angelucci, E., Malesci, A., Caprilli, R., 2008. Biological agents for
ulcerative colitis: hypes and hopes. Med. Res. Rev. 28, 201–218.
Duque, J., Díaz-Muñoz, M.D., Fresno, M., Iñiguez, M.A., 2006. Up-regulation of
cyclooxygenase-2 by interleukin-1b in colon carcinoma cells. Cell Signal. 18,
1262–1269.
Eckmann, L., Jung, H.C., Schurermaly, C., Panja, A., Morzyckawroblewska, E., Kagnoff,
M.F., 1993. Differential cytokine expression by human intestinal epithelial cell
lines – regulated expression of interleukin-8. Gastroenterology 105, 1689–
1697.
Fais, S., Capobianchi, M.R., Pallone, F., Di Marco, P., Boirivant, M., Dianzani, F.,
Torsoli, A., 1991. Spontaneous release of interferon gamma by intestinal lamina
propria lymphocytes in Crohn’s disease. Kinetics of in vitro response to
interferon y inducers. Gut 32, 403–407.
Forsythe, R.M., Xu, D.Z., Lu, Q., Deitch, E.A., 2002. Lipopolysaccharide-induced
enterocyte-derived nitric oxide induces intestinal monolayer permeability in an
autocrine fashion. Shock 17, 180–184.
Grishin, A., Wang, J., Hackam, D., Qureshi, R., Upperman, J., Zamora, R., Ford, H.R.,
2004. P38 MAP kinase mediates endotoxin-induced expression of
cyclooxygenase-2 in enterocytes. Surgery 136, 329–335.
Han, X.N., Fink, M.P., Delude, R.L., 2003. Proinflammatory cytokines cause NO center
dot-dependent and -independent changes in expression and localization of
tight junction proteins in intestinal epithelial cells. Shock 19, 229–237.
Hershko, D.D., Robb, B.W., Wray, C.J., Luo, G.J., Hasselgren, P.O., 2004.
Superinduction of IL-6 by cycloheximide is associated with mRNA
stabilization and sustained activation of p38 map kinase and NF-kappa B in
cultured Caco-2 cells. J. Cell. Biochem. 91, 951–961.
Hoffmann, E., Dittrich-Breiholz, O., Holtmann, H., Kracht, M., 2002. Multiple control
of interleukin-8 gene expression. J. Leukoc. Biol. 72, 847–855.
Jung, H.C., Eckmann, L., Yang, S.K., Panja, A., Fierer, J., Morzyckawroblewska, E.,
Kagnoff, M.F., 1995. A distinct array of proinflammatory cytokines is expressed
in human colon epithelial-cells in response to bacterial invasion. J. Clin. Invest.
95, 55–65.
Kolios, G., Wright, K.L., Jordan, N.J., Leithead, R.B., Robertson, D.A., Westwick, J.,
1999. C–X–C and C–C chemokine expression and secretion by the human
colonic epithelial cell line, HT-29: differential effect of T lymphocyte-derived
cytokines. Eur. J. Immunol. 29, 530–536.
Kolios, G., Valatas, V., Ward, S.G., 2004. Nitric oxide in inflammatory bowel disease:
a universal messenger in an unsolved puzzle. Immunology 113, 427–437.
Ligumsky, M., Simon, P.L., Karmeli, F., Rachmilewitz, D., 1990. Role of interleukin-1
in inflammatory bowel-disease – enhanced production during active disease.
Gut 31, 686–689.
Ma, T.Y., Iwamoto, G.K., Hoa, N.T., Akotia, V., Pedram, A., Boivin, M.A., Said, H.M.,
2004. TNF-alpha-induced increase in intestinal epithelial tight junction
permeability requires NF-kappa B activation. Am. J. Physiol. Gastrointest.
Liver Physiol. 286, 367–376.
Ma, T.Y., Boivin, M.A., Ye, D.M., Pedram, A., Said, H.M., 2005. Mechanism of TNF-
alpha modulation of Caco-2 intestinal epithelial tight junction barrier: role of
myosin light-chain kinase protein expression. Am. J. Physiol. Gastrointest. Liver
Physiol. 288, 422–430.
