Celiac disease
Zorro Manrique, Maria Magdalena
DOI:
10.33612/diss.122712049
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Zorro Manrique, M. M. (2020). Celiac disease: From genetic variation to molecular culprits. University of
Groningen. https://doi.org/10.33612/diss.122712049
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
CHAPTER 1
Ch
ap
te
r 1
General introduction
Celiac disease (CeD) is a common immune-mediated disorder triggered by intake of grain-derived gluten proteins that affects 1-2% of individuals in the western world. Although the precise cause of CeD is unknown, multiple environmental and genetic factors have been found to contribute to the development of this complex disease1.
In most CeD patients gluten ingestion triggers a strong immune response that provokes the activation of gluten-specific CD4+ T cells, production of anti-transglutaminase 2 (TG2)
antibodies by B cells and lymphocyte infiltration in the small intestine, and all these processes contribute to the villous atrophy and crypt hyperplasia characteristic of the disease. As a result, CeD patients have a variable degree of small intestinal inflammation and a broad range of manifestations that include diarrhea, abdominal pain and malabsorption2. In addition
to the classical gastrointestinal symptoms, CeD patients can also present extra-intestinal manifestations such as anemia3, osteoporosis4, dermatitis herpetiformis5 and neurological
disorders6. Because of this wide range of symptoms, it is thought that many cases of CeD go
undiagnosed7.
Currently, a lifelong gluten-free diet (GFD) is the only available treatment for CeD. Avoidance of gluten contributes to the recovery of the intestinal mucosa, the reduction of levels of anti-TG2 antibodies and eventually to reduction of symptoms8. In spite of its benefits, a
GFD can be challenging to maintain due to multiple factors, including social restrictions, nutrient deficiencies and high costs8,9. In addition, a small sub-group of patients (1-5%) fail
to respond to a GFD, and these individuals are at risk of developing a severe condition called refractory celiac disease (RCD) that is characterized by a remarkable infiltration of intraepithelial cytotoxic T cells (IE-CTLs) with abnormal phenotype. In RCD these intestinal abnormalities persist and can eventually contribute to the development of enteropathy-associated T cell lymphoma10.
Although gluten exposure is the most important environmental factor in CeD pathogenesis, recent evidence implicates other environmental factors in disease development. Intestinal viral infections and bacterial microbiota have been linked to CeD as possible environmental triggers1,11,12. In a recent study Bouziat et al. suggested that reovirus infection induces
a proinflammatory response with a concomitant loss of oral tolerance to gluten13, while
the involvement of the bacterial microbiome has been suggested by studies reporting gut microbiome dysbiosis in CeD patients as compared to healthy individuals14,15. Interestingly,
these CeD-associated changes in microbiota composition have been shown to affect the processing of gluten peptides16, which may affect gluten presentation to gluten-specific
CD4+ T cells and thereby increasing the inflammatory response. Thus, in addition to
gluten, environmental factors such as the gut microbiome and virome may contribute to the environmental component of the risk of developing CeD, although their respective contributions
to CeD development are still unclear1,11,13. In contrast, genetic risk factors have been estimated
to contribute approximately 50% of CeD risk, making them the major predisposing factors currently known for CeD1,17.
Genetic risk factors
The strongest genetic factor associated to CeD risk is the human leukocyte antigen (HLA) region. More than 90% of CeD patients carry either the DQ2 (DQ2.5) or the HLA-DQ8 allele, and these alleles appear to account for up to 40% of the genetic risk of developing the disease18. However, although the absence of these alleles in individuals means they will
not get the disease, their presence alone cannot predict who will develop CeD because these alleles are present in approximately 30-40% of the general population. This suggests that, while the HLA-DQ2 and -DQ8 alleles are necessary for CeD development, additional genetic factors are required19. To date, genome-wide association studies (GWAS) have identified 42
non-HLA genetic variants to be associated with CeD (Fig. 1)17,20. Due to the modest effect
of each non-HLA variant on overall disease risk, these together account for approximately 15% of heritability20. Interestingly, most of these genetic variants are also shared with other
Figure 1. Manhattan plot showing the results of association for 39 of 42 non-HLA CeD risk loci. Known loci
(black), novel loci (blue) and risk loci with multiple signals (underlined) are depicted. The vertical line represents the genome-wide significant threshold (p value 5x10-8). Adapted from Trynka, G. et a, 201220.
