Has anyone used Promethease to analyse DNA results in the determination of the IL23R and ERAP1 status?

I recently got my results and uploaded them to Promethease. I'm still trying to decipher the results but of the genes known to be associated with AS, I fall into the BAD Repute of IL23R across the board however GOOD in all 4 of the ERAP1 results.

Under rs7743761, the results show 20x increased risk of Ankylosing Spondylitis however the gene this falls under this SNP (I believe they call it) is DHFRP2 and shows diseases associated with DHFRP2 include systemic lupus erythematosus (per but no mention of AS on that site (nor have I seen DHFRP2 anywhere in relationship to AS).

Does anyone have light they could shed on understanding this and how (or if) it has relevance to the diagnosis of Spondylitis.

I ordered the kit and when the raw data was available, I uploaded it to Promethease which gave me the SNPs and Allele and genotype.
For instance, I am showing a bad repute for SNP rs7743761(A;A) which is defined as:

20x increased risk of Ankylosing Spondylitis. Greatly increased odds (but still only about 2%) of this disease. But age of onset is usually between 15 and 25, so if you are older than that, there's not much need for worry. Ankylosing Spondylitis is a rare autoimmune disease that causes inflammation of the spine and joints. Based on preliminary research.

This is attached to a gene by the name of DHFRP2 which I had never heard of.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 08/22/15 03:55 PM
I ran my snp data through Promethease a couple of years ago.

I've got the 4.6x risk allele for that SNP -;C)

For the ERAP1 related ones I've got 2 higher risk, 3 lower risk.
For the IL23R related ones, I've got 3 higher, 3 normal risk.

I posted some other info at the time here -
So would you say since both of the subset genes were identified in your profile (and mine) that it further confirms the diagnosis of AS (as per the pubmed below)?

I'm not sure if I'm reading it right to assume having the presence of the two substantiates the diagnosis or what. A little mixed with the findings and conclusion. I know I read somewhere that the two were linked directly to AS.

Under the IL23R my results were:

rs10889677(A;C) -- 1.5x increased risk for certain autoimmune diseases; 2x increased risk for Graves disease
rs11209026(G;G) -- Normal, but higher risk for certain autoimmune diseases.
rs1004819(C;T) -- 1.5x risk
rs11209032(A;G) -- 1.3x higher risk for spondylitis
rs1495965(A;G) -- 1.2x higher risk for spondylitis
rs1343151(C;C) -- normal risk
rs11465804(T;T) -- normal risk

Under ERAP1:

rs10050860(C;T) -- 0.71x lower risk for Ankylosing Spondylitis.
rs30187(C;C) -- Normal low risk (0.1%) for ankylosing spondylitis.
rs17482078(C;T) -- 0.76x lower risk for spondylitis
rs2287987(C;T) -- 0.71x lower risk for spondylitis
rs27044(C;C) -- normal risk

This was the article I was thinking of:
Hi Simply Southern
Over the years I've tried to stay updated with the latest findings regarding AS.

It might be worth contacting the makers of 'Promethase' and asking them to confirm whether or not rs7743761 which apparently corresponds to DHFRP2 (dihydrofolate reductase psi-2) is really associated with increased risk of AS.

Clearly most of the disease risk for AS currently lies within the HLA B27 gene (and to a much lesser extent HLA B40 B51 B47 and B13), followed next by ERAP1 (formerly ARTS1) then IL23R and then a string of lesser known and possibly less important genes such as CMP2 (Capillary Morphogenesis Protein 2).

Also worth noting is that the increased risk of AS of say x20 of one version of an SNP over the other version is only x20 specific to that particular SNP within that gene.

In the absence of at risk versions of HLA B27, ERAP1 and /or IL23R then the risk of AS associated with lower ranking genetic factors such as CMP2 or (presumably DHFRP2) is probably inconsequential.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 08/23/15 08:11 AM
I think David is right. Because the general chance of AS is so low, and the relative risk factors for most of the SNPs reported for AS in Promethease are between 0.7 - 1.5, I don't think they would be of any diagnostic value.

However the 20x relative risk rs7743761(A;A) allele could be an exception since it's a high odds ratio and only a very small percentage of people have that allele - looks like about 2% of Europeans from the Promethease graph.

These are the two studies that are linked to as references in Promethease for that SNP:

The first one mentions rs7743761 but I couldn't find any mention of a 20x odds ratio or see anything like that in the tables or figures. The second one doesn't seem to mention rs7743761 at all. So perhaps Promethease made an error in their reporting on that SNP.

I was diagnosed in 1986 after a Nuclear Bone Scan, HLA-B27+ and a Myelogram (and years of back and hip pain) but never kept my records.

At the time, there was no one closely knowledgeable of AS in the area and because I wasn't the worst case scenario (extreme AS case with fusion), I was dismissed by a Rheumy within even looking at any films, reports or blood work. He told me I was wasting his time that anyone with AS would be completely hunched over and unable to straighten up because they were fused. He told me to never come back as he had patients in real need.

Needless to say, I was so humiliated that @%LL would have frozen over before I went through that again so I walked away and refused to see another one but, a lot of things happened since but I've been more focused on those than AS. I'm -.02 on SLE, Bilateral Peroneal Neuropathy, 4 thoracic fractures last year, and so on.....any way....with all that somewhat stabilized, I moved to get closer to family and without established history, I've been told there is no urgency to get in with a Rheumy so I'm in an 8 month wait....for a November appointment. By then, next appointment I will be changing insurances....again. I don't like the closest one taking my insurance is 76 miles away....I could understand if I was in a remote area but....not the case.

Any how, what I find very interesting is the HLA-B27 is not on the report results but under HLA-B there were three (unevaluated) SNPs:


It seems like I have a lot of SNPs for Lupus, Rheumatoid Arthritis, Multiple sclerosis, Crohn's disease, and overall Bad Reputes for Autoimmune conditions.

Overall it seems without clear fusion, variables are so many that a diagnosis can be explained away.

Basically treatment. As of right now, all that is pending (last verbal comment in 2014 from the one visit I had with a Rheumy was she was furious that years had passed without any prevention).

Personally I don't care if I have it or not but unless there is some type of reason to suggest treatment, I've been dismissed.

Yes I was diagnosed in 1986. The symptoms over those years (outside of low activity) are the typical of low back pain, bilateral SI pain that goes back and forth (narrowing joint), severe buttock pain, peripheral issues in the shoulder, hip and foot (heel and ankle), suspected enthesitis in both feet (said by a GP), IBS, diverticulosis, history of conjunctivitis (possibly associated), history of anemia, low Vitamin D levels, iron, etc. Much of this started as a pre-teen but years of doctors that called it imaginary to jealousy of sibling to growing pains it was dismissed over and over.

As an adult it got so bad I could not walk, lay still or stand in one place. It had affected my neck and shoulder and my spine has fractured in four places (just from a fall in my home on level ground). SED and CRP are essentially normal but 50% of cases are....inflammation was clear in my neck but that was seen by my Neurosurgeon. Stiffness and severe reduction in range of motion in most just goes on and on.

THE BIGGEST PROBLEM is that unless I was under constant care before moving here, the response I've gotten is there is no urgency to be seen.
I am on meds but ones that are not specific to treat Spondylitis but for pain from a slipped disc, thoracic fractures and osteoarthritis. I have been taking anti-inflammatories for years, gradually stepping up to Celebrex but MRI are still showing inflammation so clearly it's not controlling it.

