Download The Importance of Intestinal Eotaxin

Dig Dis Sci
DOI 10.1007/s10620-016-4047-z
ORIGINAL ARTICLE
The Importance of Intestinal Eotaxin-1 in Inflammatory Bowel
Disease: New Insights and Possible Therapeutic Implications
Tomer Adar1 • Shimon Shteingart1 • Ami Ben-Ya’acov1 • Ariella Bar-Gill Shitrit1 •
Dan M. Livovsky1 • Shimrit Shmorak1 • Mahmud Mahamid1 • Bernardo Melamud1
Fiona Vernea2 • Eran Goldin1
•
Received: 13 August 2015 / Accepted: 18 January 2016
! Springer Science+Business Media New York 2016
Abstract
Background Involvement of eotaxin-1 in inflammatory
bowel disease has been previously suggested and increased
levels of eotaxin-1 have been described in both ulcerative
colitis and in Crohn’s disease. The association between
serum levels of eotaxin-1 and that within the colonic
mucosa has not been well defined, as is the potential
therapeutic value of targeting eotaxin-1.
Aims To characterize serum and intestinal wall eotaxin-1
levels in various inflammatory bowel disease patients and
to explore the effect of targeting eotaxin-1 by specific
antibodies in dextran sodium sulfate-induced colitis model.
Methods Eotaxin-1 levels were measured in colonic
biopsies and in the sera of 60 ulcerative colitis patients,
Crohn’s disease patients and healthy controls. We also
followed in experimental colitis the effect of targeting
eotaxin-1 by a monoclonal antibody.
Results Colon eotaxin-1 levels were significantly
increased in active but not in quiescent ulcerative colitis and
Crohn’s disease patients compared to healthy controls.
Levels of eotaxin-1 in the colon were correlated with eosinophilia only in tissues from active Crohn’s disease patients.
Our results did not show any statistically significant change
in serum eotaxin-1 levels among ulcerative colitis, Crohn’s
Tomer Adar and Shimon Shteingart have contributed equally to this
work.
& Tomer Adar
[email protected]
1
Digestive Diseases Institute, Shaare Zedek Medical Center,
Affiliated with the Hebrew University School of Medicine,
12 Bayit St., 91031 Jerusalem, Israel
2
Department of Pathology, Shaare Zedek Medical Center,
Jerusalem, Israel
disease and healthy controls. Moreover, we demonstrate that
in dextran sodium sulfate-induced colitis, targeting of
eotaxin-1 with 2 injections of anti eotaxin-1 monoclonal
antibody ameliorates disease activity along with decreasing
colon weight and improving histologic inflammation.
Conclusion Eotaxin-1 is increasingly recognized as a
major mediator of intestinal inflammation. Our preliminary
human and animal results further emphasize the value of
targeting eotaxin-1 in inflammatory bowel disease.
Keywords Eotaxin-1 ! Inflammatory bowel disease !
Experimental colitis ! Immune system
Introduction
The etiologies of Crohn’s disease (CD) and ulcerative
colitis (UC), the two major forms of IBD, are still not
completely elucidated. The current paradigm for the
pathogenesis of IBD involves an unregulated immune
response to resident intestinal bacteria that mediates
intestinal tissue damage in genetically susceptible hosts, all
influenced by lifestyle and environmental factors [1–3].
The disease courses of CD and UC are characterized by
exacerbations and remissions; however, several clinical
and pathological features differ between these two diseases, suggesting that they represent independent clinical
entities. The identification of inflammatory components in
IBD led to the development of more specific anti-inflammatory treatments. Studies have shown that cytokines have
a crucial role in controlling intestinal inflammation and are
associated with clinical symptoms of IBD [4]. Modulation
of cytokine function can be used for therapy, and newly
identified cytokines are becoming potential therapeutic
targets for IBD [5]. Anti-tumor necrosis factor (anti-TNF)
123
Dig Dis Sci
drugs were the first biological therapies developed for the
management of IBD; however, additional cytokines such
IL-1b, IL-27, IL-33, and IL-35, all important mediators in
IBD, are attractive therapeutic targets [6]. Studies in
experimental mouse models of IBD have indicated that the
neutralization of pro-inflammatory cytokines could be used
for prevention and therapy of IBD [5, 7, 8]. Interleukin-1b
(IL-1b), for example, has been known for more than
20 years to be highly expressed in mononuclear cells from
colonic mucosal biopsies of IBD patients [9]. Cominelli
et al. [10] have demonstrated the beneficial effect of
recombinant IL-1 receptor antagonist (IL-1ra) in a rabbit
colitis model. Similar efficacy was shown in the rat acidinduced colitis model [11]. In humans, three approved
therapies are now clinically available for blocking the IL1
pathway [6]. Current medical treatment for IBD, including
immunosuppressive and immunomodulatory drugs such
steroids, thiopurines and ‘‘biologics,’’ are all aimed at
reducing non-specific inflammation. These treatment
options are sometimes limited in terms of efficacy and side
effects, with a need for novel therapeutic approaches [12].
Eotaxins are small secreted peptides that serve as potent
chemoattractants for eosinophils and other cell types [13].
Eotaxin-1 was first described in allergic airway inflammation and was further established as a key mediator in
regulating tissue eosinophilia [14–16]. Eotaxin-1 is constitutively expressed in the lamina propria within the gastrointestinal (GI) tract [17, 18] and is the major eosinophil
chemoattractant in this organ. Additional studies have
established the relationship between an increased number
of eosinophils and their activation in IBD [19]. Increased
mRNA levels of eotaxin-1 were found in lesions of IBD
patients [16]. Indeed, many studies in animal models and in
humans have demonstrated the link between elevated
levels of eotaxin-1 and IBD [20–24]. For example, in two
animal models of UC, DSS colitis and acute oxazolone
colitis, increased eotaxin-1 levels were found in colonic
tissues from injured mice [24, 25]. In humans, increased
serum eotaxin-1 levels were found in CD and in UC
patients [22]. However, the correlation between serum and
tissue levels of eotaxin-1 in active and non-active IBD
patients is not yet established. Targeting eotaxin-1 has been
previously described in animal models and in clinical trials
of allergic conditions [26–29], but not in IBD.
