Scandinavian Journal of Gastroenterology. 2014; 49: 1191–1200
ORIGINAL ARTICLE
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Myosin IXb variants and their pivotal role in maintaining the intestinal barrier: A study in Crohn’s disease
MATTHIAS PRAGER1*, TAHIR DURMUS2*, JANINE BÜTTNER1, TAMAS MOLNAR3, DIRK J. DE JONG4, JOOST PH. DRENTH4, DANIEL C. BAUMGART5, ANDREAS STURM6, KLAUDIA FARKAS3, HEIKO WITT7,8* & CARSTEN BÜNING1* 1
Department of Gastroenterology and Hepatology, Charité, Campus Mitte, Universitätsmedizin Berlin, Berlin, Germany, Department of Radiology, Charité, Campus Mitte, Universitätsmedizin Berlin, Berlin, Germany, 31st Department of Medicine, Faculty of Medicine, University of Szeged, Szeged, Hungary, 4Department of Gastroenterology and Hepatology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands, 5Department of Gastroenterology and Hepatology, Charité, Campus Virchow-Klinikum, Universitätsmedizin Berlin, Berlin, Germany, 6Department of Internal Medicine, Krankenhaus Waldfriede, Berlin, Germany, 7Else Kröner-Fresenius-Zentrum für Ernährungsmedizin (EKFZ) & Zentralinstitut für Ernährungs- und Lebensmittelforschung (ZIEL), Technische Universität München (TUM), Gregor-Mendel-Str. 2, 85350 Freising, Germany, and 8Department of Pediatrics, Klinikum rechts der Isar (MRI), Technische Universität München (TUM), Parzivalstr. 16, 80804 Munich, Germany
2
Abstract Background. Myosin IXb (MYO9B) is involved in the regulation of epithelial barrier function. We hypothesized that MYO9B variants are associated with increased intestinal permeability measured in patients with Crohn’s disease (CD), where barrier dysfunction is crucially involved in disease development. Methods. We sequenced MYO9B and genotyped five MYO9B variants (rs1545620, rs1457092, rs2279003, rs2305764 and rs2279002) and correlated these data to measurement of intestinal permeability in German CD patients (n = 122) obtained by standard oral sugar test using the lactulose/mannitol ratio after measurement of urinary excretion. We furthermore studied MYO9B variants in three European cohorts with inflammatory bowel disease (IBD) and healthy controls : Germany (CD = 264; ulcerative colitis = 143 [UC]; HC = 372); Hungary (CD = 147; UC = 117; HC = 195), the Netherlands (CD = 157; HC = 219). Results. We found an association for four studied MYO9B variants to an increased intestinal permeability in CD patients (rs1545620, p = 0.010; rs1457092, p = 0.024; rs2279003, p = 0.003; rs2305764, p = 0.015). Furthermore, we observed significantly higher absolute values of intestinal permeability for individuals carrying risk alleles within MYO9B. Looking for an overall disease association, only the rs2305764 variant was associated with CD in the Dutch cohort (p = 0.004), but not in the German or Hungarian cohort. No association to UC or a distinct phenotype in both CD and UC patients was observed for all studied MYO9B variants. Conclusion. Our data suggest a link between MYO9B variants to an increased intestinal permeability in CD patients. This supports the influence of Myosin IXb on the integrity of the epithelial barrier. The role of MYO9B variants in the overall susceptibility to IBD, however, remains to be elucidated.
Key Words: Crohn’s disease, inflammatory bowel disease, intestinal permeability, myosin IXb, ulcerative colitis
Introduction A disturbed intestinal epithelial barrier appears to be critical in development and course of numerous
gastrointestinal and primary non-gastrointestinal diseases such as inflammatory bowel diseases (IBD) [1,2] celiac disease, non-alcoholic-steatohepatitis and diabetes mellitus [3,4]. Although the concept is
Correspondence: PD Dr. med. Carsten Büning, Department of Hepatology and Gastroenterology, Charité, Universitätsmedizin Berlin, Campus Mitte, Charitéplatz 1, 10117 Berlin, Germany. Tel: + 49 30 450614237. Fax: + 49 30 450514906. E-mail: [email protected] *M Prager, T Durmus, H Witt and C Büning contributed equally.
(Received 3 April 2014; revised 19 May 2014; accepted 19 May 2014) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2014.928903
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1192
M. Prager et al.
accepted, the underlying mechanisms remain to be elucidated for all diseases. It is uncertain whether intestinal epithelial barrier dysfunction results from genetic variation or reflects secondary changes due to local or systemic inflammation. Evidence for a genetic predisposition comes from family-based studies especially for Crohn’s disease, where clustering of increased intestinal permeability is predominantly seen in first-degree relatives compared to spouses [1,5]. The intestinal epithelial barrier has a gatekeeping role, which protects the internal milieu from unwanted pathogenic luminal contents. Tight junctions appear in the closest possible cell contact sealing the paracellular space and forming the rate-limiting step in determining epithelial integrity (Figure 1) [6]. However, regulation of epithelial barrier is complex. Myosin IXb (MYO9B) is expressed in intestinal epithelial cells and encodes a protein involved in actin remodeling in epithelial enterocytes and tight junction assembly [7–11]. Experimental studies revealed that overexpression of rat MYO9B leads to actin filamentrelated morphologic changes in epithelial cells [9]. Interestingly, MYO9B variants have been associated with celiac disease [12]. Since passing of
Paracellular transport
immunogenic gluten peptides through the epithelial barrier is characteristic for celiac disease, these observations point towards a primary epithelial defect. However, up to now there are no supportive data that suggest that MYO9B variants are associated with variation in intestinal permeability. To answer this question, we choose Crohns disease (CD) as a disease model because this reflects a disorder with inherent abnormal intestinal permeability [13]. In the first step, we investigated CD patients for changes in intestinal permeability in vivo by standard oral sugar test and correlated these data to four known MYO9B variants [14], among them three intron variants. Since we herein observed a significant association to an increased intestinal permeability, we decided in a second step to sequence all coding regions, exon/intron boundaries and promoter regions of MYO9B to look for other functional variants. Sequencing was performed in CD patients with altered intestinal permeability to detect possible additional genetic changes that could have an impact on intestinal integrity as well. In the last step, all MYO9B variants were studied in large CD cohorts from different European populations to look for an
Apical
Tight junctions
Myosin 9B Head Z0 1,2,3
Claudins Neck
Occludin Actomyosin cytoskeleton JAM1
Tail
Adherens junctions
Desmosomes F-actin
Transcellular transport
Basal
Figure 1. Schematic representation of Myosin IXB within the epithelial barrier. The paracellular transport of molecules through epithelial barrier is regulated by the tight junctions complex, adherens junctions and desmosomes. The tight junction complex is connected with the actomyosin cytoskeleton.
Myosin XIb variants and intestinal barrier
1193
smoking, or never smoked), disease localization and behavior. The following data of UC patients were obtained: age at diagnosis, gender, familial or spontaneous disease, smoking habits and disease localization (Table I). Stratification was performed according to the Montreal classification [17].
overall susceptibility. Data were compared to healthy controls and also to patients with ulcerative colitis, another chronic inflammatory bowel disease. Materials and methods
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Subjects Genotyping of the MYO9B variants. Genomic DNA was extracted from peripheral blood leukocytes. As primers we used the oligonucleotides 5¢-GCTGCC GTGTACCACCA-3¢ and 5¢CTTAGTGCCACAG GCCAGA-3¢ for rs1545620 (exon20), 5¢-GGCAT CCACCAGGCACA-3¢ and 5¢-GGCAGGGGAGA GGTGAACA-3¢ for rs1457092 (intron 20), 5¢-GC CTGGCCTTGGACAGCA-3¢ and 5¢-ACTGGCT GAGCATGGCACTA-3¢ (intron 21), 5¢-GCCA ACCTCCTCCTTGTTC-3¢ and 5¢-AGGGGTGA CTTATTCAAGACT-ACTGC-3¢ for rs2305764 (intron28) and 5¢-GGGGTGCTGAAGCAGTG-3¢ and 5¢-CCACCAACACGACCTGAGAT-3¢ for rs2279002 (intron32). We performed PCR using 0.75 U AmpliTaq Gold (Applied Biosystems), 400 mM dNTPs, 1.5 mM MgCl2 and 0.1 mM of each primer in a final volume of 25 ml. The reaction mix was denatured at 95 C for 12 min followed by 48 cycles of denaturation at 95 C for 20 s, annealing at 60 C for 40 s, elongation at 72 C for 90 s and a final extension step for 2 min at 72 C in an automated thermocycler.