I can't get the meds because they need to come from a Rheumatologist which I can't get in to see. I have tried the NSD and didn't see any improvement.

It's not that I can't see one, I have an appointment however, without a 'transfer' from an existing practice, the wait time was 8 months.

In the interim, I just have to wait.
I could IF I could get someone to take me. We don't have many in this area but I have called all, some several times asking if they would consider accepting my insurance or allowing me to pay separately but most don't want to take on a patient that is in their care that needs them and then tells them they can't pay (not that I would pull that but some have so they don't like to consider it).

Hi jroc,

Thanks for posting those two references.

I've listed a paragraph from the first paper and it reads ...
"As expected, SNPs in the MHC on chromosome 6p were strongly associated with ankylosing spondylitis (rs7743761 P = 5.0 × 10 to the power of -304). Association was evident across a very broad region surrounding the MHC, including five SNPs lying in a 153-kb region at 26.0–26.1 Mb from the p-telomere (5.4 Mb from HLA-B), which achieved P < 10−5. The most associated SNP in this region was rs3734523 (P = 1.6 × 10−6). However, conditional logistic regression analysis suggested that this was unlikely to represent a separate independent association because conditioning on five of the most significant SNPs from the MHC (rs7743761, rs2596501, rs3915971, rs2516509, rs1265112) caused the association to disappear (P = 0.27)."

This indicates that rs7743761 is on Chromosome 6 - which contains the MHC and specifically HLA B antigens. It becomes very difficult to identify independent risks factors (that is other than B27) on Chromosome 6 because of the concept of linkage disequilibrium.

Put simplistically, if there is a high risk SNP on or near the MHC on chromosome 6, the association with AS might disappear if you can somehow control for the presence or absence of HLA B27.

For a discussion on linkage disequilibrium refer to Wikipedia.
Interestingly, some of the most significant non-MHC SNPs encode variants with a loss of function that confers protection against developing disease (eg. IL23R R381Q and ERAP1 rs30187).
Currently the number of AS-associated gene loci is about 30 and intriguingly many of these genes congregate along an immunomodulatory pathway.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 08/28/15 03:00 PM
Thanks David, that's interesting to know there can be links between SNPs and different genes. I guess that's one of the main issues with having access to these sort of tests - without expert interpretation you can end up with a lot of data, but not much information.
Hi jroc,

I found this paper that you might find of interest since you've gone to the trouble of getting a genome scan.

The aim of the study was to investigate whether two ERAP1 nsSNPs non-synonomous Single Nulceotide Polymorphisms)- namely rs30187 and rs27044, influence the clinical characteristics of AS; in particular with respect to Age of Onset (arbitrarily set as <30 or >30) and whether or not the patient has Axial disease or Mixed disease (defined as axial AND peripheral manifestations)

Firstly, it is current consensus that the association of ERAP1 variants with AS is valid ONLY for HLA-B27 positive patients, but NOT for HLA-B27 negative ones.

The study found that there was association between the studied ERAP1 variants and the development of early onset AS.
So if you are HLAB-27+ but lack either of the two studied ERAP1 variants then you will probably develop late onset AS.

When restricting yourself to look at only AS sufferers who are HLA-B27+ early-onset patients, the study found that the rs30187 ERAP1 variant is associated with Axial-only group whilst the rs27044 ERAP1 variant is associated with the Mixed group (that axial AND peripheral symptoms).
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 08/29/15 07:24 AM
Thanks David, that's really interesting.

I'm HLA-B27+, early onset (20 years), mixed - axial and peripheral. And I have:

Rs30187(C;T) -;T)
Rs27044(C;G) -;G)

So if I'm reading the study correctly I have an Rs30187 allele that is associated only with the axial form, and a Rs27044 allele that is associated with the mixed form. And both of those are also associated with early onset.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 08/29/15 07:25 AM
Sorry about the thread hijack SimplySouthern. I hope you can get to see a rheumatologist soon and start an appropriate treatment plan.
JROC -- No worries -- more info is more to learn.

In response to the above, I am HLA+ and have the two as:


The above results seem odd with some of the other results like rs7743761(A;A). Mine as well started early but like many, it took over two decades to diagnose as the symptoms were more than axial.
Hi jroc and SimplySouthern,

I think you're reading that right jroc. Bad luck that you should have both the at-risk ERAP1 minor alleles for rs30187 and rs27044 - which is consistent with you being early onset mixed.
By contrast, you, SimplySouthern, have the major alleles for ERAP1 rs30187 and rs27044, that is, for these two SNPs you have the alleles that are not associated with AS.
As for rs7743761 - I think it's significance is lessened by the fact that it is on chromosome 6 - in or near HLA-B27.
Are you saying the significance is lessened (the rs30187 and rs27044) in my case because of the results being on the lower end? (sorry 4 nights of little to no sleep is making my logical side pretty lame).

On the flip side of the IL23R, of the seven, 6 are listed as Bad Reput and the other, neutral. I not sure if any of that has meaning. In your research, has anything crossed your path on that DavidP?

Hi SimplySouthern,

The study I referred to is one of the first to try and link variants of ERAP1 to the phenotype or clinical characteristics of AS patients.
Several SNPs of ERAP1 have been identified, namely, rs30187 (which means Lysine is replaced by Arginine at position 528, designated Lys528Arg); rs27044 (Gln730Glu); rs2287987 (Met349Val); rs10050860 (Asp75Asn) and rs17482078 (Arg725Gln). The authors restricted themselves to looking at rs30187 and rs27044 and whether or not these influenced Age of Onset (for a HLA-B27 positive patients) and axial Versus mixed pattern of disease for the early onset sub-group.
Even though those two ERAP1 variants are not contributing to your disease there are the other 3 ERAP1 variants to consider and as you point out the seven or so listed variants within the genetic code of the IL 23 Receptor protein.
IL23 is a very important cytokine (white cell chemical messenger) for AS patients. It is produced predominantly in the gut but unfortuneately there are a type of recently discovered lymphoid cells that are restricted to tendon and ligament attachment sites that are IL23R+ so they will respond to high levels of circulating IL23 by ramping up inflammation at those sites (rather than repairing any tendon micro tears as they are supposed to do).
Since as you have indicated you have six out of seven of the at-risk IL23R SNPs then perhaps for you the IL23/IL23R axis is more of a driver of disease than ERAP1 and may be you might be expected to have considerable issues peripherally and perhaps significant gut issues?
Hi SimplySouthern,
I went back to the first page and looked at some of your earlier posts. It clearly shows that you don't have any of the ERAP1 at-risk variants - so all good there.
But, as you pointed out you have 5 of the 7 IL23R at-risk SNPs - not good - so in conjunction with HLA-B27, IL23R must be driving your disease.
As for rs7743761 which you say Promethease locates onto DHFRP2. I think something is wrong there since a quick scan of the internet locates DHFRP2 onto chromosome 19 (I think). But this does not jell with the paper from 2010 that jroc referred to, that suggested rs7743761 is on chromosome 6 on, or near HLA-B27 - it may even be within HLAB27 - and that would lessen its significance.
I think that has been one of the hardest reasons to hold the diagnosis as when something has occurred in the peripheral areas, it has been dismissed as affiliated. I have had numerous issues in my hands, shoulders, legs and ankles (and brittleness of the spinal bones) so that would make sense. As far as gut, officially IBS and in the colon, other issues but when in flares, I really wasn't under a case that could diagnose it.
I found this that supports Chromosome 6 for DHFRP2. Both 23andMe and Promethease has it under 6 as well. It looks like DHFRP1 is under 18 but I can't find anything connected to 19.

