This educational article covers a published synthesis of lysergic acid, the precursor of the psychoactive drug lysergic acid diethylamide or LSD.
A team of chemists recently reported a synthesis of LSD in only 6 laboratory steps! We will look at the chemistry behind it and uncover some other insights – for example, how do chemists measure how trippy a molecule is?
Rationale for LSD synthesis
So these scientists, are they a bunch of Breaking Bad wannabes or why would they investigate even more chemical syntheses of LSD? Well, LSD derivatives such as bromocryptine can be pharmacologically useful for treatment of neurological, metabolic and other disorders. This means we want to get more efficient at making LSD-like scaffolds for drug discovery.
In 2020, there was an interesting structure-activity relationship study. It showed for the first time that psychedelic compounds, such as derivatives of DMT, can be engineered lose hallucinogenic side effects while retaining their useful psychoplastogenic properties. The left-hand side 5-methoxy-DMT makes you trip. The isomer with the methoxy substituent shifted by just one carbon, does not. While this might be disappointing for some of you, it’s obviously better if patients are not hallucinating weird shit after taking their pills.
If you wondered – trippy-ness can be estimated by looking at how often mice violently shake their head after administration of psychoactive drugs. This is a well-validated proxy for hallucinations and was first established already 70 years ago! You can see that while 5-methoxy-DMT leads to head twitching, the 6-methoxy isomer has no significant hallucinogenic activity. There’s actually a nice concentration dependent relationship.
Six-Step Synthesis of Lysergic Acid
So how does this super-quick route look like? This synthesis builds on a key intramolecular Heck reaction which creates the key vinyl bond that is present in LSD. This Heck-approach is not an invention of the 2023 synthesis, as it had been used in previous, longer syntheses already. However, this route efficiently traced the intermediate back to this indole containing. This starting material can be bought commercially and conveniently has the bromo group for the Heck reaction. Obviously this makes a lot more sense than unnecessarily taking apart the indole ring. Let’s take a closer look at the specifics of this synthesis.
The first step was a magnesium-halogen exchange of this iodopyridine to create a heterocyclic nucleophile. This one is happy to attack the electrophilic carbon of the functionalized aldehyde, leaving a hydroxyl group in the product. As you might remember, there is no oxygen in LSDat this position. Thus, the next step simply removed this group by reduction with triethylsilane.
The acid used in this step removed the N-Boc protecting group, so they re-installed afterwards. After this protection, the most nucleophilic group is the pyridine nitrogen – so it was methylated with methyl triflate. This gave a pyridinium salt which was reduced by sodium borohydride. Two hydride equivalents attack the ring: The first one gives the reduced tertiary amine that is part of LSD. The second hydride reduces one of the double bonds, leaving the alpha-beta unsaturated ester. All of this happened in the same reaction vessel. But still, the authors were a bit sneaky to categorize this as just one single step.
But wait – to enable the key Heck coupling reaction, the olefin actually needs to be located at the other carbon. They achieved this by using LiTMP as a strong base. The resulting isomerized anion which can be protonated in a diastereoselective manner. The desired isomer has the ester on the same side as the existing hydrogen of the 6-membered ring. While the preference isn’t great, they formed it in slight excess over the undesired one. Conveniently, they found subjecting it to the same conditions recycled some of it to the desired product.
The Heck reaction proceeded with the standard mechanism. Oxidative addition of Pd(0) allowed for olefin insertion and creation of the C-C bond in blue. Now, given there are two beta-hydrogens available, there are two pathways towards elimination. There’s the orange hydride elimination, and the pink one, which is preferred in a rough 1 to 3 ratio. Note that the stereochemistry of the ester became wobbly again the orange product. This is because the reaction occurred at a 100 degrees with mild base with some isomerization taking place.
Even though we end up with three different products, it’s no big deal. They simply added potassium hydroxide to all, and heated things up to get to lysergic acid in around 50% yield. This is double-deprotection and isomerization. Natural products are usually stable isomers so it’s not surprising that the isomerization forms the configuration present in LSD preferentially. Unfortunately, their final product is not so satisfying as they only isolated a brown solid. I don’t suggest supplying this to the dangerous dealer in the neighborhood. I’ve seen some procedures getting nice white crystals but these folks didn’t care too much about ultra-pure product.
Lastly, they showed that this synthetic route could be useful to explore and study LSD analogs – remember the methoxy-substituted DMT structures at the start? The started with a chloro-substituted indole starting material and replicated the reactions – including the Heck reaction – to create a C12-cholor-lysergic acid derivative. Theoretically, you could create different LSD analogs now by functionalizing the aryl chloride – which might help scientists find future drugs based on LSD with differentiated therapeutic profiles.
If you are interested in the synthesis of other psychedelics, check out the discussion of ibogaine, psilocybin, MDMA or THC-P.
LSD synthesis references
– Six-Step Synthesis of (±)-Lysergic Acid | J. Org. Chem. 2023, 88, 2158 – Identification of Psychoplastogenic N,N-Dimethylaminoisotryptamine (isoDMT) Analogues through Structure–Activity Relationship Studies | J. Med. Chem. 2020, 63, 1142
The thalidomide tragedy was the biggest “man-made disaster apart from war”.
This molecule’s structure looks simple and innocent, but a chiral carbon gives rise to two enantiomers with different pharmacological effects. While one isomer was a safe sedative, the other one led to limb malformations in thousands of babies. We will check out how this works exactly, but notably, the enantiomers interconvert in the body so there’s no way to control them.
Get this: Despite increasing doubts of its safety, the US distributor of this horror drug aggressively pushed for its approval simply to maximize sales over Christmas!
If you think the heart-breaking tragedy sealed thalidomide’s fate, you couldn’t be more wrong. Half a century after the tragedy, it generated up to $500million sales for a pharma company! But wait, there’s more. A slightly modified, much more expensive version of thalidomide brought in eye-watering 12 billion dollars.
How is it possible that the same drug which inflicted so much damage revived and even influenced one of the biggest acquisitions in pharma history? As we will see, the answer is glue – no joke. The story of course wouldn’t be complete without classic big pharma monopoly games and price hiking. And let’s not forget everyone’s favourites – lawsuits.
In this post, we will go through history, biology, organic chemistry and pharmacology of thalidomide and its many cousins. You will also appreciate some paradoxical and morally questionable aspects of drug development.
What is Thalidomide?
Let’s get into it. Thalidomide has a very simple chemical structure. In the 1950s, the relatively small and inexperienced pharmaceutical company Chemie Grünenthal looked for new antibiotics. Instead of antibiotic activity, thalidomide seemed to be a great sedative and help with sleep or nausea. Unfortunately, it doesn’t help the story that some of Grünenthal’s leaders were ex-Nazi scientist. The research head Mückter was involved in death camp experiments and made bank during his tenure. It started all rosy. Initial safety tests in mice and rats showed good tolerability and no side effects even at remarkably high doses. Back then, you didn’t have to understand how drugs worked at all. So, thalidomide was dubbed completely safe and aggressively marketed in 1957.
It quickly became Germany’s second best-selling pharmaceutical, just behind Bayer’s Aspirin.Ironically,safety was one of its key marketing messages. Many pregnant women used thalidomide for morning sickness. The lack of toxicity was convenient as unlike barbiturates, this agent couldn’t be misused for suicide attempts. However, over the next two years, sudden increases in cases of usually rare limb defects were detected in newborns.
It’s less known that there were prettyearly findings of teratogenicity and neurotoxicity from various researchers. Some of them, like these observations on tad poles, were shared with Grünenthal already in 1959 – with no response.
Thalidomide Side Effects: Beginnings
Because the incidence of deformations increased so unprecedently, German paediatricians suspected an environmental factor. In late 1961, thalidomide use by mothers during early pregnancy was the common factor. Only later did we find out that deformations were actually just the tip of the iceberg. Thalidomide induced many more miscarriages and less obvious defects like organ problems. After increasing noise on the issue, the German government pulled the drug off the market against the company’s wishes. The adverse impact of early use is so high that even a single tablet was enough to induce pregnancy loss or abnormalities – but why? To understand, we need to recap two topics.
First, we need to know about the ubiquitin proteasome system, basically, cellular garbage management. Some of you might remember the process from your biochemistry classes.
The process starts with very few, so-called E1 enzymes which are activated with ubiquitin, a small regulatory protein consisting of roughly 80 amino acids. This green ubiquitin tag is ultimately what directs the degradation of target proteins. As you can imagine, there are thousands of proteins that the cell might want to be able to degrade. However, it would be challenging to do this specifically if all you have are a few different E1 enzymes. This is ubiquitin groups are cascaded to a broader variety of E2 conjugation enzymes which finally put the tag on more than a thousand so called E3 ligases. These enzymes recognize specific substrates and upon addition of enough ubiquitin tags, the proteasome shreds up their targets into smaller peptides.
Second, we need to know about molecular glues. These work exactly like you would think. The glue molecules bind between two different proteins, aggregating or gluing them together. This can lead to several effects, but targeted protein degradation is the most important one.
How does this relate to thalidomide? Well, the molecule binds to the protein cereblon which is part of a E3 ubiquitin ligase complex. Once bound, it can act as a glue between cereblon and neo-substrates, innocent molecules. This exposes them to the E3 ligase machinery, so they are ubiquitinylated and degraded.
Remember that the adhesion arises from nuanced interactions between functional groups of the molecular glue, cereblon in purple, and its neo-substrate in green. We don’t know the natural targets of cereblon but amongst others, it’s critical for brain development, hence its name.
Thalidomide Side Effects: Cereblon
By acting via cereblon, Thalidomide-initiated protein degradation influences the body’s immune system (immunomodulatory drug). The mechanism is very complex, but one of the innocent casualties is SALL4. This is an important transcription factor that governs gene expression for normal limb development. Its absence results in deformations which is why thalidomide proved so dangerous for pregnant women. Actually, genetic deletion of SALL4 replicates a similar phenotype. But why was this missed by Grünenthal? A critical piece of information that was not known in the 1950s – rodents are resistant to thalidomide’s teratogenicity. This explains the absence of safety signals, even at high doses. Only later did people figure out that rabbits or chickens are more sensitive animal models. Why?
The susceptibility comes from a single amino acid difference in cereblon sequences. Primates and rabbits with a valine suffer thalidomide embryopathies – while rodents with an isoleucine did not show any safety signals. Interestingly, the bushbaby bears an isoleucine and is the only known resistant primate. Scientists demonstrated that mutant mice with an unnatural valine at the position become sensitive. Compared to wildtype mice, they showed statistically significant higher miscarriage rates. This is really fascinating – a slightly bulkier amino acid influences binding and degradation of substrates such as CK1 alpha but potentially also SALL4 or others.
If you had a chemistry class or two, you would know that by having one chiral center, thalidomide has two enantiomers. Much too late, it turned out that the (S) enantiomer is ten-fold more potent protein degrader. Giving only the safe isomer as a drug is not an option. The acidic proton at the chiral center triggers partial conversion to the bad enantiomer at pH levels of over 6. If you’ve watched my deuterium video, you will know that this interconversion is slower for deuterium due to its kinetic isotope effect.
How can the almost identical (R)-enantiomer be safe? Although it has the same molecular contacts with cereblon, its affinity is much lower due to an energetically unfavourable conformation that it needs to adopt upon binding. This twisting occurs because the glutarimide ring wants to minimize steric clashing with the binding pocket, particularly the highlighted tryptophan 383.
Thalidomide Synthesis & AFtermath
If we look at its synthesis, it becomes obvious why original thalidomide is a racemic mixture. The original Grünenthal synthesis starts with a condensation of L-glutamic acid and phthalic anhydride. Even though the amino acid used was chiral, the basic conditions and high temperature result in a racemic product due to the. To close the ring, the free acids linked via activation with acetic anhydride and a last treatment with urea introduced the nitrogen.
So, the aftermath entailed a large criminal trial, examining potential negligent behaviour by leading Grünenthal employees. The process was extremely drawn out, probably the dream of every lawyer. 600thousand pages of documents without any clear verdict. Ultimately, it was said that based on the state of science at the time, the teratogenic effects of thalidomide could not have been anticipated – so the trial was terminated and settled between Grünenthal and impacted parents. The company is still providing support to affected persons through a novel foundation, with more than $100 million Euros contributed to date.
Thalidomide in the United States
The impact in the US is another ridiculous story. Grünenthal had offered the company Smith, Kline & French – today’s GSK – to market the drug in North America. SKF ran a large clinical trial which likely also resulted in several phocomelia cases, and they declined Grünenthal’s partnership offer.
However, another company Richardson-Merrell was eager to introduce it. These guys were calling up the FDA to submit a marketing authorization application in the fall of 1960. I invite you to read this nice article which contains comments of Frances Kelsey who reviewed the application at the FDA. And yes, that’s her with President John F. Kennedy, getting a medal for Distinguished Federal Civilian Service. You see, as if this story couldn’t get worse and more capitalistic, Richardson-Merrell was pushing for an early approval prior to Christmas to maximize their sales.
In a – now recognized as heroic – move Kelsey challenged the drug’s data. Something felt off about giving enormous amounts without any toxicity – so suspicions rose that other conditions could change the drug’s absorption and unveil toxic effects. Her suspicions proved to be right – but that didn’t prevent Richardson-Merrell from giving away literally millions of thalidomide tablets for “investigational use”, at the time permissible under existing regulations. The FDA cited 17 children born in America with thalidomide-associated deformities, but the true number is surely higher.
What has science learned from this tragedy? On one hand, drug controls got stricter. Prior to 1962, drug developers only had to show that new drugs were safe – and as we just saw, even that was not a given. A new pivotal amendment required strict “proof of efficacy” from well-controlled studies, and not the bro-science which Richardson-Merrell tried to pass. Drug advertising now required accurate information about side effects, and clinical trials had to include informed consent of participants prior to the study. For us in the 21st century, this seems obvious. The FDA also launched a comprehensive assessment of drugs that were already on the market. Finally, drug testing got more robust with a requirement to use rabbits and other thalidomide-sensitive species for teratogenicity testing.
Is Thalidomide Still Used Today?
