Drug discovery and development scientists spend a lot of time wondering if their candidates are getting across the right membranes, or trying to get them to do so. Think about what happens when you take a pill containing a small-molecule drug that needs to get to an intracellular target: it dissolves in the stomach and/or small intestine, and then has to make it across into the bloodstream from the inside of the gut. There are quite a few ways for that to happen (see here and the following slides on that site), and there are quite a few ways for it not to happen as well as you’d like, either. Once in the bloodstream, your drug molecules will be floating around, probably bound to serum albumin or to some other protein, or perhaps even slipping through the membranes of the red blood cells and sitting in there. But all of these are equilibria (they’d better be), and to get to their targets these molecules are going to have to slip through the vascular epithelial layer in the capillaries and find a way to get across the cell membranes in the target tissue. There’s a whole list of ways to do that, too, naturally, and an equally long list of ways that the process can go wrong.
There’s an uncountable amount of work that’s gone into trying to understand what sorts of drug structures do well in these processes, versus ones that have problems. And there are uncountable arguments about what’s useful and what’s harmful about the recommendations that have come out of these efforts, and arguments about the arguments. To be honest, you can find exceptions to almost every absorption/membrane penetration dogma in the small molecule space, so laying down the law is a risky move. But if you move past that area of chemistry, there are some rules that are hardly ever broken. Antibodies, for example, rarely if ever are taken up inside cells. Very highly charged polymers (like modified polysaccharides, or indeed oligonucleotides) don’t really cross well, either, and there are plenty of other examples.
But that doesn’t mean we wouldn’t want to do such things from time to time, so over the years there have been similarly uncountable proposals for ways to get around these limits. A lot of these involve presenting cell membranes with something that they might not recognize as foreign, and taking advantage of the many existing import mechanisms in a Trojan-horse fashion. It’s fair to say that these ideas can work, and indeed have worked. But they need a lot of fiddling on a case-by-case basis, and we could use all the help we can get in the search for more universally successful platforms.
Here are a couple of interesting new papers along those lines from Nature Chemistry. This one has some work on engineering a known cell-penetrating peptide (Tat), which is an idea that’s had a lot of variations over the years. Tat is a viral protein originally from HIV, and it does all-too-good a job at penetrating into cells by itself. In this latest work, the authors build some chemical cores with various geometries and attach Tat peptides to the “arms” of these structures in different combinations. One that carries three Tat sequences, each of them contained in a large cyclic structure, seemed to work the best, and the paper actually demonstrates its ability to shuttle entire antibodies into cells. Even that, though, seems to be the combination of at least two different mechanisms, likely even more, with somewhat different time courses. And another big variable is what happens once the cargo crosses into the cell. A big route for cell entry is endosomal transport, but once inside the cell, these endosome “bubbles” can have very different fates. Some of them just sit there and don’t release their contents very well, while others go straight to the lysosomes and dump it there, which is usually not what you’re after. In this case, the tricyclic Tat species seems to be good at producing endosomes that just unload straight into the cytosol (“endosomal escape“).
Meanwhile, this paper is tapping into the very lively area of biomolecular condensates, the membraneless droplets that appear in several parts of living cells, appearing and disappearing under a wide range of conditions. They have found some droplet-forming peptides that can carry useful cargo along with them (including mRNAs, large peptides, and more), and seem to enter cells in a way that bypasses the endosomal pathways entirely. Their peptides include disulfide bonds that are sensitive to attack by glutathione inside the cell, which causes them to fall apart on a time scale of hours once they’ve made it into the cytosol. The sequences of the peptides are quite important – modifiying single key amino acid residues or making droplets without the unzippable disulfides give you droplets that sit around unchanged in cells for days, apparently unchanged, and not releasing anything at all. These things could even transport hefty enzymes such as beta-galactosidase (MW= 430,000) into the cytosol while maintaining its activity, which is quite a feat. The authors envision delivery of siRNA payloads, CRISPR enzymes, DNA plasmids, and all sorts of protein therapeutics. It would be interesting to try some of the heftier “small molecule” ideas that are coming along these days (large bifunctional degrader molecules and the like) to see if their membrane-penetrating activity could be improved, too.
It’s really good to see this sort of thing, and it makes a person dream of a day when membrane penetration isn’t this nasty black box that causes your drugs to fail in ways that you have trouble ever fixing. A more universal platform – better, several of them – that could make these barriers less of a barrier would surely improve our drug discovery lives in all sorts of ways. . .
