RETRACTED: Alterations in cytokine gene expression profile in colon mucosa of Inflammatory Bowel Disease patients on different therapeutic regimens

Cytokine 92 (2017) 12–19

Contents lists available at ScienceDirect

Cytokine journal homepage: www.journals.elsevier.com/cytokine

Alterations in cytokine gene expression profile in colon mucosa of Inflammatory Bowel Disease patients on different therapeutic regimens Tsvetelina Velikova a,⇑, Dobroslav Kyurkchiev a, Zoya Spassova b, Iliya Karakolev c, Ekaterina Ivanova-Todorova a, Iskra Altankova d, Spaska Stanilova c a

Department of Clinical Laboratory and Clinical Immunology, Medical University of Sofia, University Hospital St. Ivan Rilski, bul. Acad. Ivan Evst. Geshov 15, Sofia 1431, Bulgaria Department of Internal Medicine, Medical University of Sofia, Clinic of Gastroenterology, University Hospital St. Ivan Rilski, Acad. Ivan Evst. Geshov 15 Blvd., Sofia 1431, Bulgaria Department of Molecular Biology, Immunology and Medical Genetics, Medical Faculty, Trakia University, Armeiska 11 Str., Stara Zagora 6000, Bulgaria d University Hospital Lozenets, Sofia University, ul. Kozyak 1, Sofia, Bulgaria b c

a r t i c l e

i n f o

Article history: Received 5 June 2016 Received in revised form 7 January 2017 Accepted 9 January 2017

Keywords: Inflammatory Bowel Disease Cytokines IL-17 FoxP3 Th17 Tregs Immunosuppressive therapy

a b s t r a c t Inflammatory bowel disease (IBD) is assumed to be caused by genetic and environmental factors that interact together in promoting intestinal immune dysregulation where cytokines have validated role. However, the underlying intimate mechanisms in the human IBD involving cytokines still needs to be supplemented especially in the clinical context. The aim of this study was to investigate the expression of some inflammatory and regulatory cytokines (IL-17A, IL-23, IL-6, TGFb1, and IL-10) as well as of the transcription factor FoxP3 in mucosal samples of IBD and non-IBD patients. We assessed the mRNA relative quantities (RQ) of the above-mentioned cytokines and the transcription factor FoxP3 in paired colonic samples (inflamed and adjacent normal mucosa) from 37 patients with IBD and in normal mucosal tissue in 12 persons without IBD by performing a qRT-PCR assay and tested the protein levels of target cytokines in serum samples. The patients were divided into three groups: without any therapy (n = 10), on 5-ASA (n = 11) and on immunosuppressants (Azathioprine ± 5-ASA/corticosteroids) (n = 16) in order to compare the RQ values for each therapeutic group. All investigated genes were found upregulated in the inflamed mucosa of IBD patients in the following order: IL-6 > FoxP3 > TGFb1 > IL-23 > IL-17A > IL-10. We also observed that the gene expression of FoxP3 and IL-6 were substantially higher in the inflamed mucosal tissue of the IBD patients than the adjacent normal mucosa (p = 0.035, p = 0.03 respectively). Differences between higher mRNA expression of FoxP3 and IL-6 in inflamed tissue were considered significant in patients with ulcerative colitis (UC) (p = 0.011, p = 0.000 respectively) and with Crohn’s disease (CD) (p = 0.008, p = 0.000 respectively) in comparison to the normal mucosa of non-IBD persons and we found increased TGFb1 in CD patients alone (p = 0.041). Furthermore, IL-6 and TGFb1 were overexpressed (RQ > 10) in non-inflamed mucosa from IBD patients compared to the normal mucosa from the controls. When we compared the gene expression for paired mucosa in the immunosuppressive treated group with the 5-ASA treated group we observed opposite changes in IL-6 and TGFb1 expression. Additionally, we found higher serum levels of IL-23 (p = 0.008), TGFb1 and IL-6 in IBD patients compared to non-IBD patients. The obtained specific expression profile consisting of IL-6, TGFb1, IL-10 and FoxP3 may represent a transcriptional hallmark for IBD. Furthermore, we found that treatment with immunosuppressive therapy was more beneficial for driving cytokine expression to restore immune regulation in patients with IBD, unlike the 5-ASA therapy. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction

⇑ Corresponding author. Department of Clinical laboratory and Clinical Immunology, Medical University of Sofia, University Hospital St. Ivan Rilski, Laboratory of Clinical Immunology, Acad. Ivan Evst. Geshov 15 Blvd., Sofia 1431, Bulgaria. E-mail address: [email protected] (T. Velikova). http://dx.doi.org/10.1016/j.cyto.2017.01.008 1043-4666/Ó 2017 Elsevier Ltd. All rights reserved.

Inflammatory bowel disease (IBD) is assumed to be caused by genetic and environmental factors that interact together in promoting intestinal immune dysregulation [24,44]. Both entities classically referred as IBD – ulcerative colitis (UC) and Crohn’s disease (CD), are examples of complex disorders, which may possess inflammatory and autoimmune features [9]. Both innate and