Here, we report the results of an observational human
study in which eotaxin-1 levels were measured in the
serum and in the colonic tissue of IBD patients with active
and non-active UC and CD. Our results show an increase in
eotaxin-1 levels in the tissue but not in the serum of active
CD and UC patients. We further determined the therapeutic
potential of an anti-mouse eotaxin-1 monoclonal antibody
in experimental colitis. We show an amelioration of
123
dextran sulfate sodium (DSS)-induced colitis in mice
treated with anti-eotaxin-1, reflected by significant reduced
disease activity, colon weight, and marked improvement in
colon histology.
Aim
The aim was to characterize serum and intestinal wall
eotaxin-1 levels in active and non-active UC and CD
patients. Additionally, we explored the effect of targeting
eotaxin-1 by specific antibodies in DSS- induced colitis
model.
Methods
Human Subjects
Our study included healthy controls and patients with a
confirmed diagnosis of IBD that were referred for colonoscopy at the Digestive Diseases Institute in Shaare Zedek
medical center between February 2012 and June 2013. The
patient cohort consisted of 60 subjects: 10 healthy controls,
15 patients with active UC, 10 patients with UC in
remission, 16 patients with active CD and 9 patients with
CD in remission. The study included 36 males and 24
females between 24 and 77 years of age. Subjects undergoing colon cancer screening colonoscopy were recruited
for the control group. Confirmed diagnosis of UC and CD
patients was made at least 3 months prior to study enrollment according to acceptable clinical, laboratory, endoscopic, and histological criteria. Group allocation
according to disease activity was made by the attending
gastroenterologist prior to endoscopy. For UC patients, a
complete MAYO score was also calculated. Biopsies from
active UC and CD patients were taken from all segments of
the colon and from areas that were most likely to show
inflammation. Colonic biopsies from healthy controls were
obtained randomly from subjects undergoing colonoscopy
for colon cancer screening, which is recommended for
subjects who are 50 years old. All biopsies were examined
in the department of pathology at Shaare Zedek medical
center by a pathologist blinded to tissue or serum eotaxin-1
levels. Blood samples for a complete blood count and the
measurement of CRP were taken before endoscopy. The
study was performed with the approval of the Institutional
Review Board of Shaare Zedek Medical Center. Written
informed consent was obtained from all subjects for analyses of demographics, medical history and for serum and
tissue biopsies obtained at the time of endoscopic procedures as a part of the clinical trial.
Dig Dis Sci
Evaluation of Human Tissue and Serum Cytokines
Two colon biopsies from every subject were obtained at the
time of colonoscopy. Tissue for cytokine determination
was snap-frozen immediately and stored at -80 "C. After
thawing, samples were lysed in cold lysis buffer (50 mM
Tris–HCl with 2 mM EDTA, pH 7.4) with a Potter–Elvehjem homogenizer and then centrifuged for 2 min. Pellets
were discarded and supernatants were kept at -80 "C and
sent to Myriad-RBM laboratories (Austin, TX, USA) for
immunoassay. The samples were analyzed using the
Human MAP-B v1.0 panel comprised of immunoassays for
14 analytes. This panel is based upon multiplex
immunoassay technology to measure a range of inflammatory cytokines. All measurements were conducted under
blind conditions. Protein concentration was measured in
the tissue lysates using the bicinchoninic acid (BCA)
method. Human serum samples were obtained, aliquoted,
and stored at -80 "C until analysis. Human serum eotaxin1 levels were determined using a commercial ELISA kit
(R&D systems, Minneapolis, MN, USA) according to the
manufacturer’s protocol.
Eosinophil Quantification
Colon biopsies for all the subjects were fixed with 4 %
paraformaldehyde/PBS, processed using standard histological techniques and stained with hematoxylin and eosin
(H&E). The slides were then examined in a blinded manner
by a pathologist.
Samples were scored as the number of eosinophils in the
colonic mucosa per high-power field (HPF) using an
Olympus BX40 microscope with 409 objective. A total of
10 HPFs were assessed for each biopsy. Scores were
averaged for each subject.
Real-Time PCR Analysis
Fresh biopsies were immediately placed in tubes containing RNA save solution (Biological Industries, Beit Ha’emek, Israel), then moved to 4 "C over night, and finally
stored at -80 "C until RNA extraction. Total RNA was
extracted from colonic biopsies using TRI reagent (Invitrogen, CA, USA). RNA integrity during the study was kept
and controlled by spectrophotometry and by running RNA
on agarose gel stained with ethidium bromide. The
expression of mRNA for eotaxin-1 in colonic biopsies was
assessed by real-time PCR. Briefly, 1 lg of RNA was
reverse-transcribed using a high-capacity cDNA reverse
transcription kit (Applied Biosystems, CA, USA). Each
PCR reaction was performed with 5 ll of cDNA and
DyNAmo Flash SYBR Green qPCR mix (Thermo Scien-
tific, PA, USA). The sequences for human eotaxin-1 primers were as follows: (F): AGATACCCCTTCAGC
GACTAG. (R): GGTCTTGAAGATCACAGCTTTC.
Real-time PCR was performed using a StepOne Plus system (Applied Biosystems, CA, USA). The data were normalized to GAPDH gene (F): CCACTCCTCCACCTT
TGAC. (R): ACC CTG TTG CTG TAG CCA.
