The cell membrane – the fundamental architecture of living things, the foundation of how our very bodies are organized – is a major pain in the behind. I express this ungrateful opinion merely because over the years it has rejected entry to some of my best ideas for drug candidates, and I bear a grudge. I am not alone: membrane permeability is one of the most basic properties that we look for in a drug. Even if you’re going after a target on the outside of the cell, your compound has to make it to the cells in the first place, and some membrane-crossing ability will probably come in handy there, too.
There are rules of thumb. Some ranges of size and polarity are definitely more likely to make it through than others, because they have a better chance of passively diffusing across (although some disagree about that). That’s the place to start, although there’s no substitute for just throwing your compounds at the cells and seeing if they do anything or not. One big complication is that there’s active transport going on, cell-membrane proteins that are deliberately moving things in and out. So some unlikely molecules can get taken in, and some perfectly good ones can get thrown back out, and there’s often not a lot that we can do about that (it’s a major problem in treating both bacterial infections and some kinds of tumor cells, both of which have latched onto these efflux pumps as survival strategies).
Even short of drug therapy, though, it would be very useful just to get some classes of compounds into cells in the lab. Fluorescent probes, modified proteins, selective and potent synthetic molecules that just bounce off the membranes – you could learn a lot about targets and pathways if you could just reach down and jam such things into the cells, even if you’re not going to be able to do it to a patient in the clinic. A number of ways have been developed over the years to do just that kind of thing, with probably the best known being electroporation.
That’s pretty much what it sounds like: it turns out that if you expose cells to an electric field, their membranes become somewhat more permeable. My mechanistic understanding of the process is rather crude (but I’ll bet that’s true of a lot of people who use it regularly, too!) As you increase the electric field, the membrane resembles a capacitor, with opposite charges piling up on each side of the (not very conductive!) lipid bilayer. Eventually, you reach a point where defects appear – the charged species meet as the membrane folds in, making temporary watery pores that close up again on a fairly short time scale. Unless, of course, you keep jacking up the voltage and current – in that case, you can rip the membranes apart entirely.
The exact conditions needed to do this in a controlled fashion are empirical – depends on the cells, their loading, the buffer, the design of the electroporation gizmo and its electrodes, and so on. As I understand it, many mammalian cells don’t stand up to rough treatment too well, whereas bacteria (for whom rough treatment is a way of life) can be tingled more vigorously. Since the pores produced are hydrophilic, this technique works quite well for charged species, including (very conveniently) hunks of DNA and RNA. Electroporation is, in fact, a workhorse technique for introducing new genes into bacteria and other cells. One complication is that cells do not particularly enjoy being electrified and having their membranes zapped, so the technique can set off gene expression changes of its own, in a cellular “What the hell was that?” response. (The same applies, for the most part, to another similar technique: sonoporation, where ultrasound is used – carefully – to disrupt membranes for gene transfer).
That hydrophilic pore mechanism means, though, that electroporation is not necessarily so great for putting in (relatively uncharged) globular proteins or drug molecules. So there are slightly more elegant ways of playing the membrane penetration game, such as finding some sort of small molecule or peptide sequence that is optimized for some active transport protein and hitching a ride with that. Cell-penetrating peptides have been a big field of research over the years, and there are quite a few known, but the technique has a couple of downsides. You have to make a new physical conjugate of your payload with the cell-entering piece, which may or may not be feasible or even a good idea once it gets inside, and you also have to choose from a variety of possible vectors, often with little idea of which one (if any) will do the job.
In recent years, another vector-free technique has come on that works in a complementary fashion to electroporation. Here’s a new paper in ACS Chemical Biology that uses it, and it’s a perfect example of Don’t Be Afraid of the Obvious. It turns out that if you cause cells to queue up, one by one, in microfluidic channels and then force them to squeeze through tiny bottlenecks, this places their membranes under enough stress that small molecules can slip through. This goon-squad idea has been used in the past, but there was always a problem with cell death, because it’s hard to control the conditions. Microfluidic techniques, though, allow you to reproducibly put the squeeze on the cells, adjusting for the range between maximum uptake and minimum death.
The openings produced in this manner seem to be fundamentally different from the electroporation ones, because uncharged molecules get in quite well. That new paper referenced above shows a series of Pfizer JAK inhibitors (otherwise cell-impermeable bricks) crossing right over, which is nice. This is just what’s needed for better target-engagement studies in living cells, because we chemists and chemical biologists have, unfortunately, a pretty good backlog of interesting molecules that don’t go where we want them to.
Here’s a side issue, though: I was reading this new paper for more insight into the difference between this technique and electroporation, and came across something puzzling. The ACS Chem Bio paper says (links added by me) that “In line with previous results, the fluorescent signal was found to be diffuse in the cytoplasm and not in endocytic vesicles as previously seen with electroporation.” But when I go to that second reference, which is on siRNA delivery, its authors make a point of saying that “A few minutes after electrotransfer, the cellular localization of the siRNA was observed spread homogeneously throughout the cytoplasm“. There’s no mention of endocytic vesicles. I wouldn’t mind if someone with a more intimate knowledge of membrane penetration techniques could clarify this (!)