Introduction Single nucleotide polymorphisms in ERAP2 are strongly associated with ankylosing spondylitis (AS). One AS-associated single nucleotide polymorphism, rs2248374, causes a truncated ERAP2 protein that is degraded by nonsense-mediated decay. Approximately 25% of the populations of European ancestry are therefore natural ERAP2 knockouts. We investigated the effect of this associated variant on HLA class I allele presentation, surface heavy chains, endoplasmic reticulum (ER) stress markers and cytokine gene transcription in AS.
Methods Patients with AS and healthy controls with either AA or GG homozygous status for rs2248374 were studied. Antibodies to CD14, CD19-ECD, HLA-A-B-C, Valpha7.2, CD161, anti-HC10 and anti-HLA-B27 were used to analyse peripheral blood mononuclear cells. Expression levels of ER stress markers (GRP78 and CHOP) and proinflammatory genes (tumour necrosis factor (TNF), IL6, IL17 and IL22) were assessed by qPCR.
Results There was no significant difference in HLA-class I allele presentation or major histocompatibility class I heavy chains or ER stress markers GRP78 and CHOP or proinflammatory gene expression between genotypes for rs2248374 either between cases, between cases and controls, and between controls.
Discussion Large differences were not seen in HLA-B27 expression or cytokine levels between subjects with and without ERAP2 in AS cases and controls. This suggests that ERAP2 is more likely to influence AS risk through other mechanisms.
- Ankylosing Spondylitis
- Gene Polymorphism
Statistics from Altmetric.com
Ankylosing spondylitis (AS) is an immune-mediated arthritis. Genetic variants in endoplasmic reticulum (ER) aminopeptidase (ERAP) 1 are associated with AS in multiple populations.1 The ERAP1 enzyme is a component of the major histocompatibility class (MHC) I antigen presentation pathway, and the AS-associated protective variants have been shown to be loss of function variants.2 ERAP1 has been shown to destroy ligands so higher ERAP1 enzyme activity can translate into fewer ligands.3 Recent work by Reeves et al that demonstrates poor ligand generation supports this, however their work was unable to demonstrate whether this was through poor generation of ligands or destruction of ligands by increased activity, because they didn't directly measure the enzyme cleavage by ERAP1.4 ,5
Recently, we reported the association of AS with variants in two additional aminopeptidases: ER aminopeptidase 2 (ERAP2) and puromycin-sensitive aminopeptidase (NPEPPS).6 ERAP2 has now also been shown to be associated with AS in HLA-B27-positive and HLA-B27-negative patients with AS.7 There are two variants in ERAP2 that are known to affect the function of the ERAP2 enzyme. The most functionally important variant is rs2248374; multiple independent groups have shown that the G allele of this single nucleotide polymorphism (SNP) causes reduced strength of a splicing site for exon 10 and that this consequently truncated mRNA is degraded by nonsense mediated decay.8 ,9 Homozygotes for this G allele (∼25% of white European descent), which is protective for AS, therefore express no ERAP2 protein. A second functional variant, rs2549782, causes a change in the peptide trimming preference of ERAP2, however this variant is almost always carried on the same haplotype as the null-allele encoding rs2248374 (D'=1.00, r2=0.90), this abnormally functioning enzyme is almost never produced.10
Previously reported experiments on the effect of ERAP1 knockdown in HeLa cells showed increased MHC class I expression.3 Mice deficient in ERAP associated with antigen processing, the only murine resident ERAP, show 20–70% reduction in MHC class I molecule expression on splenocytes.11 Peptidome analysis of human cell lines with different ERAP1 variants demonstrated that protective (loss of functional) SNPs present longer peptides and have lower stability of the HLA-B27:peptide complex.12
These results demonstrate that changes in genetic variants in ERAP1 affect antigen presentation. We have previously shown that ER stress is not changed by AS associated ERAP1 genetic variation, or HLA-B27 status.13 In this study we used the naturally occurring human ERAP2 functional knockout model (homozygous for the G allele at rs2248374) to determine the effect of changes in ERAP2 on cell surface MHC class I presentation, cell surface and ER heavy chain formation. These effects were investigated in monocytes and mucosal associated invariant T (MAIT) cells because of their potential important role in gut related immune regulations.14 ,15 In addition, as ER stress has been suggested to play a role in AS pathogenesis, we determined cytokine and ER stress marker levels in patients with AS deficient in ERAP2.1
Case and control samples
All cases were recruited from the University of Queensland AS clinic in Brisbane, Australia and were selected on the basis of their genotype. All patients met the modified New York criteria for AS, and all cases and controls were of white European descent. For flow cytometry experiments peripheral blood mononuclear cells (PBMCs) from 22 AS cases were used. For ER stress and cytokine expression studies cDNA and serum from 51 AS cases were used.