IBD patients. We included CD or ulcerative colitis (UC) patients diagnosed upon clinical, endoscopic, radiological and histological findings [15,16]. The study was approved by the responsible ethic committees of all three institutions and informed consent was obtained from each participant. In total, 407 white IBD patients (264 with CD and 143 with UC) were recruited from Germany. In addition, we enrolled CD 147 patients and 117 UC patients from Szeged, Hungary and 157 CD patients from Nijmegen, the Netherlands. Unrelated and healthy subjects from Germany (n = 372), Hungary (n = 195) and the Netherlands (n = 219) served as controls for the genetic case-control analysis. Clinical data of IBD patients were obtained through retrospective collection from the patients clinical charts prior to genotyping. The following data of CD patients were obtained: age, age at diagnosis, gender, familial or spontaneous disease (familial disease was considered when one first- or second-degree relative was diagnosed with IBD), smoking habits (current smoking, or history of
Table I. Demographical and clinical characteristics of CD and UC patients from Germany, Hungary and the Netherlands. CD Population Gender (male:female) Age at diagnosis (median, SD) Disease duration (median, SD) Familial IBD Smoking Localization (Montreal) L1: Ileum L2: Colon L3: Ileocolon L4: Upper GI Perianal disease L1: Proctitis L2: Left-sided colits L3: Pancolits Behavior (Montreal) B1: Non-stricturing non-penetrating B2: Stricturing B3: Penetrating
UC
Germany (n = 264)
Hungary (n = 147)
Netherlands (n = 157)
Germany (n = 143)
Hungary (n = 117)
102:162 29.7 ± 11.3 10.4 ± 7.9 years 26 (10.2) 92 (34.8)
64:83 25 ± 11.8 10 ± 7.5 years 10 (6.8) 62 (42.1)
57:100 23 ± 9.5 20.0 ± 8.9 years 9 (15.8) n.d.
64:79 35.3 ± 14.1 10.8 ± 10.6 years 7 (5) 15 (10.8)
44:73 31 ± 10.4 13.5 ± 8.3 years 4 (3.4) 12 (10.3)
66 (21.3) 63 (20.3) 170 (54.8) 11 (3.5) 84 (29.7) -
22 (15.0) 61 (41.5) 56 (38.1) 8 (5.4) 41 (27.9) -
40 (26.5) 31 (20.5) 78 (51.7) 2 (1.3) 26 (40.6) -
5 (3.7) 52 (38.5) 62 (45.9)
15 (12.8) 67 (57.3) 35 (29.9)
64 (25.1) 78 (30.6) 113 (44.3)
48 (34.0) 39 (27.7) 54 (38.3)
32 (20.8) 35 (22.7) 87 (56.5)
-
-
Absolute numbers (percentages) are shown. Abbreviation: IBD: inflammatory bowel disease; n.d.: not detected; SD: standard deviation.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1194
M. Prager et al.
The detection of the mutant alleles was carried out by melting curve analysis with fluorescence resonance energy transfer (FRET) probes or SimpleProbes in the LightCycler (Roche Diagnostics). For detection of rs1545620, the sequence of the sensor fluoresceinelabeled probe was 5¢-CCCCTCCATGCGGCCTGFL and that of the anchor LC Red 640 labeled probe was 5¢-LC640-TGCCGGTAGAGCTTCCGCTG CCAGTA-ph (ph_ = phosphate). For identification of rs1457092 we used the probes 5¢-AGAAGCCCCCAGGAGGATATCAGC-FL and 5¢-LC610-GCT CCCGTCCCAGCACCCACAG-ph, for rs2305764 5¢-CCGTATATGCCGGGGGAA-FL and 5¢-LC 610-AACACACTCACACGTG-ph and for rs2279 002 5¢-TCGCTGTGACCAGCCCAG-FL and 5¢LC640-AGGCAGGACCACCCAGTTCACAAACph, for rs2279003 we used GCTXIGCCAGGGG CTGGGCTGAC—ph. All FRET and SimpleProbes were designed and synthesized by TIB MOLBIOL, Berlin, Germany. In each LightCycler run, patients as well as control samples were analyzed. Control subjects were tested for Hardy-Weinberg equilibrium. Sequencing MYO9B. Sequencing was performed in n = 54 CD patients. All CD patients had an increased permeability index (PI). All relevant regions (exons and exon/intron boundaries of exon 12 to 35) of the candidate gene were amplified by specific designed primers according to the published sequences (GenBank) using the free software ExonPrimer (http://ihg. gsf.de/ihg/ExonPrimer.html). DNA sequencing of the amplified regions were performed by cycle sequencing with fluorescent dye terminators, followed by analysis using an ABI 310 automatic sequencer (Applied Biosystems). Analysis was confirmed by sequencing in both directions. Measurement of intestinal permeability. Intestinal permeability was assessed using an oral sugar test, as previously described in detail [18]. Of note, a subgroup of individuals were recruited in previous studies [1,2]. The sugar drink test is based on measurement of urinary excretion of an orally administered non-metabolized carbohydrate. We used the lactulose/mannitol ratio PI as a marker for intestinal permeability. The upper limit of normal intestinal permeability was defined as mean value + 2SD of the control group (= 0.03), as previously described [18,19]. In total, n = 122 CD patients were included. Some of these patients were recruited for previous familybased studies [1,2]. All CD patients were in remission at time of testing (CDAI < 150). The test was carried
out at home by all persons. Predetermined exclusion criteria for all subjects were any (for CD patients: concomitant) gastrointestinal or hepatobiliary disease, severe neurological, endocrine, cardiovascular, pulmonary, or renal disease, gastrectomy, colectomy, or extensive resection of the small bowel, cancer, rheumatoid based diseases, acute infections, acute urticaria, or pregnancy. Alcohol and nonsteroidal anti-inflammatory drugs were forbidden for at least 48 h before the test.
Statistical analysis The primary end point of the study was to test for an association between intestinal permeability (defined as increased vs non-increased intestinal permeability) to n = 5 MYO9B variants. Analysis was performed by c2-test (two-sided). p-Values less than 0.01 were considered significant after Bonferronni correction for multiple testing (p = 0.05/5 = 0.01). We analyzed the following secondary end points: .
.
.
Absolute values of intestinal permeability testing were compared depending on the MYO9B genotype using the non-parametric Wilcoxon-UMann-Whitney test. Genotypes of the MYO9B variants were compared between different categories (IBD, CD and UC, and healthy controls). For the overall and the individual cohorts, allele frequencies were compared by Fisher’s exact test. Genotypes of the MYO9B were compared between different categories in CD and UC patients to different clinical characteristics (classified according to the presence/absence of e.g. fistulizing disease behavior; see above section “Clinical characteristics of IBD patients”) using a c2-test (two-sided).
Odds ratio and 95% confidence intervals were estimated using allele frequencies in 2x2 contingency tables. For all secondary end points, p values less than 0.05 were considered to be significant. The data were analyzed using the GAUSS Mathematical and Statistical System, Version 8.0.0 (Aptech Systems, Inc, Black Diamond, USA) and SPSS software (Version 19; SPSS, Chicago, USA).
Ethical considerations The study was approved by the ethics commission of the Charite, Universitätsmedizin Berlin (EA1/133/ 09). Informed consent was obtained from each participant.
Myosin XIb variants and intestinal barrier
observed significantly higher values for individuals carrying one or two risk alleles with respect to all five analyzed MYO9B variants (Figure 2).
Results Effect of MYO9B variants on intestinal permeability
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1195
Clinical characteristics of patients are shown in Table I. In the first step, we investigated whether intestinal permeability depends on the MYO9B genotype. Increased intestinal permeability was found in 57/122 (46.7%) of German CD patients. Correlation to the MYO9B genotype yielded the following results for the four variants: rs1545620, p = 0.010 OR [95%CI:1.17–3.39]; rs1457092, p = 0.024 [95%CI:1.02–3.05]; rs2305764, p = 0.015 [95%CI:1.14–3.17]; rs2279002, p = 0.051 OR 1.50 [95%CI: 0.85–2.66] (Table II and Figure 2). By sequencing, we identified a further mutation in exon 21 of MYO9B which leads to a synonymous codon substitution P1265P (rs2279003). This variant was also associated with altered intestinal permeability (p = 0.003, OR 95% CI:0.226–0.678; significant after Bonferroni correction), but herein individuals carrying the minor allele had lower values for intestinal permeability suggesting that MYO9B might also harbor variants that protect from developing epithelial barrier dysfunction. Based on these data, we defined a risk genotype for each individual MYO9B variant (depending on the presence of at least one risk allele; Figure 2). In a second step, we compared absolute values of permeability testing according to this stratification. We
Distribution of MYO9B variants in IBD patients and controls We analyzed the five MYO9B variants in IBD patients and control subjects from three different European populations. Genotypes are shown in Table III. Genotypes of all three control populations were in Hardy-Weinberg equilibrium. We observed an association of the intron 28 variant (rs2305764) to CD in the Dutch cohort (p = 0.004; OR: 1.5569, 95%CI: 1.15–2.11), but not in the German (p = 0.24; OR: 1.16, 95%CI: 0.91–1.49) or Hungarian (p = 0.31; OR: 0.8462, 95%CI: 0.62–1.15) cohort. We did not find an association of the other four variants, rs1545620, rs1457092, rs2279003 and rs2279002 (Table III) in any of the three populations studied, nor in the overall combined cohort. Effect of MYO9B variants on the clinical course of IBD We then used the clinical and demographical data of the IBD patients to perform detailed genotype phenotype analysis. These analyses, however, did not reveal a significant association with respect to all investigated clinical characteristics (data not shown).
