2014 FC CALL FOR PROPOSALS

PUBLICATION OF COMPLETED STUDIES



Grant FC 014/2014 – Correlation between HLA-DQ gene dosage, expression level of DQ alpha and beta chains and gluten-specific T cell response 

PI: Giovanna Del Pozzo
Address: Istituto di Genetica e Biofisica-CNR, Napoli


Summary of the Project 

Grant: FC 014/2014

Title: Study of gene expression of HLA-DQ alleles associated to celiac disease in anti-gluten CD4+T cell immune response

Topic: Genetic and Immunology of Celiac Disease

Duration: 15 months

Principal Investigator: Giovanna Del Pozzo, Istituto di Genetica e Biofisica del CNR di Napoli (IGB-CNR)

Partnerships: Carmen Gianfrani, Istituto di Biochimica delle Proteine del CNR (IBP-CNR)

Pubblications: J. of Autoimmunity 2016 Jun;70:63-72


Project rationale and aims

We have investigated the ability of professional Antigen Presenting Cells (APC) such as B lymphocytes (B-LCL) and dendritic cells (DC), carrying DQ2.5 genes either in homozygosis (DR3/DR3) or heterozygosis (DR1/DR3 and DR3/ DR5) to present gluten peptides to specific CD4 + T cells in relation to the mRNA and protein expression of each DQA1*05 and DQB1*02 gene. The hypothesis was that the level of stable gliadin peptide-DQ2.5 complexes on the APC surface, a key step to activate the inflammatory cascade in CD patients, may reflect not only the genetic configuration but also the effective DQA1*05 and DQB1*02 gene expression.


dal_pozzo_eng_1.jpg 

FIGURE 1 – The current gene dosage model suggests that the celiac disease risk and seriousness of pathology  are determined by the HLA class II  asset of risk genes ( homozygous versus heterozygous): this model does not explain some discrepancies found in the classification of patients into different risk classes. Our study aimed to clarify and complete the model of genetic risk by the measurements of the HLA-DQ2.5 proteins, on the surface of antigen presenting cells, and by the quantification of messengers DQA1*05 and DQB1*02 encoding the surface molecule.  

Research plan and results obtained 

1) We measured the activation of gluten peptide-specific CD4+ T cells, using as APCs either DQ2.5 homozygous (DR3/DR3) and heterozygous (DR1/DR3 and DR3/DR5) B-LCLs or DCs, obtained from either celiac and healthy controls.
The results demonstrate that the activation threshold of an anti-gluten CD4+ T cell response in subjects with CD does not strictly correlate with the DQ2.5 genotype of professional APC. These experiments strongly suggested that the achievement of the activation threshold of pathogenic CD4+ T cells is primarily dependent on the concentration of gluten immunogenic peptides.

2) We assessed the effective surface amount of DQα1*05 and DQβ1*02 single chains by specific monoclonal antibodies in B-LCLs from homozygous and heterozygous patients. These results clearly demonstrated that similar capabilities of DQ2.5 homozygous and heterozygous APCs to activate anti-gluten CD4+ T cells are due to a comparable surface expression of DQα1*05 and DQβ1*02 proteins that form the DQ2.5 heterodimer.

3) We have measured the mRNA amounts of CD-associated DQA1*05 and DQB1*02 alleles and of HLA alleles not associated with the disease. In B-LCLs and DCs heterozygous for DQA1*05, the percentage of DQA1*05 mRNA was higher than the percentage of other DQA1*01 mRNA non-CD associated and not significantly different compared to that expressed by homozygous cells. Similar results were obtained for DQB1*02 mRNA which amount is greater compared to the expression of DQB1*03, *05 and *06 mRNA of heterozygous cells.

These data demonstrated that in all types of APCs heterozygous for DQA1*05 and DQB1*02 genes, from both healthy and celiac subjects, the relative expression of risk alleles is surprisingly higher than that predicted by the genetic configuration. This molecular finding explains the enhanced surface expression of HLA-DQ2.5 heterodimer with respect to other HLA dimers. 

Experimental design and methodologies 

For functional tests we used B-LCLs or DCs as APC to present specifi gliadin peptides to gliadin-specific T cell lines and T cell clones. The CD4+T cells activation was  assessed trough IFN-γ production by ELISA test and cell proliferation assay by [3H]thymdine incorporation.
The surface expression of DQα1*05 and DQβ1*02 chains was determined by labelling with specific monoclonal antibodies and flow cytometry analysis. The mRNA quantization was performed on B-LCL and DC, after assessment of patient genotype ,by real time PCR( qRT-PCR), using allele-specific primers.

Potential pitfalls and caveats

The results of our project are accepted for publication on J of Autoimmunity. In the next future we plan to focus on the translational aspect of our results. We will work to design a kit having dual applications: 1) the genotyping of patients, necessary  to identify the  predisposition to pathology; 2) the quantization of DQA1*05 and DQB1*02 mRNA expression, as marker for a prevision of  risk level to develop CD.

CONCLUSIONS AND DISCUSSION

A direct correlation between the genetic configuration of DQ2.5 haplotypes, the expected amount DQ2.5 heterodimer and the magnitude of T cell response to gluten has been demonstrated in previous studies. However, the “gene dosage theory”, does not fully explain the correlation of predisposition of CD among certain HLA genotypes and the pathogenic CD4+ T cell reactivity.
Here, we demonstrated that HLA DQA1*05 and DQB1*02 gene expression is much higher than expression of non-CD-associated genes. This influences the protein levels and causes a comparable cell surface exposure of DQ2.5 heterodimers between DQ2.5 homozygous and heterozygous celiac patients. As a consequence, the magnitude of the anti-gluten CD4+ T cell response is strictly dependent on the antigen dose and not on the DQ2.5 gene configuration of APCs. Furthermore, our findings support the concept that the expression of DQ2.5 genes is an important risk factor in celiac disease. The preferential expression of DQ2.5 alleles provides a new functional explanation of why these genes are so frequently associated with celiac disease and with other autoimmune disorders.
dal_pozzo_eng_2.jpg 

FIGURE 1 – The current gene dosage model suggests that the celiac disease risk and seriousness of pathology  are determined by the HLA class II  asset of risk genes ( homozygous versus heterozygous): this model does not explain some discrepancies found in the classification of patients into different risk classes. Our study aimed to clarify and complete the model of genetic risk by the measurements of the HLA-DQ2.5 proteins, on the surface of antigen presenting cells, and by the quantification of messengers DQA1*05 and DQB1*02 encoding the surface molecule.  

 

top