Animals
Male, Balb/C mice (8–9 week old) were purchased from
Harlan (Jerusalem, Israel). Mice (5 per group) were housed
at the Sha’are Zedek animal facility and maintained in a
temperature-controlled room with alternating 12-h light/dark cycles. Mice had access to water and rodent laboratory chow ad libitum. All experiments were approved by
the Institutional Animal Care and Use Committee of the
Hebrew University.
Induction of DSS Colitis
Acute colitis was induced by feeding mice with 4 % (wt/
vol) dextran sodium sulfate (DSS, MW 36–50 kD, MP
Biomedicals, Ohio, USA) dissolved in drinking water for
7 days. Treated mice (n = 5) were i.p. injected with
100 lg per mouse of anti-mouse eotaxin-1 monoclonal
antibody (clone 42285, R&D Systems, MN, USA) 1 day
prior to DSS administration (on day 0) and on day 4.
Control mice (n = 5) were also i.p. injected on day 0 and
day 4 with an isotype control antibody (R&D systems Inc,
MN, USA). All mice were killed on day 8.
Evaluation of DSS Colitis
Animals were monitored daily for weight, morbidity,
diarrhea, and presence of blood at the anus. Disease
activity index (DAI) was calculated by assigning a wellestablished and validated score [30]. Colonic inflammation
was evaluated in a blind manner by estimating weight loss,
diarrhea, and bleeding. These were scored on a scale
ranging from 0 to 4 as follows: (1) weight loss (0
point = none, 1 point = 5–10 % weight loss, 2
points = 10–15 % weight loss, 3 points = 15–20 %
weight loss and 4 points- more than 20 % weight loss), (2)
diarrhea (0 points = none, 1 points = mild, 2 points = gross diarrhea), (3) bleeding (0 points = no bleeding, 2,
slight bleeding, 4 points, gross bleeding). Together, the
scores on the individual parameters could result in a total
score ranging from 0 to 10.
Colons were stained with H&E, examined, and histologically scored by light microscopy.
123
Dig Dis Sci
Increased Eotaxin-1 Levels in Colonic Tissue
from UC and CD Patients
Statistical Analysis
Quantitative data are expressed as the mean ± SEM. Statistical analysis for significant differences was performed
according to Student’s t test for unpaired data, where
p values \0.05 were considered statistically significant.
Pearson correlation analysis was used to examine the
relationship between the serum and tissue levels of
eotaxin-1.
Results
Study Population
Patient characteristics are presented in Table 1. Based on
histological evaluation of the 50 IBD patients, nine CD
patients were in remission and 16 had active disease; 10
UC patients were in remission, and 15 had active disease.
IBD patients were significantly younger than control subjects (compatible with the age group undergoing screening
colonoscopy for colon cancer). The body mass index
(BMI) of active CD patients was significantly lower
(p = 0.02) compared to healthy controls. Medical treatment is presented in Table 1. Most IBD patients in our
study received medical treatment. 80 % of active UC
patients received medications for IBD, and most of the
patients were treated with 5-aminosalicylate (5-ASA).
40 % of UC patients in remission were treated, all with
5-ASA. 62.5 % and 55.5 % of active CD and CD patients
in remission, respectively, received various treatments (5ASA, steroids, 6-mercaptopurine and biologics) (Table 1).
Table 2 shows several laboratory markers that were measured: white blood cell count (WBC), C-reactive protein
(CRP), hemoglobin (Hb), fecal calprotectin, and absolute
number of eosinophils (Eos). WBCs and calprotectin values were significantly higher in active UC patients compared to healthy controls.
Table 1 Patients’
characteristics
Several studies have demonstrated increased eotaxin-1
levels in colonic samples of IBD patients [31, 32]. In our
study, we also determined the levels of eotaxin-1 in colonic
biopsies of active and in remission patients with UC and
CD. Figure 1 shows more than a threefold increase in the
tissue level of eotaxin-1 taken from colon biopsies of active
UC and CD patients compared with controls (p = 0.0001
for controls vs. UC active and p = 0.0006 for controls vs.
UC in remission). The difference in colonic eotaxin-1
levels between UC patients with active disease and those in
remission was not statistically significant, but the changes
between active and non-active CD patients reached statistical significance (p = 0.02).
Increased mRNA Expression of Eotaxin-1 in Colonic
Tissue of Active UC and CD Patients
As the levels of eotaxin-1 were increased in the colonic
tissues of active UC and CD patients, we next assessed
whether the mRNA expression of eotaxin-1 was also elevated in these tissues. Figure 2 shows that the mRNA
expression of eotaxin-1 in tissues taken from active UC and
CD patients was indeed significantly increased compared
with healthy controls. Importantly, the difference in fold
induction of mRNA expression between UC or CD patients
in remission and healthy controls was not statistically
significant. A statistically significant (p = 0.002) difference was noted in active versus non-active UC patients but
not between active and non-active CD patients (Fig. 2).
Serum Levels of Eotaxin-1 Are Not Altered in UC
and CD Patients
Next we measured the serum levels of eotaxin-1 in UC and
CD patients. Interestingly, none of the IBD groups exhibited higher serum eotaxin-1 levels than the healthy controls
Control
UC-active
UC-rem
CD-active
CD-rem
N
10
15
10
16
9
Sex (M/F)
8/2
9/6
6/4
7/9
6/3
Age
60.1 ± 3.6
39.9 ± 3.7*
45.2 ± 4.9*
35.5 ± 3.0*
34.8 ± 2.9*
BMI
30.6 ± 3.5
24.9 ± 1.0
22 ± 1.1
19.6 ± 2.3*
22.3 ± 1.1
Treated
0/10
12/15
4/10
10/16
5/9
Treatments included: 5-amino salicylic acid, steroids, 6-mercaptopurine and anti-TNFs. Values are given as
mean ± standard error
rem remission, M male, F female, BMI basal metabolic index
* p \ 0.05 compared with control. Data was analyzed using t test
123
Dig Dis Sci
Table 2 Laboratory markers
that were measured for all
subjects
Control
UC-active
UC-rem
CD-rem
WBC
6.1 ± 0.41
5.7 ± 0.35
6.2 ± 0.4
6.5 ± 0.53
CRP
0.40 ± 0.05
1.11 ± 0.27
0.44 ± 0.06$
1.18 ± 0.41
0.83 ± 0.19
Hb
14.0 ± 0.41
13.2 ± 0.43
14.1 ± 0.36
12.4 ± 0.47*
13.6 ± 0.25$
Eos (Abs.)