Hi SimplySouthern,
I saw that you said nowhere did the results say you were HLA-B27 positive. Instead they gave just three SNPs within the HLA-B gene without comment.
HLA-B27 Human Leukocyte Antigen B27 is located on Chromosome 6.
HLA-B27 is one of 60 or so HLA-B alleles, and is a Major Histocompatibility Complex (MHC) Class I gene that encodes a peptide binding protein (in this case HLA-B27) that are expressed on all nucleated cells. The molecules consist of an α chain and β2microglobulin (not part of MHC) and present ligands (antigen fragments of 8-10 amino acids long, usually virally derived or ‘self’, but also can be derived from intracellular bacteria).
HLA-B27, like the all the HLA-B so called 'antigens' were originally defined serologically. That is, the serum of multi-parous women (multiple births) were run against panels of white cells from different people and eventually antiserum generated that could react with different groups of white cells so that the cells were designated as displaying surface markers HLA-B1 through to B60 (some are now redundant). We each have two HLA-B genes (one from each parent) so I might be HLA-B27/B14 for instance.
Since those days HLA typing has been refined and explained by different patterns of amino acids over stretches of hundreds of amino acids that comprise a single B type. This has led to further extensive subtyping within HLA-B27. This explains why a genome scan looking for SNPs will not be capable of readily identifying your 'B'status. Instead that is performed by a standard lab and reported as either HLA-B27 positive or HLA-B27 negative (eg. B14/B58)

The Major Histocompatibility Complex (MHC) is a large genomic region of 3.6Mb (3.6 million base pairs) situated on chromosome 6 containing 224 known genes, most having immune or related functions. The high allelic diversity in the MHC, particularly at HLA loci, is thought to provide a survival advantage due to the likelihood of heterozygosity increasing the range of antigens that can be presented.
The proteins encoded by the HLA genes are expressed on the surface of cells where they display both self antigens (short peptide fragments from the cell itself) or non self antigens (such as fragments of invading microorganisms) to a type of leukocyte or white blood cell called T-cells (lymphocytes) that have the capacity to kill or facilitate the killing of pathogens and infected or malfunctioning cells.
I'm HLA-B27+. I've been tested 4 times (I know why....because of doctors that wanted they own results over the course of years). Neither Promethease or 23andMe show the breakdown to that specification but list the subset genes of discussion here.

BTW - should I mention you are sooooo far advanced in research and interpretation of the papers than I.
Hi SimplySouthern,

You know its sad that so called 'experts' were able to wipe you so completely - sad because they were showing their own ignorance - perhaps they could be excused back in the 80's because then AS only applied to the worst of the worst - young men with bamboo spines.
The broader grouping of spondlyoarthropathies now encompasses lesser conditions like reactive arthritis, psoriatic arthritis and undifferentiated spondyloarthropathy (which sounds like you) - but each still overwhelming and difficult to deal with in their own right - as you and I both know - made all the worse by even close family and friend suspecting us as being borderline hypochondriacs - which probably I am!

Also sad that unless you have Grade 2 changes to the SI joints on plain X-ray one doesn't qualify for financial support for the biologics. The reality is that the mechanism of SI attack is the same as that of the hands, feet, ankles, ribs, knees, wrists, manubrio-sternal joint, etc - each is preceded by an inflammatory enthesitis.
I've got to agree with you there. I remember walking into a Rheumatology appointment and without any images or blood work, I was criticized for taking his time because if I had AS, I would be bent over with a fused spine. He was supposedly one of the best in that area at the time....wrong! Unfortunately because of those comments in my file, I couldn't get another referral.

That was in 1987 and thankfully things have changed considerably but still, that initial diagnose was breath taking and with it, I walked away after being humiliated as one, as I didn't want to believe I had it and two, didn't want to be so to speak on trial to beg for help.

As far as the Grade 2 or above, I have what appears to be Grade 3 (bilateral clear erosion and narrowing) but was told it's just OA....that with years of symptoms wasn't sufficient for the last Rheumy...lots of folks that could use better training before dismissing a diagnosis.

In 2010, I fell down a small flights of brick stairs and ripped the tendons from my left foot. Although told it would heal on it's on (6 weeks in a boot cast). It has never been since (unbearable pain that limits the time I can be on my feet, movement and I have to wear tight socks all the time to prevent the pain and it's in some of my toes as well). A GP said it sounds like enthesitis and another said, it could be. Not sure what to do with that but I do know the pain can be out of control.
Hi SimplySouthern,

I made these notes in relation to IL23R a couple of years ago and thought you might find them of interest - I hope most of it is still relevant:

IL23R Interleukin 23 Receptor is located on Chromosome 1p31.3
IL23R is a cytokine receptor present on T-cells, NK lymphocytes, monocytes, and dendritic cells. These cells identify foreign substances and defend the body against infection and disease. IL23R is a critical cytokine receptor on a subset of CD4+ T-cells called T helper 17 (Th17) cells. The receptor for IL-23 is formed by the IL12Rβ1 subunit and an IL23 specific subunit, IL23R. Note that Th17 cells express a high level of the cytokine IL-17 in response to stimulation.

The primary articular site of inflammation in the spondylarthropathies are the entheses (tendon-bone attachments) and the aortic root and valve (which are structurally similar to entheses). Recent studies have shown that IL23 acts on previously unidentified IL23R+ ROR-γt+ CD3+ CD4- CD8- Sca1+ entheseal resident T cells, which normally act to repair and remodel these sites. In response to high levels of gut derived circulating IL-23 these cells elaborate inflammatory mediators including IL-6, IL-17, IL-22 and CXCL1 with the specific and characteristic development of enthesitis and entheseal new bone formation in the initial complete absence of synovitis. Thus, dysregualtion of IL-23 results in lesions at precise and predictable anatomical sites. Moreover, these cells express the molecule PLZF, which allows them to respond to cytokines extremely rapidly, and indeed entheses respond within hours to IL23 in vitro.

At the cell surface IL23R binds with IL23 and triggers intracellular signalling that promotes inflammation. Genetic alleles of IL23R substantially affect susceptibility to AS, Crohn’s Disease, and psoriasis and this may in part explain the clinical association of these diseases. IL23R polymorphism is also associated with MS susceptibility. It is known that up to 70% of AS patients have subclinical ileal inflammation resembling Crohn’s and that about 10% of AS patients have clinical IBD. The strong association with susceptibility to disease suggests that IL23/IL23R targeted therapies may even be capable of disease prevention.

One of the alleles of IL23R appears to reduce the likelihood of developing AS. This allele results from the substitution of a single amino acid, glutamine for arginine at protein position 381, written R381Q or Arg381Gln. The same allele is protective against Crohn’s disease and psoriasis. Researchers believe that the role of IL23R in triggering inflammation in the intestinal wall may underlie its connection with Crohn’s and probably with AS. Accumulating evidence suggests that IL-23 is a likely master regulator of mucosal immunity during gastrointestinal infection and inflammation.

