A disease-associated gene desert directs macrophage inflammation through ETS2 [in IBD] 2024 Stankey et al

[Andy]

Abstract

Increasing rates of autoimmune and inflammatory disease present a burgeoning threat to human health1. This is compounded by the limited efficacy of available treatments1 and high failure rates during drug development2, highlighting an urgent need to better understand disease mechanisms. Here we show how functional genomics could address this challenge.

By investigating an intergenic haplotype on chr21q22—which has been independently linked to inflammatory bowel disease, ankylosing spondylitis, primary sclerosing cholangitis and Takayasu’s arteritis3,4,5,6—we identify that the causal gene, ETS2, is a central regulator of human inflammatory macrophages and delineate the shared disease mechanism that amplifies ETS2 expression. Genes regulated by ETS2 were prominently expressed in diseased tissues and more enriched for inflammatory bowel disease GWAS hits than most previously described pathways. Overexpressing ETS2 in resting macrophages reproduced the inflammatory state observed in chr21q22-associated diseases, with upregulation of multiple drug targets, including TNF and IL-23. Using a database of cellular signatures7, we identified drugs that might modulate this pathway and validated the potent anti-inflammatory activity of one class of small molecules in vitro and ex vivo. Together, this illustrates the power of functional genomics, applied directly in primary human cells, to identify immune-mediated disease mechanisms and potential therapeutic opportunities.

Open access, https://www.nature.com/articles/s41586-024-07501-1

 

[Andy]

Articles:

Bowel disease breakthrough as researchers make ‘holy grail’ discovery

https://www.theguardian.com/society...ease-hope-researchers-find-biological-pathway


Major cause of inflammatory bowel disease found

https://www.bbc.co.uk/news/articles/c1wwdd6v2wjo

 

[Kitty]

Increasing rates of autoimmune and inflammatory disease present a burgeoning threat to human health1. This is compounded by the limited efficacy of available treatments1 and high failure rates during drug development2, highlighting an urgent need to better understand disease mechanisms. Here we show how functional genomics could address this challenge. By investigating an intergenic haplotype on chr21q22—which has been independently linked to inflammatory bowel disease, ankylosing spondylitis, primary sclerosing cholangitis and Takayasu’s arteritis3,4,5,6—we identify that the causal gene, ETS2, is a central regulator of human inflammatory macrophages and delineate the shared disease mechanism that amplifies ETS2 expression.

https://www.nature.com/articles/s41586-024-07501-1

Also reported in The Guardian (UK):

Bowel disease breakthrough as researchers make ‘holy grail’ discovery

https://www.theguardian.com/society...ease-hope-researchers-find-biological-pathway

 

[Kitty]

Sorry for the duplicate post, I did a search before adding it but obviously picked the wrong terms.

 

[Hutan]

Resolving the biology at these loci is a formidable task, as the same DNA sequence can function differently depending on the cell type and/or external stimuli9. Most non-coding variants are thought to affect gene regulation13, but difficulties identifying causal genes, which may lie millions of bases away, and causal cell types, which may only express implicated genes under certain conditions, have hindered efforts to identify disease mechanisms. For example, although genome-wide association studies (GWASs) have identified over 240 IBD risk loci3, including several possible drug targets, fewer than 10 have been mechanistically resolved.

So, the combination of a susceptible gene, and the non-coding variant that regulates that gene and, possibly, certain environmental conditions are needed to result in disease.

One notable example is an intergenic region on chromosome 21q22 (chr21q22), where the major allele haplotype predisposes to five inflammatory diseases3,4,5,6. Such regions, which were originally termed ‘gene deserts’ owing to their lack of coding genes, often contain GWAS hits but are poorly understood. To test for a shared disease mechanism, we performed co-localization analyses and confirmed that the genetic basis for every disease was the same, meaning that a common causal variant(s) and a shared molecular effect was responsible

Gene desert = sections without coding genes, but it's increasingly understood that they aren't just junk DNA, but include gene regulation instructions

@Andy, please remind me, will DecodeME initially look at non-coding regions?