23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 10 20 –log 10 P 30 40 50 C hr om oso m e HCFC1,TMEM187, IRAK1 UBE2L3, YDJC ICOSLG UBASH3A PTPN2 SOCS1, PRM1,PRM2
CIITA and others CLK3 and others ZFP36L1
SH2B3, ATXN2 ETS1 TREH, DDX6
POU2AF1 and other
ZMIZ1 PFKFB3, PRKCQ PVT1 ELMO1 TAGAP OLIG3, TNFAIP3 PTPRK BACH2 IRF4
KIAA1109, ADAD1, IL2, IL21 LPP SCHIP1, IL12A ARHGAP31 CCR1-3, LTF CCR4, GLB1 CD28, CTLA4, ICOS STAT4 UBE2E3,ITGA4 IL18R1, IL18RAP PLEK, FBX048 PUS10 C10rf106 RGS1 FASLG,TNFSF18 RUNX3 C1orf93, MMEL1, TTC34
Ch
ap
te
r 1
immune-mediated disorders such as rheumatoid arthritis21 (RA) and type I diabetes22 (T1D),
indicating the presence of a common etiological component in immune-mediated complex diseases.
Although GWAS have been very successful in associating genomic loci with disease, there is a need for follow-up studies aimed at pinpointing causal genetic variants (single nucleotide polymorphisms, SNPs) and genes in these loci23. There are still some limitations that
prevent successful identification of causal SNPs24. First, due to linkage disequilibrium, many
adjacent SNPs are co-inherited and are likely to have a similar association or correlation, which complicates prioritisation23. Second, in only four of the CeD-associated loci – MMEL1,
SH2B3, NCF2 and IRAK – do the SNPs affect protein encoding regions25–27. Most of the risk
SNPs associated with CeD (and complex immune-mediated diseases in general) fall in non-coding regions of the genome, including intergenic regions, and their consequence is not understood28,29. In complex diseases it has consistently been found that these variants are
enriched in regulatory domains controlling gene expression, including promoter and enhancer elements characterized by regions of open chromatin and specific histone modifications30.
This indicates that, rather than altering protein function, risk SNPs associated with immune diseases control the expression of genes encoding proteins or non-coding RNAs.
To overcome the difficulties discussed above, complementary methods are applied to move from SNP associations to the downstream consequences on gene expression and the regulation of biological pathways. Zooming in on GWAS loci via fine-mapping, for instance, can identify smaller regions that encompass smaller groups of variants with the highest probability of causality24. The functional impact of fine-mapped SNPs can then be tested by
assessing the correlation between expression and genotype (quantitative trait locus (eQTL)
analysis)31; the interaction between GWAS SNPs and other genomic regions (chromatin
interaction conformation assays (3C,4C))32; the enrichment-overlap with functional elements
such as enhancers and promoters (using data generated by the ENCODE33 or Epigenome
roadmap projects)33 and the potential alterations of transcription factor binding sites23. These
approaches have confirmed that CeD-associated SNPs affect gene expression rather than change the amino acid sequence of proteins17. Additionally, application of computational
approaches such as gene set enrichment analysis and gene network analysis to genes differentially expressed in CeD can elucidate the biological pathways and tissues where these risk genes play a role in disease pathophysiology34, and the prioritized set of candidate genes
that results can then be further validated by in vitro and in vivo assays24. From SNPs to disease mechanisms
It has been hypothesized that the intestinal barrier is compromised in CeD35, thereby facilitating
the passage of gluten peptides into the lamina propria, where they are deaminated by tissue TG2. This deamination process strongly increases the binding affinity of gluten peptides to HLA-DQ2 or -DQ8 molecules on the surface of antigen presenting cells (APCs) such as
dendritic cells, B cells and macrophages36,37. The gluten peptides presented in the context
of HLA-DQ2 or -DQ8 are recognized by gluten-specific CD4+ T cells, leading to activation
of the inflammatory response characteristic of CeD. Upon activation, CD4+ T cells release
cytokines such as IFNg and IL-2138,39. IL-21 may provide inflammatory signals to B cells
and intraepithelial cytotoxic T cells (IE-CTLs). B cell activation results in the secretion of antibodies against gluten and TG240. However, the clinical implications and pathogenic role
of these antibodies are still unclear. IFNg promotes the Th-1 immune response, activation of APCs and licensing of IE-CTLs to kill intestinal epithelial cells (Fig. 2)40.