Thalidomide’s risks but also benefits continue to linger. Just shortly after its initial withdrawal, it proved efficacious in ENL, a leprosy complication. To avoid teratogenicity, access to the drug depends on so-called Risk Evaluation and Mitigation Strategies, short REMS. For instance, female patients must avoid pregnancy though regular testing and use of two or more forms of reliable contraception. In the US, thalidomide was approved for leprosy in 1998 and REMS were well regulated. However, use over decades in countries like Brazil with subpar REMS has still led to some cases of embryopathy.
Although there was research on thalidomide in cancer already in the 60s, the molecule was finally proven to have anti-cancer activity in the 90s. This time around, the company Celgene got IP rights to the drug and thoroughly interrogated its potential. A landmark trial showcased its value in multiple myeloma, a type of blood cancer. Given its unique mechanism and profile, combination with other agents was powerful. This resurrected thalidomide, turning a monster drug into a precious option for patients who relapsed or did not respond to other treatment.
Mechanistically, the anti-oncology effect arises from degradation of Ikaros and Aiolos. Unlike SALL4, these transcription factors regulate the development of B and T cells of the immune system. Ultimately, thalidomide inhibits the process of angiogenesis. As a very smart person, you will realize that lower growth of new blood vessels in turn suppresses tumor growth.
The “new early” days of thalidomide remained controversial, with a whistleblower lawsuit accusing Celgene of off-label marketing. Allegedly, they actively pushed thalidomide, which as we saw was approved for leprosy, to be off label prescribed to cancer patients prior to its approval. While these off-label prescriptions extended thousands of lives, intentional off-label marketing by companies is not compliant. The company ultimately had to settle the lawsuit for 280 million dollars.
Lenalidomide: multiple myelomA Blockbuster
Thalidomide’s legacy is even more shocking. We already know that small changes have big impacts so you shouldn’t be shocked. You see, two simple functional group modifications created lenalidomide. This is the big boy. By the way, there’s even another hybrid between the two – a bit less imaginative. Due to the new structures, these analogues enjoyed new marketing exclusivity, with great commercial success.
Lenalidomide received an orphan drug designation – a FDA incentive that gives drug developers special tax incentives and market exclusivity. With a hefty original 6-figure price tag, the drug earned Celgene double digit billions in yearly sales.
Lenalidomide didn’t sell for no reason. Molecularly, it is much more potent than thalidomide across various metrics. For instance, its IC50 inhibitory value against resistant multiple myeloma cells is orders of magnitude lower.
These molecular changes also translate into better survival outcomes for patients than thalidomide. In addition, risks of neuropathies and other adverse events were lower. This is a follow-on strategy gone well: new molecule, better efficacy and financial success.
Legal Considerations
You could also argue that given Celgene’s research is not associated with Grünenthal’s initial wrongdoings, they actually changed the world for the better. You could also call it a perverse twist – a horror drug ended up as the basis for massive profit. Some have accused Celgene of using particularly fierce ways of preventing entry of generics. Doing so, it managed to command soaring prices for the drugs, and even increase them. This includes more than a dozen of patents on their REMS system which further blocked generic competition. Remember, REMS are special activities that patients, providers and distributors need to take to prevent harm from teratogenicity.
In this case, it means that access to the drug is dependent on several criteria, such as regular pregnancy tests and surveys. This is great because it encourages safe use of the medicine, but as always, we have more potential illegal activities looming.
For many years, Celgene fought it out with the generics manufacturer Mylan. You see, to call approve a generic medicine, the FDA needs to see bioequivalence data. Essentially, we need to undoubtedly prove that the copy has the same effects. So, generic manufacturers need to buy the branded drug. Allegedly, Celgene not only refused to sell the drugs directly, but they also implemented distribution restrictions that prevented Mylan from buying thalidomide and lenalidomide. This was resolved after five years in classic fashion by paying 100 million dollars and some change to settle the claims.
As alluded to, subsequently BMS acquired Celgene and with it, the knowledge on thalidomide and friends. If you thought this was the end of the saga, think again.
Outlook:More Cereblon Modulators
The development of next-generation cereblon modulators such as iberdomide is still ongoing. At a first glance, this one might look like a simple copycat molecule.
As you can see from the crystal structure, the morpholino side chain extends into a pocket on cereblon, increasing surface interactions and binding of the molecular glue.
This enhanced affinity results in a faster protein degradation of neo-substrates, and anti-proliferative activity against multiple myeloma cell lines. In normal cells, iberdomide is much more potent – compare the red and green curves. More importantly, this analog retains activity in cells resistant to lenalidomide. Due to this higher penetration, the hope is that the drug will prove more efficacious in resistant cases. It’s currently in phase 3 trials and again, the idea is combining it with other medications to stack up the effects.
We will close with a final next-generation idea: covalent modification of cereblon. Scientists have noted that there’s a histidine residue close to the molecular glue binding site. Do you already know where this is heading? By creating analogs with electrophilic groups such as this highly reactive fluorosulfate, we can trigger a covalent bond formation with the proximal histidine.
Why is this even interesting? Well, this covalent modification triggers broader conformational changes which change cereblon’s activity. The scientists found that this covalent modulator led to the degradation of the so far elusive protein NTAQ1. Thus, such experiments might unlock even more avenues for drug discovery in different tumor types.
To not overdo it, we’ll wrap up here. As always, until next time!
Key references on thalidomide science and other information:
Frances Oldham Kelsey. FDA medical reviewer leaves her mark on history | FDA Consum 2001, 35, 24
The Thalidomide Syndrome | Scientific American 1962, 207, 29
THALIDOMIDE AND CONGENITAL ABNORMALITIES | Lancet 1962, 279, 45
The Ubiquitin Proteasome System in Neuromuscular Disorders: Moving Beyond Movement | Int J Mol Sci 2020, 21, 6429
Molecular glues modulate protein functions by inducing protein aggregation: A promising therapeutic strategy of small molecules for disease treatment| Acta Pharmaceutica Sinica B 2022, 12, 3548
Exploiting ubiquitin ligase cereblon as a target for small-molecule compounds in medicine and chemical biology | Cell Chem Biol 2021, 28, 987
Crbn I391V is sufficient to confer in vivo sensitivity to thalidomide and its derivatives in mice | Blood 2018, 132, 1535
Differentiation of antiinflammatory and antitumorigenic properties of stabilized enantiomers of thalidomide analogs | PNAS 2015, E1471
Lawsuit Blames Thalidomide for More Birth Defects | Scientific American 2011
Antitumor Activity of Thalidomide in Refractory Multiple Myeloma | NEJM 1999, 341, 1565
Immunomodulatory Drugs for the Treatment of B Cell Malignancies | Int. J. Mol. Sci. 2021, 22(16), 8572
A Cereblon Modulator (CC-220) with Improved Degradation of Ikaros and Aiolos | J Med Chem 2018, 61, 535
This post explains the question: What is total synthesis? If you read this, you might be forced to study organic chemistry 🙁 or are already interested in total synthesis. In any case, this brief commentary will explain the concept very simply. For advanced content, check out this page or my videos.
Imagine you’re a pastry chef and have stumbled across a picture of the most delicious-looking cake – but you lack the recipe. How can you re-create the culinary masterpiece? It will take some planning (which ingredients do I need?), experimentation (what baking temperature, for how long?), practical skills(how do I pipe the frosting?)…
… and luck! As we will see, some chemists had to push the ambiguity to the limit – sometimes testing a thousand reactions for one step, or realizing the shape of their glassware influenced their product!
Cake: Complex molecules
Chemists and the world are interested in complex molecules – a chemist’s cake – with intriguing properties – the cake’s flavours. For instance, natural products produced by other organisms can have biochemical activities that make them candidates for potential medicines.
What is total synthesis? Simply said, it is the complete chemical assembly of complex targets from simpler starting materials – like baking a cake!
Let’s take lissodendoric acid A as an example. This marine natural product isolated from sponges contains several connected rings and functional groups (in blue: amines, double bonds, carboxylic acid). This unique molecular setup gives the molecule anti-Parkinson’s disease activity in certain model experiments. Once chemists are sufficiently convinced of such a target‘s utility – or sometimes simply its molecular beauty – they need to figure out how to synthesize it. Note: In some cases, large-scale isolation from natural sources or biotechnological production might be feasible.
Recipe: Retrosynthetic analysis
Much like chefs writing recipes, chemists design a planen route to their masterpiece. This roadmap details the individual steps of reactions needed to transform simple starting materials into the desired molecular delicacy.
This plan is defined in a backward sense and thus called retrosynthetic analysis. This means that the first retrosynthetic “disconnection” corresponds to the finalstep of the forward/ laboratory synthesis.
Why backwards? Imagine you need to replicate a cake based on an image. Your first thought will be “How do I finish off the top and add the glazing?” (i.e., last ‘forward’ step). Later, you might think “How do I assemble the various layers of the cake?” (i.e., building the cake’s complexity).
You would not directly jump into thinking “For this cake, I need exactly 10 cups of flour and 8 eggs”. This would be a premature move! You haven’t even thought about how many cake layers you need, what their consistency and thickness should be…
How a retrosynthetic analysis looks like in practice. The target natural product (+)-Heilonine (1) is believed to be a constituent in the important Chinese herbal drug “Bei-mu”, which has traditionally been used as a sedative, antitussive, and expectorant. Source: JACS 2021, 143, 40, 16394|CC-by-4.0
It’s the same in total synthesis! We first work backwards through plausible, theoretical retrosynthetic disconnections. With this, we arrive at suggested starting materials for our laboratory endeavours in real life.
Experimentation: How To Learn Chemistry
Just like baking, chemistry is highly experimental. How can you tell if baking your cake layers at 170 °C, 180 °C, 190 °C or 200 °C will work best?
1) The first source is intuition to narrow your option space. For instance, baking the layers at 100 °C will either not work or take way too long. Baking them at 300 °C will lead to a bit too much crunch 😉 ! You don’t have to try 300 °C as the outcome is already certain.
Just like a freezing flame, thinking about the deprotonation of ketones with cyanide does not make sense. Once you have studied enough theory and reviewed practice, it becomes second nature!
As an example, we know that certain deprotonation reactions will simply not work if our base is not “strong”/basic enough. This is why pKa values are important! Chemists acquire their “common sense” by studying the fundamentals of chemical concepts, reactions and synthesis. Additional intuition is gathered by reading about and performing experiments. With time, chemists get a sense of what will likely work vs. what will not.
2) The second source is existing external research. Let’s assume other chefs reported 180 °C to work well for cakes with similar albeit different ingredients. This means starting at 180 °C is a good idea.
Let’s assume a specific Diels-Alder reaction was reported to require elevated temperature (e.g., reflux in toluene). If we want to perform a very similar reaction, we can assume that we don’t need to test cold temperatures like – 78° C. By reviewing scientific publications and patents, chemists can derive starting points for their experimentation.
3) The third source are your own experiments. If you want to bake a cake no one has ever baked before, you will need to gather completely novel data points.
Let’s pick up the example of (+)-Heilonine again. As reported by the authors, use of standard conditions (data source #2, if you will) was not fruitful. Instead, thorough investigations were need to identify adequate reaction conditions for this unique educt. Such optimization efforts also draw on intuition and inventiveness.
Is Chemistry Random?
Chemistry is only predictable to a certain (low) degree. Finding successful reaction conditions during organic synthesis can be challenging or straight-up impossible!
“Roughly 1000 experiments were conducted changing every conceivable variable from the base used to deprotonate, the solvent employed, additives, and the electrophile. Emerging from this exhaustive study was the remarkable finding that the addition of LaCl3·2LiCl to the extended sodium enolate of 3, followed by quenching with freshly prepared formaldehyde gas led to the desired adduct 11 in 84% yield as a 2:1 diastereomeric mixture favoring 11 (3 g scale).“
From: 11-Step Total Synthesis of (−)-Maoecrystal V | JACS 2016, 138, 30, 9425
Finding gold nuggets can sometimes be very tedious and serendipitous. Equally random are findings like the following:
Reproduced from Chem. Sci., 2022,13, 6181 with permission from the Royal Society of Chemistry.
What you are seeing is a reaction that only works in old borosilicate reaction flasks. This can happen in cases where the glass surface of a flask is actually where the reaction occurs.
Similar to such “vessel effects”, the source of chemicals (i.e., the supplier/ vendor) can in rare cases also make or break reactions. This can be driven by trace metal impurities which catalyse or also poison certain reactions. These impurities can vary across suppliers due to different syntheses/ purifications of these chemicals. Imagine if you could only bake your most beloved cake with a certain brand of milk!
Aside from driving chemists crazy, such unexpected effects exemplify the experimental nature of chemistry. Navigating this uncertainty while using strategic planning, creativity and practical skills – this makes total synthesis an art!
Exercise mimetics: Watch this video or read the written blog below!
Are you tired of dieting or pounding the pavement like David Goggins just to shed a few pounds? Imagine a future medicine that could mimic the benefits of literally running for days.
Weight loss meds like Ozempic have been sending shockwaves through Hollywood and Wall Street. Advocacy by famous figures went viral on social media, causing supply shortages and more recently, questions on their safety emerged. The stonks of select pharma companies exploded, with Danish Novo Nordisk’s market capitalization surpassing the country’s GDP.
Is this first wave of drugs already the be-all, end-all? Side effects like loss of hard-earned gains and pooping pants lead to many users stopping treatment.
Meet SLU-PP-332, a simple small molecule melting fat and effectively mimicking marathon training in mice – all without setting a tiny paw on a treadmill, and without eating less. We will cover this molecule’s discovery, chemical synthesis, and pre-clinical efficacy. This will enrich your interdisciplinary knowledge and give you some practise for data interpretation. We will also explain how other exercise mimetics work and cover random facts, such as taking a closer look at alleged health benefits of red wine.
How obesity Drugs Work
So regardless how we feel about it, obesity is the problem for healthcare systems, and source of many problems. By 2030, nearly half of Americans will be obese – not to mention overweight.
Why even consider drugs for weight loss? We all know that exercise and diet regimens have very low compliance in reality. People just don’t stick to them even if they know they should. Also, some patients have co-morbidities that make them exercise-intolerant. It doesn’t matter how much you want to be David Goggins – if you have chronic heart failure, you can’t run for days.