As these heterogeneous diseases are all immune mediated, we reasoned that this locus must contain a distal enhancer that functioned in immune cells. By examining H3K27ac chromatin immunoprecipitation–sequencing (ChIP–seq) data, which marks active enhancers and promoters, we identified a monocyte/macrophage-specific enhancer within the locus (Fig. 1b). Monocytes and macrophages have a key role in many immune-mediated diseases, producing cytokines that are often targeted therapeutically14.

 

[Andy]

@Andy, please remind me, will DecodeME initially look at non-coding regions?

Simple answer is I don't know if we will or won't. I suspect that as we are looking across the whole genome, some of our results will come from what are currently considered to be non-coding regions, but I have asked the question of the team, and will come back with an informed answer.

 

[rvallee]

Hmm, no I don't think they're doing it right. Where are the questionnaires that ask overlapping questions, showing how it's all anxiety, and the dubious biased trials showing how it's an "aw shucks, it's gonna be one of those days isn't it?" attitude in the morning that makes the bowels feel like they hurt?

Did they even consider the mind-body-brain-Flying spaghetti monster connection? Not even a finger-painting group therapy where they explore their thoughts and beliefs about tubes and how it's what makes their tummies hurt?

One day, the biopsychosocial model of illness will have its first ever breakthrough. That day isn't today, or ever. But one day it will show its usefulness. You'll see!

 

[Hutan]

Hmm, no I don't think they're doing it right.

Reading this paper makes me more hopeful. I don't understand a lot of this, but there's a logic to the process, things were approached from different angles to check ideas still made sense, and they clearly have great tools. It's a million miles away from the 'fill the gaps in knowledge with beliefs that the patient just needs to try harder to be well'.

ETS2 is an ETS-family transcription factor and proto-oncogene21, but its exact role in human macrophages is unclear, with previous studies using either cell lines or complex mouse models and assessing a limited number of potential targets22,23,24,25,26. This has led to contradictory reports, with ETS2 being described as both necessary and redundant for macrophage development27,28, and both pro- and anti-inflammatory22,23,24,25,26.

That paragraph illustrates the fact that, as scientific investigation gets better, conclusions change. Just because something wasn't found last year, doesn't mean that nothing will be found next year.

 

[Jonathan Edwards]

That paragraph illustrates the fact that, as scientific investigation gets better, conclusions change. Just because something wasn't found last year, doesn't mean that nothing will be found next year.

Hmm. It. might indicate that in certain areas, like this one, people are constantly drawing inappropriate conclusions from in vitro studies that get fashionable for a while and then get replaced by another fashion.

I have not read the full article but the abstract reads like an advert for these researchers' favourite molecule causing everything. I don't see any crucial new data being presented. It has been clear for 30 years that these illnesses involve genetic thresholds for inflammation (and we know the dominant ones to be different for each condition). Finding expression of a protein in tissue is a very weak piece of evidence, since inflamed tissue tends to express a whole bang shoot of things for nonspecific reasons.

 

[Jonathan Edwards]

Increasing rates of autoimmune and inflammatory disease present a burgeoning threat to human health1.

• It is a pity that their opening sentence is a reference to a putative increase in autoimmune disease when the conditions they are reporting on are not autoimmune diseases. You have to be fairly muddled to make a mistake like that.

 

[Hutan]

Nearly 5% of humans live with an autoimmune or inflammatory disease. These heterogeneous conditions, ranging from Crohn’s disease and ulcerative colitis (collectively inflammatory bowel disease (IBD)) to psoriasis and lupus, all require better therapies, but only 10% of drugs entering clinical development ever become approved treatments2. This high failure rate is mainly due to a lack of efficacy8 and reflects our poor understanding of disease mechanisms. Genetics provides a unique opportunity to address this, with hundreds of loci now directly linked to the pathogenesis of immune-mediated diseases9. Indeed, drugs that target pathways implicated by genetics have a far higher chance of being effective10.