Calprotectin
7.2 ± 0.45*
CD-active
0.14 ± 0.04
0.20 ± 0.04
0.14 ± 0.03
0.14 ± 0.03
0.11 ± 0.03
159.3 ± 56.7
947.4 ± 285.1*
165.4 ± 49.6$
724.4 ± 304.3
210.5 ± 47.1
Values are given as mean ± standard error
WBC white blood count, CRP C-reactive protein; Hb hemoglobin, Eos (Abs.), absolute count of eosinophils
* p \ 0.01 compared with control,
t test
$
p \ 0.05 compared with active UC or CD. Data was analyzed using
Colonic Eotaxin-1 m-RNA expression
(fold induction)
300
NS
250
200
**
**
150
100
50
0
Control
UC-Active UC-Remission CD-Active
*
20.0
(Fig. 3). We then evaluated the correlation between the
serum and the tissue levels of eotaxin-1 in the various
groups. A significant correlation (r = 0.75; p = 0.01) was
found between tissue and serum eotaxin-1 levels only in
the healthy controls (Fig. 4). Neither in UC nor in CD
patients was the level of eotaxin-1 in colonic tissues correlated with the serum levels.
Colonic Eotaxin-1 Levels Correlate with Tissue
Eosinophilia in Active UC Patients
Eosinophil infiltration into the GI is associated with
marked gastrointestinal pathology and augmented eotaxin1 levels [33]. We examined the correlation between both
serum and colonic levels of eotaxin-1 with tissue
eosinophilia. Figure 5a shows that eosinophilia was significantly correlated (r = 0.63; p = 0.01) with colonic
*
15.0
10.0
5.0
0.0
CD-Remission
Fig. 1 Colon tissue eotaxin-1 is significantly increased in active UC
and active CD. Snap-frozen colonic biopsies were lysed, and eotaxin1 levels were measured by using the multiplex technology (human
MAP-B, Myriad-RBM) with each sample corrected for tissue lysate
protein concentration, as described in the Methods. Each column
represents mean ± SE of the mean of the different groups. *p \ 0.05
compared with control; ***p \ 0.001 compared with CD-active
group. NS nonsignificant
*
25.0
control
UC-active UC-remission CD-active CD-remission
Fig. 2 Eotaxin-1 mRNA is highly expressed in the colon of active
ulcerative colitis and active Crohn’s disease patients. Total RNA was
extracted from frozen colonic biopsies and mRNA expression for
eotaxin-1 was assessed by real-time PCR, as described in the
Methods. Mean densitometry for real-time PCR production is
expressed as fold induction. Each column represents mean ± SE of
the mean of the different groups. *p \ 0.01 compared with control
180
Serum Eotaxin-1 (pg/ml)
Tissue Eotaxin-1 (pg/mg protein)
*
160
140
120
100
80
60
40
20
0
Control
UC-Active
UC-Remission CD-Active CD-Remission
Fig. 3 Serum eotaxin-1 levels are not increased among ulcerative
colitis and Crohn’s disease patients. Serum eotaxin-1 levels were
determined by ELISA. Data are presented as mean ± SE
eotaxin-1 levels only in tissues taken from active UC
patients. A high count of eosinophils was not correlated
with colonic eotaxin-1 levels in active CD patients (data
123
Dig Dis Sci
Serum Eotaxin-1 (pg/ml)
350.0
Assessment of Cytokines in Colonic Tissues of UC
and CD Patients
300.0
250.0
r=0.75
200.0
p=0.01
150.0
100.0
50.0
0.0
0
50
100
150
Tissue Eotaxin-1 (pg/mg protein)
(a)
120
Eosinophils count (HPF)
Fig. 4 A significant correlation between tissue and serum eotaxin-1
levels in healthy controls. Correlation of tissue eotaxin-1 with serum
eotaxin-1 levels in healthy subjects. Correlation coefficient and
p value are: r = 0.75; p = 0.01
100
80
60
r=0.63
40
p=0.01
20
0
0
50
100
150
200
250
300
350
Tissue eotaxin-1 (pg/mg protein)
Eosinophils count (HPF)
(b) 60
Cytokines play a key role in IBD [4]. In the current study,
we used the human Myriad-RBM multiplex panel to
measure the level of different cytokines in colon biopsies
taken from UC and CD patients and healthy subjects.
Table 3 shows the profile of various analytes that were
measured. Eotaxin-1 levels were significantly (p \ 0.01)
higher in active UC and CD patients. The levels of this
cytokine were not altered in non-active UC and CD
patients. Non-active CD patients exhibited significantly
lower levels of eotaxin-1 compared to active patients. The
levels of IL-1b were also significantly (p \ 0.01) elevated
in active CD and UC patients. Patients in remission presented significantly lower levels of IL-1b. IL-12p40 levels
were elevated in active UC and CD patients; however,
compared to healthy subjects, these elevations did not
reach statistical significance. IL-12p70 levels were below
detection. The levels of IL-15 and IL-17 were significantly
(p \ 0.05 and p \ 0.01, respectively) higher only in tissue
biopsies of active UC patients compared to controls, but
not in active CD patients. Interestingly, IL-23 levels were
significantly (p \ 0.05) lower in active CD patients.