Another issue: the emerging obesity drugs melt away body mass, but much of it is also muscle tissue. Drugs which can trigger fat loss, retain muscle and simulate exercise would be helpful for many people.
Obviously, with hundreds of millions of obese people, this is a massive long-term opportunity for pharmaceutical companies. This bullish outlook has resulted in strong investor interest in obesity drug developers, adding billions of dollars to their valuations.
The drugs behind this gold rush are GLP1 agonists. As mimetics of the incretin hormone GLP1, they stimulate insulin production and help manage blood sugar – this is especially key for diabetics. They also slow down movement of food in your stomach which can help patients feel fuller faster and curb hunger. Leave a comment if you want to learn more about these hyped obesity drugs in a future video – they have a massive history of research behind them. Today, we will instead check out the so-called exercise mimetics.
Science Behind Exercise Mimetics
PGC-1a is a key link between endurance exercise and physiological adaptation. Expressed in various tissues, this is the master regulator of creation of mitochondria – also known as the powerhouse of the cell – as well as other processes like glucose and lipid metabolism. Because it’s a coactivator, it interacts with other transcription factors to modulate the expression of certain genes. Looking at the example of detoxification of reactive oxygen species, we realize this gets into complex cellular biology territory. Due to this complexity, dysregulation of PGC-1 alpha disrupts physiological processes and contributes to many diseases.
Why is this relevant for exercise mimetics? Well, while various mimetics have different primary targets, most ultimately all trace back to PGC-1 alpha.
One rather famous molecule in this class is resveratrol. This polyphenol is present in many foods and wines, and it can trigger just about every effect under the sun. It likely indirectly activates the so-called SIRT1 protein, which in turn deacetylates PGC-1 alpha and ramps up beneficial activities. There’s a lot of literature on this if you want to check out the cellular biology.
Some of the first insights came from a 2006 study looking at daily intake of resveratrol in mice being fed a high-fat diet. If you are good at playing ‘spot the difference’, you will notice that fat and muscle tissues feature much denser mitochondria. It looks like these mice adapted to exercise they didn’t perform – thus, we’re calling such effects mimicry.
Shockingly, some human studies showed resveratrol actually blunted some aspects of training adaptation. Ironically, removal of reactive oxygen species by resveratrol might limit training-induced adaptations. This once more highlights that in biology, nothing is simple or black-or-white.
Based on the immature human data, the verdict on resveratrol is still open. If you check Wikipedia, you can see that no health benefits have clear evidence. Such lacking clinical data is a common theme for exercise mimetics in general, as they represent a new class of compounds.
Even big pharma companies dabbled in this space, with GSK paying 700 million dollars for a biotech working on a resveratrol formulation 15 years ago. They did not test it in obesity but rather haematological cancers. This proved to be bad luck as they killed the program after seeing increased risks of kidney failure. You can see that even introducing seemingly healthy substances like resveratrol into medical practice can be challenging.
Estrogen-related REceptors
So let’s get into exercise mimetics more deeply. To understand SLU-PP-332, we need to take a look at another investigational compound. Unlike resveratrol which targeted SIRT1, this one activates the estrogen-related receptor gamma.
This is one of three siblings of the ERR family, expressed in tissues with high energy demands. These nuclear receptors have received considerable attention for their potential value in treating metabolic diseases. As a side note, nuclear receptors are proteins chilling in the cytosol or nucleus. They can sense specific ligand molecules and in turn, regulate expression of specific genes.
As their name suggests, ERRs are structurally related to estrogen receptors (ERs). These nuclear receptors utilize estrogens as ligands and contribute to breast and other cancer types. A key ER-drug is Nolvadex – more famously used by bodybuilders to manage their gynecomastia. Back to ERRs, which despite their resemblance work via different mechanisms than ERs.
The company GSK developed the first small molecule ERR agonists already in the early 2000s. Remember for later that this hydrazone-based agent strongly activates beta and gamma, but not the alpha ERR isoform. ERR beta is a negligible player given it’s not present in skeletal muscle .
ERR gamma highly expressed in oxidative slow-twitch muscle tissues in the calves, with minimal expression in quadriceps which appear more white. Its powerful effects can be clearly seen if a spooky experiment is performed, creating transgenic mice that express ERR gamma more broadly. These super mice have deep red muscle bellies due to improved oxidative capacity, increased vascularization and bigger mitochondria. In an endurance exhaustion test, transgenic mice ran roughly 1500 meters instead of measly 600m by wildtype mice. This means without any specific training, ERR overexpression creates endurance monsters that can run more than twice as far.
We also need to look at ERR alpha, the receptor which was not significantly activated by the GSK compound. Like the related gamma isoform, it’s expressed in skeletal muscle and has similar functions.
We’ve just seen how transgenic mice expressing ERR gamma are endurance monsters. For ERR alpha, scientists also looked at the opposite model – so called knockout mice lacking this important nuclear receptor. These mice are able to live somewhat normally which means that this receptor type is not vital for life. However, if you look at the relative size of the heart and muscles compared to body weight, the knockout mice in blue have significantly lower muscle mass.
As you might expect, this means the mice have lower endurance capacity and reach exhaustion much faster. The realization here is that if lacking ERR alpha results in endurance weakness, we could be able to mimic endurance exercise by activating it with a drug.
SLU-PP-332
The first question is, how do we find an ERR alpha drug? One way is to start with the GSK compound – but wait, didn’t we say this one only activated the beta and gamma ERRs? I’ll explain. First, you have to know that the only available X-ray structure for this molecules is with ERR gamma. This tells us with high certainty what the binding mode looks like – for example, in red we can see that the phenolic hydrogen is involved in a hydrogen bond with an aspartate residue of ERR gamma. This structure can guide the simulation of how the slightly different binding pocket of ERR alpha would bind to ‘4716. As we’ve said the binding is not that strong, but we can use it as a starting point for the design of more potent drugs.
The scientists behind this research identified a crucial phenylalanine at position 328, here in pink, which is present in ERR alpha but not gamma. By engineering interactions with this unique group, we could design a drug that selectively targets alpha over gamma.
This was achieved very easily by converting the iso-propyl benzene of ‘4716 into a naphthalene ring. As you can see from the new simulation, this extended aromatic system can undergo pi-pi stacking with then phenylalanine. This simple change increases affinity for ERR alpha by more than 50-fold. Let’s compare it again to ERR gamma. As the phenylalanine is not present, the interactions are weaker here and the agent is around 4-to-1 selective for the alpha receptor.
The chemistry behind this is so easy that it can be managed by even the clumsiest undergrad . The starting materials are simple and cheap – the only thing needed is cooking them up in toluene overnight. The highly nucleophilic hydrazide adds to the electrophilic aldehyde, creating an adduct. After a proton shift, the intermediate can eliminate water, forming the hydrazone linkage of the product. As it precipitates from the solution, it can be easily separated and subsequently recrystallized to give pure SLU-PP-332.
Exercise mimetic effects
By now you are eager to hear about its effects – is it really as impressive as the clickbaity title? Let’s start from micro and go to macro. For some of these, feel free to pause and take more time to digest the info.
Upon treatment of isolated myocytes or muscle cells, the researchers observed a doubling of the maximal mitochondrial respiration rate. Obviously, more oxygen means a higher energy production. Not only are the mitochondria more productive, but there are more of them!
There were also structural differences. Here you can see stained sections from quadriceps muscle. Notice the difference? There’s significantly more green color which corresponds to myosin protein in type IIa fibers which are fast, aerobic muscles. On the other hand, there are lessred, type IIb fibers. These are muscles which act fast but use anaerobic metabolism, meaning they fatigue quickly. Interestingly, no difference was observed for slow aerobic type I muscle.
Knowing this, what do you expect regarding exercise performance? Well, mice treated for a few days with this compound showed superior endurance without any training, being able to run roughly 70% further than normal mice. Unfortunately, the experimental procedure for this assessment is less fun. You can tell because the wording says “mice were run”. If they subsequently didn’t react to electrical shocks, you know that they were legitimately exhausted.
There are additional investigations into the “why” behind this such as specific gene expression targets. I leave this topic for interested nerds to check out on their own.
One interesting finding was that extended dosing for 2 weeks led to difference in grip strength as well. Unfortunately, the authors don’t describe this in more detail. It looks like grip strength decreased over time for both the active and the control group. This might be because of accumulating fatigue or other things. It would be cool if the authors are correct – meaning if SLU-PP-332 could enhance some strength endurance, and not just pure aerobic performance.
We haven’t covered one major, obvious question yet. The molecule’s exercise mimetic effects are very intriguing, but does it have any impact on weight?
Well, in another recent study the team documented that mice treated with the drug usedmore energy while consuming the same amount of food! Numerically it looks like they even had less food.
Metabolically, ‘332 triggered a shift towards fat burning. Fatty acid oxidation increased by roughly 25%, while the use of carbs decreased reciprocally. Again, more details can be found in the paper.
The magnitude of weight loss depends on the starting point. Normal mice with healthy diet and weight did not lose weight. On the other hand, obese mice eating an unhealthy high-fat diet saw 12% weight loss after one month. The researchers also looked at a genetic model. These mice don’t produce the key metabolic hormone leptin, leading to excessive hunger and food intake. Similar to the diet-induced obese mice, these chunky fellas also dropped significant weight.
We’ve noted that providing more than just weight loss is the next frontier. SLU-PP-332 might be one step in that direction. More pre-clinical work is required to understand its long-term effects. Maybe, related, optimized molecules could be even more potent. Researchers have not seen any safety signals but tolerability, administration and translation will be key to elucidate prior to first-in-human trials. These drugs will (likely) not be launched earlier than the end of the 2030s but with the excitement around obesity, it’s definitely going to remain an interesting space.
References on exercise mimetics
PGC-1α, Inflammation, and Oxidative Stress: An Integrative View in Metabolism | Oxidative Medicine and Cellular Longevity, 2020, 1452696
Caloric restriction and exercise “mimetics’’: Ready for prime time? | Pharmacological Research 2016, 103, 158
Resveratrol improves mitochondrial function and protects against metabolic disease by activating SIRT1 and PGC-1alpha | Cell 2006, 127, 1109
Distribution and Effects of Estrogen Receptors in Prostate Cancer: Associated Molecular Mechanisms | Frontiers in Endocrinology 2022, 12, 811578
Identification and structure-activity relationship of phenolic acyl hydrazones as selective agonists for the estrogen-related orphan nuclear receptors ERRbeta and ERRgamma | J Med Chem 2005, 48, 3107
Exercise and PGC-1 alpha-Independent Synchronization of Type I Muscle Metabolism and Vasculature by ERR gamma | Cell Metabolism 2011, 13, 283
Estrogen-related receptor-α coordinates transcriptional programs essential for exercise tolerance and muscle fitness | Mol Endocrinol 2014, 28, 2060
Watch the video on YouTube or read the written post below!
When people with an IQ of 50 hate on the “evil” scientists of big pharma, they often overlook that there is relentless work and ingenious brainpower behind the discovery and optimization of medicines, including antiviral drugs. If drug discovery would be so easy, pharma companies would not spend up to billions of dollars to get just one drug to the market. Also, chemists would be balling instead of complaining about bleak career opportunities on Reddit.
In this post, we will look at the educational drug discovery journey of an antiviral drug. Just by looking at today’s molecule, you should know this is going to be a nice one – and yes, that’s a boron atom in a pharmaceutical. You will learn why using boron in drugs can be powerful, and why it’s not good if your people in clinical trials turn yellow.
Hepatitis C: Significant Innovation on Major disease burden
Hepatitis C is a severe infectious disease, leading to liver disease and serious complications. The hep C virus chronically infects over 170 million people or over 2% of the world’s population! The dilemma is that disease incidence and drug market are inversely related. The key Western Pacific, Southeast-Asian and African regions have the highest prevalence with 130 million infections and no access to HCV drugs. The US and Europe on the other hand have around 15 million infections but are the target market for drugs. HCV occurs in 6 different “genotypes” or variants across the globe which complicates treatment. High-income countries have primarily the genotype 1, which is only 10% of disease burden in low income countries. Without being a scientist, you can guess that these viral genotypes influence drug sensitivity – that’s also what we’ve seen with C19 vaccines and variants such as Omicron. Oh yeah, and there’s a problem of viral resistance.
HCV spreads via blood-to-blood contact, so injection drug use, poorly sterilized medical equipment and other pathways. In contrast to HIV or Hepatitis B, it is not a STD. An infected individual can show no symptoms for decades while increasing their risk of liver failure and cancer. The good news it that unlike infections with HIV or Hepatitis B, HCV is curable. However, while there are approved vaccines for Hep B, only few are still in development for Hep C. Early physiological studies earned the Nobel Prize in 2020, and the last two decades of saw a true revolution of HCV drug discovery.
As HCV is a viral disease, there are multiple potential inhibition points in the virus lifecycle – entry into the host cell, protein synthesis steps and packaging and release. However, if we contrast HCV drug development with HIV, it becomes evident that there was quite the slow start. While the first antiviral drug was approved 3 years after the virus was identified, it took 24 years from HCV discovery to approval. There are many reasons for this – low perceived market value, low pressure from patient associations but also a poor understanding of the nature of HCV as a disease. Without going through all of the details, advancements in structural biology, pharmacology and also clinical trial design ultimately led to the approval of various therapies.
If you squint hard enough, you might be able to read something!
Most are combinations based on interferon proteins – which work by increasing immune defense – administered by subcutaneous injection, and ribavirin – which is an oral broad-spectrum antiviral that is used against various viral fevers. This “one size fits all” had a decent efficacy of around 55% – but true improvement came only with the development of combination stacks which added a direct-acting antiviral drug. These were more targeted and robust against resistance and also came with a halved treatment duration. As the third evolution, all-oral combinations were also developed – such as the last entry.