Genome wide association studies have previously shown that a single nucleotide polymorphism (SNP) in the IL23R gene (R381Q ) is significantly higher among healthy controls than in patients with psoriasis, Crohn’s Disease, sarcoidosis or AS. The associated SNP (rs11209026) which consists of a guanine (G) to adenine (A) substitution at the DNA level results in an arginine (R) to glutamine (Q) substitution at position 381 (R381Q) within the cytoplasmic domain of IL23R.
The possible Genotypes are : GG - Typical odds of AS; AG - moderately lower odds of AS and AA - substantially lower odds of AS.

Th17 cells generated from A allele carriers and G allele carriers are capable of producing similar amounts of Th17 cytokines. However, IL-23 mediated effector function was impaired, as Th17 helper cells from the protective A allele carriers had significantly reduced IL-23 induced IL-17A production and reduced STAT3 phosphorylation compared to the common G allele carriers.

The functional consequences of carrying the protective gene variant are elaborated in the 2011 paper “Functional studies of the IBD susceptibility gene IL23R implicate reduced receptor Function in the protective genetic variant R381Q” by Pidasheva, S et al.
‘We investigated the effects of this variant in primary T cells from healthy donors carrying IL23RR381 and IL23RQ381 haplotypes. Using a proprietary anti-IL23R antibody, ELISA, flow cytometry, phosphoflow and real-time RT-PCR methods, we examined IL23R expression and STAT3 phosphorylation and activation in response to IL-23. IL23RQ381 was associated with reduced STAT3 phosphorylation upon stimulation with IL-23 and decreased number of IL-23 responsive T-cells. We also observed slightly reduced levels of proinflammatory cytokine secretion in IL23RQ381 positive donors. Our study shows conclusively that IL23RQ381 is a loss-of-function allele, further strengthening the implication from GWAS results that the IL-23 pathway is pathogenic in human disease. This data provides an explanation for the protective role of R381Q in CD and may lead to the development of improved therapeutics for autoimmune disorders like CD.’
That is, the IL23R R381Q SNP protects against multiple immune mediated diseases by impairing IL-23 mediated Th17 responses, namely, reduced IL17A production and STAT3 phosphorylation.

Early studies of Han Chinese suggested there was NO association of IL23R with AS, presumably because there was no polymorphism within the IL23R gene at rs11209026 . In a contradictory result from another study (2009) the difference in the genotypes rs11209032 and the differences in the genotypes and allele frequencies of rs6677188 between AS cases and controls in a Han population WERE significant. The two SNP’s rs11209032 and rs 6677188 were in strong linkage disequilibrium.

A more recent study has implicated rare IL23R variants with AS aetiopathogenesis, and has identified a low frequency nsSNP with predicted loss of function effects that is protectively associated with AS in Han Chinese, suggesting that decreased IL-23R function protects against AS. These findings further support an important role for IL23-signalling in AS.

A new injectible monoclonal antibody treatment, Stelara, is an Anti-IL12p40 antibody given (45mg) every 3 months (cost ~$5000 /shot) and it is currently only approved for psoriasis but may be available soon for psoriatic arthritis.

(1)The allele G frequency of rs11209032 is higher in AS groups than in the controls (A vs. G: OR = 1.173, 95% CI = 1.107–1.243, P < 0.001). (2) The allele A of rs1004819 is higher in the AS groups than in the controls in both all-pooled population (A vs. G: OR = 1.147, 95% CI = 1.022–1.287, P = 0.02) and Europe-pooled population (A vs. G: OR = 1.199, 95% CI = 1.007–1.429, P = 0.042). (3) The allele frequency T of rs1343151, G of rs10489629, and A of rs11209026 is lower in AS groups than in the controls. (4) No significant differences are found in allele frequency of rs10889677 polymorphism between cases and controls by random effects model. The conclusion is that the genetic susceptibility to AS is associated with the IL-23R gene polymorphisms. The protective SNPs include rs1343151, rs10489629, and rs11209026 while rs1004819 and rs11209032 may be the susceptibility SNPs.
Interesting article in todays edition of "The Australian" newspaper:
Cut-Price shortcut to what ails us
A shortcut developed by Australian scientists could slash the costs of genetic analysis by 95 per cent, fast-tracking cures for common illnesses such as dementia, schizophrenia and heart disease.
Queensland researchers say their new method of "imputing" genetic mutations allows DNA to be analysed for as little as $70 a pop, with only marginally less precision than full genomic sequencing that costs 20 times as much. The technique, outlined this morning in Nature Genetics, offers a cost effective way to analyse huge banks of tissue samples. "The greater the sample size the increased statistical power of the result," said lead author Jian Yang, from the Queensland Brain Institute at the University of Queensland. "When we study the genome for genes linked to diseases we need to test tens or even hundreds of thousands of samples because for most common diseases the effect of a single gene is tiny."
Advances have cut the cost of genome sequencing colossally, from $US119 million ($166m) in 2001 to about $1400. The new approach slashes costs further using tests known as "single nucleotide polymorphism arrays", which directly measure only a fraction of the genome.
Geneticists then guess the remaining DNA variants based on known genetic associations, using statistical imputation.
Last week University of Hong Kong researchers unveiled their own "user-friendly" imputation software tool and Californian company 23andMe uses imputation methods in a saliva test that costs consumers about $138.
Doubts over accuracy led the US Food and Drug Administration to stop 23andMe from using its tests for health advice in 2013, putting it out of the genetic testing business for 15 months.
But the UQ researchers say they have put these concerns to rest. In a study of more than 44,000, their imputation technique obtained results almost on par with full sequencing.
Project leader Peter Visscher said, while full sequencing would still be used to hunt for cures for rare illnesses, imputation could become favoured for some of the most common scourges.
"(They are) the diseases where we already know that there are hundreds - if not thousands - of variants that contribute to risk," he said.
I am all for "Being in the Know" but in reading some group responses on the 23andme website it is clear many are new to Spondylitis. I have learned more in groups and forums in the last two years than I have known in the last nearly 30 years since my diagnosis.

Here's wishing more of the medical community takes heart to DNA studies as well. smile as what you don't see is not always what you don't have!
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 09/04/15 01:59 PM
I had another thought/question about the odds ratios for the ERAP1 SNPs. The relative risks listed for Rs30187(C;T) and Rs27044(C;G) are 1.5x and 1.4x. If those are the odds ratios for the general population, but the increased risk is only associated with HLA-B27+ people (~8% of the population), then wouldn't the relative risk of those SNPs in an HLA-B27+ person be much higher?

you would think BUT, are the results taking into consideration you have the genes because you are positive? IDK
I'm not familiar with this publication but found it interesting (of what I could make sense of). I did check and all but one gene mentioned was on my report.
Here's a bit of background information on what ERAP1 does.
ERAP1 stands for Endoplasmic Reticulum Amino Peptidase 1
It is located on Chromosome 5q15 and is composed of 20 Exons (Transcription Length 5,488 bps Translational Length 948 aa) and 19 Introns (Transcription length ~42 kb (total 47.28 kb).