ICAM-1 levels were threefold higher in active UC patients
and were also significantly higher in active CD patients and
UC patients in remission. The levels of colonic MMP-3 and
VEGF were significantly (p \ 0.01 and p \ 0.05, respectively) increased only in active UC patients, but not in
active CD patients. Taken together, most of the cytokines
measured by the Myriad-RBM system presented noticeable
increases in active UC and CD patients.
Neutralizing Eotaxin-1 Ameliorated DSS-Induced
Experimental Colitis
50
40
30
20
10
0
Control
UC-Active UC-Remission CD-Active CD-Remission
Fig. 5 A significant correlation between tissue eotaxin-1 and tissue
eosinophils in active UC patients. a Correlation of tissue eotaxin-1
levels with eosinophil infiltration in active UC patients. Number of
eosinophils is per high-power field counts. Correlation coefficient and
p value are: r = 0.63; p = 0.01. b Colonic eosinophils are not
increased among ulcerative colitis and Crohn’s disease patients.
Number of eosinophils per high-power field (HPF). Ten HPFs were
assessed for each biopsy. Data are presented as mean ± SE
not shown). We also did not find any significant difference
in tissue eosinophilia between the various IBD patients and
healthy controls (Fig. 5b).
123
DSS treatment of mice promotes an experimental UC,
which possesses pathophysiological features similar to UC
in humans [34]. To evaluate the effect of targeting eotaxin1 in DSS-induced colitis, we used anti-mouse eotaxin-1
monoclonal antibody (anti-eotaxin-1). The administration
of anti-eotaxin-1 antibody or an isotype-matched control
IgG was started 1 day before DSS exposure (day 0), and a
second injection was administered on day 4. Weight loss in
anti-eotaxin-1 treated mice was significantly smaller than
in IgG-control mice. As shown in Fig. 6a, disease activity
index (DAI) was significantly (p = 0.04) reduced in antieotaxin-1-treated mice compared to IgG-control mice. To
further assess the effect of neutralizing eotaxin-1 on the
severity of colitis, colon length was measured in both
groups. Figure 6b shows that the colonic weight/length
ratio in anti-eotaxin-1-treated mice was 33 % smaller than
that of control mice. Accordingly, marked improvement in
Dig Dis Sci
Table 3 Colonic tissue cytokine profile
Control
UC-active
UC-rem
CD-active
CD-rem
BDNF
48.6 ± 30.9
56.6 ± 11.7
14.6 ± 5.0##
68.3 ± 15.1
42.2 ± 28
Eotaxin-1
34 ± 11.9
159.8 ± 23.2**
106.1 ± 33.0
219.6 ± 43.0**
97.8 ± 29.9&
ICAM-1
1953.8 ± 205.6
6048.4 ± 766.9**
3087.7 ± 389.6*,##
3293.1 ± 424.3**,##
3120.1 ± 808.4
12.5 ± 10.3
IL-1a
0.6 ± 0.1
42.3 ± 20.6
1.3 ± 0.4
5.0 ± 1.9*
IL-1b
7.5 ± 1.1
428.5 ± 123.9**
45.9 ± 18.4##
121.3 ± 38.4**,#
283.7 ± 227.6
IL-1ra
2912.3 ± 351.1
5734.8 ± 848.4**
2732.7 ± 659.2#
3605.6 ± 636
3312.2 ± 710.4
IL-12p40
39.7 ± 8.3
66.5 ± 10.9
27.8 ± 8.9#
33.8 ± 5.1#
62.8 ± 35.1
IL-12p70
BD
BD
BD
BD
BD
IL-15
81.5 ± 12.7
140.2 ± 17.8*
57.2 ± 10.5##
##
76.5 ± 14.1##
87.2 ± 30.0
##
IL-17
0.7 ± 0.1
10.9 ± 1.6**
2.9 ± 1.2
3.7 ± 1.8
IL-23
MMP-3
119.3 ± 14.6
702.7 ± 160.6
128.6 ± 14.3
3371.8 ± 621.2**
84.1 ± 13.8#
785 ± 387.8##
70.3 ± 10.2*,##
1402.6 ± 404.6#
74.9 ± 29.1
971.9 ± 525.4
3.3 ± 1.5
SCF
54.1 ± 6.2
40.5 ± 10.1
41.0 ± 7.0
33.6 ± 7.1*
30.7 ± 7.4*
VEGF
196.6 ± 16.6
305.3 ± 39.8*
206.2 ± 25.7
185.0 ± 19.8#
257.3 ± 57.5
Values are in pg/mg total protein. Snap-frozen colonic biopsies were lysed, and cytokine levels were measured by using the multiplex technology
(panel of 14 cytokines, human MAP-B, Myriad-RBM) with each sample corrected for tissue lysate protein concentration, as described in the
Methods. Values are given as mean ± standard error
BDNF brain-derived neutrophic factor, ICAM-1 intercellular adhesion molecule-1; IL interleukin, MMP-3 matrix metalloproteinase-3, SCF stem
cell factor, VEGF vascular endothelial growth factor
* p \ 0.05; ** p \ 0.01 compared with control;
CD-active
#
p \ 0.05;
##
p \ 0.01 compared with UC-active;
the clinical manifestations of DSS-induced colitis was
confirmed by H&E-stained sections of lamina propria of
anti-eotaxin-1-treated mice (Fig. 6c). In contrast, colons
from IgG-control mice presented with extensive crypt
destruction and edema (Fig. 6c). In conclusion, these
results indicate that neutralization of eotaxin-1 activity had
a beneficial effect in the colon of DSS-treated mice.
Taken together, administration of anti-eotaxin-1 to DSSmice resulted in clear improvement in colon inflammation
and disease course.