You might think, we have more than one blockbuster antiviral drug already on the market that cured millions of patients, why should we look further into HCV? Well, all the remaining, untreated, tens of millions of HCV infected individuals can’t afford outrageous prices of up to over $50K per treatment. Neither can the WHO, who aspires to treat 80% of diagnosed populations by 2030. The development of new direct-acting antivirals thus may serve to increase market competition and lower costs, thereby enabling more equitable access. Also, many of the drugs developed were not equally effective against the six genotypes we covered at the start – so there is more work to be done.
Design of aN antiviral drug against Hepatitis C
We already mentioned that various steps in the viral lifecycle can be potential drug targets. If the virus can’t productively infect host cells, the infection will not be sustained. In 2014, GSK published studies investigating inhibitors of theNS5b polymerase, the viral RNA printer. This is the same mechanism of action of Gilead’s Sovaldi.
Inhibitors can directly block active sites of enzymes – so where the catalysis is occurring – or instead, bind to other, allosteric sites, inducing macro-conformational changes that decrease an enzymes activity. In the case of NS5b, there are four well established allosteric sites – and the palm II site is particularly interesting because it is closest to the active site, and there are many amino acids that are highly conserved HCV genotypes – so a potential inhibitor could combat all variants equally effective.
The GSK team found an inhibitor here tagged with the number (3), which nicely docked into this palm site. This impacts the enzyme in two ways: firstly, the allosteric binding of compound 3 stabilizes a so-called closed state which is much less active. Secondly, the head group of the inhibitor also interacts with catalytic Mg2+ cofactors and thereby disrupts placement of incoming nucleotides, making polymerase initiation and propagation more challenging.
So what do these inhibitors look like? Their quest started from this scaffold – a benzofuran core with a cyclopropyl group swagging around – which was published by researchers from another company. Looking at a terminal alcohol as the “head group” at the sulfonamide, they observed strong activity against wildtype genotype 1 HCV, as well as against a common mutation which occurs at the 316 aminoacid position. The activity was assessed by looking at cell-based replication systems, as well as more “raw” data from a biochemical polymerase assay.
Boron in PharmaceuticaLS?
As mentioned previously, the apolar part of the inhibitor is bound in the grey palm site while the head group looks into the polar active site. By screening through different head groups, the GSK team found that by appending a boronic acid pushed activity into the single-digit nanomolar range, even for the critical polymorph 316N. For you morons out there, the lower the concentration, the stronger the inhibition. Aryl boronic acids were also more effective, particularly for wild type 1a. In contrast to the para isomer, the meta substituted aryl boronic acid showed reduced activity – highlighting the need for proper orientation of the head group.
What about the need for boron? Well, while substituting the boronic acid with other polar groups maintained strong IC50 in the assay in the right-column, it led to significantly lower replicon activity. This is because anionic compounds suffer from lower cell-permeability – due to the low acidity of boronic acids, this problem does not occur for this series. However, when looking at drug metabolism and pharmacokinetics, DMPK, they found that these aryl boronic acids had very low bio-availability in their rat model, and were rapidly cleared and excreted. This challenge was significantly improved by adding a fluorine substituent to the phenyl ring – increasing bio-availability 5-fold, while even further increasing activity, particularly against the 316N variant. They tried to further optimize activity by embedding the boronic acid into a ring – while this resulted in better performance in the polymerase assay, especially the 316N efficacy decreased.
Although less active, this analogue has an easy but instructive synthesis that nicely tests your understanding of fundamental reactions. First, this starting material was carbonylated – of course, this only touches the aryl bromide as fluorides usually don’t react with Palladium. Next, a Sandmeyer reaction exchanged the amine into a new aryl bromide, going through a diazonium as an intermediate. Then, radical bromination installed the benzylic bromide which was coupled with the free sulfonamide of the apolar core. An easy borylation again leveraged the aryl bromide as a functional handle. The final cyclization is triggered by reduction of the ester with Lithium borohydride – which is quite cool.
By the way, if you are wondering about the use of boron in pharmaceuticals. The application of boron in medicine dates back to the early 19th century, when boric acid, so B(OH)3, was used as a mild antiseptic. However, boron derivatives were long neglected thereafter as a result of largely unfounded claims that they are unstable and toxic. After the approval for bortezomib in 2003, there has been quite an upsurge in interest and we will further explore why it can be quite powerful.
Back to the medicinal chemistry: After feeling satisfied with the activity, the team performed crystallizations to reveal binding modes. There are various supramolecular interactions at work – quite basic apolar and polar interactions, but also cation-pi interactions of the electron-rich benzofuran with the positively charged Arginine side chain in the protein. This Arginine is conserved across all HCV genotypes and is the reason why this scaffold is well-suited to occupy this position. This even has a ripple effect as this Arginine anchors a network of hydrogen bon interactions to other parts of the inhibitor. Notably, the boronic acid did not form any covalent bonds or complexes – instead, it forms a hydrogend bonds with a bound water molecule, as well as other polar interactions that are not indicated here.
Optimization and Structure-activity relationship Of the Antiviral Drug
So hey, we got ourselves the final antiviral drug at hand already – right? Well, the molecule we flashed at the very start actually looks slightly different from the last lead compound we saw. Why aren’t we done yet? You see, when the team moved forward to human studies, they saw that the antiviral drug had a very short half-life in blood plasma of 5 hours, resulting in a higher anticipated daily dose for efficacy. Additionally – back to the importance of DMPK – they found that metabolic breakdown resulted in a major metabolite with long half life. This compound was also observed in Phase 2 of another drug, and associated with strong liver toxicity. Although there was no direct evidence of toxicity, the team wanted to avoid its formation to limit adverse reactions and improve its pharmacokinetic profile. The key step to prevent was oxidation of the benzylic carbon – quite evident given that the team could also detect the carboxylic acid product.
To maintain a similar binding mode and not risk starting from scratch, they hypothesized three major approaches that might reduce the propensity for benzylic oxidation. The first two consisted of cyclization, either onto the phenyl ring or in the direction of the sulfonamide moiety, while the third simply excised the benzylic CH2 group. This shortening of course came with the risk of severely disrupting the binding of the head group due to its positional shift. You might think – what the heck, why are we doing these complicated things – can’t we just throw a methyl group on the benzylic position and hope that steric shielding reduces the rate of oxidation? You would be correct – but unfortunately, the authors found that adding a methyl group reduced potency 50-fold. This was probably because the Methyl substituent induces a unfavorable conformational twist – and that’s the authors envisioned the two cyclization approaches to strive towards a more pre-organized conformation.
Let’s look at their results. As you can see in this table, cyclization onto the aryl ring actually reduced inhibition significantly, particularly for 316 variants. So, no good.
Within approach B, the team replaced the sulfonamide with carbonyl containing groups because it appeared that only of the oxygens made meaningful contact with the NS5b protein. The results were mixed – for some analogs, even wild-type inhibition decreased significantly. However, the oxazolidinone 25 showed very good potency. Just again demonstrating that the boronic acid is critical, the team found that removing it resulted in an over >100-fold loss of activity.
So, compound 25 was quite encouraging – and the team figured shifting the ring to an aromatic system should be even more metabolically stable, improving the clinical profile of the molecule. They found that the triazoles, such as compound 31, were basically as potent as the previous oxazolidinone. Removing one of the nitrogens decreased activity significantly as it removed a critical hydrogen bond with the protein. Notably, adding substitution to the triazole diminished activity, highlighting the steric constraints in the pocket. In summary, these aromatic designs were not really better – so the team also looked at the approach C – directly eliminating the culprit, the benzylic CH2 group.
Binding Mode of the antiviral drug
This series was right on the money. Recognizing that shortening of the head group would change the position, the team also looked at the meta-boronic acid – but the para-substitution continued to perform better. Here you can also see the dramatic impact of electron-withdrawing groups on activity, particularly on 316N and 316Y genotypes. But the chloro compound 47 was even more potent – this is why they moved away from the fluorine. The final refinement was closure to a benzoxaborole to enhance chemical and metabolic stability – and thankfully, this modification did not lower the activity too much.
Looking at crystal structures revealed an extensive network of interactions, mediated by highly ordered water molecules – the 3 red balls – which the authors did not observe with other series. Here we see the beauty and complexity of supramolecular interactions – the oxaborole moiety interacts directly with two water molecules: One contacts the backbone N–H of a glycine and the second H-bonds to another ordered water molecule bridging an arginine and asparagine. A close look at the boron reveals that its trigonal planar geometry is slightly distorted – it looks like the proximal water is well-positioned to occupy the empty p-orbital of boron and induce a more tetrahedral configuration. The distance between boron and water is 2.5 Angstrom, which is close enough for a strong interaction but not as short as predicted for a covalent bond. This setup is basically an equilibrium between a water-bound and unbound boronate complex. This interconversion of planar to trigonal binding, leading to multiple potential binding modes, is why boron is such a powerful and flexible functionality.
The final question was to check whether this chemical optimization was actually reflected in an improved pharmacokinetic profile of the antiviral drug.Looking at PK in rat, they found that compounds 33 and 49 had much lower in vivo clearance, and were much more bio-available. Looking into drug metabolism, the team also compared cytochrome p450 inhibition. These enzymes are the major route of elimination for multiple drugs and their disruption is one of the most common mechanisms leading to harmful drug-drug interactions and side effects. For example, analog 33 had a micro-molar activity versus major CYPs, raising a potential risk for clinical development. On the other hand, lead compound 49 was less active, posing significantly less risk.
This improved profile was demonstrated when in first-in-human studies, where they saw much longer drug half life and no oxidation to the potentially toxic metabolite which triggered them to re-explore their strategy. GSK then progressed this asset to phase 2, combining it with an RNA-based treatment called RG-101 of Regulus Therapeutics. After they saw two cases of seriousjaundice – so patients turning yellow – the FDA put a hold on RG-101 and GSK actually also decided to not further develop this compound. Maybe not the success story we were all expecting.
I hope you learned a thing or two from this story. See you next time!
References on Hepatitis C Antiviral Drug Discovery
Design of N-Benzoxaborole Benzofuran GSK8175—Optimization of Human Pharmacokinetics Inspired by Metabolites of a Failed Clinical HCV Inhibitor: J. Med. Chem. 2019, 62, 7, 3254
Discovery of a Potent Boronic Acid Derived Inhibitor of the HCV RNA-Dependent RNA Polymerase: J. Med. Chem. 2014, 57, 5, 1902
HCV796: A Selective Nonstructural Protein 5B Polymerase Inhibitor with Potent Anti-Hepatitis C Virus Activity In Vitro, in Mice with Chimeric Human Livers,and in Humans Infected with Hepatitis C Virus: Hepatology 2009, 49, 3, 745
If you took any chemistry classes ever, you’ve heard that benzene is particularly stable due to its aromaticity. Well, benzene is cute but it pales in comparison to the massive Kekulene. It’s so apolar and insoluble that you can only dissolve it in spicily hot solvents and need to measure your NMR spectra in custom-made solvents at 200 °C. But is Kekulene super-aromatic?
What is Super-Aromaticity?
Before we get ahead of ourselves on super-aromaticity, we need to understand aromaticity. In the 19th century, pioneering chemists started to explore the field by doing all kinds of random reactions. They were puzzled that benzene was so unreactive towards addition reactions even though it was assumed to have a high degree of unsaturation. August Kekulé first proposed the cyclohexatriene structure for benzene in 1865. You might wonder what the big deal is – but remember that at this point in history, structures for compounds were still lacking. In one of his reports, he said that a vision of ouroboros, the snake that eats it own tail, inspired him to think of the mesomeric structure.
The actual fundamentals and rationale behind aromaticity were discovered some 60 years later – with the famous Hückel rule on 4n+2 pi-electrons. His concepts were quite unrecognized for two decades – apparently also due to his lacking communication skills – but his contributions made him a cornerstone of organic and physical chemistry. Looking at molecular orbitals, it became that 4n+2 corresponded to a full set of binding molecular orbitals, resulting in higher stability of aromatic compounds.
Upping the ante, the concept of super-aromaticity envisions that macrocyclic conjugation in large, cyclic polycyclic aromatic hydrocarbons leads to an increased stabilization of the molecules. In 1951, the physical chemist McWeeny postulated the potential existence of Kekulene. In 1965, 100 years after Kekulé’s seminal work, first synthetic investigations were published and the molecule was named in Kekule’s recognition.
How Many Pi-electrons in Kekulene?
The key question for Kekulene was which electronic model best represented its structure: Is it a simple, localized structure consisting of 6 benzene rings? This is what McWeeny postulated already in 1951. Or was it rather a super-aromatic one that is based on two connected annulene rings, blue and purple, that both satisfy the 4n+2 rule?
The implications of this setup are quite significant, and one of them relates to diamagnetic anisotropy. When benzene is placed in an external magnetic field during NMR, its pi electrons circulate in the conjugated plane. This new ‘green’ magnetic field that opposes the external light blue magnetic field. If you are a benzene proton, you will feel the new magnetic field in the same direction. This leads to de-shielding and higher chemical shift in 1H-NMR. The same is true for an outer red proton in [18] annulene, where electrons are delocalized over the whole ring. What changes is however the experience for the inner, blue proton. Here, the induced magnetic field opposes the external one. This shields the proton and leads to a lower chemical shift, even negative in the case of annulene. The implications for Kekulene are clear. If global delocalization is a thing, we should see highly shielded inner protons like we see in annulenes.
Synthesis Of Kekulene (Staab, Diederich)
To answer what Kekulene looks like, we have to make it in the lab and characterize its properties. There’s only so much you can compute. We will first look at the landmark synthesis of Kekulene achieved by Staab and Diederich at the University of Heidelberg. Staab is most known for inventing the CDI reagent for hydroxyl and amine derivatization. Diederich, his PhD student at that time, majorly contributed to our current understanding of supramolecular chemistry and medicinal chemistry.
The synthesis by Staab and Diederich was based on almost 2 decades of work! It starts the nitration of meta-xylene and condensation with benzaldehyde. Next, a large-scale hydrogenation with 1.6kg of starting material in 50L of solvent, reduced the double bond. This set the stage for a nice cascade featuring a Pschorr reaction. It resembles a Sandmeyer reaction as it proceeds via oxidation of the aryl amine to the diazonium. This is reduced by copper, triggering an intramolecular cyclization reaction. The cascade happens on both sides in one pot, but yields the pentacycle in poor yield only.