The association of ERAP1 and disease is a major breakthrough in AS research. The gene product has two known functions, firstly within the Endoplasmic Reticulum of a cell, ERAP1 (in association with the closely related ERAP2) is responsible for trimming short peptides to optimal length for MHC Class I presentation. Since AS is primarily an MHC Class I mediated disease, this has implications favouring mechanisms of disease generation that involve abnormalities of peptide presentation. The second known function of ERAP1 is (when secreted), ectodomain shedding of cell surface cytokine receptors of the pro-inflammatory cytokines IL-1 (IL1R2, decoy IL-1 receptor), IL-6 (IL6Rα) and TNFα (TNFRSF1A).
So far there is no evidence for the association of ERAP1 with Crohn’s Disease or Ulcerative colitis, both of which are not Class I mediated diseases; whereas up-regulation of IL-1, IL-6, TNFα and IL-23 is a feature of these conditions. This would suggest that it is the peptide trimming function of ERAP1, and not cytokine receptor cleavage, that explains the mechanism of association with AS.
ERAP1 is a Zn containing metallopeptidase that is a member of the ‘M1/gluzincin family’ of peptidases. It is most clearly related to ERAP2 and P-LAP and together ERAP1, ERAP2 and P-LAP form the oxytocinase subfamily of M1 peptidases. All three have highly conserved C-terminal domains that allow the recognition and cleaving of peptide precursors. ERAP1 has a modular organisation that explains its molecular ruler mechanism. When a precursor peptide’s C-terminus is bound to ERAP1’s regulatory domain it induces a conformational change of ERAP1 from the open to the closed position thereby allosterically activating the catalytic zinc site 30 Angstroms distant which trims the last amino acid, provided the precursor peptide is long enough to reach the catalytic site. It has been shown that peptides shorter than 8 residues are not long enough to be efficiently trimmed by ERAP1. Longer antigenic precursors could be accommodated by bulging or zig-zagging of the middle of the peptide within the binding groove.
The structure of the soluble domain of ERAP1 has recently been determined. A central wide channel leads to the active site Zn(II) and it is the obvious candidate to accommodate the peptide substrate. The enzyme undergoes conformational changes which represent snapshots of the catalytic cycle. Interactions between ERAP1 and the peptide mainly involve hydrophobic and van der Waal contacts, plus a H-bond contact with the carboxylate end of the peptide. ERAP1 has four domains: domain I (residues 46-254); domain II (residues 255-529, which contains the catalytic site; domain III (residues 530-614 forming a β sandwich with two β-sheets) and domain IV (residues 615-940 forming a large bowl shaped α-helix domain containing 16 α-helices). The ERAP1 regulatory domain is composed of sub-domains III and IV which can interact with domain II through conformational changes to either shield or expose the catalytic site.

For proteolysis to occur domain IV must swing back over the catalytic domain, and in so doing activate the protease. After the N-terminal residue is cleaved, the cavity must open in order to release the products, which can then bind for another round of proteolysis. This is supported by peptide intermediates when monitoring the shortening of the ERAP substrates. The non-processive nature of peptide trimming is a result of N- and C-terminal anchoring, requiring active site closure, and subsequent release of the N-terminally cleaved amino acid residue. For a new round of cleavage to occur, the trimmed peptide needs also to be released to allow another round of productive binding due to the necessity to attach the free amino terminus at the S1 binding site. Note that all trimming intermediate products can be observed with different maxima at successive time points. ERAP1 functions as a molecular ruler because cleavage efficiency is significantly reduced for peptides shorter than eight residues. Proteolysis only occurs if there is a secondary binding site within the internal cavity where the C-terminal part of the substrate is bound and hence this defines the minimal product length. The internal cavity is large enough to accommodate large peptides (9-15 peptides) but peptides shorter than 8 or 9 peptides are not cleaved because both binding sites cannot be occupied. The 8-mer is still able to bind ERAP1 but is not further processed and acts as a competitor for the 9-mer.
Fascinating information DavidP. Thank you!
Yes fascinating, I don't understand a bit of it but fascinating neverless.
Glad there are much smarter people than I researching this stuffs. laugh2
I have to agree with Stormy and's well above my head but I'm grateful someone has the knowhow to dissect some of these publications.

Another one I saw today that I found interesting but for those with the 3 genes (HLA-B27, EARP1 and IL23R) holds a one and four chance of developing AS. I don't know how credible it is (in light of the advertising at the bottom of the page).
Thanks jroc and Simply Southern, you inspired me to pay the $280 (Australian, including post) to get tested by 23andme.

Basically my results are identical to Simply Southern, that is, I knew already that I was HLA-B27+, and the genetic test showed that I carried no ‘at risk’ ERAP1 alleles, but carried 6 out of 8 IL23R variants with bad repute.

According to me (and no one else), and according to the ‘ubiquitous pathogen exposure hypothesis’, HLA-B27+ individuals who also carry ‘at risk’ alleles of ERAP1, will develop a spondyloarthropathy when activated gut dendritic cells stimulate gut-mucosal memory T cells to be polarised toward a Th17 phenotype. Elevated levels of circulating IL23 will drive the disease processes such that those individuals who also carry ‘at risk’ alleles of IL23R will also suffer significant peripheral enthesitis and IBD / Crohns-like symptoms on top of the axial aspects of AS.

Individuals who are B27+ and have ‘at risk’ ERAP1 drivers, but not IL23R risk alleles might be expected to present as classic early onset AS but without significant gut or peripheral symptoms? Such cases might feel they are unaffected by diet?

In the case of someone like me, with B27 and IL23 drivers, but without ‘risk’ ERAP1 alleles, they might be expected to suffer gut and peripheral symptoms with lesser axial symptoms (no fusion).

I’d encourage more KickAS’ers to get tested.
Regards David.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 10/13/15 06:12 AM
That's cool David, thanks for sharing. That's an interesting hypothesis.

Did 23andMe provide any health related info or did you have to get that by running the raw data through Promethease? My understanding was that since the FDA shut them down a while back they are no longer allowed to offer any health interpretations of results, just ancestry information.
Hi jroc,

23andMe provide only the raw SNP data, and as you say ancestral paternal and maternal (mitochondrial) DNA typing.

As you would know, you have to pay an additional $5 to run the data through Promethease. You retain access to Promethease's results for 45 days only.

There is some value in rerunning your results through Promethease a year or 2 down the track as the information attached to each SNP might in the future change depending on scientific advances.

I'm still confused about the DHFRP2 gene SNP on Chromosome 6 which like you gives a 4.6X increased risk of AS for me - I still think that this is probably due to linkage disequilibrium with HLA-B27 ?

Regards David
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 10/13/15 09:45 AM
Ah ok, that's what I thought. I was just wondering as 23andMe has a detailed report on Crohn's risk with 12 different SNPs in their Health Risks section.

Promethease will no doubt cover all those same SNPs. It's a shame they had to shut down the health risks reporting in 23andMe as I quite liked the way they did it with showing the general population risk, your risk based on your SNPs, the relative risk factor, and info on the role of genetics vs environment for that disease, links to studies etc.

I had read the same hypothesis about the two (ERAP1 and IL23R). I'm glad you are encouraging others to do the tests as well as I honestly think there is something to this Axial (ERAP1) versus more Peripheral (IR23R) involvement and wonder about others AND, how it in massive results, it could help many seeking the proper care.