Discussion
Abnormal immune responses have been shown to play a
central role in the pathogenesis of IBD. Among the wide
range of these aberrant responses, increased levels of
eotaxin-1, a major chemotactic mediator for eosinophils,
have been observed in IBD patients [20, 22]. Nevertheless,
a clear distinction between serum and colonic levels in
active or quiescent stages of CD and UC was not present.
In our study, we demonstrate that elevated tissue levels of
eotaxin-1 may represent a hallmark of both UC and CD,
especially in the active stages of the disease.
A significant increase in eotaxin-1 taken from colonic
tissues of active UC and CD patients was further confirmed
by elevated mRNA expression of eotaxin-1 in these tissues.
These results are similar to previous reports showing an
&
p \ 0.05;
&&
p \ 0.01 compared with
increase in colonic tissues of eotxain-1 both at the protein
level, and especially at the mRNA level, particularly of UC
patients [23, 24, 35]. However, in contrast to other reports
showing an increase in eotaxin-1 in the serum of UC
patients compared to controls [22, 24], in our study, serum
levels of eotaxin-1 were merely similar among various
study groups and healthy controls.
It is important to note that age influences serum eotaxin1 levels in healthy people as a significant linear correlation
between serum eotaxin-1 levels and age was found [36].
Indeed, older control subjects who are usually being
screened for colon cancer were included in our study and
by others as well. In the study of Coburn et al. [24] the
mean age of the control group was markedly higher than
the mean age of quiescent UC and active UC patients.
However, this cohort was much larger than ours and
included 175 subjects. Therefore, we conclude that the
discrepancy is mainly attributed to lower statistical power
in our study.
Our study shows that tissue eotaxin-1 was significantly
correlated with eosinophilia only in active UC patients.
These results are in line with the study of Lampinen et al.
[35] who reported that in active and quiescent UC, patients
had an increased number of peripheral blood and rectal
eosinophils accompanied by increased levels of eotaxin-1
and IL-5 mRNA expression in colonic biopsies. Interestingly, Mir et al. [22] found elevated serum eotaxin-1 levels
in UC and CD patients and also noticed a negative
123
Dig Dis Sci
(a) 10
(b) 350
9
300
8
*
6
DAI
mg/10cm colon
7
5
4
3
*
200
150
100
2
50
1
0
250
control Ab
0
anti eotaxin-1
control Ab
anti eotaxin-1
(c)
An! eotaxin-1
Control Ab
50
50
An! eotaxin-1
Control Ab
Fig. 6 Anti-eotaxin-1 ameliorates DSS-induced colonic inflammation. Colitis was induced in mice by administration of dextran sodium
sulfate (DSS) in drinking water. i.p treatment with either 100 lg of
control Ab or of anti-eotaxin-1 was on day 0 and 4 and animals
(n = 5) were all killed 7 days after DSS administration. a Clinical
assessment of DSS-induced colitis. DAI was scored from each mouse
for weight loss, diarrhea and bleeding. b Colon weight. Data are
presented as mean ± standard error of 5 mice. c Photomicrograph of
H&E-stained paraffin section of distal colon (magnification, 9400)
from DSS colitis mice. Colitis in control Ab-treated mouse resulted in
marked bowel wall thickening and colonic inflammation, while
treatment with anti-eotaxin-1 resulted in milder inflammatory changes
correlation between serum eotaxin-1 serum level and
eosinophil counts in peripheral blood of CD patients.
Despite the established evidence that eosinophils are
functionally involved in the pathophysiology of IBD, the
exact mechanism of one of their mediators, eotaxin, in
blood and tissue may suggest different mechanism in CD
and UC. Defining the mechanisms that control recruitment
of eosinophils to tissues is essential to understanding these
disease processes. The expression of various receptors on
various leukocytes may play an important role in these
mechanisms. The CCR3–eotaxin-1 axis, for example, was
first described only in eosinophils. Now it is well accepted
that eotaxin-1 may affect various cell types via several
receptors [32]. The expression of these receptors is
123
Dig Dis Sci
influenced by previous cytokine exposure and also by
various medications taken by IBD patients, thus influencing the local inflammatory milieu.
The key role of cytokines was also highlighted by the
fact that blockade of TNF is now commonly used as a
standard therapy for IBD patients [37, 38]. In our study,
along with eotaxin-1, eight cytokines (ICAM-1, IL-1a, IL1b, IL-1ra, IL-15 and IL-7) were increased in colonic tissues from active UC patients. In active CD patients, three
cytokines (ICAM-1, IL-1a, IL-1b) were increased compared to healthy controls. For example, we have found that
IL-17 was significantly elevated in active UC patients.
These results are in line with the results of Coburn et al.
who have also shown an increased level of IL-17 in tissues
of active UC patients compared to controls. These results
are also in agreement with a recent gene expression profiling showing that IL-17A was 6000-fold and 2000-fold
higher in active UC and active CD patients, respectively
[39]. ICAM-1 is expressed in inflamed intestinal tissues.
