It is with great interest that we read the paper by Frachimont and colleagues (Gut 2004;53:987–92) in which they describe a novel association of the toll-like receptor 4 (TLR4) +896 A>G polymorphism with both Crohn’s disease (CD) and ulcerative colitis (UC), supporting the genetic influence of pattern recognition receptors (PRRs) in triggering inflammatory bowel disease (IBD). PRRs are sensors of pattern associated molecular patterns of microorganisms in the intestinal flora. Independently, we performed a similar study. However, special attention to the presence of anti-Saccharomyces cerevisiae antibody (ASCA) was taken, as Tada and colleagues1 have recently reported that the S cerevisiae mannan-LBP complex is recognised by CD14 on monocytes and signalling through TLR4 leads to the production of proinflammatory cytokines in a manner similar to that induced by lipopolysaccharide (LPS).
Patients and controls were recruited from the Outpatient Department of Gastroenterology, VU University Medical Centre, Amsterdam, the Netherlands. The group consisted of 112 CD patients and 170 unrelated Dutch Caucasian controls. Diagnosis of disease was based on clinical, histopathological, and endoscopic findings. CD patients were categorised using the Vienna classification (general patient characteristics are described elsewhere2). ASCA IgA and IgG ELISAs were performed as described previously.3 Genotyping for the CD14-260 C>T and TLR4+896 A>G single nucleotide polymorphisms (SNPs) was performed as described previously by our group.4 The CD14-260 and TLR4+896 genotypes, allele, and carrier frequencies were compared between the different clinical patient groups and controls. In addition, synergism between CD14 and TLR4 genotypes and alleles (carrier trait analyses) was studied. Vienna classification and ASCA status were included in the statistical modelling.
The results are shown in table 1 ▶. The frequency of the G allele of TLR4+896 was significantly increased in CD patients compared with controls (19% v 10%; p = 0.049; odds ratio (OR) 2.1 (95% confidence interval (CI) 1.0-4.1)). Disease phenotype was assessed in patients using the Vienna classification. Carriage of TLR4 +896*G significantly increased the risk of colonic localisation of CD compared with non-colonic localisation (43% v 12%; p = 0.0017; OR 5.5 (95% CI 1.9–15.4)). There was a clear trend (test for trend: χ2: 16, p<0.0001) when we compared the increasing frequency of the G allele of TLR4 +896 in controls (10%) to CD patients (19%) and to CD patients with colonic localisation (43%).
Table 1.
CD14−260 and TLR4+896 genotype distribution in Crohn’s disease (CD) patients and healthy controls (HC)
| Group | Vienna classification | ntotal | CD14 −260 C>T | TLR4 +896 A>G | ||||
| 1.1 (CC) | 1.2 (CT) | 2.2 (TT) | 1.1 (AA) | 1.2 (AG) | 2.2 (GG) | |||
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | |||
| HC | 170 | 48 (28) | 82 (48) | 40 (24) | 153 (90) | 16 (9) | 1 (1) | |
| CD | 112 | 35 (31) | 54 (48) | 23 (21) | 91 (81) | 19 (17) | 2 (2)* | |
| A1 | 97 | 29 (30) | 49 (51) | 19 (20) | 77 (79) | 19 (20) | 1 (0) | |
| A2 | 15 | 6 (40) | 5 (33) | 4 (27) | 14 (93) | 0 (0) | 1 (6) | |
| B1 | 43 | 14 (33) | 20 (47) | 9 (21) | 36 (84) | 6 (14) | 1 (2) | |
| B2 | 45 | 15 (33) | 21 (47) | 9 (20) | 37 (82) | 7 (16) | 1 (2) | |
| B3 | 24 | 6 (25) | 13 (54) | 5 (21) | 18 (75) | 6 (25) | 0 (0) | |
| L1 | 41 | 14 (34) | 17 (41) | 10 (24) | 36 (88) | 4 (10) | 1 (2) | |
| L2 | 23 | 6 (26) | 13 (57) | 4 (17) | 13 (57) | 9 (39) | 1 (4)† | |
| L3 | 47 | 15 (32) | 23 (49) | 9 (19) | 41 (87) | 6 (13) | 0 (0) | |
| L4 | 1 | 0 (0) | 1 (100) | 0 (0) | 1 (100) | 0 (0) | 0 (0) | |
*TLR4*G was more frequent in CD patients compared with HC (19% v 10%; p = 0.0489; odds ratio (OR) 2.076 (95% confidence interval (CI) 1.041–4.142)).
†TLR4*G was significantly associated with colonic localisation compared with non-colonic localisation (43% v 12%, p = 0.0017; OR 5.455 (95% CI 1.931–15.410)).
Vienna Classification: (age) A1: age <40 years; A2: age >40 years; (behaviour) B1: non-stricturing, non-penetrating; B2: stricturing; B3: penetrating; (localisation) L1: ileal; L2: colonic; L3: ileocolonic; L4: upper gastrointestinal tract.