If this is something found with substance, it would certainly turn my life around as for 25 years I have been told all my issues have no bearing on AS (or SpA).

Earlier this year, the FDA did ease up on the restrictions however, it doesn't seem as if they have yet changed that (since DavidP and I both got just the raw data).
Hi Simply Southern,

In the future, knowing your ERAP1 or IL23R SNPs might be helpful in the choice of drug therapy.

I found a paper in the Romanian Journal of Rheumatology for 2015 that addresses the impact of IL23R in AS - it's worth a read.

Regards David.
I'll try that address again.

Cheers again!

Promethease did have the similar type data. My 45 days has expired but I did download all of the data. While active, you could export/view results in different formats and bar graphs was one of them.

DavidP -- I'm curious how your results came out on Promethease for the Crohn's. The following were my results (sorry I have NO idea how to clean up the columns):

Name Repute Summary
rs6601764(C;T) Bad 1.16x increased risk of developing Crohn's disease
rs9469220(A;G) Bad 1.1x risk of Crohn's disease
rs10883365(A;G) Bad 1.2x increased risk for developing Crohn's disease
rs13361189(C;T) Bad 1.3x increased risk for Crohn's disease
rs4958847(A;G) Bad 1.3x increased risk for Crohn's disease
rs7807268(C;G) Bad 1.3x risk for Crohn's disease
rs2241880(C;T) Bad 1.4x increased risk for Crohn's disease in Caucasians
rs2201841(C;T) Bad 1.5x increased risk for Crohn's disease; 2x increased risk for Graves' disease
rs1000113(C;T) Bad 1.5x risk of Crohn's disease
rs9858542(A;A) Bad 1.8x risk
rs6908425(C;C) Bad 1.95x increased risk of developing Crohn's disease
rs6596075(C;C) Bad 2x risk of Crohn's disease
rs11229030(C;C) Bad higher odds of Crohn's disease
rs12567232(A;G) Bad Increased risk for Crohn's Disease
rs2631367(C;C) Bad Increased risk for Crohn's Disease
rs11209026(G;G) Bad Normal, but higher risk for certain autoimmune diseases.
rs12037606(G;G) Good Normal risk of developing Crohn's disease
rs3814570(C;C) Good Normal risk of developing Crohn's disease
rs11465770(C;C) normal Crohn's Disease risk
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 10/14/15 02:28 AM
Originally Posted By SimplySouthern

Promethease did have the similar type data. My 45 days has expired but I did download all of the data. While active, you could export/view results in different formats and bar graphs was one of them.

Thanks, that's cool. I just took a look at the promethease website and the site and the reports have changed a lot since when I did it back in 2013. Looks much more professional now and they integrate with the 23andMe through their API so you don't have to manually upload. I might have to drop another $5 on a new report.
DavidP, (or anyone else that recalls it)

I recently ran across an article that was written in very layman's terms that suggested those with emphasis of bad repute for ERAP1 were more likely affected with Axial involvement whereas if the bad repute was stronger on the Il23R side, it most likely affected peripheral areas (I know you touched on that above but there was another article that went more into it).

I'm hoping with your expertise of research you may have crossed paths with such an article. I'm terrible about reading something of interest and not printing it and then can't find it to reference it.
Hi Simply Southern,
Its hard to know what to make of Crohns Disease SNPs.
Of the ones you listed I had 8 for which CD risk was normal and 8 with slightly raised risk of CD.
Of the total of 149 CD SNPs I had 99 - Not set; 23 of Good Repute and 27 of Bad Repute.
Of the CD SNPs of Bad Repute the most interesting were those at the NOD2 locus on chromosome 16 since the NOD2 locus is considered to represent one of the strongest genetic risk factors for Crohns Disease.
They were
rs2066843 (T;T) 4.1X risk of CD
rs17221417 (G;G) 1.9X risk of CD and
rs2076756 (G;G) 1.7X risk of CD.

I don't think there are any articles which put a simple hypothesis ERAP1 bad repute SNPs -> axial disease Vs IL23R bad repute SNPs -> peripheral diseae.

You might have been thinking of this article - which is quite dated now.

Understandably, most of the AS cohorts in genetic studies have in the past enrolled probably mainly classic AS cases - and very few more atypical cases - like those with an undifferientiaed spondyloarthropthy.

Recently the focus has been on IL23 and IL23R because of the discovery of a population of entheseal-resident IL23R+ T cells. It is speculated that other unique IL23R+ T cells might populate the skin, the gut, and the eyes.
That is in fact the article I was thinking of. I had viewed it a week or so ago and for some reason moved on and could not recall where I had seen it, naturally wanting to recall it thereafter.

Thanks for stopping my insanity of searching!

I have NOT been diagnosed with Crohns or IBD but have IBS and Diverticular Disease. My Mom has suffered with Diverticulitis for years and has a strong bout of it a few times a year. For the prevention, she takes nothing but an antibiotic which in time, heads it off. I do use Miralax daily when tends to limit my IBS to a moderate occurrence. I honestly believe that has been of great help. Anyway, I've wondered if there isn't something beyond that going on -- for both my Mom and I, but as yet, no suggestion of CD.

As for the Il23R focus, skin issues have never been a concern however as a child, I had frequent conjunctivitis (a couple times as an adult) and frequent soreness/redness and dry eye. I went to a Ophthalmologist and nothing was found but had no outbreaks so he said it could be but saw no damage. I have wondered if Lazik could have masked previous clue as the procedure is out of my realm of knowledge. So many avenues of research!
ERAP1 is known to be highly polymorphic with at least 70 nsSNPs, although the majority are rare. Considering the more common SNPs, at least 15 SNPs in ERAP1 are known to be associated with AS. Until recently it was not known whether they acted independently or as disease associated haplotypes. Recent studies have shown the latter, that SNPs occur as distinct haplotypes (hereafter called allotypes) in the human population and that these allotypes encode functionally distinct ERAP1 molecules. Using a range of substrates researchers have demonstrated that each allotype has a trend bias toward ‘normal’, ‘hypofunctional’ or ‘hyperfunctional’ ERAP1.
The prevalence of specific ERAP1 allotypes is different in AS cases and healthy controls, although, at this stage the exact frequency of each allotype has not yet been accurately established for the general population or for AS cohorts. Each person has a pair of ERAP1 allotypes (one from each parent) and each of the chromosomal copies of ERAP1 are co-dominantly expressed. Most healthy controls have at least one ‘normal’ ERAP1 allotype compared with AS patients, who in general have two ‘hypoactive’ allotypes (and sometimes hyperactive allotypes). The disease associated ERAP1 allotype combinations have a reduced ability to generate peptides for presentation at the cell surface by MHC class I molecules, including HLA-B27. Hence, an ERAP1 pair composed of a ‘normal’ and a ‘hypoactive’ allotype will still produce sufficient peptide ligands, and of sufficient quality, as to allow for the normal biochemical functioning of HLA-B27.
Increased disease risk is potentially manifested in two ways. Firstly, aberrant ERAP1 activity, either hypo-functional of hyper-functional, might result in increased misfolding of HLAB-27 in the ER and the generation of B27 homodimers and an unfolded protein response; and secondly, aberrant ERAP1 activity might generate unstable peptide ligands and result in elevated cell surface B27 homodimers which could activate NK cells (through KIR3DL2 engagement) and/or Th17cells.

