Junk food or junk genome?

As evidence pointing to the importance of non-coding regions of the genome mounts, a team from Spain discover a long non-coding RNA which plays a pivotal role in coeliac disease

Coeliac disease (CD) is an immune-mediated enteropathy caused by ingested gluten that develops in genetically susceptible individuals. Nowadays, it is estimated that around 1% of the European population suffers from this disease. The main genetic factor – contributing 40% of the genetic susceptibility – maps to the HLA (human leukocyte antigen) region. However, HLA variants (DQ2 and DQ8) are not sufficient to develop coeliac disease. As with other complex diseases, in which more than one gene is implicated, the combination of many common variants together with environmental factors will make some of us prone to develop coeliac disease.

[caption id="attachment_56367" align="alignnone" width="620"] Coeliac disease affects around 1% of the European population.[/caption]

The completion of the human genome project has given us new tools to find variants that contribute to the development of common diseases like CD. One of those tools is the Genome Wide Association Studies (GWAS) that look for markers across the genome in order to find genetic variations associated with a particular disease. Around 80% of the phenotype-related loci identified by GWAS of many human diseases map to non-coding regions of the genome, so explaining why the implication of these variants in disease development has been challenging.

In the last few years a new group of RNAs, named long non-coding RNAs (lncRNAs) have been shown to be differentially expressed in several diseases and are gaining importance as functional players in their development. LncRNAs have been implicated in several aspects of cell function, including the control of DNA transcription or the response to stress, and are emerging as an important regulation layer in cell signalling. Interestingly, the link between the GWAS phenotype-related loci on lncRNA expression or function, and its implications for disease, remain uncharacterised, and open a new field of study around the functionality of GWAS hits in intergenic regions.

As with other complex diseases, in which more than one gene is implicated, the combination of many common variants together with environmental factors will make some of us prone to develop coeliac disease

To date, around 40 genomic regions associated with the risk of developing coeliac disease have been described in several GWAS studies, and many of the markers are located in noncoding regions of the genome¹. Taking into account the abundance of lncRNAs in the genome, it is meaningful to hypothesise that some of these variants may lie within functional lncRNAs and thus have an impact in the yet unexplored part of the genome. One of those intergenic single nucleotide polymorphisms (SNPs), rs917997, located 1.5 kb downstream of the protein-coding gene IL18RAP (that has been related to coeliac disease and other autoimmune pathologies) caught our attention. The equivalent region in mice presents an actively transcribed lncRNA (lnc13), so we decided to analyse this region in human cells and found transcription of human lnc13, thus discovering a novel non-coding human gene that harbours a coeliac disease associated variant, and confirming our primary hypothesis.

Once we knew that the genomic part harbouring the SNP is transcribed in humans, we needed to rule out the possibility that it was just noise from the neighbouring IL18RAP coding gene. Although both genes overlap and are transcribed from the same DNA strand, we confirmed that transcription of lnc13 and IL18RAP are independent from each other. Moreover, IL18RAP expression is induced in biopsies of coeliac disease patients and in macrophages treated with an immune response stimulator. However, under the same conditions, we found that the expression of lnc13 is reduced. Using cells from lnc13 knockout mice we could conclude that the observed reduction of lnc13 is dependent on the transcription factor NFkB, which has been known for a long time to be constitutively active in the intestinal epithelium of coeliac patients (even on a gluten-free diet)². We now have a novel lncRNA that not only harbours a disease-associated variant but that could also be implicated in the immune response, due to its relationship with the key immune regulator NFkB.

We observed that lnc13 levels were negatively correlated with the levels of various inflammatory genes, some of them significantly overexpressed in intestinal tissue from coeliac patients

The decreased lnc13 found in coeliac patients made us wonder about the mechanism responsible for its degradation. We saw that activated macrophages showed an induction of the protein Dcp2, a negative regulator of RNA stability. Using mice knockouts for Dcp2 we were able to confirm that this protein is involved in the destabilisation and consequent low levels of lnc13. Moreover, coeliac patients also present increased levels of Dcp2, due to the constitutive activation of NFkB, which in turn explains the reduction of lnc13. We were able to confirm this interaction using the newly discovered CRISPR-Cas9 gene editing technology, which was used to make Dcp2 deficient or defective cells. These cells were not able to degrade lnc13 in response to NFkB activation. This step of our research helped us analyse the upstream events of lnc13 degradation.