You may well be right that this paper doesn't have the answer to IBD etc, although I still like the effort made to find a solution. I think they are quite careful not to call IBD an autoimmune disease.

I think the point they are making in that paragraph above is that genetics can help with the understanding of both types of disease and with making better guesses about treatments.

 

[Hutan]

Step 1

To test for a shared disease mechanism [for Crohns, UC, ankylosing spondylitis, PS cholangitis, Takayasu's arteritis], we performed co-localization analyses and confirmed that the genetic basis for every disease was the same, meaning that a common causal variant(s) and a shared molecular effect was responsible

- see Fig1a - the blue spike



Step 2
They viewed all five of the disease as immune mediated, so, they looked for genetic enhancers acting on immune cells

By examining H3K27ac chromatin immunoprecipitation–sequencing (ChIP–seq) data, which marks active enhancers and promoters, we identified a monocyte/macrophage-specific enhancer within the locus (Fig. 1b).

Step 3
Find the gene regulated by the enhancer

Using promoter-capture Hi-C and expression quantitative locus (eQTL) data from human monocytes (Methods), we found that the disease-associated locus physically interacts with the promoter of ETS2—the most distant candidate gene (around 290 kb away)—and that the risk haplotype correlates with higher ETS2 expression (Fig. 1c).

(I'm not sure how they concluded the following as a supporting fact , but there is a figure explaining it that I haven't looked at )

Indeed, increased ETS2 expression in monocytes and macrophages, either at rest or after early exposure to bacteria, was found to have the same genetic basis as inflammatory disease risk (Extended Data Fig. 1c)

Step 4
Confirmation that ETS2 was causal

To directly confirm that ETS2 was causal, we used CRISPR–Cas9 to delete the 1.85 kb enhancer region in primary human monocytes before culturing these cells with inflammatory ligands, including TNF (a pro-inflammatory cytokine), prostaglandin E2 (a pro-inflammatory lipid) and Pam3CSK4 (a TLR1/2 agonist)

So, they knocked out the enhancer, stimulated the monocytes and looked to see what impact that had on the three possible genes that they thought might be involved.

we used PrimeFlow to measure the dynamics of mRNA expression and detected increased levels of all three genes (ETS2, BRWD1 and PSMG1) after TPP stimulation of unedited monocytes (Fig. 1e). Deletion of the chr21q22 enhancer did not affect BRWD1 or PSMG1 expression, but the upregulation of ETS2 was profoundly reduced (Fig. 1f), confirming that this pleiotropic locus contains a distal ETS2 enhancer.

Without the enhancer, expression of two of the genes was unchanged, the upregulation of ETS2 was 'profoundly reduced'.

I've got to stop there, but, it continues on in the same way with a lot more analyses. e.g.

Examining the genes that were consistently downregulated in ETS2-edited macrophages (adjusted P (Padj) < 0.05 for both gRNAs), we identified over 20 IBD-risk-associated genes[from GWAS], including many thought to be causal at their respective loci3,33(Extended Data Table 1).

As I said, I liked the logical way they seemed to approach the problem. I don't have the background to evaluate the methods, but the way it was written inspired my confidence in them. And they propose drug types that they believe are worth testing, so there are clear next steps.

 

[EndME]

You may well be right that this paper doesn't have the answer to IBD etc, although I still like the effort made to find a solution. I think they are quite careful not to call IBD an autoimmune disease.

I think the point they are making in that paragraph above is that genetics can help with the understanding of both types of disease and with making better guesses about treatments.

I also understood the abstract and perhaps the whole paper to be more of an advertisement for functional genomics rather than anything else and that the authors don't call IBD an autoimmune disease. I suppose they might be applying for grants in this direction.