A recently performed genetic study integrated different layers of genomic information, including eQTL analysis, cell-type-specific enhancer enrichment, functional annotation of GWAS SNPs and co-expression analyses41. In addition to confirming what was already known
about the involvement of the immune system in CeD, this study provided genetic evidence for
Figure 2. Main players in CeD pathogenesis. A compromised intestinal barrier in predisposed individuals allows the
passage of gluten peptides into the lamina propria, where the gluten peptides are deaminated by TG2, which enhances their affinity for HLA-DQ2 or HLA-DQ8 molecules on the surface of APC (dendritic cells, B cells, macrophages). Presen-tation to gluten-specific CD4+ T cells results in activation and proliferation of these cells, which then release IFNg and IL-21. These cytokines provide signals that enhance the cytolytic properties of IE-CTLs and promote the differentiation of B cells towards plasma cells that produce anti-gluten and anti-TG2 antibodies. Some of the environmental factors (gluten, microbiome, infections) that can influence the disease onset are indicated. Cytokines/inflammatory molecules expressed by intestinal epithelial cells are depicted (IL-15, IFN-1), as are intestinal epithelial cells (IEC), intraepithelial cytotoxic lymphocytes (IE-CTLs), antigen presenting cells (APC).
Microbiome Gluten TG2 IL-15 IFN-1 IE-CTL IEC IFNg IL-21 Anti-gluten Anti-TG2 HLA DQ2/8 IL-21 IFNg ? Gluten specific CD4+ T Cell APC B cell
Ch
ap
te
r 1
the involvement of the adaptive immune system via the IFNg signaling pathway (although the IFN locus itself has not been associated with CeD) and a role for B cells in CeD. Moreover, the same study prioritized several genes (LPP, C1orf106) in CeD-associated loci that might contribute to decreased intestinal barrier function41. Disturbance of intestinal permeability
could not only facilitate the passage of gluten into the lamina propria but also that of infectious agents. This would boost the presentation of gluten peptides to CD4+ T cells, resulting in
the release of pro-inflammatory signals, and these processes would contribute to a stronger immune response16. Although this evidence suggests that barrier dysfunction can contribute
to CeD onset35,41, it is still unclear whether this is a primary defect and a cause that contributes
directly to disease onset or of it is a consequence of the inflammatory environment in the gut of CeD individuals42.
In addition to dysregulation of the adaptive immune system and adaptive cytokines, innate cytokines such as IL-15 and IFN type I (IFN-1) are upregulated in intestinal epithelial cells of CeD patients43,44. These cytokines enhance the cytolytic and proinflammatory properties
of CTLs and dendritic cells43–45, respectively. Simultaneously, an induction of stress-induced
non-classical major histocompatibility complex class I molecules is observed on the surface of epithelial cells, which are recognized by natural killer receptors expressed on IE-CTLs. This interaction licenses IE-CTLs to kill intestinal epithelial cells, thus contributing to villous atrophy46. These findings demonstrate the involvement of both adaptive cytokines (IFNg and
IL-21) and tissue-derived cytokines (IL-15 and IFN-1) in activation of CTLs. To date, little is known about the signaling processes elicited in IE-CTLs by these cytokines47.