13

T. Velikova et al. / Cytokine 92 (2017) 12–19

adaptive immune cells have been identified as the key participants in IBD. The innate immune system takes part mainly at the very outset of an inflammatory response, whereas the adaptive immune system maintains the inflammation [11]. The inflamed mucosa in IBD patients is heavily infiltrated with distinct T-helper cells: Th1, Th2 [14] and also Th17 cells [6,17], which lymphocytes produce a various array of cytokines [13]. Cytokines are the crucial mediators of cells dynamic interplay [24,44], but they are also key regulators of naïve T cells differentiation. Th17 cells express the transcriptional factors RORct and RORa only in the simultaneous presence of two cytokines-transforming growth factor b-1 (TGFb1) and IL-6 [2] and are characterized by a certain cytokine profile (17A, 17F, 17AF), IL-21, IL-22 and IL-26 [3,27], generally found to be proinflammatory [6,16]. They strongly perpetuate the intestinal inflammation during IBD. Meanwhile, Th17 cells play an important role at mucosal interfaces in the defense against bacteria and fungi [36]. The maintenance and expansion of Th17 cells depend also on the presence of IL-23, a heterodimeric cytokine which shares its p40 subunit with IL-12 [2,30,35]. Moreover, the IL-23/Th17 pathway has been proposed to contribute to local chronic inflammation in intestines, whereas the IL-12/Th1 pathway is possibly connected with the systemic inflammation in IBD patients [43,46]. Meanwhile, there is some evidence for another regulatory role of Th17 cells [19] – in the absence of IL-23 they can produce IL10 and this mechanism may play negative feedback control of the inflammation [16]. The function of regulatory T lymphocytes (Tregs), phenotypically described as CD4 + CD25+/-FoxP3+ is to drive the suppressive immune response. However, their role in IBD is not completely understood. Naturally occurring Tregs (nTregs) are generated if TGFb1 is predominantly expressed and inducible Tregs (iTergs) derived from the differentiation of naïve T cells, named as nTregs, stimulated by IL-10, TGFb1, and vitamin D3 in the periphery [18]. It has been suggested that Tregs interact with dendritic cells and consequently suppress the immune response in the mucosa, including the pathogenic effects of IL-17 – producing cells through IL-10 [18]. Tregs capable of producing IL-17 have been also described in IBD and colon cancer [28]. Although the pathophysiology of both CD and UC is better understood, the underlying intimate mechanisms in the human IBD involving cytokines still needs to be supplemented especially in the clinical context [15,24]. IBD – related tissue damages are determined by an integrated response of the innate, adaptive and regulatory immune response against intestinal challenges [10]. Since there are strong proofs that the inflammatory processes take part in the intestinal mucosal damages, the cytokines are the main source of biomarkers in IBD [10]. The quantitative real-time polymerase chain reaction (qRT-PCR) could be successfully employed in the gene expression analyses of IBD-related cytokines, chemokines, adhesion molecules and their respective receptors, revealing their participation in the pro- and anti-inflammatory processes [10,24,32,45]. The involvement of cytokines in the intestinal inflammation of IBD is intensively studied recently and their implication has been proven in experimental animal models [24,44]. Specific gene expression profile, represented the molecular events in the inflamed mucosa of IBD patients, could be of benefit as a future diagnostic and a follow-up tools for these patients. Lack of adequate investigations in humans designates for the need for predictive and prognostic markers that can be easily obtained from the patients samples and based on the molecular fingerprinting technologies of IBD in every single patient [10]. For this reason we decided to design our study by firstly investigating a panel of cytokines related to Th17 and Tregs, as well as the transcription

factor FoxP3, in order to specify the gene and protein expression profile in inflamed and in the adjacent visibly normal colonic samples obtained from Bulgarian IBD patients. Of particular interest for us was to discover if there were differences in gene expression profile in patients with UC in contrast to CD patients. We also questioned whether the type of therapy was able to change the gene expression profile and if it was true in what direction. To address these questions, we purposed to evaluate specific cytokine expression profiles by examining mRNA and protein expression of the cytokines IL-17A, IL-23, IL-6, TGFb1, and IL-10 and the transcription factor FoxP3 in paired colonic mucosa samples of IBD patients and to compare them with the profiles obtained from the control group of non-IBD patients. We also assigned a goal to investigate the differences in mRNA expression profile in IBD patients on different therapeutic regimens and to compare the protein levels of the target cytokines of the IBD and non-IBD patients‘ sera. 2. Material and methods 2.1. Subjects We investigated colonoscopically obtained biopsy samples of paired mucosal samples (inflamed and adjacent normal tissue) from 37 patients with IBD (23 with UC and 14 with CD) at mean age 40 ± 16 years. An overwhelming proportion of the patients (30/37) was in a state of activity which was assessed by Mayo Clinic score of activity for UC patients and by Crohn’s disease activity Index (CDAI) for CD patients. The biopsies were taken during a routine colonoscopy after written confirmed consent signed as following: 3 samples from inflamed and 3 samples from adjacent normal mucosa from each study patient, where the total number of biopsies was 6. Patients with IBD were divided into three groups: without any therapy (n = 10), on 5-ASA (5-aminosalycilates) alone (n = 11) and on immunosuppressive drugs (Azathioprine ± 5-ASA/ corticosteroids) (n = 16). The patients without any therapy were mostly newly diagnosed, whereas the patients on therapy were adherents to the standard weight-adjusted dose of 5-ASA or Azathioprine ± 5-ASA/corticosteroids for a minimum of 3 months. The descriptive statistics of the IBD patients is shown in Table 1. The control group included 12 age and sex matched persons without IBD. They were diagnosed with chronic colitis or anemia during endoscopic and histological evaluation. All patients and control subjects were found negative for autoimmune disease markers (such as anti-nuclear antibodies, rheumatoid factor, anti-neutrophil cytoplasmic antibodies). The study was approved by the Ethic Committee of the Medical University of Sofia and University Hospital St. Ivan Rilski. All patients were informed about the purpose of the study. Table 1 Descriptive characteristics of IBD patients. Characteristics

Number (%) or Mean ± SD

Number UC patients CD patients Age (years) Female Activity of the disease State of activity State of remission Therapy Without On 5-aminosalycilstes (5-ASA) On 5-ASA ± Immunosuppressants

37 23 (62%) 14 (38%) 40 ± 16 20 (54%) 30 (81%) 7 (19%) 10 (27%) 11 (29.8%) 16 (43.2%)

All investigated genes, in accordance with their RQ value, were found upregulated in the inflamed mucosa of IBD patients in the following order: IL-6 > FoxP3 > TGFb1 > IL-23 > IL-17A > IL-10 (Table 2). The mRNA expression of FoxP3 and IL-6 were found substantially higher in the inflamed mucosa than in the adjacent normal mucosa within the IBD patients (p < 0.01). We also found a borderline significant difference for IL-10 shown in Table 2 (p = 0.05). The cytokines IL-23, IL-17A and TGFb1 were also upregulated in inflamed mucosa but this observation did not reach statistical significance (p > 0.05). We also ascertained significant differences between higher mRNA expression of IL-6 in inflamed tissue in UC patients (p = 0.025) compared to the adjacent normal mucosa within the same patients (RQ = 6.3) and increased IL-23 gene expression in CD patients in inflamed versus adjacent normal mucosa (RQ = 28.09, p = 0.046).