I have presented here only a simplified summary of a paper entitled “Functionally distinct ERAP1 allotype combinations distinguish individuals with Ankylosing Spondylitis” by Reeves, Colebatch-Bourn, Elliot, Edwards and James – 2014.
Regards to all – David.
Hi jroc and Simply Southern,

If you fit our 23andme ERAP1 data to the table in the above paper we should be able to guess what allotypes we have.

The five main SNP's in 23and me are rs2287987 (M349V); rs30187 (K528R); rs10050860 (D575N); rs17482078 (R725Q) and rs27944 (Q730E)
It's pretty simple for me and Simply Southern - we have two of the normal functioning *002 allotypes, that is amino acids MKDRQ for both our ERAP1 versions.

jroc on the other hand has one and possibly two hypofunctional alleles.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 10/28/15 05:18 AM
Hi David

That sounds interesting. How do I plug my data into the table to figure out my allotype - were you meaning this table? -
Hi jroc,

Yeh, that's the table I was refering to.

This table is from an earlier paper by the same authors.
The equivalent DNA bases that are reported in 23andme or Promethease are in lower case.

Let us know what you get.

Regards David
Call me dumb, call me tired.....I am totally lost. I have opened both links and see no where to enter data. What am I missing??
Hi jroc and Simply Southern,

It's difficult to work backward from the amino acid change, to the antisense mRNA codon, then to the sense DNA strand - sometimes the DNA quoted in 23andme is FWD, other times Reversed.

By convention M349V stands for M being the major allele amino acid and V the minor allele amino acid.

M349V is equivalent M=T and V=C where M and V are a.a.'s and T and C are the DNA bases quoted in Promethease.
K528R K=C and R=T
D575N D=C and N=T
R725Q R=C and Q=T
Q730E Q=C and E=G

For me Promethease gives
rs2287987 (349) T;T -> M;M
rs30187 (528) C;C -> K;K
rs10050860 (575) C;C -> D;D
rs17482078 (725) C;C -> R;R
rs27044 (730) C;C -> Q;Q

Put another way I have two allotypes MKDRQ and MKDRQ - that is two of the wild type or normal functioning ERAP1 allotypes

By the way, 23andme gives only the main nsSNPs in ERAP1 - the papers quoted give a few extras which help to delineate their 13 allotypes but not necessarily being disease causing?
It helps if you put a highlighter down each of the columns 349, 528, 575, 725, and 730 and mentally block out the rest.

I knew my products and customers but science and chemistry were not my forte'

Based on the break down to figure it out, here are my results:

rs2287987(C;T) 349 V;M
rs30187(C;C) 528 K;K
rs10050860(C;T) 575 D;N
rs17482078(C;T) 725 R;Q
rs27044(C;C) 730 Q;Q
Hi Simply Southern,
Based on your results the most likely combination of allotypes for you would be MKDRQ and VKNQQ
That would correspond with the *002 allotype which is an 'efficient' trimmer and the other as *010 allotype which is one of the disease associated 'hypoactive' trimmers.

That combination of allotypes, because of their codominant expression, would most likely not develop AS - but you would probably have a greater risk of AS than me with my two *002 allotypes.

Interestingly, you have one of the minor alleles at rs10050860 -> the N amino acid at 575 - which, at least on it's own, is considered protective against AS.
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 10/30/15 12:39 AM
Ah I see, I think I get it now. These are my results:

rs2287987 (CT) > VV
rs30187 ( CT) > KR
rs10050860 (CT) > DN
rs17482078 (CT) > RQ
rs27044 (CG) > QE

And if I've understood correctly that would mean:

Allotypes: VKDRQ (*012) & VRNQE (*001)

Haplotypes: M349V & 5SNP
Hi jroc,

Did you really mean you were heterozygote C;T at rs2287987 because that would correspond with M;V

For sure you are going to have the *001 allotype, which is fairly common in controls (21%) and even more abundant in AS cases (44%), bearing in mind that their sampled groups contained just 19 controls and 17 AS patients.

If you are homozygote C;C at rs2287987 then your analysis as *001 and *012 allotype pair is correct - which is a disease associated allotype pair.
If you search on the term ERAP1 allotypes you can come up with some references to Behcet’s disease. Behcet’s Disease (sometimes called Behcet’s Syndrome or Silk Road Disease) is an immune-mediated small-vessel systemic vasculitis that often presents with recurrent mucous membrane ulceration and ocular lesions. It is described as a triple-symptom complex of recurrent oral aphthous (stomatitis) ulcers, genital ulcers, and uveitis. The disease can also involve visceral organs such as the gastrointestinal tract, pulmonary, musculoskeletal, cardiovascular and neurological systems. Skin lesions include erythema-nodosum-like lesions, erythema multiforme-like lesions and Sweet’s Syndrome-like lesions and the inflammatory infiltrating cells are predominantly macrophages.

Behcet’s disease, like the spondyloarthropathies, has a strong contribution to disease by variants of the ERAP1 gene, in fact, the very same ERAP1 allotypes that are linked to AS are seemingly involved in Behcet’s Disease. But whereas AS is associated with the MHC class I molecule HLA-B27, Behcet’s disease is associated with HLA-B51 and HLA-B5. Whilst HLA-B51 is generally below 1% in Caucasians its prevalence can be up to 20% in Turkish populations, for instance.

Why do AS and Behcet’s present so differently – presumably B27 and B51/5 activate and polarise the innate immune system in aberrant, but, different directions. AS is polarised toward the Th17/IL23 axis whilst traditionally BD is regarded as a predominantly Th1-mediated inflammatory disease (with possibly Th17 as well?).
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 10/30/15 02:03 AM
Originally Posted By DavidP

Did you really mean you were heterozygote C;T at rs2287987 because that would correspond with M;V

Nice catch, yes I did mean CT for rs2287987 but then I screwed up and put VV instead of MV. So it would actually be:

Allotypes: MKDRQ (*002) & VRNQE (*001)

Haplotypes: WT & 5SNP
Hi jroc,

Allotypes *001 and *002 (previously referred by the authors of the earlier paper as 5SNP and WT haplotypes) would be the most common allotypes (*002 more than *001) for healthy controls, remembering of course the tiny sample sizes. The combination *001 and *002 allotype pair would normally not be associated with AS, but clearly you are an exception.

Usually *002, being an efficient trimmer would nullify the effects of *001 which is a 'hypoactive' trimmer.

Regards David

PS. I think the same concept when applied to IL23R will yield useful results. In the case of ERAP1 allotype pairs, cumulative loss of function is a risk factor for AS; whereas for IL23R, loss of function is protective against AS, because if IL23 can't engage functionally or properly with IL23R then inflammation will be shut down?
Hi jroc and Simply Southern,

At this point it's worth mentioning that 23andme only measure a small proportion of the quoted SNPs using SNP microarrays then use computer programs to impute (a sophisticated form of guessing) the rest. That's why the price has come down from $1000 to $150. They do state up front that their results can't reliably be used for research, for that, more exact and detailed fine mapping might be required.

