Big New Genome Wide Association Study On Androgenetic Alopecia - Preprint

[Beowulf]

Plus I figured you guys might find this fun, it tells you about the knock on effects of using CRISPR to knock out a certain gene, the only reason I haven't used it is because I can't figure out the region and Genomic sequence of the AR gene.

http://www.gt-scan.net/submit

Plus with the AR gene in the DNA, is the idea that the gene is basically a blueprint of how to make an androgen receptor and our problem is that our body is creating over sensitive androgen receptors? Surely there must be some sequence in the DNA, or some other signalling element which tells the primordial stem cells that there should be in androgen receptors in the scalp for whatever reason.

 

[InBeforeTheCure]

I hope this is somehow helpful or at least relevant.

Awesome, it's working now -- both on their site and in R.

Those unidentified SNPs aren't in the HapMap database, which is what they use, so I can just replace those with nearby SNPs that are in the HapMap database and it should be fine. Then after adding SNPs from the regions identified in MAGMA, there will be 60-something gene regions, so it could take a while to run.

 

[Beowulf]

Awesome, it's working now -- both on their site and in R.

Those unidentified SNPs aren't in the HapMap database, which is what they use, so I can just replace those with nearby SNPs that are in the HapMap database and it should be fine. Then after adding SNPs from the regions identified in MAGMA, there will be 60-something gene regions, so it could take a while to run.

I'm glad you know what you're doing, despite how far we are from knowing the cause of all this, it's very exciting. I'm surprised people aren't piling in here just to praise you for all the amazing work you're doing!

 

[InBeforeTheCure]

Plus I figured you guys might find this fun, it tells you about the knock on effects of using CRISPR to knock out a certain gene, the only reason I haven't used it is because I can't figure out the region and Genomic sequence of the AR gene.

http://www.gt-scan.net/submit

Plus with the AR gene in the DNA, is the idea that the gene is basically a blueprint of how to make an androgen receptor and our problem is that our body is creating over sensitive androgen receptors? Surely there must be some sequence in the DNA, or some other signalling element which tells the primordial stem cells that there should be in androgen receptors in the scalp for whatever reason.

For the AR gene (or any other gene), the gene is transcribed from DNA into messenger RNA (mRNA) by RNA polymerase, and then is translated into the AR protein itself by ribosomes. This is what Francis Crick called the central dogma of molecular biology. As far as A.G.A., I have seen a couple studies which found that short CAG repeats confer higher risk of hair loss, although I also recall a study that found no correlation. CAG repeat length is inversely correlated with AR-mediated transcription of its target genes. People with Kennedy's disease have longer CAG repeats (like 40+) and they have lower androgen sensitivity, and lower rates of A.G.A. This region looks like...

CAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAGCAG

and so on. CAG codes for the amino acid glutamine, so in the AR protein there will be a long string like...

Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu-Glu

Other than that, the level of AR expression is important, and AR expression is higher in DPCs from baldies than in fullheads, as well as higher in bald scalp than in non-bald occipital scalp. Transcription of DNA into mRNA and translation of mRNA into proteins happens constantly, as well as degradation of mRNA and protein, so expression of AR mRNA and of AR protein depends on the rates of production and of removal of AR mRNA/protein. Most variants identified in GWAS for complex polygenic conditions (like A.G.A.) are thought to affect level of expression rather than protein structure. Examples of how this happens are:

- There's some sequence upstream of the gene which is a binding site for some transcription factor that increases the rate of transcription for that gene. If there are differences in this sequence that make it more likely that this transcription factor binds there, it could increase the rate of transcription of the gene and then SNPs tagging that variant might show up in a GWAS.
- A variant within a gene might affect binding of some microRNA to the mRNA. microRNAs increase degradation of mRNA or prevent it from being translated.

Oh, and you can find the DNA sequence for the AR gene here.

 

[InBeforeTheCure]

I'm glad you know what you're doing, despite how far we are from knowing the cause of all this, it's very exciting. I'm surprised people aren't piling in here just to praise you for all the amazing work you're doing!

Thanks. But it's better we don't have a cheerleading squad in here, so the last point is not such a bad thing.

 

[Beowulf]

Sorry to do this, but I've just got some questions to see that I'm following all this correctly, if you'd rather just point me to a text book that's fine.
I'm a psychology student who took physics in high school, so the last I heard of biology was in middle school.