RQ Inflamed (IBD) vs. normal tissue (non-IBD patients)

2.15 (0.05–10.72) 13.97 (0.43–155.98) 16.11 (0.02–122.78) 6.17 (0.11–45.12) 0.81 (0.08–2.26) 2.88 (0.04–11.38) 5.69 (0.06–35.63) 29.11* (0.98–233.57) 15.99 (0.005–123.63) 17.91** (0.19–125.24) 2.67 (0.09–30.87) 6.96 (0.05–46.18) 7.06 (0.007–20.53) 8.03* (0.000006–61.65 13.65 (0.009–310.83) 12.28* (0.11–151.17) 4.88*** (0.22–36.10) 3.96 (0.07–27.95) 19.48 ± 2.68 (14.83–21.74) 20.79 ± 0.69 (20.10–21.73) 9.21 ± 3.23 (5.21–16.71) 16.94 ± 1.65 (13.54–20.40) 16.67 ± 3.61 (9.22–21.38) 17.93 ± 1.73 (15.01–19.49) Statistical significance: * p<0.01, ** p<0.05, *** p=0.05

3.1. Gene expression profile of inflamed versus adjacent normal mucosa in IBD patients

19.97 ± 2.48 (16.06–23.96) 19.80 ± 2.33 (13.51–21.99) 8.99 ± 4.41 (2.27–16.71) 15.86 ± 2.41 (11.45–20.01) 17.55 ± 1.42 (15.49–20.3) 18.02 ± 2.63 (14.42–22.76)

3. Results

18.45 ± 3.41 (7.05–23.64) 17.09 ± 2.44 (10.53–20.83) 8.43 ± 4.32 (2.26–16.97) 14.64 ± 2.54 (9.97–19.37) 16.83 ± 1.89 (11.72–20.16) 17.58 ± 3.80 (5.42–22.53)

The results for mRNA expression were obtained by performing the comparative DDCt method. The mean DCt for each target gene was obtained by normalization Ct value to endogenous control. The data of adjacent normal mucosal samples of IBD patients and normal mucosa of non-IBD patients were used as calibrators for relative quantitative (RQ = 2 DDCt) analyses which represented an n-fold mean difference relative to a calibrator. The latter was calculated using 7500 system SDS software. The row data were statistically evaluated by parametric and non-parametric tests using SPSS v.19. Differences in mRNA and protein expression of the target genes were considered significant if the p-value was less than 0.05.

IL-17A IL-6 TGFb1 FoxP3 IL-10 IL-23

2.5. Statistical methods

RQ Inflamed vs. adjacent normal tissue (IBD patients)

The protein levels of IL-17A, IL-6, TGFb1, IL-10 and IL-23 in the patients‘ sera were measured by performing ELISA (Human IL17A ELISA kit, Human TGFb1 ELISA kit, Human IL-6 ELISA kit, Human IL-10 ELISA kit and Human IL-23 ELISA kit, all from Diaclone, Gene probe).

Non-IBD patients (controls) Mean DCt ± SD (Range)

2.4. Enzyme immunoassays

Adjacent normal tissue

We used a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) to perform a quantitative real-time polymerase chain reaction with the following validated PCR primers and 6FAM-labeled TaqMan MGB probes: FoxP3 (Hs00203958_m1), IL-10 (Hs00174086_m1), IL-23A (Hs00372324_m1) - Applied Biosystems, UK, and IL-17A (NM_002190), IL-6 (NM_000600), TGFb1 (NM_000660) Primerdesign, UK. Eukaryotic 18S ribosomal RNA (Hs99999901_s1, Applied Biosystems, UK) was used as endogenous control.

Inflamed tissue

2.3. Quantitative real-time polymerase chain reaction

IBD patients Mean DCt ± SD (Range)

Total RNA from biopsy samples was isolated using a columnbased RNA isolation kit (GeneJET RNA Purification kit, Fermentas, Thermo Scientific). Synthesis of cDNA was accomplished with First Strand cDNA Synthesis kit (Fermentas, Thermo Scientific).

Target gene

2.2. RNA extraction and reverse transcription

RQ Adjacent normal (IBD) vs. normal tissue (non-IBD patients)

T. Velikova et al. / Cytokine 92 (2017) 12–19 Table 2 Gene expression levels in inflamed colonic mucosa samples of patients with IBD and normal colonic mucosa in patients without IBD. Results are expressed as mean DCt ± SE with min-max range and as n-fold mean relative quantification (RQ) with min-max range.

14

15

T. Velikova et al. / Cytokine 92 (2017) 12–19

3.3. Correlation between paired genes We found a significant correlation between paired target genes in inflamed and in adjacent normal tissue from IBD patients for all pairs except FoxP3-IL-10 pair in the inflamed tissue (Table 3). The correlation coefficients varied from moderate (r = 0.436) to strong (r = 0.859). 3.4. Differences in gene expression in IBD patients on different therapeutic regimens We detected that in the patient group on 5-ASA treatment IL-6 (RQ = 16.63 vs. 5.75), IL-10 and IL-17A expression were enhanced, whereas IL-23 and TGFb1 (RQ = 0.64 vs. 1.54) were downregulated compared to the non-treated group (Table 4). On the contrary, in the immunosuppressive treated patients, we found a significant downregulation of IL-6 (RQ = 1.24 vs. 5.75) and IL-17A simultaneously with upregulation of TGFb1 (RQ = 40.41 vs. 1.54) and IL-10