Blocking ICAM-1 and VCAM-1 was suggested to have a
therapeutic benefit in CD [40]. In our study, we found a
significant elevation of ICAM-1 in colonic tissues from
both UC and CD patients. These results are in line with
those of Smyth et al. [41] who showed that ICAM-1
mRNA was increased sevenfold in UC patients. Also the
increases in IL-1a, IL-1b and IL-12p40 are well established in IBD patients [4]. In our study, we demonstrated
significant elevations in colonic levels of IL-a and IL-b
among UC- and CD-active patients. This is in agreement
with studies showing increased activation of the IL-1 system in areas of actively inflamed colonic mucosa among
IBD patients [9, 42, 43]. Members of the IL-12 family
(such IL-12, IL-23, IL-27 and IL-35) are produced by
APCs during intestinal inflammation and are known to play
a role in IBD [44, 45]. In our study, among active UC
patients, markedly elevated levels of IL-12p40 were
observed. We further noticed a decrease in IL-12p40 and in
IL-23 in UC patients in remission but not in CD patients in
remission. Studies in various mouse models of IBD have
shown that affecting cytokine function can be used for
therapy and have identified new cytokines as potential
therapeutic targets for chronic intestinal inflammation [5,
46]. Based on these studies, inflammatory cytokines
became targets in IBD therapy in human [4, 6]. Administration of recombinant anti-inflammatory cytokines or the
neutralization of pro-inflammatory cytokines in experimental mouse models of IBD could be used for both the
prevention and therapy of chronic intestinal inflammation
[7, 8]. Current therapies for IBD patients include various
immunosuppressive and immune regulatory drugs. Loss of
effect, serious side effects, and intolerance commonly
develops toward some of these drugs and represent a
persistent major unmet need. The similarities between
DSS-induced experimental UC and UC in humans suggest
that both diseases could be explained by common underlying pathological mechanisms. The critical role for
eotaxin-1 in DSS-induced colitis was already demonstrated
by Forbes and colleagues [47], and eotaxin-1 is strongly
suggested to play an important role in the development of
mucosal inflammation in IBD. In our study, we examined
whether targeting eotaxin-1 in a murine model for UC may
affect the disease course. Administration of monoclonal
antibody against eotaxin-1 in DSS-induced colitis resulted
in decreased disease activity index and improved colon
histology. However, further studies are warranted to better
establish the therapeutic potential of anti-eotaxin-1 in these
mice, thereby lowering eosinophil infiltration to the GI.
Taken together, our results support the importance of
eotaxin-1 in both CD and UC. However, it is possible that
this mediator has a different effect in these two clinical
conditions. In addition to the observational human study,
our interventional animal study in a DSS colitis model
marks eotaxin-1 as a potential therapeutic target.
Acknowledgments
Pharmaceuticals.
This study was partially supported by Immune
Compliance with ethical standards
Conflict of interest Eran Goldin is a consultant for Immune Pharmaceuticals. Immune Pharmaceuticals supported Tomer Adar for
travel to a clinical conference.
References
1. Kaser A, Zeissig S, Blumberg RS. Inflammatory bowel disease.
Annu Rev Immunol. 2010;28:573–621.
2. Mukhopadhya I, Hansen R, El-Omar EM, et al. IBD-what role do
Proteobacteria play? Nat Rev Gastroenterol Hepatol. 2012;9:
219–230.
3. Scharl M, Rogler G. Inflammatory bowel disease pathogenesis:
What is new? Curr Opin Gastroenterol. 2012;28:301–309.
4. Neurath MF. Cytokines in inflammatory bowel disease. Nat Rev
Immunol. 2014;14:329–342.
5. Powrie F, Leach MW, Mauze S, et al. Inhibition of Th1 responses
prevents inflammatory bowel disease in scid mice reconstituted
with CD45RBhi CD4? T cells. Immunity. 1994;1:553–562.
6. McLean MH, Neurath MF, Durum SK. Targeting interleukins for
the treatment of inflammatory bowel disease-what lies beyond
anti-TNF therapy? Inflamm Bowel Dis. 2014;20:389–397.
7. Neurath MF, Fuss I, Kelsall BL, et al. Antibodies to interleukin
12 abrogate established experimental colitis in mice. J Exp Med.
1995;182:1281–1290.
8. Strober W, Fuss IJ, Blumberg RS. The immunology of mucosal
models of inflammation. Annu Rev Immunol. 2002;20:495–549.
9. Casini-Raggi V, Kam L, Chong YJ, et al. Mucosal imbalance of
IL-1 and IL-1 receptor antagonist in inflammatory bowel disease.
A novel mechanism of chronic intestinal inflammation. J Immunol. 1995;154:2434–2440.
123
Dig Dis Sci
10. Cominelli F, Nast CC, Duchini A, et al. Recombinant interleukin1 receptor antagonist blocks the proinflammatory activity of
endogenous interleukin-1 in rabbit immune colitis. Gastroenterology. 1992;103:65–71.
11. Thomas TK, Will PC, Srivastava A, et al. Evaluation of an
interleukin-1 receptor antagonist in the rat acetic acid-induced
colitis model. Agents Actions. 1991;34:187–190.
12. Yarur AJ, Abreu MT, Deshpande AR, et al. Therapeutic drug
monitoring in patients with inflammatory bowel disease. World J
Gastroenterol WJG. 2014;20:3475–3484.
13. Baggiolini M. Chemokines in pathology and medicine. J Intern
Med. 2001;250:91–104.
14. Jose PJ, Adcock IM, Griffiths-Johnson DA, et al. Eotaxin: cloning
of an eosinophil chemoattractant cytokine and increased mRNA
expression in allergen-challenged guinea-pig lungs. Biochem
Biophys Res Commun. 1994;205:788–794.
15. Rothenberg ME, Luster AD, Leder P. Murine eotaxin: an eosinophil chemoattractant inducible in endothelial cells and in
interleukin 4-induced tumor suppression. Proc Natl Acad Sci
USA. 1995;92:8960–8964.
16. Garcia-Zepeda EA, Rothenberg ME, Ownbey RT, et al. Human
eotaxin is a specific chemoattractant for eosinophil cells and
provides a new mechanism to explain tissue eosinophilia. Nat
Med. 1996;2:449–456.
17. Mishra A, Hogan SP, Lee JJ, et al. Fundamental signals that
regulate eosinophil homing to the gastrointestinal tract. J Clin
Investig. 1999;103:1719–1727.
18. Kitaura M, Nakajima T, Imai T, et al. Molecular cloning of
human eotaxin, an eosinophil-selective CC chemokine, and
identification of a specific eosinophil eotaxin receptor, CC chemokine receptor 3. J Biol Chem. 1996;271:7725–7730.