In the work described in this thesis, we have studied how CeD-associated genetic variation can be translated to molecular culprits (genes, pathways and relevant cell types). One of the most exciting discoveries of genomics research using next generation sequencing methodologies has been the existence of a novel class of genes that are transcribed but not translated: long non-coding RNAs (lncRNAs). Our group found genetic evidence that, in loci associated with
Figure 3. Overview of the cell types and methods used in this thesis. A) Human-derived primary and immortalized
cell lines and culture systems (2D and 3D). B) Tools to target genes and mimic inflammation. C) Lab approaches. D) Data
analysis strategies, in silico tools, datasets and population cohorts employed to conduct this study Cell types and culture
systems Gene and environmental perturbations Conventional and novel lab methods Methods and cohortsIn silico
IELs Gut Biopsies T cells 2D/3D Cultures Monocytes Caco-2 PBMCs Cell lines Biopsy-derived intestinal cells Anti CD3/CD28 Stimulation Cytokine stimulation Rescue SiRNA shRNA WB qPCR Olink FACS sorting Single cell RNA-seq Chip-seq Bioinformatic analysis Co-expression Analysis Data clustering Available datasets Population Cohorts ACGTA CGTAA GCTTA Bulk RNA-seq
complex diseases including CeD, lncRNAs are affected by disease-associated SNPs48. Thus,
it is not only non-coding regulatory elements, but also lncRNAs, that may contribute to the deregulation caused by CeD-associated variants.
In the research leading to this thesis I made use of in silico approaches in combination with in vitro experiments in human-derived cells exposed to conditions that resemble the inflammatory environment in the intestine of CeD patients in order to prioritize genes and lncRNAs potentially involved in CeD and to further understand their role in disease onset and pathogenesis (Fig. 3).
Aim and outline of the thesis
The aim of this thesis is to gain further insights into the function of some of the CeD candidate genes and the molecular pathways and mechanisms that play a role in CeD. The candidate genes have been selected by in silico and omics approaches in cell types known to play a role in CeD pathophysiology.
As was described above, it has been hypothesized that CeD patients have an inherent barrier function defect that could facilitate gluten transport into the lamina propria. Previously, LPP has been identified as one of multiple CeD-associated genes that may be involved in cell-cell interaction and intestinal barrier homeostasis. In CeD biopsies, LPP expression is reduced when compared to normal biopsies. In Chapter 2, we examined whether a reduction in LPP does lead to decreased barrier homeostasis using the Caco-2 cell line, which is widely used in barrier function and pharmacology research, as an in vitro model. We generated a stable LPP knockdown cell line of the parental Caco-2 cell line and evaluated the effect on proliferative capacity, permeability and lumen formation in 2D and 3D culture environments. Moreover, we evaluated the transcriptional response in this cell line under standard culture conditions and upon challenge with IFNg, a cytokine known to be involved in the pathogenesis of CeD. During the course of my thesis research, population cohort studies became available that could be used for eQTL analysis. Additionally, novel statistical genetics approaches were developed in our lab or published by others that could be applied to the data generated in the cohort studies to prioritize culprit SNPs and genes in disease-associated loci. In Chapter 3 we applied a systematic approach to integrate eQTL data from the BIOS cohort (total RNA transcriptomics from whole blood of 4000 participants from general population cohorts)49
with CeD association data derived from the most recent CeD GWAS meta-analysis50. We
applied four different in silico approaches (LD-overlap, Bayesian co-localization, Mendelian randomization and DEPICT) to prioritize potential causal genes, resulting in the identification of 126 positional and functional candidate genes. Co-expression and pathway analysis were applied to prioritize the main cell types and biological pathways in which these genes are most likely to play a role. TRAFD1, one of the prioritized genes, was selected for functional follow-up.