For the AR gene (or any other gene), the gene is transcribed from DNA into messenger RNA (mRNA) by RNA polymerase, and then is translated into the AR protein itself by ribosomes

So DNA is basically just a cookbook of protein. RNA polymerase choose different selections of genes and transcribe them and they end up making proteins which make up dermal papilla cells. Using stem cells the dermal papilla cells construct the hair follicles. For some reason dermal papilla cells have lots of proteins which react to androgens, and since we have some type of correlation between hair loss and DHT we're pretty sure that for some reason the androgens destroy the dermal papilla cells and maybe like two other cell types that are also involved in the replication and maintenance of hair follicles.

There's some sequence upstream of the gene which is a binding site for some transcription factor that increases the rate of transcription for that gene. If there are differences in this sequence that make it more likely that this transcription factor binds there, it could increase the rate of transcription of the gene and then SNPs tagging that variant might show up in a GWAS.

But it might not be the fault of the RNA polymerase for making too much AR protein, another gene that's some how specific to the hair on our scalps might be forcing it to.

 

[Beowulf]

So then how did they measure the expression of AR in the GWAS?

If they took a bunch of people with hairloss, looked at the level of AR expression in their blood cells, saw that it was over expressed in comparison with full hair, but that there was no correlation with any other phenotype/illness, wouldn't that suggest that over expression of AR would only be associated with Androgenetic Alopecia and nothing else? Which doesn't make any sense since inhibiting the AR gene can cure forms of prostate cancer.

Surely you'd have to measure the expression of AR in the scalp, DPC's and so forth.

 

[Nadester]

You said Lithium Chloride can turn the inhibtion of DHT??

 

[InBeforeTheCure]

Ran the "Prix Fixe" analysis. I used this list of SNPs:

rs12565727
rs1997927
rs1333144
rs13021718
rs9287638
rs7642536
rs10212796
rs929626
rs12203592
rs6936030
rs1450596
rs1534775
rs661555
rs2435212
rs538628
rs8085664
rs201594
rs201593
rs12558842
rs5919427
rs7061504
rs4914293
rs17216820
rs6664846
rs6681289
rs11803731
rs1948922
rs7605725
rs6714212
rs12053317
rs1430657
rs4684775
rs1798802
rs9850626
rs7624941
rs9824780
rs4690296
rs2051756
rs4836465
rs1324532
rs9487643
rs1997927
rs9492767
rs9768991
rs11779445
rs3808434
rs4743039
rs1898094
rs1248696
rs1055256
rs4128868
rs174548
rs11054724
rs7955911
rs805512
rs11830322
rs16977243
rs12594457
rs7165761
rs226008
rs12927959
rs7206570
rs10853751
rs6087565
rs12624863
rs5762924

I ran it with an LD threshold of 0.25 and upstream/downstream padding both 500kbp. Results are here.


I think I made better picks myself.

 

[InBeforeTheCure]

So DNA is basically just a cookbook of protein.RNA polymerase choose different selections of genes and transcribe them and they end up making proteins which make up dermal papilla cells.

RNA polymerase transcribes a section of DNA -- this section is a gene -- into RNA. And then sections of that RNA molecule is translated into a protein by ribosomes. Here is a section from a textbook (Molecular Biology of the Gene) which you'll hopefully find helpful.

Using stem cells the dermal papilla cells construct the hair follicles. For some reason dermal papilla cells have lots of proteins which react to androgens, and since we have some type of correlation between hair loss and DHT we're pretty sure that for some reason the androgens destroy the dermal papilla cells and maybe like two other cell types that are also involved in the replication and maintenance of hair follicles.

Very few proteins react to androgens, but many proteins interact with the androgen receptor protein, which is activated when androgens bind to it. Activated AR can:

- bind DNA and change the rates of transcription of hundreds of genes (i.e. AR is a transcription factor)
- interact with other proteins and alter their function. So for example, AR is known to bind to the transcription factor beta-catenin and inhibit its activity or alter what genes it regulates. Since beta-catenin is essential for hair follicle regeneration, and AR has been shown to bind beta-catenin at a higher rate in dermal papilla cells from people with A.G.A. than in dermal papilla cells from fullheads, this interaction could be important.

Ultimately though, the mechanism of A.G.A. is unknown.

But it might not be the fault of the RNA polymerase for making too much AR protein, another gene that's some how specific to the hair on our scalps might be forcing it to.

Well, RNA polymerase is what actually transcribes the genes into RNA. A transcription factor can bind to DNA near a gene and sort of "open it up" to RNA polymerase. This will make RNA polymerase more likely to bind to the gene and transcribe it, therefore it will be more highly expressed. A transcription factor can also "close off" the sequence to RNA polymerase, and then the gene will be downregulated.

