I’ve been thinking a lot about disulfide bonds off and on over the last few months. Normally the first thing people think of when they hear disulfide bonds is protein folding, stability, and tertiary structure. But when I hear them I think of one disulfide bond in particular: the one in my protein sitting at the bottom of a binding pocket.
Ever since I saw this disulfide bond I wanted to take advantage of it. I wanted to develop a way to interact with it. So naturally I went to the literature. Turns out people don’t do this, perhaps for some reasons I’ll get to, all there is concerning disulfide bonds is what you’d expect to find: folding, stability, tertiary structure, enzymes, and of course in vitro reduction. Yes, there’s more there than that but not surrounding my interest…or so I think at least.
All the hype is surrounded around cysteine, which was highlighted just today by Derek Lowe concerning covalent drugs. And this was one of my initial thoughts. Let’s find a way to reduce the disulfide in vivo and then have an electrophile like a Michael acceptor around to generate a covalent bond. But of course that is fantasy land because the pocket only allows for something slightly larger than a benzene ring. That would be demanding too much in such a small space with absolutely no specificity. Not to mention how difficult it is to reduce a disulfide bond in vivo.
So, I’ve concentrated my thought efforts towards a nucleophilic bomb, so to speak. The idea is being able to selectively place a great nucleophile in that pocket where it can attack the disulfide in a manner that ideally creates a irreversible inhibitor. However, I’d even accept something that creates an interaction with the disulfide. What do I mean by “interaction”? Take a look at glutathione and other enzymes. There is often a cysteine residue in the active site. Often a disulfide is reduced and then another cysteine creates a new disulfide. If unregulated the reverse reaction can occur and regenerate the original disulfide. The kinetics of this has been well studied. But this would still be an “interaction” even if it wasn’t an irreversible bond formation. Once again we come to this issue of needing the disulfide to be reduced first, however, what if you had something more nucleophilic than cysteine? Say selenocysteine? Well now things might get interesting.
Again, this comes back to being able to selectively place that nucleophile in that precise pocket. But for arguments sake, let’s say I can do that (I’m actually very confident I can do that for reasons I can’t discuss here). My question then is will this nucleophile have off target effects on it’s way to this protein both in vitro and in vivo. I would bet yes. Lucky for me our lab specializes in drug delivery, so I have some things I can try to get a targeted delivery of whatever my final compounds look like. Thus, as a proof of concept the only burden that lies on me is showing good binding affinity (i.e. better binding affinity than whatever else I can place in that pocket).
From the other spectrum I’ve also thought about stabilizing the disulfide bond. Remember, all I want is an “interaction” and stabilizing counts. So far I’ve only seen metals such as zinc and copper that do any stabilization and most of that has been shown within an enzyme that coordinates the metal and disulfide with several other amino acids as ligands. Maybe there’s something there, but I start to worry about toxicity. Speaking of toxicity, mercury also interacts with disulfides. I thought about that for a hot minute and even figured out some synthetic routes to incorporate mercury into my molecules before abandoning that thought.
So, I don’t know. This is still something I’m playing with and trying to find more literature on interactions with disulfides that don’t involve in vitro reduction and enzyme cleavage/usage. If I’m missing something obvious feel free to send it my way. But I’ve never seen an example of this either in a known drug or a molecular tool. Until then I think I’ll start diving into the bioconjugate techniques textbooks for more fundamental (and applicable) insights.