Ch
ap
te
r 1
The pro-inflammatory response elicited by gluten in CeD patients ultimately converges on IE-CTLs that are consequently licensed to kill epithelial cells. Although several key CeD-associated cytokines are known to affect IE-CTLs, little is known about the transcriptional programs triggered by these cytokines. In Chapter 4 we generated TCRab+ CD8+ cytotoxic T
cell lines from human small intestinal epithelium to study the dynamics of the transcriptional response and of genome-wide H3K27 acetylation (H3K27ac) changes in response to stimulation with tissue-derived cytokines, also known as alarmins (IFNb and IL-15) or a T cell–derived cytokine (IL-21). These three cytokines have not only been associated with tissue destruction in CeD, but also with other autoimmune disorders with tissue-specificities such as RA, inflammatory bowel disease (IBD) and T1D. The data we generated was analyzed in depth to describe the biological pathways that are triggered in IE-CTLs in response to tissue-derived cytokines versus T cell–tissue-derived cytokines. We further studied the relation between gene expression and epigenomic (H3K27ac) profiles to get a better understanding of the potential mechanism of gene regulation. Finally, we tested the potential enrichment of genes responding to cytokine stimulation in risk loci associated with autoimmune diseases (GWAS data) to identify genes that might contribute to immune deregulation in IE-CTLs.
The next generation sequencing and genomics revolution has led to the discovery of novel classes of genes and novel insights into disease biology. A significant portion of the SNPs associated with complex immune-mediated diseases have been shown to overlap with DNA motifs that control the expression or binding sites of micro-RNAs (miRNAs) or to intersect gene regulatory motifs of lncRNAs. Chapter 5 is an overview of the general features of these two classes of non-coding RNAs in terms of synthesis, structure and the potential mechanism by which they modulate gene expression or other processes in the cell. This review focusses on the role of these genes in different cells of the adaptive and the innate immune system, and on their involvement in immune mediated disorders such as CeD, IBD and multiple sclerosis. Chapter 6 focuses on a single lncRNA, lncRNA RP11-291B21.2, that is strongly modulated in response to TCR activation in CD8+ T cells, including IE-CTLs. We describe
the expression pattern of this lncRNA in different CD8+ T cell populations derived from blood
and infer its potential biological function using single cell RNA-seq data and co-expression network analysis. Finally, knockdown experiments performed in IE-CTL cell lines and RNA sequencing data were interrogated to provide additional clues for pinpointing this lncRNA’s function in IE-CTLs, which are the effector cell type in CeD.
Chapter 7 summarizes the major findings of this thesis research project and sets them in a broader perspective by discussing their implications in the context of CeD and immune-mediated disorders in general. The main drawbacks and limitations of our experimental approaches are described, as are directions and suggestions for how to dig deeper into the biological contributions of the risk genes in the pathogenesis of complex diseases.
References
1. Lionetti, E. & Catassi, C. The role of environ-mental factors in the development of celiac disease: what is new?. Dis. 3, 282–293 (2015). 2. Caja, S., Mäki, M., Kaukinen, K. & Lindfors, K.
Antibodies in celiac disease: Implications Be-yond Diagnostics. Cell Mol Immunol. 8, 103-9 (2011).
3. Freeman, H. J. Iron deficiency anemia in celiac disease. World J. Gastroenterol. 21, 9233–8 (2015).
4. Mustalahti, K., Collin, P., Sievänen, H., Salmi, J. & Mäki, M. Osteopenia in patients with clini-cally silent coeliac disease warrants screening. Lancet. 354, 744-5 (1999).
5. Fry, L. Dermatitis Herpetiformis. Baillieres. Clin. Gastroenterol. 9 (2), 371–93 (1995). 6. Hadjivassiliou, M. et al. Gluten sensitivity: from
gut to brain. Lancet Neurol. 9, 318-30 (2010). 7. Fasano, A. et al. Prevalence of celiac disease
in at-risk and not-at-risk groups in the united states: a large multicenter study. Arch. Intern. Med. 163, 286-92 (2003).
8. Makharia, G. K. Current and emerging therapy for celiac disease. Front. Med. 1-6 (2014). 9. Leffler, D. A. et al. Factors that influence
adher-ence to a gluten-free diet in adults with celiac disease. Dig. Dis. Sci. 53, 1573–81 (2008). 10. Rubio-Tapia, A. & Murray, J. A. Classification
and management of refractory coeliac disease. Gut 59, 547–57 (2010).