So then how did they measure the expression of AR in the GWAS?

If they took a bunch of people with hairloss, looked at the level of AR expression in their blood cells, saw that it was over expressed in comparison with full hair, but that there was no correlation with any other phenotype/illness, wouldn't that suggest that over expression of AR would only be associated with Androgenetic Alopecia and nothing else? Which doesn't make any sense since inhibiting the AR gene can cure forms of prostate cancer.

Surely you'd have to measure the expression of AR in the scalp, DPC's and so forth.

When you do a GWAS, this is what you do.

1) Take a bunch of people from a certain population, in this case British men.
2) Find what alleles (A, G, C, T) these people carry at a bunch of SNPs (single nucleotide polymorphisms) - let's say 1 million SNPs for example. These are single-letter parts of the genome for which there is variation in that population.
3) Split your population into two groups, one corresponding to one condition (in this case "bald") and one corresponding to the other (in this case "not-bald").
4) Find which SNPs are more commonly found in one group than the other. These are likely found near genes that contribute to the development of that condition.

So GWAS doesn't look at gene expression (mRNA/protein), it only looks at variation at the DNA level.

You said Lithium Chloride can turn the inhibtion of DHT??

Lithium chloride is a GSK3-beta inhibitor, which sort of mimics canonical Wnt signaling.

 

[Beowulf]

- bind DNA and change the rates of transcription of hundreds of genes (i.e. AR is a transcription factor)
- interact with other proteins and alter their function. So for example, AR is known to bind to the transcription factor beta-catenin and inhibit its activity or alter what genes it regulates. Since beta-catenin is essential for hair follicle regeneration, and AR has been shown to bind beta-catenin at a higher rate in dermal papilla cells from people with A.G.A. than in dermal papilla cells from fullheads, this interaction could be important.

Theoretically that sounds like a perfectly good explanation of A.G.A. At this stage it seems like the only real way to prove the mechanism would be with 3D printed skin. But from what I've read dermal papilla cells do not divide, rather when the dermal papilla rises up to the bulge, the dermal papilla cells communicate to the epithelial cells through Wnt signalling to replenish the dermal papilla cells and then start to regrow the follicle (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3115771/).

According to Olsen 2003 during miniturisation the dermal papilla gets smaller and smaller, suggesting that hair follicle size depends on the number of dermal papilla cells. Further Cotsarelis found that only CD34hi and CD200hiITGAS6 were missing in balding areas, yet KRT15hiITGAS6hi was still present (and the implication I've gotten so far is that it can make the CD200hiITGAS6 cells)

https://www.ncbi.nlm.nih.gov/pubmed/15617565

If signalling between the dermal papilla and bulge was diminished then the dermal papilla cells would have no way to proliferate, the less dermal papilla cells the less of an ability they have to communicate with the bulge to create more cells.

That would also explain DHT inhibition regrowth. Suddenly the DP can communicate with the bulge again and send a signal for more cells before it dies. It also explains how SM04554 works since it's a B-Catenin Agonist. It must tell the bulge to create more dermal papilla cells.

I don't know enough about minoxidil to factor that in, but I don't think anyone really does.

Oh yeah, and I can't believe how insidious AR is, I mean could it upregulate it's own gene? That would be crazy!

1) Take a bunch of people from a certain population, in this case British men.
2) Find what alleles (A, G, C, T) these people carry at a bunch of SNPs (single nucleotide polymorphisms) - let's say 1 million SNPs for example. These are single-letter parts of the genome for which there is variation in that population.
3) Split your population into two groups, one corresponding to one condition (in this case "bald") and one corresponding to the other (in this case "not-bald").
4) Find which SNPs are more commonly found in one group than the other. These are likely found near genes that contribute to the development of that condition.

So GWAS doesn't look at gene expression (mRNA/protein), it only looks at variation at the DNA level.

Oh okay so by looking at our DNA they found a way to predict who would and who wouldn't have balding with 82% accuracy. So to increase the accuracy you could consider the actual connection between the genes, and look for potential SNP/Genes which they just might not have considered.

Does that mean we all have general Wnt signalling problems?

But it does imply that there is an innate issue with our genome which is causing us to lose hair.

Thanks for the chapter, I think I'm starting to get this stuff now. Before I just kept searching the genes you mentioned in my uni's database and didn't get anything.