To add more rings, they bromomethylated the pentacycle via an electrophilic aromatic substitution. Then, the benzylic bromide was converted into a thiol in two steps through nucleophilic substitution with thiourea and basic cleavage of the adduct.
To build the Kekulene scaffold, they coupled the two halves through double nucleophilic substitution. This reaction was performed under high-dilution conditions with only 1 mM concentration. This favored intramolecular closure to the ring instead of successive intermolecular reactions and led to a high yield of 60%.
The sulfur was a useful group to build the scaffold – but at some point, you have to remove it. To prepare this, the team methylated the dithiacyclophane, and then subjected it to base-mediated stevens rearrangement. This led to isomers of ring-contracted thio ethers.
Sulfur-ExtrusionReactions
To remove the sulfur completely, the team looked at different methods. The first approach was methylation to the sulfonium which is a leaving group, and can thus be eliminated. However, they had to explore another route as by-products were very difficult to separate from the desired product. This was accomplished by oxidation to the disulfoxide which was then pyrolysed at 450 °C.
As a side note, they also looked into the exotic Ramberg-Bäcklund contraction to give the olefin. It proceeds via alpha-halogenation of the sulfone and subsequent intramolecular substitution – leading to a three membered ring that can again eliminate SO2. It’s quite a miracle they even managed to isolate the 2 mgs or 1% yield of this reaction by preparative TLC. Obviously, they opted for the other approach instead of this one.
Finally, the product was photo-cyclized to yield octrahydro Kekulene. Here they found that using the saturated starting material was vital for success. They initially tried to install the double bond first and then perform the cyclization last, but this was unfruitful. Probably, creating the fully planar system makes it too rigid – instead, the aliphatic CH2 groups add some flexibility that is needed to enable photochemical reactivity.
Last, DDQ oxidized the octahydro derivative. Just to solubilize the reagents and achieve the reaction, they used trichlorobenzene as a solvent and let the reaction run for 3 days at 100 °C. They found Kekulene to be so insoluble that they had to recrystallize it from boiling, 400 °C hot triphenylene by “slowly” cooling to 300 °C – whatever slowly means here.
Is Kekulene super-aromatic?
Growing Kekulene crystals allowed them to investigate its molecular structure. They found very low variation in bond lengths, even for the inner protons. By looking at bond lengths, they derived that on the basis of X-ray analysis, Kekulene appeared not to have globally delocalized electrons and super-aromaticity.
What about NMR? Well, due to its low solubility, the team had to resort to creating deuterated trichlorobenzene and recording NMR spectra of saturated solutions at 200 °C to get anything usable. When they finally measured the compound, they showed the inner protons to be extremely de-shielded. Basically, the contrary of the delocalized annulene example we talked about at the start. The protons look like benzene protons, suggesting that the super-aromatic structure was incorrect.
Modern Synthesis Of Kekulene
So much for the oldschool chemistry. Remember the initial Pschorr reaction with low yield? The team of Perez envisioned a clever Diels-Alder short-cut by using this commercially available bistriflate. Addition of fluoride triggers elimination towards a triple bond, which can engage in a 4+2 cycloaddition with styrene to form a six-membered ring. After re-aromatization and release of the second triflate, another Diels-Alder reaction yields two isomers – one desired “cis”-like isomer, and a trans one. After some optimization, they managed to increase the yield to 28% with a 2:3 mixture of cis to trans. This means the net yield is also just around 11% – but you save yourself all other steps in the beginning of the synthesis.
Staab’s and Diederich’s synthesis truly stood the test of time, as the team also looked into synthetic alternatives after this step. It seems even modern methodologies could not improve or even re-create the original synthesis. The point of the Perez team by the way was to perform ultra high-resolution atomic force microscopy – basically making nice pictures of single molecules.
So, is Kekulene super-aromatic? Based on their findings and calculations, they also concluded that Kekulene does not have delocalized pi-electrons, and that the Clar model with 6×6 pi systems is the most appropriate one.
Chemists create even funkier Kekulene-like molecules like Septulene (google it’s structure!). But this is where we will stop for today. Catch you in the next one!
Watch the video on YouTube or read the written blog!
Molecules take very intriguing forms. You might know cubane, the literal molecular cube. Now, scientists recently reported the synthesis and characterization of 1-azahomocubane, one of the first of hopefully many more cubane analogues. Similar to weird species like Dewar benzene, these molecules can offer chemists new insights into effects and limits of ring strain.
Today we’ll start with a trulyelegantoldschool synthesis of simple cubane, and then dive into azahomocubane’s modern synthesis and properties. Doing so, we also learn: 1) how people make the absolutely mad octanitrocubane; 2) how chemists can use the unstable antiaromatic cyclobutadiene as a reaction partner; 3) why azide groups are very useful for rearrangement reactions and last, 4) why you should never throw away your old products.
Hard to believe, but cubane was first synthesized already in 1964. Philip Eaton, part of the first cubane synthesis, continued to be a driving force of cubane research. Almost 4 decades years later, he also led the first synthesis of octanitro-cubane. As you can imagine, this required some unpleasant reaction conditions – and of course, this thing goes boom. The introduction of nitro-groups worked via so-called interfacial nitration of a cubyl anion at the melting interface of frozen THF and N2O4. I’ll gladly pass on this. Also, the last step requires you to think: “mhm ah yes, let’s do an ozonolysis of a nitrosyl-heptanitro-cubane”.
Octanitrocubane was hypothesized to a potentially best-in-class non-nuclear explosive based on theory – but its experimental density was shown to lower. There are no public records of larger-scale synthesis and testing, so it’s just researched from a computational point and remains elusive.
Back to normale cubane – mindblowingly, just after 2 years, an incredibly efficient synthesis was published. As promised, it uses cyclobutadiene as a starting material, which seems crazy but also very logical given cubane is full of squares. You will know that this is an anti-aromatic compound and very unstable, so it is not possible to store it in a bottle. However, oxidizing this iron complex, which can be handled better, leads to release of cyclobutadiene within the reaction vessel – in this case, allowing for a [4+2] cycloaddition with this diketone which bears two bromo groups and another olefin – both essential to this efficient synthesis.
Note: The Favorskii reaction proceeds in a “quasi-Favorskii” mechanism. This is because the alpha position is not enolizable due to the high ring strain of the bridged system (Bredt’s rule).
After the endo cycloaddition, the two double bonds are well positioned to engage in a [2+2] photoaddition, obviously one of the key reactions available for construction of cyclobutane rings. Now we have this basket-like intermediate, which can undergo a base-mediated double quasi-Favorskii reaction, leading to ring contraction of the cyclopentane rings. This creates the cubane system, and now you can simply remove the acid groups in some step via decarboxylation. The yields are suspiciously consistent at 80% but even if they would be lower, it would not take away from the nice sequence and use of cyclobutadiene.
Very timely, there was a super recent publication on another, metal-free way to liberate cyclobutadiene – quite nice work and much better than creating some toxic iron compounds. This works via a retro [2+2] addition to release nitrogen and create reactive cyclobutadiene. [J. Am. Chem. Soc. 2023, 145, 10, 5631]
So the little cubane boy is done – but how do we create the much more complicated azahomocubane? Well, a logical intermediate would be a simple cubane-amine. Because it’s not 1964 anymore, we don’t have to create cubanes bottom-up, rather, we can simply buy them because there are crazy folks performing cubane synthesis on kilo scale.
This di-ester-cubane is not cheap and although it comes with the cubane, it also has two esters. Because we need the mono-substituted amino-cubane, we need to get rid of one of them. This is achieved via hydrolysis with just one equivalent of hydroxide and subsequent Barton-decarboxylation of the activated acid.
So, how do we get the amine from the ester? After hydrolysis, the acid was converted to the azyl azide. This prepares the Curtius-rearrangement which is facilitated by the strong energy gain of N2 release. The intermediate isocyanate can be trapped with various nucleophiles, in this case, with tert-butanol to give the N-Boc carbamate product.
Simple acidic exposure releases the cubyl amine. What now? It’s azide time again! Exposure to triflyl azide, a terrifyingly reactive azide transfer reaction, creates cubyl azide via substitution. Both of these azides come with a hefty explosion warning, so they are best handled only in solution and behind a blast shield. Adding some acid now, you guessed it, leads to another Curtius-like rearrangement with migration of the alkyl rest to displace N2. The resulting carbocation is then trapped by the acetic acid to give this product. After all of this, we now have a single nitrogen in the correct oxidation state.
Because we need the nitrogen in a corner position, there is more to be done. Once again, let’s do a sweet rearrangement. N-chlorination creates the opportunity for a new C-N bond formation, initiated by the fragmentation of the acetate. This is reminiscent of the Favorskii ring contraction we saw in normal cubane’s synthesis, just this time, the halogen leaving group is bound to nitrogen. The intermediary amide can be hydrolyzed and voila, we finally positioned nitrogen in a cyclobutane. Now, we just need to create the final linkage of the extra carbon and nitrogen. We have the ester carbon to work with, so the final steps used the methyl ester. Now it’s just a simple orgo freshman sequence of ester reduction, chlorination and intramolecular SN2 to get to aza-homocubane. As this final product is volatile, they prepared the salt form for easier handling.
So does azahomocubane go boom like octanitrocubane? No, clearly not – just because it looks strained, does not it can release nitrogen or CO2 as part of an explosion. It’s decomposition is much less exciting actually, exposure to acid or simple storage in the refrigerator led to ring opening of the cubane. This step was irreversible, a testament to the high instability of azahomocubane. It also makes you think that the last SN2 with the primary leaving group is probably one of the only reasonable ways to create the system in the first place.
In terms of geometry, the team obviously expected the product to be different than your generic tertiary amine. DFT calculations indicated that this was definitely less than ideal sp3 geometry, and the crystal structure – funnily found via serendipitous discovery – was consistent with this. You can see that the five-membered ring distorts the picture quite a bit, so homocubanes look more like baskets or houses, instead of cubes – if that makes sense.
What is the impact on basicity? The 1-azahomocubane nitrogen is not happy with strain energy being an order of magnitude higher – but this means that basicity is more than 10-fold lower! You might think that because there is more strain, the nitrogen might be happier to be present in some other configuration. However, looking nitrogen NMR chemical shift analysis showed that the nitrogen in azahomocubane is less electron rich compared to the other frameworks, which is aligned with the basicity trend. Notably, the hypothetical azacubane has 45 kcal/mol more strain energy than azahomocubane, which makes you wonder how long it will take to synthesize it in the lab.
Finally, they looked into hypothetical atom exchange and hypohomodesmotic reaction calculations. What is that? It’s basically a nerdy theory-crafting method of computing bond separation and formation energies, and the name is due to different sets of reaction conditions qualifying for different reaction types. For example, the hypohomodesmotic reactions shown here can be used to compute strain in cycloalkanes. Basically, you’re taking heats of formation of the individual molecules – which are known values – and figure out what is the enthalpy value Q needed to balance out the hypothetical reaction. You can see that while cyclohexane has a Q value of almost zero – because it is not strained at all – Q is much higher for cyclopentane, cyclobutane and cyclopropane as strain increases.
Applying this methodology to our question, we see that azahomocubane is significantly more stable than all-carbon homocubane. But why is that?
To shed some light on this, they modelled nitrogen lone-pair sigma star interactions with the carbon framework. This revealed that there is substantial hyperconjugation from the nitrogen with both adjacent cubic and basket handle C-C bonds, somewhat stabilizing the system compared to a normal homocubane. Both hyperconjugative effects and general orbital re-configuration are at play. For example, Bent’s rule describes orbital re-orientation once you add in electronegative substituents like fluorine or nitrogen. In such settings, orbitals with s character are pointed towards more electropositive substituents like hydrogen or carbon, so this leads to some changes in bond geometry and lengths.
This work doesn’t answer all our questions, but it shows that aza-variants of strained systems will be interesting playground for chemists. Catch you in the next one!
Did you know that the active chemical in magic mushrooms psilocybin might rewire and make the human brain more flexible, translating into major therapeutic benefit in depression and other conditions? This has led to Australia approving specific medicinal use of psychedelics as the first country.
You will be surprised to hear that psilocybin can be synthesized in just 4 steps leveraging quite simple chemical reactions. It was the legendary Albert Hofmann, who synthesized it first, and just like with his discovery of LSD reported a self-experiment. This didn’t just make him trip big time, but also unveiled some salient characteristics this molecule.
This story is not commonly talked about, so I hope you are in the mood for some 60-year-old German texts. Oh, and just like heroin, psilocybin was sold as a legit pharmaceutical (lol). If you enjoyed the discussion of history and syntheses of psilocybin, LSD, THCP or ibogaine, you will like this one too!
Today we will 1) review psilocybin’s richhistory, chemical structure, and mechanism of action, 2) look at its chemical synthesis, 3) dive into the scientific evidence behinds its therapeutic potential, and 4) last, explore its effect on brain networks.
History and Context
Psilocybin is a so-called tryptamine alkaloid found in a variety of mushrooms, but most potently in the genus Psilocybe. Due to its hallucinogenic and mystical effects, human use of psilocybin for medicinal and religious purposes dates to pre-historic times. Some notable artefacts include a Spanish cave painting – though maybe this person simply liked champignons – or more telling mushroom figurines and even statues from pre-historic Columbia. Supposedly, the mushrooms were called “God’s flesh” in an Aztec language, so you can be sure they were tripping big time.
Supposedly, the oldest indication of human use was found in an Algerian cave, estimated to date back to an insane 7000-9000 BC. It shows what we probably all drew at one point in our life: transcendental, bee-faced mushroom-shamans! Jokes aside, do you think that the interpretation on the far left is sensible compared to the original? Let me know in the comments! Some folks argue that it’s very easy to over-interpret these types of drawings and immediately think that magic mushrooms are encrypted everywhere. Also, don’t forget what exposure to the elements over thousands of years can do. What looks like a mushroom to us, might have represented something else. By the way, the esoteric drawings in the Algerian cave naturally make it also a prime site for ancient aliens enthusiasts.