Table II. Intestinal permeability measurement depending on MYO9B genotype in CD patients. MYO9B Genotype AA rs1545620 (exon 20) PI: 0.03 8 GG rs1457092 (intron 20) PI: 0.03 11 CC rs2279003 (exon 21) PI: 0.03 31 rs2305764 (intron 28) PI: 0.03 9 rs2279002 (intron 32) PI: 0.03 19
%
AC
%
CC
%
n
p(c2)
OR
[95%CI]
40.9 15.1
26 32
42.6 60.4
10 13
16.4 24.5
61 53
0.010
1.99
[1.17–3.39]
2
%
GT
%
TT
%
n
p(c )
OR
[95%CI]
47.6 22.9
24 29
38.1 64.4
9 8
14.3 16.7
63 48
0.024
1.76
[1.02–3.05]
2
%
CT
%
TT
%
n
p(c )
OR
[95%CI]
32.3 54.4
29 24
44.6 42.1
15 2
23.1 3.5
65 57
0.003
0.39
[0.226–0.678]
38.46 15.79
31 35
47.69 61.40
9 13
13.85 22.81
65 57
0.015
1.90
[1.14–3.17]
58.7 38.8
18 25
28.6 51.0
8 5
12.7 10.2
63 49
0.051
1.50
[0.85–2.66]
Genotypes of the five MYO9B variants (rs1545620, rs1457092, rs2279003, rs2305764, rs2279002) according to normal/increased intestinal permeability (Permeability Index, PI; for definition see section “materials and methods”) in German CD patients (Chi2-test). Shown are absolute numbers. Abbreviations: CD: Crohn’s disease; PI: permeability index.
1196
M. Prager et al. B: rs1457092 (intron 20)
0.14
0.14
0.12
0.12 Permeability Lac/Man ratio
0.10 p = 0.015
0.08 0.06 0.04 0.02
0.10 p = 0.023
0.08 0.06 0.04 0.02
0.00
0.00 AA
AC CC
GG
0.14
0.12
0.12
0.10
0.10
Permeability Lac/Man ratio
0.14
p = 0.022
0.08
GT TT
D: rs2305764 (intron 28)
C: rs2279003 (intron 21)
Permeability Lac/Man ratio
0.06 0.04
p = 0.011
0.08 0.06 0.04 0.02
0.02
0.00
0.00 CC CT
TT
CC
CT TT
E: rs2279002 (intron 32) 0.14 0.12 Permeability Lac/Man ratio
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Permeability Lac/Man ratio
A: rs1545620 (exon 20)
0.10 p = 0.024
0.08 0.06 0.04 0.02 0.00 CC
CT TT
Figure 2A-E. Boxplot analysis of absolute values of intestinal permeability in CD patients according to MYO9B risk genotype (added is a solid line that markes the cut-off value for increased intestinal permeability, PI > 0.03). CD: Crohn’s disease; PI: permeability index.
Discussion In this study, we observed that numerous exon and intron MYO9B variants are associated with altered intestinal permeability in CD patients despite clinical remission. We therefore provide functional data that suggest a crucial involvement of MYO9B in epithelial barrier integrity.
Among the analyzed MYO9B variants, of special interest might be the rs1545620 variant within exon 20, which leads to an amino acid exchange in codon 1011 (p.A1011S). This position is involved in the binding of calmodulin [10,20]. Calmodulin regulates the motor activity of MYO9B on actin filaments and might therefore influence the velocity of MYO9B. We also found an association for rs2305764 within intron
Myosin XIb variants and intestinal barrier
1197
Table III. MYO9B variants in IBD patients.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
rs1545620 (exon 20)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
AA
AC
CC
MAF (%)
OR
95% CI
179 87 266 201
253 118 371 321
91 45 136 99
41.6 41.6 41.6 41.8
0.99 0.99 0.99
[0.84–1.17] [0.81–1.22] [0.85–1.15]
0.93 0.96 0.94
69 48 117 75
119 65 184 118
39 29 68 26
43.4 43.3 43.4 38.8
1.20 1.20 1.20
[0.93–1.58] [0.89–1.63] [0.95–1.54]
0.174 0.250 0.142
52 39 91 64
68 53 121 95
26 16 42 31
41.1 39.3 40.4 41.3
0.99 0.92 0.96
[0.73–1.35] [0.66–1.30] [0.73–1.26]
1.00 0.66 0.78
58 62
66 108
26 42
39.3 45.2
0.78
[0.58–1.06]
0.13
p-Value#
rs1457092 (intron 20)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
GG
GT
TT
MAF (%)
OR
95% CI
223 109 332 290
228 106 334 316
74 33 107 83
35.8 34.7 35.5 35.0
1.04 0.99 1.02
[0.88–1.23] [0.80–1.22] [0.88–1.19]
0.70 0.91 0.82
88 57 145 114
108 62 170 148
28 19 47 22
36.6 36.2 36.4 33.8
1.13 1.11 1.12
[0.87–1.47] [0.82–1.50] [0.89–1.41]
0.35 0.49 0.35
67 52 119 73
55 44 99 96
25 14 39 26
35.7 32.7 34.4 37.9
0.91 0.80 0.86
[0.66–1.24] [0.56–1.13] [0.65–1.13]
0.58 0.22 0.29
68 103
65 72
21 35
34.7 33.8
1.04
[0.76–1.42]
0.81
p-Value#
rs2279003 (exon 21)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls
CC
CT
TT
MAF (%)
OR
95% CI
204 81 285 234
251 136 387 278
92 36 128 100
39.8 41.1 40.2 39.1
1.03 1.09 1.05
[0.87–1.21] [0.88–1.35] [0.90–1.22]
0.73 0.45 0.56
88 40 128 105
113 68 181 121
36 19 55 46
39.0 41.7 40.0 39.2
1.28 1.51 1.30
[0.77–1.28] [0.82–1.51] [0.83–1.30]
1.00 0.54 0.77
58 41 99 44
65 68 133 59
30 17 47 17
40.8 40.5 40.7 38.8
1.09 1.07 1.08
[0.77–1.54] [0.75–1.54] [0.79–1.48]
0.66 0.71 0.64
p-Value
1198
M. Prager et al.
Table III. (Continued). rs2279003 (exon 21)
The Netherlands CD Controls
CC
CT
TT
58 85
73 98
26 37
MAF (%)
OR
95% CI
39.8 39.1
1.03
[0.77–1.39]
p-Value 0.88
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
rs2305764 (intron28)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
CC
CT
TT
MAF (%)
OR
95% CI
207 96 303 282
266 124 390 317
100 42 143 101
40.7 39.7 40.4 37.1
1.16 1.11 1.15
[0.99–1.36] [0.91–1.37] [0.99–1.33]
0.066 0.29 0.063
89 52 141 100
137 76 213 158
42 23 65 29
41.2 40.4 40.9 37.6
1.16 1.12 1.15
[0.91–1.49] [0.85–1.49] [0.92–1.42]
0.24 0.42 0.22
62 44 106 64
60 48 108 96
27 19 46 34
38.3 38.7 38.5 42.3
0.85 0.86 0.85
[0.62–1.15] [0.62–1.21] [0.65–1.12]
0.31 0.44 0.27
56 118
69 63
31 38
42.0 31.7
1.56
[1.15–2.11]
0.004
p-Value
rs2279002 (intron32)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
CC
CT
TT
MAF (%)
OR
95% CI
251 129 380 326
208 88 296 291
58 27 85 53
31.3 29.1 30.6 29.6
1.08 0.97 1.05
[0.91–1.29] [0.78–1.22] [0.89–1.23]
0.39 0.86 0.57
110 74 184 129
94 48 142 126
21 16 37 12
30.2 29.0 29.8 28.1
1.11 1.05 1.08
[0.84–1.46] [0.76–1.44] [0.85–1.39]
0.48 0.81 0.53
68 55 123 95
52 40 92 82
21 11 32 17
33.3 29.2 31.6 29.9
1.17 0.97 1.08
[0.84–1.63] [0.67–1.40] [0.81–1.44]
0.35 0.93 0.61
73 102
62 83
16 24
31.1 31.3
0.99
[0.72–1.36]
1.00
p-Value#
Genotypes of the five MYO9B variants (rs1545620, rs1457092, rs2279003, rs2305764 and rs2279002) in IBD patients from Germany, Hungary and the Netherlands. Absolute numbers are illustrated. Genotypes were compared using the Fisher’s exact test (#). Odds ratio and confidence intervals were estimated using allele frequencies in 2x2 contingency tables. CD versus controls, CU versus controls, IBD versus controls. Abbreviations: CD: Crohn’s disease; IBD: inflammatory bowel disease; MAF: minor allele frequency; UC: ulcerative colitis.
28. Interestingly, this variant was found to completely explain the association of MYO9B to celiac disease [12], but a direct link to functional data of an intestinal barrier defect has not been provided yet. Interestingly, an association for the rs2305764 variant
to both Barrett’s esophagus (BE) and esophageal adenocarcinoma (EAC) has been described suggesting that MYO9B may play a role in the etiology of BE and EAC by increasing the permeability of the epithelial barrier [21].