35 30

RQ value

When we compared gene expression of the target cytokines obtained in paired mucosal samples of IBD patients to those obtained from the mucosa of non-IBD patients we received the following results. Differences between higher gene expression of FoxP3 and IL-6 in the inflamed tissue of IBD patients compared to the normal mucosa from persons without IBD were considered significant (Table 2). Furthermore, IL-6, TGFb1, and FoxP3 genes were overexpressed (RQ > 15) in the inflamed IBD mucosa in comparison to their expression in normal mucosa from non-IBD persons. The rest of the genes (IL-10, IL-23, IL-17A, and TGFb1) were also upregulated in patients with IBD but without statistical significance (p > 0.05). The cytokine gene with the least increased expression was IL-10 (RQ = 2.67) (Table 2). There were not significant differences between the gene expressions in adjacent normal mucosa obtained from IBD patients and the gene expressions in the mucosal samples obtained from nonIBD patients. The RQ values of all genes, excluding TGFb1, in inflamed mucosa of IBD patients, calibrated to tissue from nonIBD ones, were higher than the RQ values of these genes in the adjacent normal samples from IBD patients calibrated to normal non-IBD tissue (Table 2). We also compared gene expression levels in the inflamed colonic mucosa of UC and CD patients, respectively, to the normal colonic mucosa in patients without IBD according to their DCt and RQ. We found significant differences between increased gene expression of FoxP3 and IL-6 in the inflamed tissue of both UC (RQ = 17.01, p = 0.011 and RQ = 33.07, p = 0.000, respectively) and CD patients (RQ = 19.39, p = 0.008 and RQ = 21.63, p = 0.000, respectively), compared to the expression in the normal mucosa of persons without IBD. We observed that both diseases share similar expression profiles with a significant increase of TGFb1 expression in CD patients alone (RQ = 22.09, p = 0.041). The mean RQ values of target gene expression in inflamed mucosal samples of patients with IBD, UC, and CD, respectively, compared to adjacent normal tissue in the same patients are shown in Fig. 1A. We found higher RQ values for IL-17A in UC patients and higher RQ value of TGFb1 in CD patients with borderline significance (p = 0.067 and p = 0.058, respectively). On Fig. 1B, we present the comparison between the mean RQ values of target genes in inflamed mucosal samples of patients with IBD, UC, and CD, and the mean RQ values of target genes in normal tissue from patients without IBD. We found a higher RQ values only for TGFb1 in CD patients but with borderline significance (p = 0.077).

A 40

25

IBD patients UC patients CD patients

20 15 10 5 0

B

IL-17A

IL-6

TGFβ1 FoxP3

IL-10

IL-23

40 35 30

RQ value

3.2. Gene expression profile of IBD patients compared to non-IBD patients

25

IBD patients

20

UC patients CD patients

15 10 5 0

IL-17A

IL-6

TGFβ1

FoxP3

IL-10

IL-23

Fig. 1. Mean n-fold relative quantification (RQ) of gene expression levels in inflamed colonic mucosa samples of patients with IBD, UC and CD, respectively, compared to adjacent normal tissue in the same patients (A) and compared to normal colonic mucosa from patients without IBD (B).

Table 3 Correlation between target genes couples in inflamed and adjacent normal tissue in IBD patients. Target genes couples

Inflamed tissue

Adjacent normal tissue

FoxP3 – IL-10 IL-23 – IL-17A IL-6 – IL-23 IL-10 – TGFb1 IL-23 – TGFb1 IL-6 – TGFb1 TGFb1 – IL-17A IL-6 – IL-17A FoxP3 – TGFb1

r = 0.345 r = 0.525** r = 0.681** r = 0.506** r = 0.725* r = 0.699* r = 0.662* r = 0.669** r = 0.436**

r = 0.701* r = 0.701** r = 0.805** r = 0.661* r = 0.859* r = 0.798** r = 0.627** r = 0.737** r = 0.663*

Significant correlations. * p < 0.01. ** p < 0.05.

Table 4 Mean RQ values of target gene expression in inflamed vs. normal adjacent tissue from IBD patients on different therapeutic regimens.

FoxP3 IL-10 IL-17A IL-23 IL-6 TGFb1 *

Without therapy

On 5-ASA

On 5-ASA ± immunosuppressive drugs

11.74 1.77 6.11 5.74 5.75* 1.54*

3.91 5.56 9.39 3.37 16.63 0.64*

20.24 7.22 1.43 44.36 1.24* 40.41

Significant RQ value (p < 0.05)

16

T. Velikova et al. / Cytokine 92 (2017) 12–19

Fig. 2. Differences in gene expression in IBD on different therapeutic regimens.

Table 5 Serum protein levels of target cytokines in IBD and non-IBD patients, presented as Mean ± SE pg/ml. Target gene

IBD patients

Non-IBD patients

P value

IL-17A IL-6 TGFb1 IL-10 IL-23

0.46 ± 0.22 8.47 ± 5.46 13976.84 ± 1147.75 1.27 ± 0.38 3.79 ± 1.12

0.16 ± 0.14 0.03 ± 0.02 10028.82 ± 2250.15 0.68 ± 0.05 0.33 ± 0.03

0.473 0.057 0.092 0.246 0.008

compared to the non-treated group. The results are shown on Fig. 2. 3.5. Protein expression of target genes in IBD and non-IBD patient’s sera We found significantly higher serum levels of IL-23 (p = 0.008), TGFb1 and IL-6 (p = 0.057) in IBD patients compared to non-IBD patients (Table 5). The protein levels of IL-17A and IL-10 were determined to be similar in both groups of patients. 4. Discussion Numerous approaches are available to explore the intestinal mucosa and to determine the cytokine expression profile which could help to elucidate the local immune responses during intestinal inflammation. Revealing the patterns of gene expression in specific tissues, as well as various proteins made in normal or diseased tissues or serum and their relative abundance and interactions, can be helpful in identifying the molecular fingerprints underlying any particular status of health and disease [9]. Our investigation of the cytokine environment in mucosal lesions of IBD patients revealed that affected areas displayed a high expression pattern of proinflammatory and regulatory cytokines in the following order: IL-6 > FoxP3 > TGFb1 > IL-23 > IL-17A > IL-10. The gene expression profile of non-IBD persons is arranged slightly different: TGFb1 > IL-10 > FoxP3 > IL-23 > IL-17A > IL-6. It is visible that the immune suppressive cytokines TGFb1 and IL-10, as well as FoxP3, the transcription marker for Tregs, are at higher expression in non-IBD persons, and we could speculate that these findings might be connected to the predominant mucosal immune tolerance in the normal intestinal mucosa. Moreover, the great difference between IBD and non-IBD patients concerns the IL-6 expression.