11. Sánchez, E. et al. Influence of environmental and genetic factors linked to celiac disease risk on infant gut colonization by bacteroides species. Appl. Environ. Microbiol. 77, 5316–23 (2011).
12. Verdu, E. F., Galipeau, H. J. & Jabri, B. Novel players in coeliac disease pathogenesis: role of the gut microbiota. Nat. Rev. Gastroenterol. Hepatol. 12, 497 (2015).
13. Bouziat, R. et al. Reovirus infection triggers in-flammatory responses to dietary antigens and development of celiac disease. Science. 356, 44–50 (2017).
14. Sánchez, E., Donat, E., Ribes-Koninckx, C., Fernández-Murga, M. L. & Sanz, Y. Duode-nal-mucosal bacteria associated with celiac disease in children. Appl. Environ. Microbiol. 79, 5472–5479 (2013).
15. Wacklin, P. et al. The duodenal microbiota composition of adult celiac disease patients is associated with the clinical manifestation of the disease. Inflamm. Bowel Dis. 19, 934–941 (2013).
16. Caminero, A. et al. Duodenal bacteria from pa-tients with celiac disease and healthy subjects distinctly affect gluten breakdown and immu-nogenicity. Gastroenterology 151, 670–683 (2016).
17. Withoff, S., Li, Y., Jonkers, I. & Wijmenga, C. Understanding celiac disease by genomics. Trends Genet. 31, 295-308 (2016).
18. Liu, E., Rewers, M. & Eisenbarth, G. S. Genet-ic testing: who should do the testing and what is the role of genetic testing in the setting of celiac disease?. Gastroenterology 128, 33-7 (2005).
19. Jabri, B. & Sollid, L. M. T cells in celiac dis-ease. J. Immunol. 198, 3005-3014 (2017). 20. Trynka, g. Et al. Dense genotyping identifies
and localizes multiple common and rare vari-ant association signals in celiac disease. Nat. Genet. 43, 1193–201 (2011).
21. Zhernakova, A. et al. Meta-analysis of ge-nome-wide association studies in celiac dis-ease and rheumatoid arthritis identifies four-teen non-HLA shared loci. Plos Genet. 2, e1002004 (2011).
22. Walker, N. M. et al. Shared and distinct genetic variants in type 1 diabetes and celiac disease. N. Engl. J. Med. 359, 2767-77 (2008). 23. Gallagher, M. D. & Chen-Plotkin, A. S. The
post-GWAS era: from association to function. Am. J. Hum. Genet. 102, 717–730 (2018). 24. Edwards, S. L., Beesley, J., French, J. D. &
Dunning, M. Beyond GWAS: Illuminating the dark road from association to function. Am J Hum Genet. 93, 779-97 (2013).
25. Bonvouloir, N., Lemieux, N., Crine, P., Boileau, G. & Desgroseillers, L. Molecular cloning, tis-sue distribution, and chromosomal localization of MMEL2, a gene coding for a novel human member of the neutral endopeptidase-24.11 family. DNA Cell Biol. 20, 493–498 (2001). 26. Mori, T. et al. Lnk/SH2B3 controls the
produc-tion and funcproduc-tion of dendritic cells and regu-lates the induction of IFN-y–producing T cells. J. Immunol. 193, 1728–1736 (2014).
27. Singh, A. et al. The IRAK-ERK-p67phox-nox-2 axis mediates TLR4, 2-induced ROS produc-tion for IL-1β transcripproduc-tion and processing in monocytes. Cell. Mol. Immunol. 13, 745–763 (2016).
28. Kumar, V. et al. Human disease-associat-ed genetic variation impacts large intergenic non-coding RNA expression. Plos Genet. 9, e1003201 (2013).
29. Zhang, F. & Lupski, J. R. Non-coding genetic variants in human disease. Hum. Mol. Genet. 24, R102–R110 (2015).
Ch
ap
te
r 1
30. Maurano, M. T. et al. Systematic localization of common disease-associated variation in regu-latory DNA. Science. 337, 1190–1195 (2012). 31. Nicolae, D. L.et al. Trait-associated SNPs are
more likely to be eQTLs: annotation to en-hance discovery from GWAS. Plos Genet. 6, e1000888 (2010).