Martin, G.R., Wallace, J.L., 2006. Gastrointestinal inflammation: a central
component of mucosal defense and repair. Exp. Biol. Med. 231, 130–137.
Martin-Venegas, R., Roig-Perez, S., Ferrer, R., Moreno1, J.J., 2006. Arachidonic acid
cascade and epithelial barrier function during Caco-2 cell differentiation. J. Lipid
Res. 47, 1416–1423.
McCormack, G., Moriarty, D., O’Donoghue, D.P., McCormick, P.A., Sheahan, K., Baird,
A.W., 2001. Tissue cytokine and chemokine expression in inflammatory bowel
disease. Inflamm. Res. 50, 491–495.
Mudter, J., Neurath, M.F., 2007. IL-6 signaling in inflammatory bowel disease:
pathophysiological role and clinical relevance. Inflamm. Bowel Dis. 13, 1016–
1023.
Neuman, M.G., 2007. Immune dysfunction in inflammatory bowel disease. Transl.
Res. 149, 173–186.
Ogle, C.K., Guo, X.L., Hasselgren, P.O., Ogle, J.D., Alexander, J.W., 1997. The gut as a
source of inflammatory cytokines after stimulation with endotoxin. Eur. J. Surg.
163, 45–51.
Ou, G., Baranov, V., Lundmark, E., Hammarström, S., Hammarström, M.L., 2009.
Contribution of intestinal epithelial cells to innate immunity of the human gut –
studies on polarized monolayers of colon carcinoma cells. Scand. J. Immunol.
69, 150–161.
Panja, A., Goldberg, S., Eckmann, L., Krishen, P., Mayer, L., 1998. The regulation and
functional consequence of proinflammatory cytokine binding on human
intestinal epithelial cells. J. Immunol. 161, 3675–3684.
Parikh, A.A., Salzman, A.L., Fischer, J.E., Szabo, C., Hasselgren, P.O., 1997. Interleukin-
1 beta and interferon-gamma regulate interleukin-6 production in cultured
human intestinal epithelial cells. Shock 8, 249–255.
Pinto, M., Robineleon, S., Appay, M.D., Kedinger, M., Triadou, N., Dussaulx, E.,
Lacroix, B., Simonassmann, P., Haffen, K., Fogh, J., Zweibaum, A., 1983.
Enterocyte-like differentiation and polarization of the human-colon
carcinoma cell-line caco-2 in culture. Biol. Cell. 47, 323–330.
Raub, Y., Sandberg, C., Hallgren, R., 1995. Mucosal synthesis and release of PGE2
from activated eosinophils and macrophages in ulcerative colitis. Am. J.
Gastroenterol. 90, 614–620.
Reimund, J.M., Wittersheim, C., Dumont, S., Muller, C.D., Kenney, J.S., Baumann,
R., Poindron, P., Duclos, B., 1996. Increased production of tumour necrosis
factor-alpha, interleukin-1 beta, and interleukin-6 by morphologically
normal intestinal biopsies from patients with Crohn’s disease. Gut 39,
684–689.
Reinecker, H.C., Steffen, M., Witthoeft, T., Pflueger, I., Schreiber, S., Macdermott, R.P.,
Raedler, A., 1993. Enhanced secretion of tumor-necrosis-factor-alpha, IL-6, and
IL-1 beta by isolated lamina propria mononuclear-cells from patients with
ulcerative-colitis and Crohns-disease. Clin. Exp. Immunol. 94, 174–181.
Romier, B., Van De Walle, J., During, A., Larondelle, Y., Schneider, Y.J., 2008.
Modulation of signaling NF-jB activation pathway by polyphenols in human
intestinal Caco-2 cells. Br. J. Nutr. 100, 542–551.
Romier-Crouzet, B., Van De Walle, J., During, A., Joly, A., Rousseau, C., Henry, O.,
Larondelle, Y., Schneider, Y.J., 2009. Inhibition of inflammatory mediators by
polyphenolic plant extracts in human intestinal Caco-2 cells. Food Chem.
Toxicol. 47, 1221–1230.