Given Europeans prohibited mushrooms and other non-alcoholic intoxicants in the 16th century, their use was driven underground – so much so, that during the early 20th century western academics weren’t even sure if psychoactive mushrooms existed at all. They didn’t see the convincing Algerian shamanic cave yet, so they are excused. This changed with Gordon Wasson, a banker at JP Morgan and investigational mushroom enthusiast, who took some trips to remote Mexico with his wife in 1953. Albert Hofmann wrote that it took more than one trip for Wasson to participate in a mushroom ceremony instead of just watch, and that he might have been one of the first white folks to ever take them. A few years later in 1957, Wasson published the first ever broadly distributed article on magic mushrooms. I invite you to look through it yourself – the link is in the description. Not only is it nice to read through his personal expeditions and experiences; but the flipping through this 70-year-old magazine itself is quite hilarious, partly because it has nice oldschool ads for things like canned meatball pasta – or the inviting flavor feast of canned pork and beans. In 1957, the chemistry and biology behind mushrooms was still a mystery – but even though Wasson thought chemists have a long road to go for isolation and synthesis, this would change very soon.
Wasson was connected to the botanist Roger Heim, who as Wasson wrote, was a man with vast experience in the field of mushrooms. Heim managed to grow the mushrooms in his laboratory in Paris – so he’s like the founding father of shroom farming – and shared samples with the Swiss pharma company Sandoz where the legendary Albert Hofmann was still active, having already discovered LSD more than 10 years earlier in 1943. Hofmann soon isolated psilocybin and its active metabolite psilocin – but how?
He first tried something I mentioned in an old video talking about the mind-blowing isolation of 0.35mg of ciguatoxin, a marine natural product, from 125kg of moray eel guts. During that effort, the scientists injected their chromatographic fractions into mice and determined presence of ciguatoxin based on if the mouse would drop dead or not.
Hofmann’s Self Experiment with Psilocybin
After the unsatisfactory animal tests, the temptation got too big. He ate 32 medium-sized mushrooms – talk about not being cautious at all – and the ensuing effects left no doubts regarding the potency of these fungi.
After half an hour, he started to experience some serious sensory disruptions and hallucinations, apparently accepting the fact that his supervising doctor might sacrifice him to some ancients Aztec gods.
His trip peaked 1.5h after eating the mushrooms, and the entire psychedelic dream lasted about 6 hours. Clearly, he came back feeling quite euphoric.
This made him realize that potency of the mushrooms was not the problem, but that animals are less sensitive and clear in terms of their response to psychedelics. You can also see this in today’s research in animals, which uses doses that are usually more than 2-3 times higher than human doses. So, with this confidence and insight, Albert Hofmann consistently performed human testing of extracts to guide the rest of isolation, which worked quite well even though these were diluted samples and not dried mushrooms.
Which of Hofmann’s self-experiments do you like better: the LSD bicycle story or the mushroom dream? Let me know in the comments!
Wrapping up on history before we go to chemistry, Hofmann also rapidly developed the first synthesis of psilocybin in 1959. Thereafter, Sandoz distributed 2mg dosed tablets, nucleating several clinical studies in the 1960s and 1970s on mental disorders and other areas. However, as part of the “war on drugs” to control abuse of psychedelics, psilocybin was classified as a highly controlled Schedule 1 substance. Only after strict governmental controls were somewhat lifted by the 2000s, psilocybin became the subject of clinical investigations again; and recently picked up even more steam with the FDA granting psilocybin a breakthrough therapy designation in 2018.
structure and biochemical effects of psilocybin/ psilocin
Before we talk about these studies, we need to take a closer look at psilocybin. Comparing their chemical structures, you will realize that psilocybin and its active metabolite, psilocin, share the tryptamine core with various other compounds – for example with the endogenous neurotransmitter serotonin which modulates mood, learning and other things in humans; but also the hormone melatonin which modulates our sleep and circadian rhythm, and of course other psychedelics such as DMT. It’s always impressive to see that simple changes in chemical structure lead to massively different physiological mechanisms of action.
But wait, what’s the connection between psilocybin and psilocin? Well, just like aspiring for example, psilocybin is a pro-drug which is rapidly dephosphorylated under acidic conditions in the intestine and liver. After first metabolism, the resulting psilocin is much less hydrophilic and only now can cross the blood-brain barrier. Once in the brain, psilocin is believed to selectively bind to the serotonin 2A receptor.
With psilocin looking quite like serotonin, it engages in similar interactions in the active site of the serotonin 2A receptor. This receptor is highly expressed in certain brain regions and part of a broader family of 14 sub-types of receptors which drive the incredibly broad biological functions of serotonin. Psilocin is believed to get its therapeutic effect from its selectivity for the 2A receptor over all other serotonin receptor types. However, there is also some research showing that affinities for 2A, 2C and 1A receptors are in the same order of magnitude – so clearly, the picture is much more complex.
These receptors are so called G-protein coupled receptors or GPCRs, which if activated lead to an array of complex downstream signaling cascades. These molecular mechanisms, most of them not well understood and involving other neurotransmitters like glutamate as well, result in structural and functional cellular changes which can translate into enhanced neuroplasticity.
Let’s remember, a typical psilocybin session is just 4-6 hours long with a peak of acute subjective effects after 60 to 90 minutes. Due to the high expression of serotonin 2A receptors in the visual cortex, individuals experience visual hallucinations – at high doses, even with eyes closed. This heterogeneous receptor expression in brain regions is something you should remember for later by the way. This is really interesting, because on top of these acute effects, there are these cellular changes we’ve mentioned earlier which happen on a much slower but also more durable timescale.
Neuroplasticity refers to the brain’s ability to change throughout life and can be driven by changes in cell structure but also changes in the efficacy of synaptic transmissions. From these different mechanisms, I just wanted to highlight one study on dendritogenesis.
This team of researchers looked at structural effects of psilocybin on mice. To validate effects and dosage, we once again have the good old head-twitch response. Within the frontal cortex, they found a significant density and size increase of dendritic spines, which are neuron protrusions which help signal transmission across the nervous system. Strikingly, a fraction of these new dendritic spines was still present after a month and seemed no different than normal spines. This is clear evidence for structural change.
But now comes another interesting thing. They pre-treated other mice with Ketanserin. This compound, you can see that its left half looks like a tryptamine analog, is also a strong 2A receptor antagonist which leaves many receptors inaccessible to psilocin activation. This blocking or knockdown is reflected in the lack of a head-twitch response, as you can see here. However, the structural remodeling took place nevertheless. It might be that this happens already at much lower concentrations of psilocybin, or potentially also proceeds via other mechanisms beyond serotonin 2A. Now that we understand psilocybin’s rich history and biochemical mechanisms, we can check out some chemistry.
The research we will talk about is a synthesis of psilocybin by scientists at the Usona Institute which runs early research and clinical trials for psilocybin and 5-methoxy DMT. This is a non-profit medical research organization which is pioneering the application of psychedelics to neurological disorders. They also have an investigational supply program of the psilocybin that they synthesize themselves. Let’s check out how this works.
Organic chemistry: Synthesis of Psilocybin and Psilocin
Psilocybin can be made from psilocin, a slightly simpler molecule, by essentially reversing what would happen in the body. Albert Hofmann’s first synthesis published in 1958 and most large-scale routes accomplish this in two steps. First, psilocin nucleophilically attacks an activated pyrophosphate. Interestingly, one of the benzyl groups hanging on the phosphate oxygens intramolecularly shifts to the nitrogen, creating a zwitter-ionic species. The nucleophilic substitution on phosphorous would have been ineffective without the benzyl or other ester groups, but now a hydrogenation is requiredto deprotect the intermediate to psilocybin. This approach has quite a few challenges, most notably that atom economy is quite bad. Starting from 700g of zwitterion, just 100g of psilocybin can be isolated. In addition, the benzyl migration step to nitrogen is quite poorly understood and behaves in funky ways which makes in-process monitoring tricky. Obviously, you don’t want to start randomly guessing reaction endpoints when you are doing a reaction on kilo-gram scale. This transformation is very dependent on reaction volume and temperature control; the authors highlight than in one reaction “gone bad”, they had to filter the product over 6 days.
Instead, the Usona team discovered a more efficient way. Let’s look at the full synthesis in just 4 steps from the commercially available starting material. 4-acetoxyindole undergoes a regioselective electrophilic aromatic substitution with oxalyl chloride to introduce the two-carbon chain. In a second step, the remaining acyl chloride facilitates an amide-formation with dimethylamine. Adding lithium aluminum hyride at forcing conditions, we exhaustively reduce both carbonyls on the nitrogen chain – liberating the tryptamine – as well as the acetyl group to give the free phenol in psilocin. Now, under optimized conditions, the authors found that using POCl3 followed by immediate hydrolysis introduce the phosphate. They were able to scale this very well, delivering a whopping 1.2kg of psilocybin in one go.
This phosphorylation proceeds by stepwise hydrolysis of the di-chlorointermediate. But as you can see, the reaction setup and work-up is quite complex at scale – and this step remains the most problematic one. For instance, they had to add 3 kilograms of celite powder to prevent the formation of sticky precipitate that messed up the stirring. The hydrolysis itself was achieved by quenching the di-chloro product into a cold THF/water mixture containing excess triethylamine. Keeping the slurry at sub-0 degrees for roughly 60min seemed to work well – in case of longer hold times, they observed decomposition to psilocin again. After some more work-up, they performed a recrystallization to isolate 1.2 kg of highly pure psilocybin.
Selected clinical data on psilocybin
So probably you’ve heard various people, including drug abusers themselves, throwing around the idea of psychedelics as the be-all end-all super medicines. Indeed, there is a growing body of evidence for psilocybin, but the focus is currently on depression – in particular, cancer-related and treatment-resistant depression, as well as anxiety. Most of these studies use one or two 25mg psilocybin doses separated by around 3 weeks; and the treatment is supported by targeted psychotherapy sessions as well – so these patients are not just taking psilocybin to their liking and magically getting better.
Also, all credible research is looking at pure psilocybin, instead of magic mushroom bites – obviously, this way the effective dose of the primary active ingredient and safety for patients can be controlled. You don’t want to give your patient some random mushroom and hope it was some good stuff. However, to give credit to nature where it’s due, there is the so called entourage effect which postulates that the sum of the parts in psychedelics might lead to a greater synergistic effect compared to the individual compounds. Some research has shown that magic mushrooms might be a more potent source of psilocybin than pure psilocybin itself – likely because there are other compounds and psilocybin-derivatives that have synergistic interactions and amplify efficacy.
There is really strong evidence from the largest psilocybin depression study to date published a few months back – and this research has triggered quite some broader public attention on psilocybin. Before we go there, I wanted to briefly showcase that some evidence in other disorders is quite promising as well:
One randomized clinical study compared two cohorts of around 50 heavy drinkers undergoing psychotherapy – compared to a tobacco use study which only had 15 patients, this sample is really not that bad. These black arrows indicate two sessions, where they either received psilocybin or a sedative which served as a quasi-placebo and blinding control. What they found was that the psilocybin group had much lower alcohol use even half a year later, also translating into fewer alcohol-related issues like physical problems or impulse control.
Let’s switch gears now, looking at this large, eye-catching study on treatment-resistant depression. Resistance in depression is a huge problem, as remission rates across courses of therapy drop by a large margin. The success rate of a first anti-depressant course is around 37% – which to be honest is shockingly low – but drops to the mid-teens after 2 or more courses. Once someone failed two different courses, their depression is classified as treatment-resistant.
This Phase 2 study included 233 patients and tested three doses: a high dose of 25mg – which isn’t crazy high really; a medium dose of 10mg and a 1mg dose which they call control – not really a true placebo but probably also too low to show any effect. Because the acute effects of psychedelics are very strong, the value of placebos and double blinding itself is questionable in these trials anyways, as each participant and staff member realize who’s actually tripping or not.
I said it a few times already – psychedelics can also be dangerous or counterproductive in some cases. If you are interested, you can read through the exclusion criteria for this study. Also, the study had participants discontinue anti-depressants and other prohibited medications during a wash-out period.
Looking across the ingoing randomized groups, 60-80% of patients were severely depressed as measured with MADRS depression scores, and around 80% had failed two treatments for their current depressive episode. Notably, less than 10% had already used psilocybin in their life – this might be important because some researchers have hypothesized that previous psychedelic exposure might impact trial enrollment and efficacy on individual patients.
The treatment consisted of a one-time psilocybin dose which was administered in a calm supervised environment, with patients chilling out to a nice playlist and laying around in introspective for a few hours.
Let’s look at the results: The medium dose of 10mg did not bring any changes compared to the low 1mg control dose – however, the 25mg group a mean drop of 12 points in the depression score, which was significantly lower than the other two.
Looking at the evolution over time, we can see steep acute drop followed by a slight rebound; however, the separation of the three groups seemed to persist over time. At 12 weeks, the changes in scores we saw translated into roughly 30% of patients dropping below a score of 10, which means they are depression-free. Considering we are talking about treatment-resistant depression here, this remission rate is very meaningful – also, we are just talking about a single psilocybin dose here.
Looking at safety, the 25mg group had few more adverse events – but this was primarily driven by headaches, nausea, and not serious AEs like suicidal ideation.
I hope this gave you a more detailed look at why some folks are getting excited about psilocybin in depression – but bigger and longer trials are needed to replicate these results at scale, and bring more insights into durability of response.
Neurostructural mechanism behind psilocybin
Finally, let’s play neuroscientists. One potential reason for psilocybin’s activity is connected to large-scale brain networks which govern all our thoughts and actions. There are many different networks depending on what model and definitions you consider, but three are important for us. The default mode network is a set of interconnected brain regions that are active when a person is not engaged in any specific task or actively thinking about something. This network is often referred to as the “resting state” network, as it is active when the brain is not focused on external stimuli – and thought to be involved in self-reflection, future planning as well as social cognition – so thinking about the mental states of other people. The executive network, on the other hand, is involved in goal-directed behavior, decision-making, and working memory. This network is active when a person is engaged in a task that requires concentration and effort, such as solving a complex problem or completing a difficult math equation. The salience network is another important network in the brain, which is involved in detecting and responding to important stimuli in the environment. This network is active when a person is presented with a novel or unexpected environmental or emotional stimulus, such as a loud noise or a sudden movement. These interconnected networks are associated with cognitive control and flexibility, ultimately governing processes like learning and task switching. In depression or disorders exhibiting cognitive inflexibility such as autism or OCD disorders, the interdependency and switching of these networks is impaired.
Research published in Nature Medicine in 2022 showed that two doses of psilocybin with psycho-therapy swiftly decreased symptoms in severely depressed patient, as shown by reductions in the BDI score. Nothing shocking for us – but what’s more interesting is that psilocybin showed strong significant decrease in brain network modularity as shown by fMRI brain imaging. In the bottom right chart, you can see that default mode network activation decreased while its integration with the executive and salient networks increased. In plain English, this kinda made their brain more flexible. They also had a control arm which administered a selective serotonin reuptake inhibitor antidepressant, which lacked this integration-inducing effect.
The researchers postulate that this decreased modularity or increased flexibility of brain networks are key to psilocybin’s mechanism of action. Depression is characterized by abnormally constricted network connectivity – psilocybin might ameliorate this by broadening mental states, in line with the liberating emotions some people feel. This effect differentiates psilocybin from other SSRIs, likely due to its more targeted effect on cortical 5-HT2A receptors which are highly expressed in regions linked to these networks. There are probably some additional mechanisms at play here like reduction of amygdala response, which is another pivotal area implicated in depression.
Another area of interest is the claustrum, a thin sheet of neurons located in the central cortex, which also shows a high level of 5-HT2A receptor expression. In other research, psilocybin was demonstrated to decrease activity or turndown this area which is highly interconnected and involved in setting attention and switching tasks. You can see that there are some different effects from psilocybin on the right claustrum, decreasing connectivity with some networks, but also increasing others. Put simply, shutting down the claustrum might explain why psychedelics help with re-setting rigid though patterns, as well as unveil new psychological insights through an effective re-wiring of brain networks.
A final interesting point is related to different people experiencing different types of trips, if you will. There is some degree of association between brain structure and personality dimension as well as risk factors or pharmacological treatment response. In one 2020 study, researchers found that the thickness of the anterior cingulate, a brain compartment connected to other areas known to be important for emotion or memory, can predict the emotional experience that people get from psilocybin, in terms of feeling strong emotions of bliss, unity or others. This might be one of dozens of other factors that dictate why people react quite differently to psychedelics or drugs.
Thank you for reading and hope you learned something new today!
Key references:
Single-Dose Psilocybin for a Treatment-Resistant Episode of Major Depression | NEJM 2022, 387, 1637
Therapeutic use of psilocybin: Practical considerations for dosing and administration | Front. Psychiatry, 2022, 13, 1040217
Psychedelics and Neuroplasticity: A Systematic Review Unraveling the Biological Underpinnings of Psychedelics | Front. Psychiatry, 2021, 12, 724606
Increased global integration in the brain after psilocybin therapy for depression | Nature Medicine 2022, 28, 844
Percentage of Heavy Drinking Days Following Psilocybin-Assisted Psychotherapy vs Placebo in the Treatment of Adult Patients With Alcohol Use Disorder | JAMA Psychiatry 2022, 79, 953
DARK Classics in Chemical Neuroscience: Psilocybin; ACS Chem. Neuroscience 2018, 9, 2438
Psilocybin acutely alters the functional connectivity of the claustrum with brain networks that support perception, memory, and attention | NeuroImage 2020, 218, 116980
Seeking the magic mushroom | Wasson’s 1957 publication in Life Magazine |
Die psychotropen Wirkstoffe der mexikanischen Zauberpilze (Albert Hofmann) | Verhandlungen der Naturforschenden Gesellschaft in Basel 1960, 71, 239
Research on Acute Toxicity and the Behavioral Effects of Methanolic Extract from Psilocybin Mushrooms and Psilocin in Mice; Toxins 2015, 7, 1018
Watch the video on YouTube or read the blog below!
You probably know that drugs can get very expensive. What do you think – what percentage of new drugs actually provide significant therapeutic benefit compared to existing treatments? And do you think that drugs with a low clinical benefit are indeed priced any cheaper than highly beneficial ones?
Beyond drug prices, we will also talk about cancer “approvals gone wrong”. These were shown only after their regulatory approval to lead to equal or even worse outcomes for patients. One of these debated drugs ultimately led to bankruptcy of the company who developed it. However, the molecule has some nice chemistry behind it. You know the channel – we will take any opportunity to study organic synthesis that we get.
Healthcare Spending and Drug Prices
Let’s face the facts – healthcare spending has grown much faster than gross domestic product – in the US and everywhere else. Drug costs are one driver behind this. The most extreme example is perhaps cancer, where we can observe almost exponential growth of drug costs over the last 50 years.
The US spent $4.2 trillion on healthcare in 2021. Who paid the bill, and on what? As you can see, government spending on Medicare and Medicaid is about 40%, whereas patients paid out-of-pocket for 10% of costs. On the other hand, hospital care received the bulk of spend with 31%, with prescription drugs coming in at 9%. Increasing drug costs are problematic, but as we see, they are not the only reason behind high healthcare spend.
Initially expensive branded drugs are becoming much cheaper after entry of generic drugs. Fortunately, this leads to more accessible, vital treatment options for patients. This is where we have the crux: While standards of care improved across all diseases, new medicines need to continuously drive improvement in therapeutic benefit.
Let’s imagine we have a new drug in a disease that affects the liver, demonstrating it slows loss of liver function by 25%. Sounds good in theory, but let’s assume this was only tested against placebo and already approved treatments have shown similar results in their trials. If we ignore any benefits on things like safety or durability of response, this means the relative therapeutic benefit of this therapy would be rather low. On the other hand, a new drug showing superior efficacy to other already approved treatments would be much better. Instead of “me-too” drugs, we would want any new drugs to significantly improve on existing drugs. The problem is that this has not been the case in past years.
The Therapeutic value of new drugs
The authors of the first piece of research looked at the last decade’s new drug approvals and their ratings of therapeutic value from health technology assessment bodies in France and Germany. These agencies assess the added benefits of a drug’s approved indication compared with existing therapies. This tells us whether there is major, considerable, minor, or no benefit of a new drug. The underlying criteria are slightly different across countries but capture the drug’s effect on reduction of disease duration or side effects, for example. If any of the two agencies rated a drug as providing considerable or major benefit, the authors consider it a high value drug in their analysis.
The sad result is that only 40 to 50% of drugs approved by the US FDA or European EMA regulator are high value drugs. But check out the second set of bars. These refer to approved uses in additional diseases or indications after a drug’s first approval. In these settings, are even less likely to be quote-unquote better drugs.
Drugs with multiple indications
But why can drugs have more than one indication? As most of you know, a single mechanism of action can apply to different diseases. For instance, immuno-oncology drugs like PD1 inhibitors can be applied across solid tumors. Keytruda, the soon-to-be world’s best-selling medicine, is approved in more than a dozen different tumor types. Beyond oncology, we have complement inhibitors like Soliris which you might remember if you watched the previous video on the world’s most expensive drugs. Because the complement immune pathway is involved in many diseases, it is possible that the same drug which saves kidneys might save eyes as well.
On one hand, patients benefit because they get one more option which might help them go into disease remission (not always). It also makes sense for pharmaceutical companies to get rewarded by bringing innovation to more patients. Because clinical development costs billions, they maximize product revenues during their limited window of exclusivity.
But the reality shatters this pleasant theory: Only 40% of drugs in supplementary indications bring significant new benefit. If we look at the relative rate, drugs are less and less likely to bring new therapeutic benefit in every follow-on indication.
Some caveats remain, like questions behind the precise logic of defining something as high therapeutic benefit. There are specific local factors, like France using their own therapeutic strategy as a criterion. But more important is that while a drug might not be high value on average, sub-segments of patients with specific mutations or other characteristics might respond very well. We know who might respond best for some drugs but for most, this is less clear.
Value of Accelerated approvals & Friends
This research confirms what others have found before. An interesting second analysis differentiated the therapeutic value of normal drugs and those with expedited approval. The exception is awarded for drugs which major promise in diseases with high unmet (e.g., certain cancers, rare diseases). The logic is obvious: imagine there is a drug which shows drastic tumor shrinkage in an early phase 2 clinical trial. Instead of waiting 3-4 years for larger and longer-term data, an accelerated approval might be lifesaving for many patients. Accelerated approvals require drug developers to run confirmatory trials in parallel. The hope is that the drug will show full efficacy and safety retrospectively. In a minute, we will talk about two cases where this unfortunately was not the case.
The publication at hand also included other expedited regulatory mechanisms in their analysis such as fast track and priority review, as well as breakthrough therapy designations.
So, what do you think is value-add of these expedited programs? As you might expect and can see in the much higher purple band, drugs under expedited mechanisms are more likely to have high therapeutic value.
Most program types are in a similar range, with breakthrough designations coming in highest – 60% are indeed high value. While normal approvals are at a crappy 13%, drugs with 3 or more designations are at 65%. These designations are awarded during development, so before the clinical product profile become clear. Thus, it’s obvious that we will never get to let’s say 90% for breakthroughs, simply because drugs which appear very promising in phases 1 or 2 might prove out to be only mediocre after a larger scale phase 3 trial. Still, regulatory agencies should give additional explanations or disclaimers to set more realistic expectations for patients and doctors.
Are Drugs getting More expensive?
Finally, let’s see what a 2020 study has to say, covering 65 cancer drugs across countries. The sad insight: low-benefit drugs are just as expensive as high-benefit medicines. You can see that the blue and red distributions are awfully close to each other, with no statistically significant differences between them. From this chart, you can also immediately see why the US is the major pharmaceutical market globally, with drug costs roughly 1.3-times higher than in the European countries.
In summary, new drugs do not always mean innovation, breakthroughs are not always proven to be true breakthroughs, and a high price does not mean high value. Did any of these statistics surprise you? On an optimistic side, we should remember that this is just theory and does not reflect the real-world use of these drugs. Patients will not switch to a new drug which works just as well as what they are already on, and payers would restrict coverage for high-price drugs with dubious efficacy.
Approvals gone wrong – Avastin in metastatic Breast Cancer
However, as we alluded to at the start, there are unfortunately cases where drugsare proven to be quote unquote useless. Let’s check out two examples, and then dive into some chemistry.
The first one is Avastin, which is a VEGF-targeting antibody which suppresses the growth of new blood vessels. This hinders supply nutrients to the cancer, slows its growth, and therefore slows cancer progression.
The FDA approved Avastin in a broad set of tumors. So, we can see the theme of supplemental indications mentioned in the first part. The problem really was around its accelerated approval in metastatic breast cancer (2008). At the time, the FDA’s Oncology Advisory committee, so an outside advisory panel, had actually recommended 5 to 4 against the approving Avastin in metastatic breast cancer. This mixed opinion was because Avastin at the time did not extend life.
We look at the details of the clinical study comparing Avastin as an add-on to chemotherapy. Overall survival – the time from trial entry to patient death – is the gold standard primary endpoint in cancer trials but takes more time and patients to collect. Avastin missed the mark on this important metric. Avastin’s OS is numerically higher, 26.7 months vs. 25.2 months, but this is not statistically significant as indicated by a p value of higher than 0.05.
However, progression free survival improved significantly. PFS is the time between treatment start and first evidence of disease progression, or also death. PFS data is available earlier than OS, so it serves as a surrogate endpoint. So essentially, Avastin reduced the speed of progression by more than 5 months, but patients were ultimately not living longer. Also, their subjective quality of life did not increase either. You might think: If the drug slows down progression, should we approve it nevertheless?
What was The Problem?
An additional problem was the overall 20% increase in adverse events with Avastin. Toxic effects like hypertension or infections increased significantly. This resulted in 6 deaths due to Avastin. So, considering all this evidence, the accelerated approval was quite optimistic but understandable given the lack of other treatments that showed similar PFS benefit.
As we learned, accelerated approvals require parallel studies which intend to confirm the drug’s benefit. However, the opposite happened.
Looking at the AVADO trial comparing two doses of Avastin, the extended PFS was still there. However, with just under 1 month difference between trial arms, this was significantly lower than expected. More importantly, patients on Avastin had numerically shorter OS. Because of the high p value, we can’t really say if this means Avastin is indeed worse or just equal to chemo. This led the FDA to withdraw the metastatic breast cancer indication for Avastin.
Interestingly, the European regulator EMA did not follow suit. In their view, benefits outweigh risks given positive albeit small PFS benefit and lack of statistically significant detrimental effect on OS.
Approvals gone wrong – rucaparib in 3rd line ovarian cancer
The second case study is Rucaparib, which was indicated, among other uses, for third-line treatment in ovarian cancer. It belongs to the class of PARP inhibitors which interfere with DNA repair mechanisms.
Like the results for avastin, rucaparib extended the median progression-free survival by 5 months. Patients with DNA repair deficient tumors responded even better as they are more prone to PARP inhibition. This seemed like a solid benefit at the time – but can you guess the problem?
Four years after its approval, now in 2022, more mature data on overall survival came in. The harsh reality – across different data cuts, patients on treatment again did not live longer. Again, we need to check the p value to see that even in the BRCA mutated sub-group, the potential strong responders, the seemingly beneficial effect is not statistically significant. This led to a voluntary withdrawal in the third-line indication by the company. Unfortunately, a somewhat desperate filing of rucaparib as a first-line therapy by the company did not work out either.
If you watched my previous video, you would remember that pharmaceutical companies burn a lot of money on development and operations. Clovis tried to cut costs through lay-offs and raise additional money, but ultimately had to throw in the towel and file for bankruptcy.
In summary, PFS benefit does not necessarily translate into true survival benefit – which is ultimately most important. Also, we saw that the FDA and EMA can reach different scientific conclusions looking at the same data.
organic synthesis of rucaparib
Finally, it’s chemistry time. Here we have two things to check out. First, we will look at just one part of Rucaparib’s original large scale process route, and second, go through a more efficient synthesis in full.