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Myosin XIb variants and intestinal barrier The precise mechanisms how MYO9B variants regulate the epithelial barrier remain to be elucidated. It is suggested that a genetic variant in the 3¢ part of MYO9B leads to an impaired interaction with RhoA, thereby perturbing tight junction gate and fence function in epithelial cells [12]. A recent study showed that loss of MYO9B function resulted in a nearly complete disruption of tight junction protein localization in an epithelial cell line. Herein, the tight junction complex proteins were mislocalazied and the intestinal barrier was compromised, permitting TRTC-dextran to readily pass through a monolayer in cell culture where MYO9B function was altered [22]. Having this in mind, there could be a combinatorial effect of the submucosal immune system with unregulated immune cell migration [23], and a disregulated intestinal permeability due to loss of MYO9B function that results in inflammation. We furthermore tested whether MYO9B variants resemble susceptibility factors for CD or lead to a distinct phenotype. We found an association of the intron 28 variant (rs2305764) to CD in the cohort from the Netherlands (p = 0.004), which could not be replicated in the German and Hungarian cohort. Van Bodegraven and colleagues [14] observed a significant association of MYO9B SNPs with IBD in three independent cohorts of IBD patients from UK, the Netherlands and Canada/Italy, with a stronger association in a UC subgroup. A similar association with UC was replicated in a Spanish cohort by Nunez et al. [24]. In contrast, Wolters et al. [25] analyzed Canadian patients and reported an association of the SNP rs1457092, rs1545620 and rs2305764 with ileal-only CD, although no overall association to CD or UC could be shown. Beyond that they observed an association in the opposite direction with inverse allelic association compared with the four European IBD studies [14,24,26,27]. We did not observe an association to any of the studied phenotypes. However, these conflicting data do not allow the conclusion that MYO9B variants resemble susceptibility variants for either CD or UC. These discrepancies reflect a typical problem in IBD genetic studies. This might be due to either different ethnic backgrounds, or collection of patients from either populations-based cohorts or tertiary referral centers, or small samples sizes. Our study has several limitations. At first, the number of patients with data on intestinal permeability in CD is limited (n = 122). We choose strict inclusion criteria for analysis of intestinal permeability including disease activity and concurrent medical treatment, for example by measuring patients in clinical remission, in order to detect changes in epithelial barrier function independent of systemic inflammation. All these factors put limitations on patient
1199
recruitment. However, IBD centers with large numbers of patients with both DNA and data on in vivo analysis of intestinal permeability are very limited. In this respect, it is of interest that an Italian group found a trend towards an abnormal intestinal permeability in Italian CD carrying the rs1545620 risk allele supporting our data [27]. Second, we did not study a cohort of healthy individuals to detect an association between MYO9B and altered intestinal permeability. Due to the fact that an increased intestinal permeability was not found in 96 healthy controls in our previous study [1], showing that an altered intestinal permeability is a very rare event in healthy individuals, it would require a very large cohort to obtain robust statistical data. Third, we observed an overall association of the rs2305764 variants only in the Dutch cohort and not in the German and Hungarian CD groups. This might be due to phenotypic differences among the three groups. The presence of perianal and fistulizing disease was remarkably higher in the Dutch cohort compared to the other two groups. In agreement with our findings, Bodegraven et al. observed the most significant association to MYO9B variants in the Dutch cohort. In summary, by using CD as a model, we provide for the first time a direct link between in vivo analysis of intestinal permeability to different and independent variants in a gene crucially involved in the maintenance of the epithelial barrier, MYO9B.
Acknowledgement The study was supported by a grant from the Eli and Edythe Broad Foundation (C.B.), by a grant of the German Ministry for Education and Research (BMBF) - Competence Network “Inflammatory Bowel Disease”: 01GI0284 TP 1.17 and by the Forschungsförderung of the Charité, Universitätsmedizin Berlin. We thank all patients who participated in this evaluation. The work is part of the doctoral thesis of Tahir Durmus.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Buhner S, Buning C, Genschel J, Kling K, Herrmann D, Dignass A, et al. Genetic basis for increased intestinal permeability in families with Crohn’s disease: role of CARD15 3020insC mutation? Gut 2006;55:342–7.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1200
M. Prager et al.
[2] Prager M, Buttner J, Haas V, Baumgart DC, Sturm A, Zeitz M, et al. The JAK2 variant rs10758669 in Crohn’s disease: altering the intestinal barrier as one mechanism of action. Int J Colorectal Dis 2012;27:565–73. [3] De Gottardi A, McCoy KD. Evaluation of the gut barrier to intestinal bacteria in non-alcoholic fatty liver disease. J Hepatol 2011;55:1181–3. [4] Vaarala O. Leaking gut in type 1 diabetes. Curr Opin Gastroenterol 2008;24:701–6. [5] Soderholm JD, Olaison G, Lindberg E, Hannestad U, Vindels A, Tysk C, et al. Different intestinal permeability patterns in relatives and spouses of patients with Crohn’s disease: an inherited defect in mucosal defence? Gut 1999; 44:96–100. [6] John LJ, Fromm M, Schulzke JD. Epithelial barriers in intestinal inflammation. Antioxid redox signal 2011;15: 1255–70. [7] Bruewer M, Hopkins AM, Hobert ME, Nusrat A, Madara JL. RhoA, Rac1, and Cdc42 exert distinct effects on epithelial barrier via selective structural and biochemical modulation of junctional proteins and F-actin. Am J Physiol Cell Physiol 2004;287:C327–35. [8] Matter K, Balda MS. Signalling to and from tight junctions. Nat rev Mol Cell Biol 2003;4:225–36. [9] Muller RT, Honnert U, Reinhard J, Bahler M. The rat myosin myr 5 is a GTPase-activating protein for Rho in vivo: essential role of arginine 1695. Mol biol cell 1997;8: 2039–53. [10] Post PL, Tyska MJ, O’Connell CB, Johung K, Hayward A, Mooseker MS. Myosin-IXb is a single-headed and processive motor. J biol chem 2002;11679–83. [11] Wirth JA, Jensen KA, Post PL, Bement WM, Mooseker MS. Human myosin-IXb, an unconventional myosin with a chimerin-like rho/rac GTPase-activating protein domain in its tail. J Cell Sci 1996;109:653–61. [12] Monsuur AJ, de Bakker PI, Alizadeh BZ, Zhernakova A, Bevova MR, Strengman E, et al. Myosin IXB variant increases the risk of celiac disease and points toward a primary intestinal barrier defect. Nat Genet 2005;37:1341–4. [13] Irvine EJ, Marshall JK. Increased intestinal permeability precedes the onset of Crohn’s disease in a subject with familial risk. Gastroenterology 2000;119:1740–4. [14] van Bodegraven AA, Curley CR, Hunt KA, Monsuur AJ, Linskens RK, Onnie CM, et al. Genetic variation in myosin IXB is associated with ulcerative colitis. Gastroenterology 2006;131:1768–74. [15] Dignass A, Eliakim R, Magro F, Maaser C, Chowers Y, Geboes K, et al. Second European evidence-based consensus
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
on the diagnosis and management of ulcerative colitis part 1: definitions and diagnosis. J Crohns colitis 2012;6:965–90. Van Assche G, Dignass A, Panes J, Beaugerie L, Karagiannis J, Allez M, et al. The second European evidence-based Consensus on the diagnosis and management of Crohn’s disease: Definitions and diagnosis. J Crohns colitis 2010;4:7–27. Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005;19:Suppl A):5A–36A. Buhner S, Reese I, Kuehl F, Lochs H, Zuberbier T. Pseudoallergic reactions in chronic urticaria are associated with altered gastroduodenal permeability. Allergy 2004;59: 1118–23. Wyatt J, Vogelsang H, Hubl W, Waldhoer T, Lochs H. Intestinal permeability and the prediction of relapse in Crohn’s disease. Lancet 1993;341:1437–9. Post PL, Bokoch GM, Mooseker MS. Human myosin-IXb is a mechanochemically active motor and a GAP for rho. J Cell sci 1998;111:941–50. Menke V, Van Zoest KP, Moons LM, Pot RG, Siersema PD, Kuipers EJ, et al. Myo9B is associated with an increased risk of Barrett’s esophagus and esophageal adenocarcinoma. Scand J Gastroenterol 2012;47:1422–8. Chandhoke SK, Mooseker MS. A role for myosin IXb, a motor-RhoGAP chimera, in epithelial wound healing and tight junction regulation. Mol biol cell 2012;23:2468–80. Hanley PJ, Xu Y, Kronlage M, Grobe K, Schon P, Song J, et al. Motorized RhoGAP myosin IXb (Myo9b) controls cell shape and motility. Proc Natl Acad Sci USA 2010;107: 12145–50. Nunez C, Oliver J, Mendoza JL, Gomez-Garcia M, Pinero A, Taxonera C, et al. MYO9B polymorphisms in patients with inflammatory bowel disease. Gut 2007;56:1321–2. Wolters VM, Xu W, Zhao X, Walters TD, Griffiths AM, Silverberg MS, et al. Replication of genetic variation in the MYO9B gene in Crohn’s disease. Hum immunol 2011;72: 592–7. Cooney R, Cummings JR, Pathan S, Beckly J, Geremia A, Hancock L, et al. Association between genetic variants in myosin IXB and Crohn’s disease. Inflamm Bowel Dis 2009; 15:1014–21. Latiano A, Palmieri O, Valvano MR, D’Inca R, Caprilli R, Cucchiara S, et al. The association of MYO9B gene in Italian patients with inflammatory bowel diseases. Aliment pharmacol ther 2008;27:241–8.
