Multi-omic insights from a multi-ancestry genome-widemeta-analysis of ankylosing spondylitis ..., 2024, Brown et al.

T

tralfamadorian97

Established Member
Title: Multi-omic insights from a multi-ancestry genome-wide meta-analysis of ankylosing spondylitis reveal novel pathways of disease susceptibility

Authors:
Matthew Brown, Zhixiu Li, Xin Wu, Nicholas Harvey, Jose Garrido-Mesa, Xiaobing Wang, Zhihao Xu, Geng Wang, Helena Marzo-Ortega, Dennis McGonagle, Ann Morgan, Nurullah Akkoc, Gokce Kenar Artin, Gary Macfarlane, Gareth Jones, Linda Bradbury, Paul Leo, Kate Zimmerman, Emma Duncan, Julia Brown, Tony Merriman,Simon Stebbings, Mahdi Mahmoudi, Ahmadreza Jamshidi, Elham Farhadi, Nigil Haroon, Robert Inman, Maxime Breban, Gensler Lianne, Michael Ward, B. Wordsworth, Michael Weisman, David Evans, Tony Kenna, Tae-Hwan Kim, John Reveille, Huji Xu

Abstract:
We report the largest genome-wide association study meta-analysis in ankylosing spondylitis (AS) to date (25,645 cases, 71,224 controls), identifying 27 novel loci and 86 independent genetic associations. Variations in FUT2 (non-secretor status) and ABO (blood group A) increase AS risk, with Mendelian randomisation (MR) linking non-secretor status to increased AS risk from reduced gut carriage of Ruminococcus torques. Associations with three telomerase maintenance genes (TERT, TERC, RTEL1), and MR analysis, suggest increased telomere length causally increases AS susceptibility. Fine-mapping prioritised likely causal variants at multiple loci.

Transcriptome- and proteome-wide association studies implicated 644 genes, highlighting immune-related pathways. Lower genetically-determined IL-6 and IL-12, and similar IL-23, levels were found in AS cases, offering a genetic explanation for the failure of IL-6, IL-12, and IL-23 inhibition in AS treatment. Finally, multi-omic analyses showed chromosome 2p15 association acts via reduced B3GNT2 expression. These findings deepen understanding of AS pathogenesis, highlighting new pathways and therapeutic opportunities.

Link
 
Last edited by a moderator:

​

My Thoughts
This is not directly relevant to ME, but is perhaps a good recent example of the kind of insights a GWAS can generate

This paper reports the results of a large meta-GWAS for the rheumatological disease Ankylosing Spondylitis (AS). It also includes smaller proteome and transcriptome studies of AS.

I was impressed by the sheer variety of biologically interesting findings. Here are some examples that stood out:
  • A genetic variant that reduces TYK2 function is protective against AS. This finding may justify trials of TYK2 inhibitors in AS.
  • A genetic variant known to lead to retention of the Il-23 receptor in a cell’s endoplasmic reticulum is protective against AS. Since the Il-23 receptor is upstream of Il-17 production, this is consistent with the effectiveness of Il-17 inhibition in AS.
  • A genetic variant that affects certain T-cell costimulatory molecules is associated with increased AS risk, consistent with T-cell-centric models of AS pathogenesis.
There was also a section that used mendelian randomization to make an argument for a causative role for the microbiome in AS pathogenesis. I did not fully understand this section, but it seemed an interesting example of the broader utility of GWAS data.

I will be happy if DecodeME produces results of comparable quality and quantity. It will be particularly exciting if DecodeME identifies protective genetic variants that are suggestive of beneficial therapies.
 
Last edited:
An impressive range of genes involved.

What intrigues me is that we still do not have a clear explanation for the B27 link. Matt Brown and Paul Bowness and I had conversations 25 years ago on how to pin that down. Looking at B27 subtypes seemed an important option but I never heard more. Paul had found that B27 heavy chains could dimerise but that story never seemed to move forward.

Nevertheless, it became clear that cytokine/receptor interaction thresholds were enough to make the difference between symptomatic disease and wellbeing so good progress was made on treatment.
 
tralfamadorian97 said:
There was also a section that used mendelian randomization to make an argument for a causative role for the microbiome in AS pathogenesis. I did not fully understand this section, but it seemed an interesting example of the broader utility of GWAS data.
Click to expand...
If I'm interpreting it correctly, it looks like they evaluated several loci which are causal for presence of certain bacterial strains (determined by MR in another study). Consistent with another study, they found that that a particular mutation which reduces levels of Ruminococcus torques is associated with disease in HLA-B27+ AS cases compared to HLA-B27+ controls. So the claim is that absence of this particular strain increases risk of developing AS in people who have the HLA-B27 allele and have another mutation affecting secretion of ABO markers in mucosal tissue.

Since they only reported the p-value I don't know how strong of an association it is. It's interesting that the same finding came up twice, at least, but I won't put much stock in it at this point--it's a convoluted 3-way relationship without anything mechanistic to back it up.

I like the general direction it's going, though. 90% of AS cases have the HLA-B27 mutations, but it's also common in 10-25% of the population (depending on where you're looking) most of whom never develop disease. So clearly HLA-B27 is central to the disease somehow, but there must be an important secondary factor in addition to HLA-B27. If it is some unknown interaction with microbiota that predisposes to AS, this finding would be direct evidence against molecular mimicry theories, since it's the lack of Ruminococcus torques that increases AS risk.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom