Skip to main content


Slipping Into the Brain

We’ve got a lot of longstanding technical difficulties in the drug development business – things that we’d really like to be able to do, but can’t quite manage to find any general solutions to. Measuring drug concentrations inside cells (or parts of cells), that’s one. Predicting human toxicology for a new compound is another, for sure. And there’s also the problem of getting things past the blood-brain barrier that the body doesn’t want to let past.

For those outside the field, the actual BBB isn’t like some sort of steel plate around the bottom of the skull – it’s actually the capillaries that penetrate every region of the brain. It’s a rather heavily perfused organ, on the absolute scale – 15% of cardiac output going into about 2% of the body’s volume. But inner layer of cells in all those blood vessels are exceptionally closely packed, with help from astrocyte and pericyte cells. The end result is that entry into the brain’s fluid space is tightly guarded. Substances that fall outside of the preferred zones for molecular weight, size, and polarity generally don’t make it in at all (and the ones that do tend to have specific active transport pathways tailored for them, as witness the large number of glucose transporter proteins up there). Active transport is involved in the related problem of things being pumped right back out of the brain as well, even if they get in.

You wouldn’t want to just permeabilize the BBB overall, because most of that stuff that’s being excluded is being excluded for a reason. But we would like to get various drugs, proteins, antibodies, oligonucleotides and other species across in a more targeted fashion, and the literature over the last few decades is full of various solutions to that problem. I’ve written here before about attempts to use ultrasound to temporarily make the junctions looser, but as far as I know that technique is still rather experimental, if indeed it’s workable at all. Another popular technique has been to try to hitch a ride on those active transport proteins – if they could be persuaded to let in other species that have the right “passport” features hooked on to one end of them somehow, for example. These things can work, but I’m not aware of any that have really achieved wide usage, because every case is different.

There’s also been a lot of research into packaging such agents into new formulations, physical forms that have an easier time getting through – nanocarriers, for instance. The body’s own membrane-packaged exosomes have been researched for this purpose, too. I was talking the other day about lipid nanoparticles (LNPs), but those don’t seem to make it through. A new paper, though, has a trick that might allow them to. The team (from Tufts) synthesized lipid-like molecules that have neurotransmitter molecules (tryptamine, e.g.) attached to one end, via good old synthetic organic chemistry, and make lipid nanoparticles out of these. The resulting “neurotransmitter lipioid” (NT)-doped formulations actually seem to get taken up by the active transporters that recognize the neurotransmitters themselves.

This is demonstrated with several interesting examples. Amphotericin B, for example, is a very important systemic antifungal drug of last resort, but it doesn’t cross the BBB. Formulated this way, though, it gets in. The paper goes on to show delivery of an antisense oligonucleotide to the Tau protein gene, and demonstrates that after such dosing that Tau mRNA does indeed go down in the brain tissue. A gene-editing Cre construct gets delivered as well. This is demonstrated in a mouse line that has a red fluorescent “Tomato” protein’s gene which is “floxed“, a technique that allows such a gene to be specifically targeted by Cre recombinase protein. In this case, it’s a “flox-stop-flox” system, which means that if the Cre recombinase is successfully delivered, the stop is removed and the tdTomato gene gets expressed. Sure enough, neurons of the treated animals fluoresce red when checked several days after treatment with the Cre formulation (cortex and cerebellum show more strongly than the hippocampus, for reasons unknown). There are some of the cells glowing at left – the team used this technique to evaluate several different formulations.

There are a lot of possibilities here. The paper takes on a number of different known lipid carrier formulations and just dopes them with their NT-lipidoid species, so it’s easy to try out a number of different ideas. And both the Cre and ASO examples demonstrate that at least some of these can go on to intracellular delivery once they get past the blood-brain barrier. I look forward to seeing how this gets developed!

20 comments on “Slipping Into the Brain”

  1. Toni says:

    that’s really fantastic.

  2. Bob Seevers says:

    Very nice work. Spin-off company in 3…2…1

  3. Colin O'Carroll says:

    The old Genentech CNS group who are now at Denali have taken the Trojan horse approach a little further. Here’s their latest in monkeys

  4. Metaphysician says:

    Out of curiosity, what are some of the clinical effects of a severe general permeation of the bbb, and is there anything known that can cause such? Quick googling only turned up technical papers beyond my limit, and sketchy websites.

  5. James Millar says:

    Will the NT carrier bind to receptors, or does being attached to the lipid change it enough that it’s inactive?