Th17 cells and their representative cytokines provide a disparate pathway in the interaction between the innate and the adaptive immunity during inflammation [21]. Fujino et al. in 2003 firstly described CD4 + cells as a source of IL-17 in IBD with the identification of IL-17+ T cells and elevated serum levels of IL-17 in IBD patients compared to controls [13]. As would be expected, high levels of IL-17A mRNA expressions were found in the inflamed mucosa of both major forms of IBD [13], compared to the non-inflamed mucosa. We observed increased gene expressions of mRNA IL-17 in the inflamed mucosa from IBD patients (7.06-fold) compared to the adjacent normal tissue in the same IBD patients (5.69-fold) and compared to the normal tissue in non-IBD persons, respectively. However, these differences did not rich statistical significance. Our results are in concordance with many investigators reported of increased gene and protein expression of IL-17 in UC patients [12], as well as increased Th17 cells in the peripheral blood and elevated levels of IL-17A in the intestinal mucosa of IBD patients [7,8]. Th17 recruitment in these circumstances is under a strict control of regulatory immune cells. We found IL-17 expression also in the adjacent normal tissue of IBD patients, as well as in normal tissue of non-IBD patients (with similar levels of mRNA IL-17 and non-significant RQ = 2.15). The necessary environment for human Th17 cells differentiation is unclear in humans. IL-1b and IL-6 are discussed as inducers of Th17 from T cells already converted into memory cells, while IL21 and TGFb1 are thought to be involved in Th17differentiation from naïve T cells [48]. Another cytokine critical in Th17 development is IL-23 and it is known to play a role in the maintenance of Th17 cells [1]. We observed increased relative mRNA levels of IL-23 (RQ = 6.96), although without a statistical significance. Other investigators did not found changes in IL-23 expression too [8]. It is accepted that IL-12 and IL-23 are mainly produced by dendritic cells but macrophages may also contribute to intestinal inflammation by secreting both of them as well as TNFa [22]. It has been proven by experimental models that IL-12/IL-23 cytokine group participates in the development of CD [33]. The IL-12/IL-23 axis is also involved in Th1 and Th17 response critically important in human IBD [34]. The cytokine TGFb1 participates in the development of both Tregs and Th17 cells and its function might be dose-dependent [47]. Low levels of TGFb1 and high levels of IL-6 in conjunction drive the naïve T cells to differentiate into Th17, while the development of Tregs requires the presence of high levels of TGFb1. Bettelli et al. using mice models have shown that IL-6 is able to inhibit completely the formation of Foxp3 + Tregs induced by TGFb1 [2].

T. Velikova et al. / Cytokine 92 (2017) 12–19

In our study, we found both TGFb1 and IL-6 upregulated in patients’ mucosa. Considering the high level of measured mRNA IL-6 and TGFb1 expression simultaneously in the patient’s mucosa, we suggest that in this panel of IBD patients IL-6 drives differentiation of naïve T cells towards Th17 cells and a Th17 effector immune response. Kimura et al. also described the pleiotropic role of IL-6 which can regulate the balance between Th17 cells and Tregs [25]. Indeed, in our investigations, this assumption is supported by the higher levels of mRNA IL-17A in inflamed compared to normal tissue. On the other hand, we obtained evidence for Tregs presence through detected significant overexpression of transcriptional factor FoxP3 in the inflamed tissue versus the adjacent normal tissue of IBD patients and versus the normal tissue in non-IBD persons (RQ = 12.28 and RQ = 17.91, respectively). The highly expressed FoxP3 in the intestinal mucosa of IBD patients suggests that Tregs may be actively produced in the intestinal mucosa in an endeavor to suppress the rising inflammation [8]. Various types of Tregs have evolved in recent years and are characterized mainly by different cytokine profiles including the predominant production of immunomodulatory cytokines such as TGFb1 and IL-10 [15]. In line with this, we found both TGFb1 (RQ = 13.65) and IL-10 (RQ = 4.88) upregulated in the inflamed mucosa. IL-10 is viable to attenuate the mucosal inflammation [41], a suggestion established after the accumulation of many reports for the development of spontaneous colitis in IL-10-/- mice mainly by shifting macrophages differentiation to a more proinflammatory subset [23]. Furthermore, so-called regulatory B cells could be a different source of TGFb1 and IL-10 [40]. We found significant correlations between functionally paired target genes in the inflamed and adjacent normal tissue samples from IBD patients for all pairs except the FoxP3-IL-10 pair in the inflamed tissue. It is well known that the investigated cytokines interact continuously especially during active intestinal inflammation [4,17]. The coefficients of correlation varied from moderate (r = 0.436) to strong (r = 0.859) and the strongest correlation was established for the IL-23 – TGFb1 pair in normal tissue. This is in consist with the suggestion that in the regulation of intestinal immune tolerance many regulatory cells (Tregs, IL-10-producing cells, Th3 cells), other cells and the cytokine milieu contribute and not necessarily at the same time [34,37,41]. Under healthy conditions, regulatory T cells may suppress the function of effector T cells [5,39] but during the inflammation, Tregs may be numerically and functionally impaired [29,31]. Some explanations of these phenomena are that Tregs may not be fully functional or they could convert to Th17 in the presence of high levels of IL-6. Previous studies have described functionally suppressed Tregs in vitro which were isolated from the gutassociated lymphoid tissue of IBD patients, although the capacity of Tregs to suppress proliferation and effector activity of Th17 cells may be limited ex vivo by reason of a proinflammatory cytokine milieu [31]. Moreover, in the presence of IL-1b and IL-6, Tregs may not only be with blocked suppressive capacity but may also convert to Th17 cells [26], which even contribute to disease. In summary, the balance between Tregs and Th17 cells assures a state of health, but changes in the cytokine environment could promote predominantly the development of Tregs or Th17 cells which could lead to the establishment of a state of disease. Comparing of the cytokine gene expression profiles of adjacent normal tissue from IBD patients to the normal mucosal samples from non-IBD persons revealed that expression of TGFb1 and IL-6 was upregulated (RQ = 16.11 and RQ = 13.97, respectively) even though not significantly (Table 2). Both cytokines are prerequisites for the Th17 development. This suggests to us the possibility for activation of a proinflammatory program of the intestinal cells at visibly normal mucosa towards inflammation. Thus, transcriptomics can reveal that the endoscopically and histologically normal