ORIGINAL ARTICLE
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Myosin IXb variants and their pivotal role in maintaining the intestinal barrier: A study in Crohn’s disease
MATTHIAS PRAGER1*, TAHIR DURMUS2*, JANINE BÜTTNER1, TAMAS MOLNAR3, DIRK J. DE JONG4, JOOST PH. DRENTH4, DANIEL C. BAUMGART5, ANDREAS STURM6, KLAUDIA FARKAS3, HEIKO WITT7,8* & CARSTEN BÜNING1* 1
Department of Gastroenterology and Hepatology, Charité, Campus Mitte, Universitätsmedizin Berlin, Berlin, Germany, Department of Radiology, Charité, Campus Mitte, Universitätsmedizin Berlin, Berlin, Germany, 31st Department of Medicine, Faculty of Medicine, University of Szeged, Szeged, Hungary, 4Department of Gastroenterology and Hepatology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands, 5Department of Gastroenterology and Hepatology, Charité, Campus Virchow-Klinikum, Universitätsmedizin Berlin, Berlin, Germany, 6Department of Internal Medicine, Krankenhaus Waldfriede, Berlin, Germany, 7Else Kröner-Fresenius-Zentrum für Ernährungsmedizin (EKFZ) & Zentralinstitut für Ernährungs- und Lebensmittelforschung (ZIEL), Technische Universität München (TUM), Gregor-Mendel-Str. 2, 85350 Freising, Germany, and 8Department of Pediatrics, Klinikum rechts der Isar (MRI), Technische Universität München (TUM), Parzivalstr. 16, 80804 Munich, Germany
2
Abstract Background. Myosin IXb (MYO9B) is involved in the regulation of epithelial barrier function. We hypothesized that MYO9B variants are associated with increased intestinal permeability measured in patients with Crohn’s disease (CD), where barrier dysfunction is crucially involved in disease development. Methods. We sequenced MYO9B and genotyped five MYO9B variants (rs1545620, rs1457092, rs2279003, rs2305764 and rs2279002) and correlated these data to measurement of intestinal permeability in German CD patients (n = 122) obtained by standard oral sugar test using the lactulose/mannitol ratio after measurement of urinary excretion. We furthermore studied MYO9B variants in three European cohorts with inflammatory bowel disease (IBD) and healthy controls : Germany (CD = 264; ulcerative colitis = 143 [UC]; HC = 372); Hungary (CD = 147; UC = 117; HC = 195), the Netherlands (CD = 157; HC = 219). Results. We found an association for four studied MYO9B variants to an increased intestinal permeability in CD patients (rs1545620, p = 0.010; rs1457092, p = 0.024; rs2279003, p = 0.003; rs2305764, p = 0.015). Furthermore, we observed significantly higher absolute values of intestinal permeability for individuals carrying risk alleles within MYO9B. Looking for an overall disease association, only the rs2305764 variant was associated with CD in the Dutch cohort (p = 0.004), but not in the German or Hungarian cohort. No association to UC or a distinct phenotype in both CD and UC patients was observed for all studied MYO9B variants. Conclusion. Our data suggest a link between MYO9B variants to an increased intestinal permeability in CD patients. This supports the influence of Myosin IXb on the integrity of the epithelial barrier. The role of MYO9B variants in the overall susceptibility to IBD, however, remains to be elucidated.
Key Words: Crohn’s disease, inflammatory bowel disease, intestinal permeability, myosin IXb, ulcerative colitis
Introduction A disturbed intestinal epithelial barrier appears to be critical in development and course of numerous
gastrointestinal and primary non-gastrointestinal diseases such as inflammatory bowel diseases (IBD) [1,2] celiac disease, non-alcoholic-steatohepatitis and diabetes mellitus [3,4]. Although the concept is
Correspondence: PD Dr. med. Carsten Büning, Department of Hepatology and Gastroenterology, Charité, Universitätsmedizin Berlin, Campus Mitte, Charitéplatz 1, 10117 Berlin, Germany. Tel: + 49 30 450614237. Fax: + 49 30 450514906. E-mail: [email protected] *M Prager, T Durmus, H Witt and C Büning contributed equally.
(Received 3 April 2014; revised 19 May 2014; accepted 19 May 2014) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2014.928903
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1192
M. Prager et al.
accepted, the underlying mechanisms remain to be elucidated for all diseases. It is uncertain whether intestinal epithelial barrier dysfunction results from genetic variation or reflects secondary changes due to local or systemic inflammation. Evidence for a genetic predisposition comes from family-based studies especially for Crohn’s disease, where clustering of increased intestinal permeability is predominantly seen in first-degree relatives compared to spouses [1,5]. The intestinal epithelial barrier has a gatekeeping role, which protects the internal milieu from unwanted pathogenic luminal contents. Tight junctions appear in the closest possible cell contact sealing the paracellular space and forming the rate-limiting step in determining epithelial integrity (Figure 1) [6]. However, regulation of epithelial barrier is complex. Myosin IXb (MYO9B) is expressed in intestinal epithelial cells and encodes a protein involved in actin remodeling in epithelial enterocytes and tight junction assembly [7–11]. Experimental studies revealed that overexpression of rat MYO9B leads to actin filamentrelated morphologic changes in epithelial cells [9]. Interestingly, MYO9B variants have been associated with celiac disease [12]. Since passing of
Paracellular transport
immunogenic gluten peptides through the epithelial barrier is characteristic for celiac disease, these observations point towards a primary epithelial defect. However, up to now there are no supportive data that suggest that MYO9B variants are associated with variation in intestinal permeability. To answer this question, we choose Crohns disease (CD) as a disease model because this reflects a disorder with inherent abnormal intestinal permeability [13]. In the first step, we investigated CD patients for changes in intestinal permeability in vivo by standard oral sugar test and correlated these data to four known MYO9B variants [14], among them three intron variants. Since we herein observed a significant association to an increased intestinal permeability, we decided in a second step to sequence all coding regions, exon/intron boundaries and promoter regions of MYO9B to look for other functional variants. Sequencing was performed in CD patients with altered intestinal permeability to detect possible additional genetic changes that could have an impact on intestinal integrity as well. In the last step, all MYO9B variants were studied in large CD cohorts from different European populations to look for an
Apical
Tight junctions
Myosin 9B Head Z0 1,2,3
Claudins Neck
Occludin Actomyosin cytoskeleton JAM1
Tail
Adherens junctions
Desmosomes F-actin
Transcellular transport
Basal
Figure 1. Schematic representation of Myosin IXB within the epithelial barrier. The paracellular transport of molecules through epithelial barrier is regulated by the tight junctions complex, adherens junctions and desmosomes. The tight junction complex is connected with the actomyosin cytoskeleton.
Myosin XIb variants and intestinal barrier
1193
smoking, or never smoked), disease localization and behavior. The following data of UC patients were obtained: age at diagnosis, gender, familial or spontaneous disease, smoking habits and disease localization (Table I). Stratification was performed according to the Montreal classification [17].
overall susceptibility. Data were compared to healthy controls and also to patients with ulcerative colitis, another chronic inflammatory bowel disease. Materials and methods
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Subjects Genotyping of the MYO9B variants. Genomic DNA was extracted from peripheral blood leukocytes. As primers we used the oligonucleotides 5¢-GCTGCC GTGTACCACCA-3¢ and 5¢CTTAGTGCCACAG GCCAGA-3¢ for rs1545620 (exon20), 5¢-GGCAT CCACCAGGCACA-3¢ and 5¢-GGCAGGGGAGA GGTGAACA-3¢ for rs1457092 (intron 20), 5¢-GC CTGGCCTTGGACAGCA-3¢ and 5¢-ACTGGCT GAGCATGGCACTA-3¢ (intron 21), 5¢-GCCA ACCTCCTCCTTGTTC-3¢ and 5¢-AGGGGTGA CTTATTCAAGACT-ACTGC-3¢ for rs2305764 (intron28) and 5¢-GGGGTGCTGAAGCAGTG-3¢ and 5¢-CCACCAACACGACCTGAGAT-3¢ for rs2279002 (intron32). We performed PCR using 0.75 U AmpliTaq Gold (Applied Biosystems), 400 mM dNTPs, 1.5 mM MgCl2 and 0.1 mM of each primer in a final volume of 25 ml. The reaction mix was denatured at 95 C for 12 min followed by 48 cycles of denaturation at 95 C for 20 s, annealing at 60 C for 40 s, elongation at 72 C for 90 s and a final extension step for 2 min at 72 C in an automated thermocycler.