32. Dekker, J., Rippe, K., Dekker, M. & Kleckner, N. Capturing chromosome conformation. Sci-ence 295, 1306–11 (2002).
33. Consortium, T. E. P. An integrated encyclope-dia of DNA elements in the human genome. Nature 489, 57–74 (2012).
34. Himmelstein, D. S. & Baranzini, S. E. Hetero-geneous network edge prediction: a data inte-gration approach to prioritize disease-associ-ated genes. PLOS Comput. Biol. 11, e1004259 (2015).
35. Van Elburg, R. M., Uil, J. J., Mulder, C. J. J. & Heymans, H. S. A. Intestinal permeability in patients with coeliac disease and relatives of patients with coeliac disease. Gut. 3, 354-7 (1993).
36. Molberg, Ø. et al. Tissue transglutaminase se-lectively modifies gliadin peptides that are rec-ognized by gut-derived t cells in celiac disease. Nat. Med. 4, 713-7 (1998).
37. Chladkova, B. et al. Gliadin fragments promote migration of dendritic cells. J. Cell. Mol. Med. 15, 938–48 (2011).
38. Bodd, M. et al. HLA-DQ2-restricted gluten-re-active T cells produce IL-21 but not IL-17 or IL-22. Mucosal Immunol. 3, 594–601 (2010). 39. Nilsen, E. M. et al. gluten specific, HLA-DQ
restricted t cells from coeliac mucosa produce cytokines with Th1 or Th0 profile dominated by Interferon gamma. Gut. 37, 766-76 (1995). 40. Lindfors, K. et al. Coeliac Disease. Nat. Rev.
Dis. Prim. 5, 3 (2019).
41. Kumar, V. et al. Systematic annotation of celiac disease loci refines pathological pathways and suggests a genetic explanation for increased Interferon-gamma levels. Hum. Mol. Genet. 24, 397–409 (2015).
42. Schumann, M., Siegmund, B., Schulzke, J. D. & Fromm, M. Celiac disease: role of the epi-thelial barrier. Cell Mol Gatroenterol Hepatol. 3, 150-162 (2017).
43. Mention, J. J. et al. Interleukin 15: a key to dis-rupted intraepithelial lymphocyte homeostasis and lymphomagenesis in celiac disease. Gas-troenterology. 125, 730-45 (2003).
44. Monteleone, G. et al. Role of interferon alpha in promoting T helper cell type 1 responses in the small intestine in coeliac disease. Gut. 48, 425–9 (2001).
45. Mattei, F., Schiavoni, G., Belardelli, F. & Tough, D. F. IL-15 is expressed by dendritic cells in re-sponse to type I IFN, double-stranded RNA, or lipopolysaccharide and promotes dendritic cell activation. J. Immunol. 167, 1179-87 (2001). 46. Jabri, B. & Sollid, L. M. Tissue-mediated
con-trol of immunopathology in coeliac disease. Nat. Rev. Immunol. 9, 858-70 (2009).
47. Jabri, B. & Abadie, V. IL-15 functions as a dan-ger signal to regulate tissue-resident T cells and tissue destruction. Nat Rev Immunol. 15, 771-83 (2015).
48. Ricaño-Ponce, I. et al. Refined mapping of au-toimmune disease associated genetic variants with gene expression suggests an important role for non-coding RNAs. J. Autoimmun. 68, 62–74 (2016).
49. Zhernakova, D. V. et al. Identification of con-text-dependent expression quantitative trait loci in whole blood. Nat. Genet. 49, 139-145 (2017).
50. Ricaño-Ponce, I. et al. Immunochip meta-anal-ysis in european and argentinian populations identifies two novel genetic loci associated with celiac disease. Eur. J. Hum. Genet. (2019).
2Department of Cell Biology, University Medical Center Groningen, University of Groningen, the Netherlands.
3Depts. of Gastroenterology, Infectious Diseases and Rheumatology, Charité–Universitätsmedizin, Berlin, Germany.