17

mucosa of UC patients in clinical remission is not really normal, as its gene expression pattern is still different from that of the normal mucosa of healthy controls or unaffected UC mucosa [38]. Although CD and UC have distinct clinical presentations and probably they differ also in their underlying genetics, pathophysiology, and molecular mechanisms, our results showed that the affected areas from both major IBD forms shared a similar cytokine gene expression profile (Fig. 1). However, we found higher IL-17A in UC patients and higher TGFb1 in CD patients (p = 0.067 and p = 0.058, respectively) in inflamed compared to adjacent normal tissue and higher TGFb1 in inflamed tissue of CD patients alone compared to the normal tissue in non-IBD patients (p = 0.077). This is not surprising, especially having in mind the proven role of TGFb1 in tissue modification (clinically presented as fistulas, strictures, etc.) which is more typical for CD patients. Thereby, the increased expression of TGFb1 in inflamed intestines might represent a distinction marker of CD. The choice of an appropriate therapy and the prediction of the ultimate outcome are ones of multiple still unanswered issues in regard to IBD [9]. The available therapeutic modalities for patients with IBD – 5-aminosalycilic acid agents (5-ASA), steroids and immunosuppressive drugs used at present are all directed to decrease the generation of proinflammatory cytokines and mediators in the inflamed mucosa. We aimed to investigate the differences in mRNA expression of key cytokines related to Th17 and Tregs in paired mucosal samples from IBD patients on different therapeutic regimen (without any therapy, on 5-ASA, and on immunosuppressants: Azathioprine ± 5-ASA/corticosteroids). We found main differences for TGFb1 and IL-6 expression, cytokines both crucial for the development of pathogenic Th17 lineage cells, in the inflamed IBD tissue versus normal adjacent tissue in the two treated groups compared to the non-treated group. We observed opposite changes in IL-6 expression and TGFb1 for paired mucosa in the immunosuppressive treated group compared to 5-ASA treated group. Immunosuppressive treatment significantly downregulated expression of IL-6, IL-17A, and IL-23 but upregulated FoxP3 and anti-inflammatory cytokine IL-10 compared to the nontreated group. Thus, by decreasing the prerequisite for Th17 developing IL-6 and the maintaining Th17 subpopulation, the immunosuppressants lowered the IL-17 levels. In 5-ASA treated patients, we also found modulated cytokine gene expression but in opposite direction – upregulation of IL-6 and downregulation of TGFb1 which could lead to Th17 differentiation. This suggestion is supported by our results for the upregulated IL-17A expression in these patients. In conclusion, treatment with immunosuppressive therapy is more beneficial for driving cytokine expression to restore immune regulation in patients with IBD, particularly by changing significantly the gene expression of IL-6 and TGFb1, unlike the 5-ASA therapy. Investigating cytokine gene expression profiles by transcriptomics would allow developing and applying for more effective therapies to individual patients with CD and UC. Obtained mucosal gene signatures in CD and UC patients could be fully applicable to personalized medicine approach in IBD and may predict response to treatment [9]. In order to pursue the eventual transcription and secretion of the target cytokines, we investigated the protein levels of cytokines in the patient’s serum and compared them to the serum levels of non-IBD patients. We found that the levels of the target cytokines in serum did not correlate with their equivalent mRNA levels in the mucosa of the patients. When we compared IBD patients to nonIBD patients, we found significantly higher serum levels of IL-23, TGFb1 and IL-6 in IBD patients but similar protein levels of IL17A and IL-10 in both groups of patients. One reason for that may be due to not a full translation of mRNA in proteins, or the relatively short lifespan of mRNA. Another possible reason is linked to the assumption that intestinal inflammation is a local process with

18

T. Velikova et al. / Cytokine 92 (2017) 12–19

no equal system reflection in circulation. A limitation of our study is the lack of enough direct data of the protein cytokines levels in the mucosa. We started a pilot study with 4 patients with IBD and 2 healthy persons (data not shown) and so far we found much higher amounts of tested cytokines in mucosa than in serum of the IBD patients. Our results for mucous cytokine levels so far are inconclusive but we think that more data in this respect might be very interesting. On account of the greatest challenges for the gastroenterologists to choose the most effective therapeutic strategy for their IBD patients, some monoclonal antibodies were recently developed. Ustekinumab, a human monoclonal antibody, blocks both IL-12 and IL-23, and Secukinumab targets IL-17. While anti-IL-23 antibody has demonstrated some efficacy [42], the anti-IL-17 antibody has not shown any clinical benefit so far in the treatment of CD patients [20]. We did find higher protein levels of IL-23 in the patient serum and antibody against it could be beneficial in patients who have increased IL-23 in their serum. In respect to our finding, we could suppose also the potential benefit of antiIL-6 antibodies as a therapeutic target in IBD. 5. Conclusion We obtained a specific expression profile consisting of upregulated IL-6 and TGFb1, needed for Th17 development, simultaneously with the enhanced expression of transcription factor FoxP3 and IL-10, defining Tregs, which profile may represent a transcriptional hallmark for IBD. Additionally, we observed that the two distinct IBD forms share similar expression profiles with differences regarding higher IL-17A in UC patients but higher TGFb1 in CD patients. Since there were mild differences between the gene expressions of TGFb1 and IL-6 in adjacent visibly ‘‘normal” mucosa obtained from IBD patients and in the non-IBD mucosa, this suggested to us the possibility for activation of a proinflammatory program towards inflammation in still non-affected IBD mucosa. Furthermore, we found that treatment with immunosuppressive therapy was more beneficial for driving cytokine expression to restore immune regulation in patients with IBD, unlike the 5-ASA therapy. Taken together our results confirm that the cytokine balance between proinflammatory and regulatory cytokines and their corresponding cells is a very important condition which provides the state of health or disease in the gut. For the clinical practice individualized specific expression cytokine profile–based approach in gastroenterology could be of benefit to provide the best possible care to every IBD patient. Acknowledgment This work was supported by Medical University of Sofia [Grant №22/2012-2013] and the Fund for Scientific and Mobile project from Faculty of Medicine at the Trakia University of Stara Zagora [Grant №4/2012-2013]. References [1] S. Aggarwal, N. Ghilardi, M.H. Xie, F.J. de Sauvage, A.L. Gurney, Interleukin-23 promotes a distinct CD4 T cell activation state characterized by the production of interleukin-17, J. Biol. Chem. 278 (3) (2003) 1910–1914. [2] E. Bettelli, Y. Carrier, W. Gao, T. Korn, T.B. Strom, M. Oukka, H.L. Weiner, V.K. Kuchroo, Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells, Nature 441 (7090) (2006) 235–238. [3] E. Bettelli, T. Korn, V.K. Kuchroo, Th17: the third member of the effector T cell trilogy, Curr. Opin. Immunol. 19 (6) (2007) 652–657. [4] F. Caprioli, F. Pallone, G. Monteleone, Th17 immune response in IBD: A new pathogenic mechanism, J. Crohn’s & Colitis 2 (4) (2008) 291–295. [5] T.L. Denning, Y.C. Wang, S.R. Patel, I.R. Williams, B. Pulendran, Lamina propria macrophages and dendritic cells differentially induce regulatory and interleukin 17-producing T cell responses, Nat. Immunol. 8 (10) (2007) 1086–1094.