Ethical approval was obtained and written informed consent was received from all participants. Whole blood was taken and serum separated and stored at −80°C and PBMCs were extracted using Ficoll density gradient separation and stored under liquid nitrogen in fetal calf serum and 10% dimethyl sulfoxide as previously described.16 Sample genotypes were determined with the Illumina Immunochip.6 HLA-B27 typing was determined by using the rs116488202 HLA-B27 tag SNP.6
Patients with AS and controls were selected based on homozygosity for ERAP2 SNP rs2248374, either GG (ERAP2 protein absent, AS protective) or AA (ERAP2 protein present, AS risk), and further grouped according to HLA-B27 carriage or absence, making a total of four patient groups.
The following antibodies were sourced from Beckman Coulter: CD14Pacific Blue, CD19ECD, HLA-A,B,C Cy7-PE, Vα7.2 PE and CD161 PerCP-Cy5.5 were purchased from Biolegend. Class I heavy chains were assessed with the HC10 (IgG2a) antibody, a kind gift of Professor Paul Bowness, University of Oxford.17 HC10 was conjugated to FITC inhouse using the SureLINK Fluorescein (FITC) labelling Kit (Kirkegaard & Perry Laboratories Inc; KPL). HLA-B27 (ME1) antibodies were conjugated to AlexaFluor 647 using the Invitrogen AlexaFluor 647 Monoclonal Antibody Labelling Kit (Invitrogen). The conjugation efficiency (dye-to-antibody molar ratio) for HC10-FITC=3.2 fluorophores per immunoglobulin and EP4-AF647= 3.9 fluorophores per immunoglobulin.
PBMCs were prepared and stained as described previously.16 Monocytes were specifically analysed by gating on CD14+CD3-CD19-cells. MAIT cell analysis was performed by analysing Vα7.2+CD3+ cells. Intact HLA-B27 was identified by ME1 staining and MHC class I free-heavy chains by HC10 staining. HLA-B27 species were examined on the cell surface and intracellularly. For intracellular staining cells were fixed with fixation buffer and permeabilised in Perm/Wash buffer prior to staining at 4°C for 30 min. Dead cells were excluded using Aqua Live/Dead Fixable Dead Cell Stain kit (Invitrogen). To account for the intrinsic variability among human samples, the HLA-ABC mean fluorescence intensities (MFIs) were standardised relative to CD14 for monocytes and Vα7.2 for MAIT cells. Cells were acquired on a Gallios 10-colour-3 lasers (488 nm, 638 nm and 405 nm) Flow Cytometer (BeckmanCoulter), and staining was analysed using Kaluza software (BeckmanCoulter).
To assess whether AS-associated ERAP2 variants alter levels of ER stress, we compared the relative expression of ER stress genes GRP78 and CHOP in patients carrying either the risk or protective genotype of rs2248374. Gene expression was normalised to the expression of the constitutive ribosomal gene RPL32.
All data are presented as the mean±SEM. Data were compared by the Mann-Whitney test for non-parametrical data using GraphPad Prism 5. Power calculations were completed in R package ‘pwr’.