I did send them an email asking which of the SNPs rs2287987, rs30187, rs10050860, rs17482078 and rs27044, was actually 'read' and which was imputed - don't know if I'll get a reply?
Originally Posted By DavidP
PS. I think the same concept when applied to IL23R will yield useful results. In the case of ERAP1 allotype pairs, cumulative loss of function is a risk factor for AS; whereas for IL23R, loss of function is protective against AS, because if IL23 can't engage functionally or properly with IL23R then inflammation will be shut down?

Based on my results, the ERAP1 is not my strong suit but lends more 'bad' rupute towards IL23R. Do you have a method for the IL23R?
At this stage I don't think anyone has attempted to define any IL23 haplotype associations with AS - not sure if it is even possible.

Regards again.

So IYO (in your opinion), what are the results really saying about a person's results?
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 11/01/15 08:56 AM
Originally Posted By DavidP
Hi jroc,

Allotypes *001 and *002 (previously referred by the authors of the earlier paper as 5SNP and WT haplotypes) would be the most common allotypes (*002 more than *001) for healthy controls, remembering of course the tiny sample sizes. The combination *001 and *002 allotype pair would normally not be associated with AS, but clearly you are an exception.

Usually *002, being an efficient trimmer would nullify the effects of *001 which is a 'hypoactive' trimmer.

Hi David

That's cool, thanks for interpreting that. Good to know that my ERAP1 trimming shouldn't be too wayward. I like how they are looking at the combination of different SNPs and how they affect immunological processes. Seems like that approach would yield much more useful information compared to just looking at the odds ratios for individual SNPs, as it's likely to be a combination of different genes and SNPs that lead to particular pathologies.

Hi jroc and Simply Southern,

Naturally occurring ERAP1 molecules are polymorphic, existing as haplotypes (also referred to as allotypes) which consist of multiple combinations of AS-associated SNPs, and that they have functional differences resulting from the different amino acid specificities. Indeed, naturally occurring ERAP1 allotypes have been shown to alter the repertoire of peptides presented by HLA-B27 and that in general ‘hypoactive’ ERAP1 allotypes (and on occasion ‘hyperactive’ allotypes) are preferentially associated with the diseased state.

Just to show that, even in scientific research, nothing is set in concrete.

Bettencourt et al. (2103) showed that haplotype analysis of four SNPs in ERAP1 identified two haplotypes of variants encoding residues 349-528-575-725 that significantly affected AS risk: V349-R528-N575-Q725 (VRNQ) was an AS protective haplotype and M349-K528-D575-R725 (MKDR) was an AS-risk haplotype. They also confirmed that the strongest association was with the AS-risk variant K528. By comparison the functional studies of the protein products of ERAP1 allotypes by Reeves et al. (2014) showed the opposite, VRNQE (349-528-575-275-730) was a risk allotype and MKDRQ was protective, and according to Ombrello, Kastner and Reimmers (2015), and quoting from the latter’s paper
“... allotypes within their collection of 17 AS cases and 19 healthy controls were surprisingly inconsistent with existing AS literature. Specifically, the AS-risk allele K528 and risk allotype (*002) were more common among their controls than cases, whereas the AS-protective allele (E730) and protective allotype (*001) were more common among their cases than controls. The source of this discrepancy is unclear, and this issue is unlikely to be resolved without an analogous, sequencing-base study of the ERAP1 locus in a substantially larger AS case-control collection.
It has also been hypothesized that AS-associated variants or allotypes may influence disease risk by affecting ERAP1 expression. Constantino et al. (2015) recently reported that the K528-D575-R725 AS-risk haplotype was strongly correlated with reduced ERAP1 mRNA levels, more so than was any individual AS-associated variant. This raises the possibility that AS-associated ERAP1 allotypes may contain both coding variants that influence AS risk through changes in enzymatic function and non-coding variants that influence AS risk through alterations in gene expression.”

Earlier researchers also found that the results of imputation analysis showed a group of six SNPs in the 5'-upstream region and two SNPs in intron 19 of ERAP1 that reached a higher level of significance than rs30187 or any of the other single SNP. Researchers were perplexed because these novel SNPs, when considered individually, did not achieve AS-association and they appeared to localise to parts of the molecule unlikely to have any functional impact. It became apparent that the role of ERAP1 SNPs associated with AS work at a level more complex than individual SNPs.
Reeves et al. (2014), when considering SNPs relating to amino acid positions 82, 102, 115, 199, 581 and 737 found two distinct sequences that seemed to be co-inherited (along with positions 349, 528, 575, 725 and 730); and phylogenetic analysis confirmed that these positions (82, 102, 115, 199, 581 and 737) formed a backbone in almost all allotypes, either VILFLV or ILPLSA, the first being more likely to be disease-associated than the second; and this suggested of an earlier (more ancient) evolutionary branching. The SNPs that caused altered ERAP1 function (349, 528, 575, 725 and 730) originated at some later times, but nevertheless, remain in strong linkage disequilibrium with the older backbone SNPs.

Regards David
Posted By: jroc Re: DNA Promethease Report and IL23R and ERAP1 - 11/05/15 05:41 AM
Wow, that's complex stuff!

Are you up with the play on the latest HLA-B27 research with regards to arthritogenic peptide vs unfolded protein response vs cell surface homodimers? I haven't kept up to date with the research from the last couple of years, but back when I was reading all the papers it seemed to be a 3 horse race, and figuring out ERAP1's involvement looked like it was going to be the key to figuring the pathogenic mechanism of B27. From the look of this abstract it doesn't look like there have been any major breakthroughs -
Hi jroc,

Sorry to have been a bit slow off the mark in replying to your question.

The classic role of HLA-B27 is (in combination with beta2-microglobulin) to present short peptides from intracellular pathogens for recognition by the T cell receptor of CD8 T-cells. B27 can also be expressed as cell surface beta2-microglobluin free homodimers. In addition to binding to the TCR, MHC class I molecules can also bind to immunoglobulin-like receptors (KIR). Natural Killer (NK) cells and some T-cells expressing the immune receptor KIR3DL2 can bind to B27 homodimers and are known to be expanded in B27+ AS patients.

The paper quoted put the hypothesis that ERAP1 allotypes form three functional groups; “normal”, “hypo-” or “hyper-” trimmers of peptides and that the prevalence of different ERAP1 allotype combinations has been shown to be different between AS cases and controls. This idea that low or high ERAP1 trimming activity might lead to the restricted supply of optimal peptides has the effect of unifying the mechanisms of homodimer formation and activation of innate and/or Th17 cells through the engagement KIR3DL2 receptors.

It is postulated that sub-optimal peptides (B27-ligands) generated by aberrant ERAP1 activity are able to bind with B27 sufficiently to pass intracellular quality control but that they may dissociate rapidly at the cell surface leading to increased levels of aberrant forms of B27. The authors were not able to rule out the possibility that different ERAP1 variants could generate specific arthritogenic peptides - so long story short - the question of unfolded protein response in the ER Vs cell surface B27 homodimers Vs arthritogenic peptide - remains unresolved.

Regards David.

Please will you re-explain all of that in lay language? My brain shuts down trying to read these sometimes.


Warm hugs,
Sadly I have to agree with Kat. Unless from a Scientific background, it all runs David....


Me Too! eek2