[6] C. Dong, Differentiation and function of pro-inflammatory Th17 cells, Microbes Infect./Inst. Pasteur. 11 (5) (2009) 584–588. [7] N. Eastaff-Leung, Regulatory T Cells, Th17 Effector Cells and Cytokine Microenvironment in Inflammatory Bowel Disease and Coeliac Disease, 2009. [8] N. Eastaff-Leung, N. Mabarrack, A. Barbour, A. Cummins, S. Barry, Foxp3+ regulatory T cells, Th17 effector cells, and cytokine environment in inflammatory bowel disease, J. Clin. Immunol. 30 (1) (2010) 80–89. [9] C. Fiocchi, Integrating omics: the future of IBD?, Dig Dis. Sci. 32 (Suppl 1) (2014) 96–102. [10] J. Florholmen, W. Fries, Candidate mucosal and surrogate biomarkers of inflammatory bowel disease in the era of new technology, Scand. J. Gastroenterol. 46 (12) (2011) 1407–1417. [11] S. Foersch, M.J. Waldner, M. Neurath, Innate and adaptive immunity in inflammatory bowel diseases, Dig. Dis. 31 (3–4) (2013) 317–320. [12] G. Fonseca-Camarillo, J. Furuzawa-Carballeda, L. Llorente, J.K. YamamotoFurusho, IL-10– and IL-20–expressing epithelial and inflammatory cells are increased in patients with ulcerative colitis, J. Clin. Immunol. 33 (3) (2011) 640–648. [13] S. Fujino, A. Andoh, S. Bamba, A. Ogawa, K. Hata, Y. Araki, T. Bamba, Y. Fujiyama, Increased expression of interleukin 17 in inflammatory bowel disease, Gut 52 (1) (2003) 65–70. [14] I.J. Fuss, Is the Th1/Th2 paradigm of immune regulation applicable to IBD?, Inflamm Bowel Dis. 14 (Suppl 2) (2008) S110–S112. [15] R.R. Gerner, A.R. Moschen, H. Tilg, Targeting T and B lymphocytes in inflammatory bowel diseases: lessons from clinical trials, Digest. Dis. (Basel, Switzerland) 31 (3-4) (2013) 328–335. [16] C. Gonzalez-Garcia, F.M. Martin-Saavedra, A. Ballester, S. Ballester, The Th17 lineage: answers to some immunological questions, Immunologia 28 (1) (2009) 32–45. [17] L.E. Harrington, P.R. Mangan, C.T. Weaver, Expanding the effector CD4 T-cell repertoire: the Th17 lineage, Curr. Opin. Immunol. 18 (3) (2006) 349–356. [18] hirumitsu. [19] Z. Hovhannisyan, J. Treatman, D.R. Littman, L. Mayer, Characterization of interleukin-17-producing regulatory T cells in inflamed intestinal mucosa from patients with inflammatory bowel diseases, Gastroenterology 140 (3) (2011) 957–965. [20] W. Hueber, B.E. Sands, S. Lewitzky, M. Vandemeulebroecke, W. Reinisch, P.D. Higgins, J. Wehkamp, B.G. Feagan, M.D. Yao, M. Karczewski, J. Karczewski, N. Pezous, S. Bek, G. Bruin, B. Mellgard, C. Berger, M. Londei, A.P. Bertolino, G. Tougas, S.P. Travis, Secukinumab, a human anti-IL-17A monoclonal antibody, for moderate to severe Crohn’s disease: unexpected results of a randomised, double-blind placebo-controlled trial, Gut 61 (12) (2012) 1693–1700. [21] G. Hundorfean, M.F. Neurath, J. Mudter, Functional relevance of T helper 17 (Th17) cells and the IL-17 cytokine family in inflammatory bowel disease, Inflamm. Bowel Dis. 18 (1) (2011) 180–186. [22] N. Kamada, T. Hisamatsu, S. Okamoto, H. Chinen, T. Kobayashi, T. Sato, A. Sakuraba, M. Kitazume, A. Sugita, K. Koganei, K. Akagawa, T. Hibi, Unique CD14 intestinal macrophages contribute to the pathogenesis of Crohn disease via IL23/IFN-gamma axis, J. Clin. Invest. 118 (6) (2008) 2269–2280. [23] N. Kamada, T. Hisamatsu, S. Okamoto, T. Sato, K. Matsuoka, K. Arai, T. Nakai, A. Hasegawa, N. Inoue, N. Watanabe, K. Akagawa, T. Hibi, Abnormally differentiated subsets of intestinal macrophage play a key role in Th1dominant chronic colitis through excess production of IL-12 and IL-23 in response to bacteria, J. Immunol. 175 (10) (2005) 6900–6908. [24] A. Kaser, S. Zeissig, R.S. Blumberg, Inflammatory bowel disease, Annu. Rev. Immunol. 28 (2010) 573–621. [25] A. Kimura, T. Kishimoto, IL-6: regulator of Treg/Th17 balance, Eur. J. Immunol. 40 (7) (2010) 1830–1835. [26] H.J. Koenen, R.L. Smeets, P.M. Vink, E. van Rijssen, A.M. Boots, I. Joosten, Human CD25highFoxp3pos regulatory T cells differentiate into IL-17producing cells, Blood 112 (6) (2008) 2340–2352. [27] T. Korn, E. Bettelli, M. Oukka, V.K. Kuchroo, IL-17 and Th17 Cells, Annu. Rev. Immunol. 7 (2009) 485–517. [28] L. Li, V. Boussiotis, The role of IL-17-producing Foxp3+ CD4+ T cells in inflammatory bowel disease and colon cancer, Clin. Immunol. 148 (2) (2013) 246–253. [29] S. Makita, T. Kanai, Y. Nemoto, T. Totsuka, R. Okamoto, K. Tsuchiya, M. Yamamoto, H. Kiyono, M W. Intestinal lamina propria retaining CD4+CD25+ regulatory T cells is a suppressive site of intestinal inflammation, J. Immunol. 178 (8) (2007) 4937–4946. [30] K.J. Maloy, The Interleukin-23/Interleukin-17 axis in intestinal inflammation, J. Intern. Med. 263 (6) (2008) 584–590. [31] J. Maul, C. Loddenkemper, P. Mundt, E. Berg, T. Giese, A. Stallmach, M. Zeitz, R. Duchmann, Peripheral and intestinal regulatory CD4+ CD25(high) T cells in inflammatory bowel disease, Gastroenterology 128 (7) (2005) 1868–1878. [32] M.G. Neuman, Immune dysfunction in inflammatory bowel disease, Transl. Res. 149 (4) (2007) 173–186. [33] M. Neurath, IL-23: a master regulator in Crohn disease, Nat. Med. 13 (1) (2007) 26–28. [34] M.F. Neurath, Cytokines in inflammatory bowel disease, Nat. Rev. Immunol. 14 (5) (2014) 329–342. [35] B. Oppmann, R. Lesley, B. Blom, J.C. Timans, Y. Xu, B. Hunte, F. Vega, N. Yu, J. Wang, K. Singh, F. Zonin, E. Vaisberg, T. Churakova, M. Liu, D. Gorman, J. Wagner, S. Zurawski, Y. Liu, J.S. Abrams, K.W. Moore, D. Rennick, R. de WaalMalefyt, C. Hannum, J.F. Bazan, A.K. Robert, Novel p19 protein engages