  6. Barry says:

    Extraordinary that the active porter for lipophilic Dimethyltryptamine (mw 188 amu) should provide access for e.g. GFP. Perhaps the liposome and payload are endocytosed and don’t pass through the NT porter at all?

  7. milkshake says:

    Please be very, very cautious about the therapeutic nanoparticle stuff. Investing your retirement savings into a biotech company pursuing therapeutic nanoparticle research means you may never retire. The nanoparticle literature is polluted with irreproducible results, data manipulation and misrepresentation, and even with outright fakery – and wishful thinking seems to be modus operandi in the subfield. Look at how many companies doing therapeutic nanoparticles went bust.

    Maybe this particular paper results are real but I will believe it when it is independently reproduced by another group, and even better, when it demonstrates effect in the phase II clinical trial.

    There is so far just single nanoparticle-formulated drug on the market, abraxane, and the reason why it works better than older formulation of paclitaxel has nothing to do with the nanoparticles per se, but with replacement of toxic excipient cremofor with albumin excipient – the nanoparticle stuff is completely incidental

    1. Derek Lowe says:

      The caution is justified, for sure. I like this idea and have no reason not to believe the results, but it’s still a long way from human application. And a long way from being part of a stock portfolio!

    2. HFM says:

      Biotech startups are like Vegas – never put money on the table you can’t afford to lose. That said, yeah, I agree that the house’s odds are a bit high for my liking in this area. Too many flim-flam artists chasing buzzwords in a space where we don’t have solid intuition on what is or isn’t plausible. Still has potential, though, and I hope it works out someday and makes people who aren’t me very rich.

      1. milkshake says:

        it is not just about the buzzwords. The bigger problem is in the lack of precise characterization with nanoparticles. PK is tricky enough on its own, but nanoparticles and their interaction with plasma proteins and their release of cargo, sequestering in organs, passive penetration through the vessels etc ads extra level of complexity. Fly by night companies promoting their nanoparticle technology will tell you about their tremendous success, showing you improved area under the curve (AUC, = integrated exposure = concentration multiplied by time) fivefold. But they do not tell you that they used HPLC method for analysis, and during their sample extraction from plasma they destroyed the nanoparticles and are looking at the total compound in the plasma. Or they will claim to prevent CNS dose limiting toxicity by showing their compound does not get partitioned into brain after a single dose, and withhold the information that the CNS toxicity shows up after repeated dosing, just little delayed in time because the nanoparticles are relatively long-lived, so it takes about 1-2 days for the compound to get out of nanoparticles and build up in brain.

        So it is about elementary scientific integrity – it is unfortunately absent from this subfield. Which is the reason why so little came out of it, since nature does not care about PR fluff and does its own thing. Unfortunately, the drug development cycle is so slow that charlatans promoting nanoparticle formulated drugs can cash out early, and ruin the chances of success for everyone else in the field. So we need better research and review standards, and more retractions, and more public shaming of the hucksters.

    3. Andreas Åslund says:

      One should always be catious of new treatment modalitites before they have been into clinical trials, but to claim that there is only one nanoparticle based medicine on the market is grossly wrong. Eventhough nanomedicine have been strugling (as all new technologies does), it has in recent years shown unique capabilities in clinic. And for the record Doxil was approved 10 years before Abraxane.

      A recent perspective on nanomedicine:

  8. Barry says:

    Didn’t CellGate claim to have solved this year’s ago with poly-Arg tails?

  9. anon the II says:

    That synthetic chemistry is reversible. Is this not a great way to deliver acrylates into the brain?

  10. JasonP says:

    I have always been interested in the work by Berislav V Zlokovic and his contention that the BBB is indeed quite “leaky” (dysfunction) in some, especially in chronic neurodegenerative disorders.

    1. Brainless in Seattle says:

      Right… and that’s how antibodies cross into brain parenchyma….. /s

  11. Howard Rosenberg says:

    Derek – Any update on Antivirals and MAbs for Covid-19? I looked and it’s been a while. Thanks for keeping us informed.

  12. tlp says:

    Has anyone considered opposite – attaching glucose to the payload so it gets in via glucose transporters? Just curious.

    1. tlp says:

      never mind, found some examples

  13. Just asking says:

    I wonder if anyone can comment on your experience on the effectiveness of using perfusion “with saline to wash away any free LNPs or DiR remaining in the blood vessels”.

    It is my understanding that this method can leave up to 50% of blood remaining in the very small brain capillaries, even with water soluble compounds. Probably even less effective with these nanoparticles?

    Thank you.

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.

This site uses Akismet to reduce spam. Learn how your comment data is processed.