 

[Armando Jose]

Theoretically that sounds like a perfectly good explanation of A.G.A. At this stage it seems like the only real way to prove the mechanism would be with 3D printed skin. But from what I've read dermal papilla cells do not divide, rather when the dermal papilla rises up to the bulge, the dermal papilla cells communicate to the epithelial cells through Wnt signalling to replenish the dermal papilla cells and then start to regrow the follicle (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3115771/).

According to Olsen 2003 during miniturisation the dermal papilla gets smaller and smaller, suggesting that hair follicle size depends on the number of dermal papilla cells. Further Cotsarelis found that only CD34hi and CD200hiITGAS6 were missing in balding areas, yet KRT15hiITGAS6hi was still present (and the implication I've gotten so far is that it can make the CD200hiITGAS6 cells)

https://www.ncbi.nlm.nih.gov/pubmed/15617565

If signalling between the dermal papilla and bulge was diminished then the dermal papilla cells would have no way to proliferate, the less dermal papilla cells the less of an ability they have to communicate with the bulge to create more cells.

That would also explain DHT inhibition regrowth. Suddenly the DP can communicate with the bulge again and send a signal for more cells before it dies. It also explains how SM04554 works since it's a B-Catenin Agonist. It must tell the bulge to create more dermal papilla cells.

I don't know enough about minoxidil to factor that in, but I don't think anyone really does.

Oh yeah, and I can't believe how insidious AR is, I mean could it upregulate it's own gene? That would be crazy!




.

Thank you Beowulf for your post

When I read these interesting thoughts, I ever think in sebum, More clear, in hardened sebum in this area. We all know that bulge and the duct of sebaceous gland are in the same "district", where sebum is released in this area, and it is possible degenerate and oxidate in hardened sebum, ..., so it is possible that the comunication between dermal papilla and stem cells in bulge area is alterated.

This is an important key in my theory, to try a explanation of common hair loss, where exist miniaturized hairs before the lost of them. Current theory is not able to explain it.

 

[Beowulf]

When I read these interesting thoughts, I ever think in sebum, More clear, in hardened sebum in this area. We all know that bulge and the duct of sebaceous gland are in the same "district", where sebum is released in this area, and it is possible degenerate and oxidate in hardened sebum, ..., so it is possible that the comunication between dermal papilla and stem cells in bulge area is alterated.

Yeah I noticed in the literature that there's a big debate about the sebaceous and the bulge. The thing about sebum though is that most people use Nizoral, so if sebum is a piece of the puzzle. It might be an environmental factor that sets it off to begin with.

 

[Armando Jose]

Nizoral can eliminate the sebum of hair and scalp but not the sebum inside of the skin and near the duct of the sebaceous gland.

 

[Beowulf]

Nizoral can eliminate the sebum of hair and scalp but not the sebum inside of the skin and near the duct of the sebaceous gland.

I just figured it did since people were using it as an antiandrogen. To be effective it would have to get to the bottom of the hair follicle.

What about salicylic acid? They use that to cure sebaceous hyperplasia apparently.

 

[Armando Jose]

The problem, IMO, is not the quantity of sebum produced by SG but that it remain a lot of time in the vicinity of the gland duct and it gets oxidized and rancid. In other words that sebum flow could be blocked for any reason.

 

[Beowulf]

The problem, IMO, is not the quantity of sebum produced by SG but that it remain a lot of time in the vicinity of the gland duct and it gets oxidized and rancid. In other words that sebum flow could be blocked for any reason.

I'm more wondering how you could clean the remaining sebum out.

 

[Nadia1972]

Nizoral can eliminate the sebum of hair and scalp but not the sebum inside of the skin and near the duct of the sebaceous gland.

My body produces too much sebum which results in lacrimal glands that clog. I think my Androgenetic Alopecia and my sebum problem to the eyes are the same disease
I think the process of the Androgenetic Alopecia is that the hair smothers and the root dies by an excess of sebum.
In fact, I do not produce too much sebum. My sebum is too thick and hardens in the lacrimal glands which clog them

 

[Nadia1972]

I'm more wondering how you could clean the remaining sebum out.

To clean my lacrimal glands blocked by sebum, I have to press with the fingers to remove the sebum
For the Androgenetic Alopecia, I must use mechanical means such as massage or dermabrasion

 

[Armando Jose]

Massage can help without doubt,
My bet is a penetrating lotion used during hours.

BTW this post is about big-new-genome-wide-association-study-on-androgenetic-alopecia.
Maybe a new post is more appropiate