T. Velikova et al. / Cytokine 92 (2017) 12–19

[36] [37] [38]

[39] [40]

[41]

[42]

IL-12p40 to form a cytokine, IL-23, with biological activities similar as well as distinct from IL-12, Immunity 13 (5) (2000) 715–725. W. Ouyang, J.K. Kolls, Y. Zheng, The biological functions of T helper 17 cell effector cytokines in inflammation, Immunity 28 (4) (2008) 454–467. K.A. Papadakis, S.R. Targan, Role of cytokines in the pathogenesis of inflammatory bowel disease, Annu. Rev. Med. 51 (2000) 289–298. N. Planell, J.J. Lozano, R. Mora-Buch, M.C. Masamunt, M. Jimeno, I. Ordas, M. Esteller, E. Ricart, J.M. Pique, J. Panes, A. Salas, Transcriptional analysis of the intestinal mucosa of patients with ulcerative colitis in remission reveals lasting epithelial cell alterations, Gut 62 (7) (2013) 967–976. M. Rescigno, Dendritic cells in oral tolerance in the gut, Cell. Microbiol. 13 (9) (2009) 1312–1318. G. Salinas, F. Braza, S. Brouard, P. Tak, D. Baeten, The role of B lymphocytes in the progression from autoimmunity to autoimmune disease, Clin. Immunol. Immunopathol. 146 (1) (2013) 34–45. F. Sanchez-Munoz, A. Dominguez-Lopez, J.K. Yamamoto-Furusho, Role of cytokines in inflammatory bowel disease, World J. Gastroenterol. 14 (27) (2008) 4280–4288. W. Sandborn, C. Gasink, L. Gao, M. Blank, J. Johanns, C. Guzzo, B. Sands, S. Hanauer, S. Targan, P. Rutgeerts, S. Ghosh, W. de Villiers, R. Panaccione, G.

[43] [44] [45]

[46]

[47] [48]

19

Greenberg, S. Schreiber, S. Lichtiger, B. FeaganCERTIFI, Ustekinumab induction and maintenance therapy in refractory Crohn’s disease, N. Engl. J. Med. 367 (16) (2012) 1519–1528. M. Sarra, F. Pallone, T.T. Macdonald, G. Monteleone, IL-23/IL-17 axis in IBD, Inflamm. Bowel Dis. 16 (10) (2010) 1808–1813. W. Strober, I. Fuss, P. Mannon, The fundamental basis of inflammatory bowel disease, J. Clin. Invest. 117 (3) (2007) 514–521. M.I. Torres, A. Rios, Current view of the immunopathogenesis in inflammatory bowel disease and its implications for therapy, World J. Gastroenterol. 14 (13) (2008) 1972–1980. H.H. Uhlig, B.S. McKenzie, S. Hue, C. Thompson, B. Joyce-Shaikh, R. Stepankova, N. Robinson, S. Buonocore, H. Tlaskalova-Hogenova, D.J. Cua, F. Powrie, Differential activity of IL-12 and IL-23 in mucosal and systemic innate immune pathology, Immunity 25 (2) (2006) 309–318. Y. Wan, R. Flavell, ’Yin-Yang’ functions of transforming growth factor-beta and T regulatory cells in immune regulation, Immunol. Rev. 220 (2007) 199–213. J. Yang, L. Jia, L. Li, C.Y. Wu. [The differentiation and regulation of antigenspecific Th17 cells], Xi bao yu fen zi mian yi xue za zhi = Chinese J. Cell. Mol. Immunol. 24(3) (2008) 213–216.