Rosenstiel, P., Fantini, M., Braeutigam, K., Waetzig, G., Kuehbacher, T., Seegert, D.,
Schreiber, S., 2003. TNF-alpha and IFN-gamma regulate the expression of the
NOD2 (CARD15) gene in human intestinal epithelial cells. Gastroenterology
124, 1001–1009.
Russel, M.G., 2000. Changes in the incidence of inflammatory bowel disease: what
does it mean? Eur. J. Intern. Med. 11, 191–196.
Saklatvala, J., Dean, J., Clark, A., 2003. Control of the expression of inflammatory
response genes. Biochem. Soc. Symp. 70, 95–106.
Schling, P., Rudolph, C., Heimerl, S., Fruth, S., Schmitz, G., 2006. Expression of tumor
necrosis factor alpha and its receptors during cellular differentiation. Cytokine
33, 239–245.
Schlottmann, K., Wachs, F.P., Grossmann, J., Vogl, D., Maendel, M., Falk, W.,
Scholmerich, J., Andus, T., Rogler, G., 2004. Interferon gamma downregulates IL-
8 production in primary human colonic epithelial cells without induction of
apoptosis. Int. J. Colorectal. Dis. 19, 421–429.
Schuerer-Maly, C.C., Eckmann, L., Kagnoff, M.F., Falco, M.T., Maly, F.E., 1994. Colonic
epithelial cell lines as a source of interleukin-8: stimulation by inflammatory
cytokines and bacterial lipopolysaccharide. Immunology 81, 85–91.
Singer, I.I., Kawka, D.W., Schloemann, S., Tessner, T., Riehl, T., Stenson, W.F., 1998.
Cyclooxygenase 2 is induced in colonic epithelial cells in inflammatory bowel
disease. Gastroenterology 115, 297–306.
Subbaramaiah, K., Yoshimatsu, K., Scherl, E., Das, K.M., Glazier, K.D., Golijanin, D.,
Soslow, R.A., Tanabe, T., Naraba, H., Dannenberg, A.J., 2004. Microsomal
prostaglandin E synthase-1 is overexpressed in inflammatory bowel disease –
evidence for involvement of the transcription factor Egr-1. J. Biol. Chem. 279,
12647–12658.
Suzuki, M., Hisamatsu, T., Podolsky, D.K., 2003. Gamma interferon augments the
intracellular pathway for lipopolysaccharide (LPS) recognition in human
intestinal epithelial cells through coordinated up-regulation of LPS uptake
and expression of the intracellular Toll-like receptor 4-MD-2 complex. Infect.
Immun. 71, 3503–3511.
Tanabe, T., Tohnai, N., 2002. Cyclooxygenase isozymes and their gene structures and
expression. Prostaglandins Other Lipid Mediat. 68–69, 95–114.
Van De Walle, J., Romier, B., Larondelle, Y., Schneider, Y.J., 2008. Influence of
deoxynivalenol on NF-[kappa]B activation and IL-8 secretion in human
intestinal Caco-2 cells. Toxicol. Lett. 177, 205–214.
Vignoli, A.L., Srivastava, R.C., Stammati, A., Turco, L., Tanori, M., Zucco, F., 2001.
Nitric oxide production in Caco-2 cells exposed to different inducers, inhibitors
and natural toxins. Toxicol. In Vitro 15, 289–295.
 J. Van De Walle et al. / Toxicology in Vitro 24 (2010) 1441–1449 1449

Vitkus, S.J.D., Hanifin, S.A., McGee, D.W., 1998. Factors affecting Caco-2 intestinal
epithelial cell interleukin-6 secretion. In Vitro Cell Dev. Biol. Anim. 34, 660–
664.
Wang, F.J., Schwarz, B.T., Graham, W.V., Wang, Y.M., Su, L.P., Clayburgh, D.R.,
Abraham, C., Turner, J.R., 2006. IFN-gamma-induced TNFR2 expression is
required for TNF-dependent intestinal epithelial barrier dysfunction.
Gastroenterology 131, 1153–1163.
Wright, K.L., Weaver, S.A., Patel, K., Coopman, K., Feeney, M., Kolios, G., Robertson,
D.A., Ward, S.G., 2004. Differential regulation of prostaglandin E biosynthesis by
interferon-gamma in colonic epithelial cells. Br. J. Pharmacol. 141, 1091–1097.