IBD patients. We included CD or ulcerative colitis (UC) patients diagnosed upon clinical, endoscopic, radiological and histological findings [15,16]. The study was approved by the responsible ethic committees of all three institutions and informed consent was obtained from each participant. In total, 407 white IBD patients (264 with CD and 143 with UC) were recruited from Germany. In addition, we enrolled CD 147 patients and 117 UC patients from Szeged, Hungary and 157 CD patients from Nijmegen, the Netherlands. Unrelated and healthy subjects from Germany (n = 372), Hungary (n = 195) and the Netherlands (n = 219) served as controls for the genetic case-control analysis. Clinical data of IBD patients were obtained through retrospective collection from the patients clinical charts prior to genotyping. The following data of CD patients were obtained: age, age at diagnosis, gender, familial or spontaneous disease (familial disease was considered when one first- or second-degree relative was diagnosed with IBD), smoking habits (current smoking, or history of
Table I. Demographical and clinical characteristics of CD and UC patients from Germany, Hungary and the Netherlands. CD Population Gender (male:female) Age at diagnosis (median, SD) Disease duration (median, SD) Familial IBD Smoking Localization (Montreal) L1: Ileum L2: Colon L3: Ileocolon L4: Upper GI Perianal disease L1: Proctitis L2: Left-sided colits L3: Pancolits Behavior (Montreal) B1: Non-stricturing non-penetrating B2: Stricturing B3: Penetrating
UC
Germany (n = 264)
Hungary (n = 147)
Netherlands (n = 157)
Germany (n = 143)
Hungary (n = 117)
102:162 29.7 ± 11.3 10.4 ± 7.9 years 26 (10.2) 92 (34.8)
64:83 25 ± 11.8 10 ± 7.5 years 10 (6.8) 62 (42.1)
57:100 23 ± 9.5 20.0 ± 8.9 years 9 (15.8) n.d.
64:79 35.3 ± 14.1 10.8 ± 10.6 years 7 (5) 15 (10.8)
44:73 31 ± 10.4 13.5 ± 8.3 years 4 (3.4) 12 (10.3)
66 (21.3) 63 (20.3) 170 (54.8) 11 (3.5) 84 (29.7) -
22 (15.0) 61 (41.5) 56 (38.1) 8 (5.4) 41 (27.9) -
40 (26.5) 31 (20.5) 78 (51.7) 2 (1.3) 26 (40.6) -
5 (3.7) 52 (38.5) 62 (45.9)
15 (12.8) 67 (57.3) 35 (29.9)
64 (25.1) 78 (30.6) 113 (44.3)
48 (34.0) 39 (27.7) 54 (38.3)
32 (20.8) 35 (22.7) 87 (56.5)
-
-
Absolute numbers (percentages) are shown. Abbreviation: IBD: inflammatory bowel disease; n.d.: not detected; SD: standard deviation.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1194
M. Prager et al.
The detection of the mutant alleles was carried out by melting curve analysis with fluorescence resonance energy transfer (FRET) probes or SimpleProbes in the LightCycler (Roche Diagnostics). For detection of rs1545620, the sequence of the sensor fluoresceinelabeled probe was 5¢-CCCCTCCATGCGGCCTGFL and that of the anchor LC Red 640 labeled probe was 5¢-LC640-TGCCGGTAGAGCTTCCGCTG CCAGTA-ph (ph_ = phosphate). For identification of rs1457092 we used the probes 5¢-AGAAGCCCCCAGGAGGATATCAGC-FL and 5¢-LC610-GCT CCCGTCCCAGCACCCACAG-ph, for rs2305764 5¢-CCGTATATGCCGGGGGAA-FL and 5¢-LC 610-AACACACTCACACGTG-ph and for rs2279 002 5¢-TCGCTGTGACCAGCCCAG-FL and 5¢LC640-AGGCAGGACCACCCAGTTCACAAACph, for rs2279003 we used GCTXIGCCAGGGG CTGGGCTGAC—ph. All FRET and SimpleProbes were designed and synthesized by TIB MOLBIOL, Berlin, Germany. In each LightCycler run, patients as well as control samples were analyzed. Control subjects were tested for Hardy-Weinberg equilibrium. Sequencing MYO9B. Sequencing was performed in n = 54 CD patients. All CD patients had an increased permeability index (PI). All relevant regions (exons and exon/intron boundaries of exon 12 to 35) of the candidate gene were amplified by specific designed primers according to the published sequences (GenBank) using the free software ExonPrimer (http://ihg. gsf.de/ihg/ExonPrimer.html). DNA sequencing of the amplified regions were performed by cycle sequencing with fluorescent dye terminators, followed by analysis using an ABI 310 automatic sequencer (Applied Biosystems). Analysis was confirmed by sequencing in both directions. Measurement of intestinal permeability. Intestinal permeability was assessed using an oral sugar test, as previously described in detail [18]. Of note, a subgroup of individuals were recruited in previous studies [1,2]. The sugar drink test is based on measurement of urinary excretion of an orally administered non-metabolized carbohydrate. We used the lactulose/mannitol ratio PI as a marker for intestinal permeability. The upper limit of normal intestinal permeability was defined as mean value + 2SD of the control group (= 0.03), as previously described [18,19]. In total, n = 122 CD patients were included. Some of these patients were recruited for previous familybased studies [1,2]. All CD patients were in remission at time of testing (CDAI < 150). The test was carried
out at home by all persons. Predetermined exclusion criteria for all subjects were any (for CD patients: concomitant) gastrointestinal or hepatobiliary disease, severe neurological, endocrine, cardiovascular, pulmonary, or renal disease, gastrectomy, colectomy, or extensive resection of the small bowel, cancer, rheumatoid based diseases, acute infections, acute urticaria, or pregnancy. Alcohol and nonsteroidal anti-inflammatory drugs were forbidden for at least 48 h before the test.
Statistical analysis The primary end point of the study was to test for an association between intestinal permeability (defined as increased vs non-increased intestinal permeability) to n = 5 MYO9B variants. Analysis was performed by c2-test (two-sided). p-Values less than 0.01 were considered significant after Bonferronni correction for multiple testing (p = 0.05/5 = 0.01). We analyzed the following secondary end points: .
.
.
Absolute values of intestinal permeability testing were compared depending on the MYO9B genotype using the non-parametric Wilcoxon-UMann-Whitney test. Genotypes of the MYO9B variants were compared between different categories (IBD, CD and UC, and healthy controls). For the overall and the individual cohorts, allele frequencies were compared by Fisher’s exact test. Genotypes of the MYO9B were compared between different categories in CD and UC patients to different clinical characteristics (classified according to the presence/absence of e.g. fistulizing disease behavior; see above section “Clinical characteristics of IBD patients”) using a c2-test (two-sided).
Odds ratio and 95% confidence intervals were estimated using allele frequencies in 2x2 contingency tables. For all secondary end points, p values less than 0.05 were considered to be significant. The data were analyzed using the GAUSS Mathematical and Statistical System, Version 8.0.0 (Aptech Systems, Inc, Black Diamond, USA) and SPSS software (Version 19; SPSS, Chicago, USA).
Ethical considerations The study was approved by the ethics commission of the Charite, Universitätsmedizin Berlin (EA1/133/ 09). Informed consent was obtained from each participant.
Myosin XIb variants and intestinal barrier
observed significantly higher values for individuals carrying one or two risk alleles with respect to all five analyzed MYO9B variants (Figure 2).
Results Effect of MYO9B variants on intestinal permeability
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1195
Clinical characteristics of patients are shown in Table I. In the first step, we investigated whether intestinal permeability depends on the MYO9B genotype. Increased intestinal permeability was found in 57/122 (46.7%) of German CD patients. Correlation to the MYO9B genotype yielded the following results for the four variants: rs1545620, p = 0.010 OR [95%CI:1.17–3.39]; rs1457092, p = 0.024 [95%CI:1.02–3.05]; rs2305764, p = 0.015 [95%CI:1.14–3.17]; rs2279002, p = 0.051 OR 1.50 [95%CI: 0.85–2.66] (Table II and Figure 2). By sequencing, we identified a further mutation in exon 21 of MYO9B which leads to a synonymous codon substitution P1265P (rs2279003). This variant was also associated with altered intestinal permeability (p = 0.003, OR 95% CI:0.226–0.678; significant after Bonferroni correction), but herein individuals carrying the minor allele had lower values for intestinal permeability suggesting that MYO9B might also harbor variants that protect from developing epithelial barrier dysfunction. Based on these data, we defined a risk genotype for each individual MYO9B variant (depending on the presence of at least one risk allele; Figure 2). In a second step, we compared absolute values of permeability testing according to this stratification. We
Distribution of MYO9B variants in IBD patients and controls We analyzed the five MYO9B variants in IBD patients and control subjects from three different European populations. Genotypes are shown in Table III. Genotypes of all three control populations were in Hardy-Weinberg equilibrium. We observed an association of the intron 28 variant (rs2305764) to CD in the Dutch cohort (p = 0.004; OR: 1.5569, 95%CI: 1.15–2.11), but not in the German (p = 0.24; OR: 1.16, 95%CI: 0.91–1.49) or Hungarian (p = 0.31; OR: 0.8462, 95%CI: 0.62–1.15) cohort. We did not find an association of the other four variants, rs1545620, rs1457092, rs2279003 and rs2279002 (Table III) in any of the three populations studied, nor in the overall combined cohort. Effect of MYO9B variants on the clinical course of IBD We then used the clinical and demographical data of the IBD patients to perform detailed genotype phenotype analysis. These analyses, however, did not reveal a significant association with respect to all investigated clinical characteristics (data not shown).
