Well, I think that I finally found one piece of what I'm looking for with regards to the link between androgens and TGF-beta. The study below involves the prostate, but it's what I needed to know with regards to androgens, the androgen receptor, TGF-beta, and Smads. This demonstrates the link between all these molecules. It's a different tissue, but the receptor should work the same.
Here's the abstract:
http://www.jbc.org/cgi/content/full/277/2/1240
The Androgen Receptor Represses Transforming Growth Factor- Signaling through Interaction with Smad3*
J. Biol. Chem., Vol. 277, Issue 2, 1240-1248, January 11, 2002
In the prostate, androgens negatively regulate the expression of transforming growth factor- (TGF-) ligands and receptors and Smad activation through unknown mechanisms. We show that androgens (dihydrotestosterone and R1881) down-regulate TGF-1-induced expression of TGF-1, c-Fos, and Egr-1 in the human prostate adenocarcinoma cell line, LNCaP. Moreover, 5-dihydrotestosterone (DHT) inhibits TGF-1 activation of three TGF-1-responsive promoter constructs, 3TP-luciferase, AP-1-luciferase, and SBE4BV-luciferase, in LNCaP cells either with or without enforced expression of TGF- receptors (TRI and TRII). Similarly, DHT inhibits the activation of Smad-binding element (SBE)4BV-luciferase by either constitutively activated TRI (T204D) or constitutively activated Smad3 (S3*). Activation of SBE4BV-luciferase by S3* in the NRP-154 prostatic cell line, which is androgen receptor (AR)-negative but highly responsive to TGF-1, is blocked by co-transfection with either full-length AR or AR missing the DNA binding domain. Immunoprecipitation and GST pull-down assays show that AR directly associates with Smad3 but not Smad2 or Smad4. Electrophoretic mobility shift assays indicate that the AR ligand binding domain directly inhibits the association of Smad3 to the Smad-binding element.
In conclusion, our data demonstrate for the first time that ligand-bound AR inhibits TGF- transcriptional responses through selectively repressing the binding of Smad3 to SBE.
This is an angry study filled with a lot of biochemistry, but the bolded part is the most important aspect of this study.
Now, I would need to see a study demonstrating what I'm about to type, but this is how my mind works.
Normally, the androgen receptor inhibits TGF transcriptional activity which means that androgens normally inhibit all downstream effects of TGF-beta. However, in an individual with an abnormal androgen receptor in the scalp, which is what happens in male pattern baldness baldness with triplet repeats creating a dysfunctional receptor, TGF-beta transcriptional activity is no longer repressed and all downstream effects can continue including fibrosis and collagen deposition through fibroblast activation and connective tissue growth factor expression.
It's short and it's sweet, but I think that might be what actually happens. But, remember that this is in-vitro, so I still need to see a study that links a faulty androgen receptor to not only increased TGF-beta (which we have), but also to the removal of any repression of the binding of Smad3 to SBE
Something else that needs to be addressed is the how 5AR inhibitory drugs actually grow hair. To be honest, I don't know. It would seem to me that DHT inhibition wouldn't have a role either way in this pathway as it's the androgen receptor that's the problem, not DHT. Androgen binding to a faulty androgen receptor shouldn't matter unless androgen binding still causes the activation/inhibition of another process along with the faulty receptor failing to inhibit TGF-beta's processes.
Also, I still can't explain why this only happens in the scalp other than embryological patterning and that still doesn't explain why all androgen receptors aren't faulty.