19. Rothenberg ME, Mishra A, Brandt EB, et al. Gastrointestinal
eosinophils in health and disease. Adv Immunol. 2001;78:291–328.
20. Chen W, Paulus B, Shu D, et al. Increased serum levels of eotaxin
in patients with inflammatory bowel disease. Scand J Gastroenterol. 2001;36:515–520.
21. Jeziorska M, Haboubi N, Schofield P, et al. Distribution and activation of eosinophils in inflammatory bowel disease using an improved
immunohistochemical technique. J Pathol. 2001;194:484–492.
22. Mir A, Minguez M, Tatay J, et al. Elevated serum eotaxin levels
in patients with inflammatory bowel disease. Am J Gastroenterol.
2002;97:1452–1457.
23. Ahrens R, Waddell A, Seidu L, et al. Intestinal macrophage/
epithelial cell-derived CCL11/eotaxin-1 mediates eosinophil
recruitment and function in pediatric ulcerative colitis. J Immunol. 2008;181:7390–7399.
24. Coburn LA, Horst SN, Chaturvedi R, et al. High-throughput
multi-analyte luminex profiling implicates eotaxin-1 in ulcerative
colitis. PLoS One. 2013;8:e82300.
25. Coburn LA, Gong X, Singh K, et al. L-arginine supplementation
improves responses to injury and inflammation in dextran sulfate
sodium colitis. PLoS One. 2012;7:e33546.
26. Gonzalo JA, Lloyd CM, Kremer L, et al. Eosinophil recruitment
to the lung in a murine model of allergic inflammation. The role
of T cells, chemokines, and adhesion receptors. J Clin Invest.
1996;98:2332–2345.
27. Ding C, Li J, Zhang X. Bertilimumab Cambridge Antibody
Technology Group. Curr Opin Investig Drugs. 2004;5:1213–1218.
123
28. Dent G, Hadjicharalambous C, Yoshikawa T, et al. Contribution
of eotaxin-1 to eosinophil chemotactic activity of moderate and
severe asthmatic sputum. Am J Respir Crit Care Med.
2004;169:1110–1117.
29. Main S, Handy R, Wilton J, et al. A potent human anti-eotaxin1
antibody, CAT-213: isolation by phage display and in vitro and
in vivo efficacy. J Pharmacol Exp Ther. 2006;319:1395–1404.
30. Shteingart S, Rapoport M, Grodzovski I, et al. Therapeutic potency
of IL2-caspase 3 targeted treatment in a murine experimental
model of inflammatory bowel disease. Gut. 2009;58:790–798.
31. Manousou P, Kolios G, Valatas V, et al. Increased expression of
chemokine receptor CCR3 and its ligands in ulcerative colitis: the
role of colonic epithelial cells in in vitro studies. Clin Exp
Immunol. 2010;162:337–347.
32. Adar T, Shteingart S, Ben Ya’acov A, et al. From airway inflammation to inflammatory bowel disease: eotaxin-1, a key regulator of
intestinal inflammation. Clin Immunol. 2014;153:199–208.
33. Rothenberg ME, Mishra A, Brandt EB, et al. Gastrointestinal
eosinophils. Immunol Rev. 2001;179:139–155.
34. Swidsinski A, Loening-Baucke V, Herber A. Mucosal flora in
Crohn’s disease and ulcerative colitis—an overview. J Physiol
Pharmacol. 2009;60:61–71.
35. Lampinen M, Waddell A, Ahrens R, et al. CD14 ? CD33 ? myeloid cell-CCL11-eosinophil signature in ulcerative colitis. J Leukoc
Biol. 2013;94:1061–1070.
36. Targowski T, Jahnz-Rozyk K, Plusa T, et al. Influence of age and
gender on serum eotaxin concentration in healthy and allergic
people. J Investig Allergol Clin Immunol. 2005;15:277–282.
37. Danese S, Fiocchi C. Ulcerative colitis. N Engl J Med.
2011;365:1713–1725.
38. Fiorino G, Szabo H, Fries W, et al. Adalimumab in Crohn’s
disease: tips and tricks after 5 years of clinical experience. Curr
Med Chem. 2011;18:1230–1238.
39. Christophi GP, Rong R, Holtzapple PG, et al. Immune markers
and differential signaling networks in ulcerative colitis and
Crohn’s disease. Inflamm Bowel Dis. 2012;18:2342–2356.
40. Ley K, Burns C. Adhesion molecules in lymphocyte trafficking
and colitis. Gastroenterology. 2001;121:1008–1010.
41. Smyth CM, Akasheh N, Woods S, et al. Activated eosinophils in
association with enteric nerves in inflammatory bowel disease.
PLoS One. 2013;8:e64216.
42. Mahida YR, Wu K, Jewell DP. Enhanced production of interleukin
1-beta by mononuclear cells isolated from mucosa with active
ulcerative colitis of Crohn’s disease. Gut. 1989;30:835–838.
43. Ludwiczek O, Vannier E, Borggraefe I, et al. Imbalance between
interleukin-1 agonists and antagonists: relationship to severity of
inflammatory bowel disease. Clin Exp Immunol. 2004;138:
323–329.
44. Langrish CL, McKenzie BS, Wilson NJ, et al. IL-12 and IL-23:
master regulators of innate and adaptive immunity. Immunol Rev.
2004;202:96–105.
45. Neurath MF. IL-23: a master regulator in Crohn disease. Nat
Med. 2007;13:26–28.
46. Neurath MF, Finotto S, Glimcher LH. The role of Th1/Th2
polarization in mucosal immunity. Nat Med. 2002;8:567–573.
47. Forbes E, Murase T, Yang M, et al. Immunopathogenesis of
experimental ulcerative colitis is mediated by eosinophil peroxidase. J Immunol. 2004;172:5664–5675.