Table II. Intestinal permeability measurement depending on MYO9B genotype in CD patients. MYO9B Genotype AA rs1545620 (exon 20) PI: 0.03 8 GG rs1457092 (intron 20) PI: 0.03 11 CC rs2279003 (exon 21) PI: 0.03 31 rs2305764 (intron 28) PI: 0.03 9 rs2279002 (intron 32) PI: 0.03 19
%
AC
%
CC
%
n
p(c2)
OR
[95%CI]
40.9 15.1
26 32
42.6 60.4
10 13
16.4 24.5
61 53
0.010
1.99
[1.17–3.39]
2
%
GT
%
TT
%
n
p(c )
OR
[95%CI]
47.6 22.9
24 29
38.1 64.4
9 8
14.3 16.7
63 48
0.024
1.76
[1.02–3.05]
2
%
CT
%
TT
%
n
p(c )
OR
[95%CI]
32.3 54.4
29 24
44.6 42.1
15 2
23.1 3.5
65 57
0.003
0.39
[0.226–0.678]
38.46 15.79
31 35
47.69 61.40
9 13
13.85 22.81
65 57
0.015
1.90
[1.14–3.17]
58.7 38.8
18 25
28.6 51.0
8 5
12.7 10.2
63 49
0.051
1.50
[0.85–2.66]
Genotypes of the five MYO9B variants (rs1545620, rs1457092, rs2279003, rs2305764, rs2279002) according to normal/increased intestinal permeability (Permeability Index, PI; for definition see section “materials and methods”) in German CD patients (Chi2-test). Shown are absolute numbers. Abbreviations: CD: Crohn’s disease; PI: permeability index.
1196
M. Prager et al. B: rs1457092 (intron 20)
0.14
0.14
0.12
0.12 Permeability Lac/Man ratio
0.10 p = 0.015
0.08 0.06 0.04 0.02
0.10 p = 0.023
0.08 0.06 0.04 0.02
0.00
0.00 AA
AC CC
GG
0.14
0.12
0.12
0.10
0.10
Permeability Lac/Man ratio
0.14
p = 0.022
0.08
GT TT
D: rs2305764 (intron 28)
C: rs2279003 (intron 21)
Permeability Lac/Man ratio
0.06 0.04
p = 0.011
0.08 0.06 0.04 0.02
0.02
0.00
0.00 CC CT
TT
CC
CT TT
E: rs2279002 (intron 32) 0.14 0.12 Permeability Lac/Man ratio
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Permeability Lac/Man ratio
A: rs1545620 (exon 20)
0.10 p = 0.024
0.08 0.06 0.04 0.02 0.00 CC
CT TT
Figure 2A-E. Boxplot analysis of absolute values of intestinal permeability in CD patients according to MYO9B risk genotype (added is a solid line that markes the cut-off value for increased intestinal permeability, PI > 0.03). CD: Crohn’s disease; PI: permeability index.
Discussion In this study, we observed that numerous exon and intron MYO9B variants are associated with altered intestinal permeability in CD patients despite clinical remission. We therefore provide functional data that suggest a crucial involvement of MYO9B in epithelial barrier integrity.
Among the analyzed MYO9B variants, of special interest might be the rs1545620 variant within exon 20, which leads to an amino acid exchange in codon 1011 (p.A1011S). This position is involved in the binding of calmodulin [10,20]. Calmodulin regulates the motor activity of MYO9B on actin filaments and might therefore influence the velocity of MYO9B. We also found an association for rs2305764 within intron
Myosin XIb variants and intestinal barrier
1197
Table III. MYO9B variants in IBD patients.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
rs1545620 (exon 20)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
AA
AC
CC
MAF (%)
OR
95% CI
179 87 266 201
253 118 371 321
91 45 136 99
41.6 41.6 41.6 41.8
0.99 0.99 0.99
[0.84–1.17] [0.81–1.22] [0.85–1.15]
0.93 0.96 0.94
69 48 117 75
119 65 184 118
39 29 68 26
43.4 43.3 43.4 38.8
1.20 1.20 1.20
[0.93–1.58] [0.89–1.63] [0.95–1.54]
0.174 0.250 0.142
52 39 91 64
68 53 121 95
26 16 42 31
41.1 39.3 40.4 41.3
0.99 0.92 0.96
[0.73–1.35] [0.66–1.30] [0.73–1.26]
1.00 0.66 0.78
58 62
66 108
26 42
39.3 45.2
0.78
[0.58–1.06]
0.13
p-Value#
rs1457092 (intron 20)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
GG
GT
TT
MAF (%)
OR
95% CI
223 109 332 290
228 106 334 316
74 33 107 83
35.8 34.7 35.5 35.0
1.04 0.99 1.02
[0.88–1.23] [0.80–1.22] [0.88–1.19]
0.70 0.91 0.82
88 57 145 114
108 62 170 148
28 19 47 22
36.6 36.2 36.4 33.8
1.13 1.11 1.12
[0.87–1.47] [0.82–1.50] [0.89–1.41]
0.35 0.49 0.35
67 52 119 73
55 44 99 96
25 14 39 26
35.7 32.7 34.4 37.9
0.91 0.80 0.86
[0.66–1.24] [0.56–1.13] [0.65–1.13]
0.58 0.22 0.29
68 103
65 72
21 35
34.7 33.8
1.04
[0.76–1.42]
0.81
p-Value#
rs2279003 (exon 21)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls
CC
CT
TT
MAF (%)
OR
95% CI
204 81 285 234
251 136 387 278
92 36 128 100
39.8 41.1 40.2 39.1
1.03 1.09 1.05
[0.87–1.21] [0.88–1.35] [0.90–1.22]
0.73 0.45 0.56
88 40 128 105
113 68 181 121
36 19 55 46
39.0 41.7 40.0 39.2
1.28 1.51 1.30
[0.77–1.28] [0.82–1.51] [0.83–1.30]
1.00 0.54 0.77
58 41 99 44
65 68 133 59
30 17 47 17
40.8 40.5 40.7 38.8
1.09 1.07 1.08
[0.77–1.54] [0.75–1.54] [0.79–1.48]
0.66 0.71 0.64
p-Value
1198
M. Prager et al.
Table III. (Continued). rs2279003 (exon 21)
The Netherlands CD Controls
CC
CT
TT
58 85
73 98
26 37
MAF (%)
OR
95% CI
39.8 39.1
1.03
[0.77–1.39]
p-Value 0.88
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
rs2305764 (intron28)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
CC
CT
TT
MAF (%)
OR
95% CI
207 96 303 282
266 124 390 317
100 42 143 101
40.7 39.7 40.4 37.1
1.16 1.11 1.15
[0.99–1.36] [0.91–1.37] [0.99–1.33]
0.066 0.29 0.063
89 52 141 100
137 76 213 158
42 23 65 29
41.2 40.4 40.9 37.6
1.16 1.12 1.15
[0.91–1.49] [0.85–1.49] [0.92–1.42]
0.24 0.42 0.22
62 44 106 64
60 48 108 96
27 19 46 34
38.3 38.7 38.5 42.3
0.85 0.86 0.85
[0.62–1.15] [0.62–1.21] [0.65–1.12]
0.31 0.44 0.27
56 118
69 63
31 38
42.0 31.7
1.56
[1.15–2.11]
0.004
p-Value
rs2279002 (intron32)
All cohorts CD UC IBD Controls Germany CD UC IBD Controls Hungary CD UC IBD Controls The Netherlands CD Controls
CC
CT
TT
MAF (%)
OR
95% CI
251 129 380 326
208 88 296 291
58 27 85 53
31.3 29.1 30.6 29.6
1.08 0.97 1.05
[0.91–1.29] [0.78–1.22] [0.89–1.23]
0.39 0.86 0.57
110 74 184 129
94 48 142 126
21 16 37 12
30.2 29.0 29.8 28.1
1.11 1.05 1.08
[0.84–1.46] [0.76–1.44] [0.85–1.39]
0.48 0.81 0.53
68 55 123 95
52 40 92 82
21 11 32 17
33.3 29.2 31.6 29.9
1.17 0.97 1.08
[0.84–1.63] [0.67–1.40] [0.81–1.44]
0.35 0.93 0.61
73 102
62 83
16 24
31.1 31.3
0.99
[0.72–1.36]
1.00
p-Value#
Genotypes of the five MYO9B variants (rs1545620, rs1457092, rs2279003, rs2305764 and rs2279002) in IBD patients from Germany, Hungary and the Netherlands. Absolute numbers are illustrated. Genotypes were compared using the Fisher’s exact test (#). Odds ratio and confidence intervals were estimated using allele frequencies in 2x2 contingency tables. CD versus controls, CU versus controls, IBD versus controls. Abbreviations: CD: Crohn’s disease; IBD: inflammatory bowel disease; MAF: minor allele frequency; UC: ulcerative colitis.
28. Interestingly, this variant was found to completely explain the association of MYO9B to celiac disease [12], but a direct link to functional data of an intestinal barrier defect has not been provided yet. Interestingly, an association for the rs2305764 variant
to both Barrett’s esophagus (BE) and esophageal adenocarcinoma (EAC) has been described suggesting that MYO9B may play a role in the etiology of BE and EAC by increasing the permeability of the epithelial barrier [21].
