Friday, December 07, 2007

is heme a hormone ?

Well, the short answer is: probably not.

But this is a long story, and an intriguing one (at least for people interested in protein biochemistry). I spent much of this week researching, writing, rewriting etc. the news item which is based on a paper in Nature Structural Biology, this press release, and communications with five different experts in the relevant fields. Essentially, the researchers found that heme (or haem), the cofactor in haemoglobin and other important proteins, binds to the receptors REV-ERBalpha and REV-ERBbeta, which were known to influence circadian rhythm, but were "orphan" receptors, meaning that nobody knows which hormone controls them.

In the paper, the researchers just report that heme binds the receptors, suggesting it may have a signalling role linking metabolism and the biological clock. No objections so far. In the significantly sexed-up press release, however, heme is referred to as a "hormone" and all kinds of wonderful medical applications are promised. The trouble is, hormones normally travel around and convey information from one cell type to another, and there is no evidence of heme doing that. This aspect is discussed in detail in my news piece.

After the piece was finished I hit on another fly in the ointment. I found out that there are two crystal structures of the REV-ERBbeta ligand binding domain in the Protein Data Bank, which the authors of the NSB paper hadn't mentioned to me. According to the crystallographers, these structures include the hormone binding site. Now the authors of the heme paper say that the crystal structures don't include the heme binding site which they confirmed by mutagenesis.

So if heme binds in a place that is distant from the hormone binding site, to me that says loud and clear that heme is not the hormone that regulates REV-ERBbeta. If anything, it is a co-regulator, and the receptors are still semi-orphaned.

Oh, and the claim for medical applications isn't much better. I wouldn't want to take a drug that competes for heme binding sites. I guess the take-home lesson is never to trust a press release ...


PS, the whole heme story reminded me of a classic Irving Geis illustration, where heme is shown as the light source illuminating the protein cytochrome c from within. The picture is shown here as Fig. 2.

No comments: