VDR is the nuclear receptor in question. The activation of this receptor is complex and its regulated, as are many other steroid activated nuclear receptors, by a co-activator, AP-1.
See:
http://endo.endojournals.org/cgi/content/full/140/1/63
This sentence, from the opening paragraph in the introduction section, really says it all:
"The AP-1 class of transcription factors can inhibit the activation of the receptors for glucocorticoid, progesterone, estrogen, androgen, thyroid, and retinoids."
AP-1 itself is activated by anabolic steroid analogs (AAS). A quickie search on Google using the terms AP-1 and anabolic steroids confirmed this suspicion.
Just how much these anabolic steroids affect AP-1 transcriptional regulational will vary somewhat within individuals due to differences with receptor polymorphisms for various targets (named above). They read like a who's who list responsible for the desired primary and undesired secondary tissue specific effects of anabolic steroids.
Note that we now have 3 levels of AAS action (to produce the side effect):
AAS steric binding blockade of VDR receptor activation
AAS initiation of AP-1 transcriptional regulation
AAS steric binding to estrogen, glucocorticoid and retinoid receptors, all of which also co-active VDR in bone cells.
Because regulation can go either way - activation or inhibition, depending on the degree of ligand "fit" - physical / chemical match between the natural activator and the wannabe (such as AAS) into a receptor binding pocket - there will be some AAS agents that may have a positive effect on bone (osteoclast) function.
The National Institutes of Health apparently was thinking along these lines, in this 2002 request for research proposals (RFP):
http://grants.nih.gov/grants/guide/pa-files/pa-03-008.html
So I can't really answer your questions on how long to estabilish potential feedback regulatory effect on mature and developing bone cells. I can't even tell you if all AAS have negative consequences. I guess the SERMs and some SARMs (estrogen- and androgen-like agents, respectively) have a potential positive effect as PCT. BUT, whether they can compete with other AAS on-cycle? You would need studies (like the ones proposed with the RFP) to answer that question.
What I have given you, and other who will have the patience to read this response, is a moment of pause, as you come to appreciate the potential for long term effects from AAS use.
You got the signals for densification from structural compressive forces during heavy exertion (training). And you have probable alteration of the receptor that answsers that signal. For some steroids, its a positive effect (nadrolone acting through estrogen receptor), for some, acting on the glucocorticoid and vitamin D receptors, it maybe a negative effect.
Can't say for sure how long these AAS compounds take to exert their effects. In various studies I looked at, the net gain in those with osteoporosis (men and women) was 3% increase in density per annum (thats without the mechanical induction signals for bone densification), expressed with the first few months of treatment with AAS. Negative impact on liver from anabolic steroids is the primary negative action that dictates the period of use as a theraputic agent; that is manifest in a matter of a few months of steroid use.
Reversibility of side effects induced by these agents depends on a variety of factors.
Sorry, not the answer you wanted, Pengers and dg, but its what you get.