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
Myosin XIb variants and intestinal barrier The precise mechanisms how MYO9B variants regulate the epithelial barrier remain to be elucidated. It is suggested that a genetic variant in the 3¢ part of MYO9B leads to an impaired interaction with RhoA, thereby perturbing tight junction gate and fence function in epithelial cells [12]. A recent study showed that loss of MYO9B function resulted in a nearly complete disruption of tight junction protein localization in an epithelial cell line. Herein, the tight junction complex proteins were mislocalazied and the intestinal barrier was compromised, permitting TRTC-dextran to readily pass through a monolayer in cell culture where MYO9B function was altered [22]. Having this in mind, there could be a combinatorial effect of the submucosal immune system with unregulated immune cell migration [23], and a disregulated intestinal permeability due to loss of MYO9B function that results in inflammation. We furthermore tested whether MYO9B variants resemble susceptibility factors for CD or lead to a distinct phenotype. We found an association of the intron 28 variant (rs2305764) to CD in the cohort from the Netherlands (p = 0.004), which could not be replicated in the German and Hungarian cohort. Van Bodegraven and colleagues [14] observed a significant association of MYO9B SNPs with IBD in three independent cohorts of IBD patients from UK, the Netherlands and Canada/Italy, with a stronger association in a UC subgroup. A similar association with UC was replicated in a Spanish cohort by Nunez et al. [24]. In contrast, Wolters et al. [25] analyzed Canadian patients and reported an association of the SNP rs1457092, rs1545620 and rs2305764 with ileal-only CD, although no overall association to CD or UC could be shown. Beyond that they observed an association in the opposite direction with inverse allelic association compared with the four European IBD studies [14,24,26,27]. We did not observe an association to any of the studied phenotypes. However, these conflicting data do not allow the conclusion that MYO9B variants resemble susceptibility variants for either CD or UC. These discrepancies reflect a typical problem in IBD genetic studies. This might be due to either different ethnic backgrounds, or collection of patients from either populations-based cohorts or tertiary referral centers, or small samples sizes. Our study has several limitations. At first, the number of patients with data on intestinal permeability in CD is limited (n = 122). We choose strict inclusion criteria for analysis of intestinal permeability including disease activity and concurrent medical treatment, for example by measuring patients in clinical remission, in order to detect changes in epithelial barrier function independent of systemic inflammation. All these factors put limitations on patient
1199
recruitment. However, IBD centers with large numbers of patients with both DNA and data on in vivo analysis of intestinal permeability are very limited. In this respect, it is of interest that an Italian group found a trend towards an abnormal intestinal permeability in Italian CD carrying the rs1545620 risk allele supporting our data [27]. Second, we did not study a cohort of healthy individuals to detect an association between MYO9B and altered intestinal permeability. Due to the fact that an increased intestinal permeability was not found in 96 healthy controls in our previous study [1], showing that an altered intestinal permeability is a very rare event in healthy individuals, it would require a very large cohort to obtain robust statistical data. Third, we observed an overall association of the rs2305764 variants only in the Dutch cohort and not in the German and Hungarian CD groups. This might be due to phenotypic differences among the three groups. The presence of perianal and fistulizing disease was remarkably higher in the Dutch cohort compared to the other two groups. In agreement with our findings, Bodegraven et al. observed the most significant association to MYO9B variants in the Dutch cohort. In summary, by using CD as a model, we provide for the first time a direct link between in vivo analysis of intestinal permeability to different and independent variants in a gene crucially involved in the maintenance of the epithelial barrier, MYO9B.
Acknowledgement The study was supported by a grant from the Eli and Edythe Broad Foundation (C.B.), by a grant of the German Ministry for Education and Research (BMBF) - Competence Network “Inflammatory Bowel Disease”: 01GI0284 TP 1.17 and by the Forschungsförderung of the Charité, Universitätsmedizin Berlin. We thank all patients who participated in this evaluation. The work is part of the doctoral thesis of Tahir Durmus.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Buhner S, Buning C, Genschel J, Kling K, Herrmann D, Dignass A, et al. Genetic basis for increased intestinal permeability in families with Crohn’s disease: role of CARD15 3020insC mutation? Gut 2006;55:342–7.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Ulster at Jordanstown on 12/04/14 For personal use only.
1200
M. Prager et al.
[2] Prager M, Buttner J, Haas V, Baumgart DC, Sturm A, Zeitz M, et al. The JAK2 variant rs10758669 in Crohn’s disease: altering the intestinal barrier as one mechanism of action. Int J Colorectal Dis 2012;27:565–73. [3] De Gottardi A, McCoy KD. Evaluation of the gut barrier to intestinal bacteria in non-alcoholic fatty liver disease. J Hepatol 2011;55:1181–3. [4] Vaarala O. Leaking gut in type 1 diabetes. Curr Opin Gastroenterol 2008;24:701–6. [5] Soderholm JD, Olaison G, Lindberg E, Hannestad U, Vindels A, Tysk C, et al. Different intestinal permeability patterns in relatives and spouses of patients with Crohn’s disease: an inherited defect in mucosal defence? Gut 1999; 44:96–100. [6] John LJ, Fromm M, Schulzke JD. Epithelial barriers in intestinal inflammation. Antioxid redox signal 2011;15: 1255–70. [7] Bruewer M, Hopkins AM, Hobert ME, Nusrat A, Madara JL. RhoA, Rac1, and Cdc42 exert distinct effects on epithelial barrier via selective structural and biochemical modulation of junctional proteins and F-actin. Am J Physiol Cell Physiol 2004;287:C327–35. [8] Matter K, Balda MS. Signalling to and from tight junctions. Nat rev Mol Cell Biol 2003;4:225–36. [9] Muller RT, Honnert U, Reinhard J, Bahler M. The rat myosin myr 5 is a GTPase-activating protein for Rho in vivo: essential role of arginine 1695. Mol biol cell 1997;8: 2039–53. [10] Post PL, Tyska MJ, O’Connell CB, Johung K, Hayward A, Mooseker MS. Myosin-IXb is a single-headed and processive motor. J biol chem 2002;11679–83. [11] Wirth JA, Jensen KA, Post PL, Bement WM, Mooseker MS. Human myosin-IXb, an unconventional myosin with a chimerin-like rho/rac GTPase-activating protein domain in its tail. J Cell Sci 1996;109:653–61. [12] Monsuur AJ, de Bakker PI, Alizadeh BZ, Zhernakova A, Bevova MR, Strengman E, et al. Myosin IXB variant increases the risk of celiac disease and points toward a primary intestinal barrier defect. Nat Genet 2005;37:1341–4. [13] Irvine EJ, Marshall JK. Increased intestinal permeability precedes the onset of Crohn’s disease in a subject with familial risk. Gastroenterology 2000;119:1740–4. [14] van Bodegraven AA, Curley CR, Hunt KA, Monsuur AJ, Linskens RK, Onnie CM, et al. Genetic variation in myosin IXB is associated with ulcerative colitis. Gastroenterology 2006;131:1768–74. [15] Dignass A, Eliakim R, Magro F, Maaser C, Chowers Y, Geboes K, et al. Second European evidence-based consensus
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
on the diagnosis and management of ulcerative colitis part 1: definitions and diagnosis. J Crohns colitis 2012;6:965–90. Van Assche G, Dignass A, Panes J, Beaugerie L, Karagiannis J, Allez M, et al. The second European evidence-based Consensus on the diagnosis and management of Crohn’s disease: Definitions and diagnosis. J Crohns colitis 2010;4:7–27. Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005;19:Suppl A):5A–36A. Buhner S, Reese I, Kuehl F, Lochs H, Zuberbier T. Pseudoallergic reactions in chronic urticaria are associated with altered gastroduodenal permeability. Allergy 2004;59: 1118–23. Wyatt J, Vogelsang H, Hubl W, Waldhoer T, Lochs H. Intestinal permeability and the prediction of relapse in Crohn’s disease. Lancet 1993;341:1437–9. Post PL, Bokoch GM, Mooseker MS. Human myosin-IXb is a mechanochemically active motor and a GAP for rho. J Cell sci 1998;111:941–50. Menke V, Van Zoest KP, Moons LM, Pot RG, Siersema PD, Kuipers EJ, et al. Myo9B is associated with an increased risk of Barrett’s esophagus and esophageal adenocarcinoma. Scand J Gastroenterol 2012;47:1422–8. Chandhoke SK, Mooseker MS. A role for myosin IXb, a motor-RhoGAP chimera, in epithelial wound healing and tight junction regulation. Mol biol cell 2012;23:2468–80. Hanley PJ, Xu Y, Kronlage M, Grobe K, Schon P, Song J, et al. Motorized RhoGAP myosin IXb (Myo9b) controls cell shape and motility. Proc Natl Acad Sci USA 2010;107: 12145–50. Nunez C, Oliver J, Mendoza JL, Gomez-Garcia M, Pinero A, Taxonera C, et al. MYO9B polymorphisms in patients with inflammatory bowel disease. Gut 2007;56:1321–2. Wolters VM, Xu W, Zhao X, Walters TD, Griffiths AM, Silverberg MS, et al. Replication of genetic variation in the MYO9B gene in Crohn’s disease. Hum immunol 2011;72: 592–7. Cooney R, Cummings JR, Pathan S, Beckly J, Geremia A, Hancock L, et al. Association between genetic variants in myosin IXB and Crohn’s disease. Inflamm Bowel Dis 2009; 15:1014–21. Latiano A, Palmieri O, Valvano MR, D’Inca R, Caprilli R, Cucchiara S, et al. The association of MYO9B gene in Italian patients with inflammatory bowel diseases. Aliment pharmacol ther 2008;27:241–8.