At this point we still did not know if the low levels of lnc13, and the associated polymorphism, have any involvement in disease development or they were just a consequence of inflammation. We observed that lnc13 levels were negatively correlated with the levels of various inflammatory genes, some of them significantly overexpressed in intestinal tissue from coeliac patients. Several experimental approaches allowed us to confirm that the function of lnc13 is to inhibit the expression of this subset of inflammatory genes in cells in which the immune response is not active. As had been previously observed for other known lncRNAs, lnc13 is mainly located in the cell nucleus and silences gene expression by binding to chromatin factors?,?. Specifically, lnc13 forms a complex with the protein hnRNPD (heterogeneous nuclear ribonucleoprotein D) and the enzyme histone deacetylase 1 (HDAC1). In non-stimulated cells, the repressive complex formed by lnc13 is localised to the promoters of the inflammatory genes, keeping them transcriptionally inactive. We were able to show how lnc13, hnRNPD and HDAC1 are bound to the promoters of certain genes but it is still unclear whether this lncRNA can target more genes and where the specificity of the binding lies.

Knowing how lnc13 works, we wanted to assess the influence of the SNP on its function. We already knew that the activity of this lncRNA is controlled by its interaction with the protein hnRNPD. The secondary structure of the lncRNA was predicted to change according to the genotype of the associated SNP, so we tested whether this interaction is dependent on the SNP allele. We found that in the presence of the risk allele rs917997*T, binding of the lncRNA to the protein is weaker, which in turn could reduce its ability to regulate the expression of target genes. People that carry the risk allele will have higher basal levels of certain coeliac-related inflammatory genes and this will give them certain predisposition to develop the disease.

This discovery underlines the importance of the three-dimensional structure of RNA molecules, and suggests that small changes in RNA sequence can have an important functional impact. Moreover, it demonstrates how variants that are intergenic can affect the function of a noncoding transcript, and opens the door to a new field of study for those variants that are located between coding genes. In brief, this study shows how lnc13, a lncRNA that was previously unknown, is altered in coeliac disease and depicts how a SNP that is associated with a complex disease can directly affect the function of a lncRNA. Other SNPs that have been associated with risk of developing common diseases are likely to function in a similar way but we cannot yet predict how these non-coding RNA molecules function. Further experimental studies on lncRNA function will help decipher the molecular principles of these RNAs and will allow the design of new therapeutic approaches for other complex genetic diseases.

References 1. Trynka, G. et al. Dense genotyping identifies and localizes multiple common and rare variant association signals in coeliac disease. Nat Genet 43, 1193-1201, doi:10.1038/ng.998 (2011). 2. Fernandez-Jimenez, N. et al. Coregulation and modulation of NFkappaB-related genes in coeliac disease: uncovered aspects of gut mucosal inflammation. Hum Mol Genet 23, 1298-1310, doi:10.1093/hmg/ddt520 (2014). 3. Li, Y., Song, M. G. & Kiledjian, M. Transcript-specific decapping and regulated stability by the human Dcp2 decapping protein. Mol Cell Biol 28, 939-948, doi:10.1128/MCB.01727-07 (2008). 4. Huarte, M. et al. A large intergenic noncoding RNA induced by p53 mediates global gene repression in the p53 response. Cell 142, 409-419, doi:10.1016/j.cell.2010.06.040 (2010). 5. Carpenter, S. et al. A long noncoding RNA mediates both activation and repression of immune response genes. Science 341, 789-792, doi:10.1126/science.1240925 (2013).

Authors

Jose Ramon Bilbao is Associate Professor of Medical Genetics at the University of the Basque Country School of Medicine.

Ainara Castellanos-Rubio did her postdoc about lncRNA involvement in autoimmunity at Columbia University, and now is a Juan de la Cierva researcher at University of the Basque Country.