We have power in excess of 80% to detect 10% differences in gene expression and cytokines levels between genotypes. We also have excellent power (>80%) to detect 20% differences in total HLA class I levels on monocytes (see online supplementary table S1). We have reduced power (<80%) to detect differences in cell surface markers on MAIT cells, or at modelled differences in effect size less than 20%.
The AS risk associated, rs2248374 allele of ERAP2 does not alter cell-surface HLA-Class I antigen presentation
Comparing rs2248374 AA and GG genotype carriers, there were no significant differences in HLA-Class I expression or free HLA-Class I heavy chain levels either extracellularly (figure 1 and see online supplementary table S2) or intracellularly (not shown). The ME1:HC10 ratio, which corrects for total intact HLA-B27, was not altered by the ERAP2 genotype (data not shown).18 Total class I (HLA-ABC) expression was also unaffected by the ERAP2 genotype. Standardised HLA-ABC MFIs were not affected by the ERAP2 genotype (figure 1). While HLA-B27 expression was higher in cases than in controls, no difference in HLA-ABC or free HLA Class I heavy chain levels was noted between AS cases and healthy controls in any cell type studied. In HLA-B27-negative patients there again were no differences observed (figure 1).
rs2248374 risk allele of ERAP2 does not lead to altered levels of ER stress
There were no differences in the relative expression levels (ΔCT) of GRP78 or CHOP between AS HLA-B27+ ERAP2−/−, HLA-B27− ERAP2−/−, HLA-B27+ ERAP2+/+ and HLA-B27− ERAP2+/+ individuals (figure 2A, B and see online supplementary table S3).
Proinflammatory cytokine gene expression is unaffected by ERAP2 variants
IL17A, IL22, tumour necrosis factor (TNF), and IL6 gene expression levels were not different between patients with AS that were HLA-B27-positive ERAP2 AA or GG, or HLA-B27-negative ERAP2 AA or GG (figure 2C–E and see online supplementary table S3).
We have demonstrated that ERAP2 variation does not have a significant effect on PBMC intracellular and cell surface HLA-class I expression and heavy chain formation, markers of ER stress or inflammatory cytokine production.
Numerous theories have been proposed as to the aetiolopathogenesis of AS and the mechanism by which HLA-B27 operates to cause the disease.19 These encompass effects of HLA-B27 on antigen presentation, and on ER stress and the interaction of heavy chains and homodimers with innate immune cells. It has also been proposed that aminopeptidases like ERAP1, ERAP2 and NPEPPS may alter the presented ligands by destroying key ligands, or altering the ligand volume.14 The current study suggests that in AS cases, absence of functional ERAP2 does not influence HLA-B27 expression or ER-stress, or affect the transcription of key AS-associated cytokines. This leaves the potential for its influence to be through other mechanisms such as ligand alteration. The type and volume of ligand could potentially be altered by changing aminopeptidase function.12
Relevant to the results of this study, ERAP2 variants may only be functionally relevant in specific cell types or tissues, in particular immunological environments, or in the presence of certain micro-organisms or antigenic peptides. Further work is required to fully investigate the ERAP2 association in AS, including investigation of different cell types and immunological milieu.
The authors thank the subjects who participated in this study.
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Data supplement 1 - Online tables
Handling editor Tore K Kvien
MAB and TJK contributed equally.
Contributors Study design: PCR, TJK, MAB, GPT; Patient recruitment: PCR, LAB, MAB; Data collection: PK, EL, MCL, TJK Writing and approving the paper: PCR, EL, MCL, PK, LAB, GPT, TJK, MAB.
Funding This study was supported by Australian National Health and Medical Research Council (NHMRC) grant #569830. TJK was supported by an AFA-ARA Heald Fellowship, PCR was supported by a NHMRC scholarship and the ARA-RACP-Starr Fellowship. MAB was supported by NHMRC Senior Principal Research Fellowship APP1024879.
Competing interests None declared.
Ethics approval Metro South Health District/University of Queensland.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement All data collected was presented except for the parts with ‘(not shown)’ stated. All this information not shown is available to anyone who requests it.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.