We also assessed if ASCA status was correlated with carriage of the TLR4 G allele. However, there was no difference between TLR4 G allele carriage in ASCA positive and ASCA negative patients (23% v 14%; p = 0.33) (data not shown) and there was no difference between TLR4 G allele carriage in ASCA positive and negative CD patients with colonic localisation (40% v 46%; p = 1.00) while the frequency of G allele carriage was identical to that of CD patients with colonic localisation (43%) without correcting for ASCA status.
Several studies have described both TLR4+896 A>G and CD14–260 C>T in CD. Klein et al have described a German population and found an increased incidence of CD14 −260 heterozygous and homozygous mutants in CD patients compared with healthy controls.5 This association could not be confirmed in our population. Preliminary data by Braat et al demonstrated an increased risk of suffering from CD in a Dutch population carrying the TLR4 +896 SNP,6 confirming our results. Franchimont and colleagues (Gut 2004;53:987–92) corroborated the results of Braat et al. In contrast with Franchimont et al, we found a clear association between the G allele of TLR4+896 and disease phenotype (colonic localisation). In contrast with the aforementioned studies and results, Arnott et al were unable to demonstrate an association between susceptibility to CD and the TLR4 and CD14 SNPs in a Scottish and Irish population.7
The association between TLR4 and CD underscores the role of impaired innate immunity in CD. TLR4 signalling is based on both exogenous (for example, LPS) and endogenous (for example, human HSPs) agonists, and as heterozygous carriership of the TLR4+896 A>G does not seem to impair LPS signalling,8,9 further agonist identification to elucidate the microorganisms involved in CD and especially in colonic localisation is essential to obtain insight into both the pathophysiological and immunogenetic aspects of CD. This insight may be helpful in developing strategies for the prevention and treatment of CD.
The association we demonstrated between TLR4 and CD is most likely not strongly based on the S cerevisiae mannan-LBP-CD14-TLR4 pathway but, as we have demonstrated, on the ASCA data in our group. It would be interesting to know whether Franchimont et al tested for ASCA in their CD patients and whether or not an association between ASCA and TLR4 was observed.
Conflict of interest: None declared.
References
- 1.Tada H, Nemoto E, Shimauchi H, et al. Saccharomyces cerevisiae-and Candida albicans-derived mannan induced production of tumor necrosis factor alpha by human monocytes in a CD14- and Toll-like receptor 4-dependent manner. Japan. Microbiol Immunol 2002;46:503–12. [DOI] [PubMed] [Google Scholar]
- 2.Linskens RK, Mallant-Hent RC, Murillo LS, et al. Genetic and serological markers to identify phenotypic subgroups in a Dutch Crohn’s disease population. Netherlands. Dig Liver Dis 2004;36:29–34. [DOI] [PubMed] [Google Scholar]
- 3.Linskens RK, Mallant-Hent RC, Groothuismink ZMA, et al. Evaluation of serological markers to differentiate between ulcerative colitis and Crohn’s disease: pANCA, ASCA and agglutinating antibodies to anaerobic coccoid rods. England. Eur J Gastroenterol Hepatol 2002;14:1013–18. [DOI] [PubMed] [Google Scholar]
- 4.van der Paardt M, Crusius JBA, De Koning MHMT, et al. No evidence for involvement of the toll-like receptor 4 (TLR4) A896G and CD14 C-260T polymorphisms in the susceptibility to ankylosing spondylitis. Ann Rheum Dis 2005; (in press). [DOI] [PMC free article] [PubMed]
- 5.Klein W, Tromm A, Griga T, et al. A polymorphism in the CD14 gene is associated with Crohn disease. Norway. Scand J Gastroenterol 2002;37:189–91. [DOI] [PubMed] [Google Scholar]
- 6.Braat H, Dijkgraaf M, Curvers W, et al. A functional single polymorphism of the TLR4 gene is correlated with Crohn’s disease but not with ulceratice colitis. Gastroenterology 2004;124:A367. [Google Scholar]
- 7.Arnott IDR, Nimmo ER, Drummond HE, et al. NOD2/CARD15, TLR4 and CD14 mutations in Scottish and Irish Crohn’s disease patients: evidence for genetic heterogeneity within Europe? England. Genes Immun 2004;5:417–25. [DOI] [PubMed] [Google Scholar]
- 8.von Aulock S, Schröder NWJ, Gueinzius K, et al. Heterozygous toll-like receptor 4 polymorphism does not influence lipopolysaccharide-induced cytokine release in human whole blood. United States. J Infect Dis 2003;188:938–43. [DOI] [PubMed] [Google Scholar]
- 9.Erridge C, Stewart J, Poxton IR. Monocytes heterozygous for the Asp299Gly and Thr399Ile mutations in the toll-like receptor 4 Gene show no deficit in lipopolysaccharide signalling. United States. J Exp Med 16June2003;197:1787–91. [DOI] [PMC free article] [PubMed]