Apart from the genetic work, my naive understanding is that the authors implicated that pharmacologically MEK inhibition could be fruitful, but that they didn't actually produce any meaningful data. But their approach to from GWAS to pharmacological targeting seems sensible to me.

The authors seem to implicate that once the GWAS work is done, functional genomics can be helpful in understanding those leads and what role non-coding genetic associations could play. I can't claim to understand any of what is going on, but perhaps an avenue to possibly explore in Project ideas for DecodeME data if DecodeME unveils something.

 

[Andy]

@Andy, please remind me, will DecodeME initially look at non-coding regions?

Simple answer is I don't know if we will or won't. I suspect that as we are looking across the whole genome, some of our results will come from what are currently considered to be non-coding regions, but I have asked the question of the team, and will come back with an informed answer.

My suspicion was correct, the Genetics Delivery Team confirm that DecodeME will "perform a genome-wide association study (GWAS) which covers both coding and non-coding regions of the genome (the later representing more 98% of the genome) rather than looking at specific regions. Actually, the majority of GWAS signals are located in non-coding regions. If we found a significant association within a region , coding or not, it will be investigated following the post-GWAS analysis plan."

 

[Hutan]

I also understood the abstract and perhaps the whole paper to be more of an advertisement for functional genomics

Thanks EndME, yes.

My suspicion was correct, the Genetics Delivery Team confirm that DecodeME will "perform a genome-wide association study (GWAS) which covers both coding and non-coding regions of the genome (the later representing more 98% of the genome) rather than looking at specific regions. Actually, the majority of GWAS signals are located in non-coding regions. If we found a significant association within a region , coding or not, it will be investigated following the post-GWAS analysis plan."

Fantastic @Andy, it's getting exciting.

 

[Jonathan Edwards]

I think the point they are making in that paragraph above is that genetics can help with the understanding of both types of disease and with making better guesses about treatments.

Yes, well we have known that since the 1970s, when Brewerton sneakily made the first report of HLA-B27 being associated with Ank Spond. (he had seen some other people's data it seems).

In the paragraph they say "This high failure rate is mainly due to a lack of efficacy8 and reflects our poor understanding of disease mechanisms."

In reality the inefficacy mostly reflects very blinkered thinking by the fashionable dogma. The problem was not 'guesses' about treatments but organised PR campaigns by people, also with interests in grant money. Dozens of anti-T cell therapies were thrown at RA based on a theory that misinterpreted some data of mine on T cells being present in joints. It took me another five years to work out why it was B cells but it was obvious that nothing pointed to T cells.

And when we worked out why an anti-B cell approach might be good, it was based on detailed analysis of tissue specific control of macrophage maturation-exactly what these people think is a wheel they have re-invented, except that we had a logical sequence of arguments as to why specific pathways would link up. All these people seem to have done is measure some mediators in tissue (like my original observation of some T cells in tissue) and done some mouse manipulation, which doesn't look to me to be specific enough to be very helpful.

Point being that you need to understand the genetics and have a very detailed understanding of tissue processes. And we knew that in 1995. I wrote a review in Immunology Today about the role of tissue specific events in pathogenic mechanisms in Ank Spond that I suspect is still valid. These people seem to be blissfully ignorant of such things.

I put in grants for looking at the non-coding genetic basis for connective tissue cell maturation pathways in 1996. Nobody was interested because it wasn't T cells. The regulator they are interested in may well be part of the story but without taking note of all the stuff we already know I would be sceptical that it changes things that much. If it is relevant to half a dozen different diseases in different ways I suspect that it would be more useful to target the things that are specific to each disease rather than block some very basic macrophage maturation event, but maybe it will prove useful. I am not sure that blocking transcriptional regulators directly is that easy though.

Maybe they are just finally getting around to catching up with what happened 25 years ago!! I am just off to a lecture on killing B cells with CAR-T cells again. Same story, just now in fashion!!