The process route, albeit long, has an interesting reaction early on. It starts with a nitration of this benzoic acid – which selectively nitrates meta to the acid. So far, nothing special – most of you should know the selectivity. Doing a nitration with 23L concentrated nitric acid is nothing so sneeze at, but we are more eager to see what happens after this step, and a simple esterification. It’s a Leimgruber-Batcho indole synthesis which leverages the moderate acidity at the benzylic position of these nitroarenes. In the first step, the benzyl anion is formed and attacks the reagent, dimethylformamide dimethylacetal. This adds one carbon to the system and after spontaneous elimination of methanol, a conjugated link is formed.
The second step of the indole synthesis is the reduction of the nitro-group. The liberated nucleophilic aniline can intramolecularly cyclize with the iminium, and again eliminate dimethylamine to create the aromatic indole. Although you might expect that immediate intramolecular attack in a 5-endo-trig fashion would give an anion that is very much stabilized through the ester group, I think that’s kinetically disfavoured work due to Baldwin’s rules.
Second synthesis of rucaparib
The remaining steps are not revolutionary so we will instead look at the very direct synthesis published in 2022. The starting material is already highly functionalized, but you can buy this commercially so it’s fair game. The first step is a Heck-reaction with the highly reactive aryl iodide. Next, the aryl amine was condensed with an aldehyde bearing the other half of rucaparib. At this point, we essentially already have all atoms that we need already, it’s just about linking them and getting to the right oxidation states.
To create the indole, the chemists used a cyanide-catalyzed imino-Strecker reaction – you might remember this one as a prime example of Umpolung chemistry. The mechanism starts with nucleophilic addition of cyanide to the aldimine. The negative charge first sits on nitrogen, but rapidly tautomerizes alpha to the nitrile. Now, we have the typical Stetter 1,4-conjugate addition onto the vinyl nitrile. Finally, just like we’ve seen in the previous Leimgruber-Batcho synthesis, elimination delivers the indole. This regenerates the catalytic cyanide which is why the reaction works with just 20 mol% of sodium cyanide.
Well, what do we do next with the nitrile? You’ve guessed it – we need to reduce it to the amine, and link it to the ester. Because typical hydrogenation did not result in any reaction, the authors looked for other reducing agents. Interestingly, the generation of nickel boride from nickel chloride and sodium borohydride worked smoothly and chemoselectively, leaving the ester group in peace. The lactamization occurred spontaneously in situ, so that was quite convenient as well. The ultimate step was a simple deprotection of the amine sitting at the other side of the molecule, completing the total 5-step synthesis.
That’s it for this time. As always, I will catch you in the next one.
Key References:
Therapeutic value of first versus supplemental indications of drugs in US and Europe (2011-20): retrospective cohort study: BMJ 2023, 382, e074166
Association between FDA and EMA expedited approval programs and therapeutic value of new medicines: retrospective cohort study: BMJ 2020, 371, m3434$
rices and clinical benefit of cancer drugs in the USA and Europe: a cost-benefit analysis: Lancet Oncology 2020, 21, 664
The US FDAs withdrawal of the breast cancer indication for Avastin (bevacizumab): Saudi Pharm J 2012, 20, 381
Efficacy and safety of bevacizumab in combination with docetaxel for the first-line treatment of elderly patients with locally recurrent or metastatic breast cancer: Results from AVADO: Eur J Cancer 2011, 47, 2387
Paclitaxel plus Bevacizumab versus Paclitaxel Alone for Metastatic Breast Cancer: N Engl J Med 2007, 357, 2666
Multikilogram Scale-Up of a Reductive Alkylation Route to a Novel PARP Inhibitor: OPRD 2012, 16, 1897
Watch the video on YouTube or read the blog below!
What is the most expensive drug in the world? Imagine a single dose of medicine priced at $3.5 MILLION. Since the approval of the gene therapy Hemgenix in late 2022, this is a reality. This eye-popping price is easy to scrutinize, especially if you know the truth behind most newly approved drugs. Drug manufacturers think these prices are still fair, with some even winding down operations due to pricing disagreements. Also, it couldn’t be more fitting that one of the companies we will talk about, who is selling a very simple drug for one million a year, very recently saw itself forced to lay off 25% of its staff, maintain loans and preserve cash burn to simply keep up its operations for another year.
After reading this post, you will know:
Why soaring prices do not always equal large profits. You will not understand why some therapies, as crazy as it sounds, cannot be sold for prices under hundreds of thousands of dollars
The answer to the question “what is the most expensive drug”?, as well as top contenders
The chemical synthesis of the most expensive small molecule drug, used for an ultra-rare disease occurring in 1 out of 40 million people
How to estimate drug sales for gene therapies
most expensive drug in the world – what is the bar?
Hemgenix approval dethroned the gene therapies Skysona and Zynteglo, both developed as rare disease therapies by the company BlueBird Bio. With costs per dose of 3 and 2.8 million respectively, they currently land on number 2 and 3. The Institute for Clinical and Economic Review, a non-profit organization publishing clinical and cost-effectiveness analyses of treatments, deemed the true value of Skysona to not be too far off. The company itself postulated normal standard of care consisting of regular transfusion treatments cumulate to total healthcare costs of $6million over a patients life time. Comparing this with Soliris, which had been the world’s most expensive drug for some years in the past decade, we can see that in certain diseases such as myasthenia gravis, a cost-effective price would require a more than 95% discount on the annual price.
So, million-dollar gene therapies are actually not bad in terms of value provided. We have also seen some companies offer outcome-based deals, where for example 80% of the price is refunded if the patient does not respond to therapy.
Gene therapies are also less-overpriced if we look at the cost that it takes to produce them. Producing the annual supply of the antibody Soliris, costs less than 1% of its price. In contrast, gene therapies require bespoke manufacturing and supply chains, with fully loaded manufacturing costs reaching over 200 thousand dollars. But maybe you’re wondering why companies wouldn’t sell them at a lower price let’s say, half a million, and still make a decent profit? First of all, these variable costs do not include the large capital investments required for manufacturing facilities.
Drug R&D Costs
But as anyone who had more than 30min of business school will tell you, there are many more direct and indirect costs behind drugs.
Currently, it costs roughly 2.3 billion dollars to research, develop and launch a new drug. As you can see, R&D costs have been trending up, roughly doubling during the last decade. Some drugs are even higher than 3-5 billion. These large upfront investments are why a company will typically command as high of a price as it can reasonably get, to try to maximize internal return on the dollars.
But if we look at the peak sales potential of drugs, we can see that this decreased instead of increased in recent years. Basically, development is getting more expensive while returns are getting smaller.
Don’t forget that there are other costs beyond the manufacturing cost we highlighted. Companies deploy various functional teams to drive uptake of drugs – the easiest to think about are sales and marketing. On a company level, selling, general and administrative cost are usually around 15-30% of net revenues. On a product level, this obviously depends on the lifecycle stage, with new drugs having much more investments behind them than established products which are more like cash cows.
Additional Factors: Discounts and risk
Another important thing to note are gross-to-net price discounts. You see, the list price you read in the news refers to a theoretical price which is actually never paid. In the extremely convoluted US healthcare system, manufacturers give discounts to various stakeholders involved in access, distribution, or other things. These typically untransparent gross-to-net discounts translate into actual net sales for companies which can be less than 50% of the original list price.
One of the underlying problems behind high development costs is the low probability of clinical success. Science and drug development is inherently high risk, high-reward. While big companies with more than a hundred projects will always get something out of their R&D funnel, smaller companies can get wiped out if several drugs turn out to simply not work as hoped. This is why no company will ever price drugs at minimal margins, because prices need to be high to compensate for large R&D costs AND offer protection against critical pipeline setbacks.
Lonafarnib: another expensive drug
This molecule was originally discovered by Schering-Plough more than 2 decades ago as an investigational cancer drug. After development for oncology was discontinued due to lack of efficacy, and Schering-Plough was acquired by Merck, a deal was struck with the small biotech Eiger Pharmaceuticals. Eiger originally wanted to develop the molecule for hepatitis D, but was also introduced to the Progeria Research Foundation by Merck. This non-profit research organization found that the mechanism underlying lonafarnib’s activity was also involved in progeria.
This heart-breaking disorder causes rapid-aging in children. With a prevalence rate of 1 case in 20 million individuals, this is an ultra-rare disease. Patients die at an average age of 13 years, perversely from heart attack or stroke – so the unmet need is massive.
This always fatal disease is caused by mutation in the gene coding for a protein called lamin A. A single cytosine-to-thymine mis-spelling is the most common mutation, and has critical effects.
Normal lamin A is modified with a farnesyl group which helps direct it to the nuclear lamina (a shipping tag). The tag is subsequently cut-off from normal lamin A. However, the progerin cannot be defarnesylated – meaning the un-natural shipping tag stays on the protein – and this interferes with a myriad of different cellular mechanisms. Because lonafarnib is a farnesyl transferase inhibitor, it prevents the addition of farnesyl groups to progerin to start with. Due to the missing shipping tag, the proteins do not reach their destination as easily but also do not accumulate and mess up the nuclear lamina either.
Although far from a cure, the simple farnesylation-block has significant clinical effects. Treated patients have significantly lower mortality risk and over the course of 11 years, live roughly 2.5 years longer. It also had a good safety profile with small number of discontinuations.
How Expensive is it?
As this is a significant improvement in ridiculously small patient population, we can start to understand why this medicine costs around one million a year, despite being a simple small molecule. This makes it the most expensive small molecule drug. There are not that many sales to be made – as many of the 400 children living with progeria worldwide were already dosed in your clinical trials. Of course, Eiger could theoretically sell it for cheaper, but you also must consider cost of the several clinical trials which lonafarnib was tested in, including its oncology history. So even though the price looks Machiavellian, the company actually lost 200 million dollars in the last three years.
The current R&D spend is focused on the development of lonafarnib in hepatitis D, which is a much bigger commercial opportunity but will also come with lower pricing. They also have additional pipeline projects, so all this cash drain resulted in them recently communicating a 25% workforce reduction. In summary, we see that having the most expensive small molecule drug doesn’t necessarily make life easy.
Chemical Synthesis
Lonafarnib was synthesized on a up to mind-blowing 100kg scale when it was planned as an oncology drug. If you compare the key intermediate to the target, you will notice that we need to introduce an aryl bromide group. How is this done selectively? If you ever had some electrophilic aromatic substitution theory, you would know that other positions are more electronically activated due to the present chloride substituent.
The team approached this starting with a single nitration, giving rise to two nitro regioisomers. Regardless of their position, the effect is the same. A novel reduction system with catalytic iodide proved to reduce the nitro group as well as ketone, saving a step. A mix of hypo-phosphorous and phosphorous acid avoided halogen side reactions while reducing both groups. However, they also observed severe foaming at the beginning of the reaction. For more control, they first added H3PO3 to reduce the nitro group, and then H3PO2 to reduce the ketone. This system proved better than hydrogenation or metal catalysis which would have decomposed the aryl halide groups and introduced potential contaminants.
With the amine in place, the desired bromination product is favored through either ortho-direction in one of the isomers, or para-direction in the other.After removal of the amino groups, the most acidic site is between the aryl rings. The addition of LDA, quinine and this electrophile led to a chiral alkylation, proceeding with impressive 95% ee. Gratifyingly, they were able to recycle the quinine – quite important if you run things on a double-digit kilo scale. Next, a diastereomer salt formation enriched ee to 99%. Last, a final amide formation introduced the missing part – and a one-pot Boc deprotection and urea formation gave the product.
Hemgenix in hemophilia B: The world’s most expensive drug
Now that the chemists are happy, we can pivot to the most expensive drug. For this, we need to understand a disease called hemophilia B, an inherited bleeding disorder caused by a deficiency of the coagulation factor IX. This disorder is characterized by frequent and recurrent bleeding into joints or soft tissue, leading to chronic pain, disability, and impaired quality of life.
The current standard-of-care for haemophilia B includes life-long on-demand and prophylactic replacement therapy with FIX concentrates. This replacement therapy is effective at reducing bleeding episodes and is well tolerated – although some patients develop antibodies, making them resistant to replacement therapy. These repeated injections become extremely expensive over a patients lifetime, with most estimates well beyond $10 million. ICER says 3 million dollars would be a very reasonable price – so Hemgenix is not too far off. So clearly this one-time treatment has a lot of impact, but how does this work biochemically?
How do Gene therapies work?
Gene therapies use engineered viruses as carriers or so-called vectors of genetic information to correct a patient’s genetic code, ultimately restoring the proper functions of vital proteins. For Hemgenix, this vector is adeno-associated virus or AAV-based. It’s like a train or truck delivering cargo. As viruses are doing virus things, they can enter cells and integrate or transduct the missing gene into the cell.
Compared to normal viruses, these AAV are modified to lose their replicative abilities, rendering them safe as therapies for humans. In addition, a DNA regulatory element called promoter limits the transgene expression in only the desired tissue. In this case, it’s the liver, as factor 9 is produced in hepatocytes and released into circulation from there. The actual factor 9 gene being incorporated is basically a factor 9 protein on steroids, which is 7-times more activity than wildtype due to a single mutation. This allows for a lower AAV vector dose which in turn decreased averse immune reactions by patients.
As you might expect, Hemgenix was superior to replacement therapy in a mid-sized trial with 54 patients. This led to a significantly lower rate of bleeding events as the primary endpoint. 96% of patients were able to stop replacement therapy. This transformational benefit shows in the increased factor IX activity of patients. Severe hemophilia is characterized with untreated factor IX activity below 5%. This also approved, with immediate and sustained increase post-treatment. Given the adult population, the approval is only in patients over 18 years. In summary, instead of continuously replacing factor 9 exogenously multiple times a week, incorporation of the missing gene solves the problem for good – or at least for more than 20 years.
How can we estimate drug sales?
Hemgenix is the world’s most expensive drug – so will it lead to the highest profits?
You might know that total yearly sales equal sales per year multiplied with the cost per patient. We already know the latter – but what is the true number of patients that will get treatment?
Well, there’s only approximately 6000 people with hemophilia B in the US. Let’s assume this population grows by 1% each year. Do you remember if we can treat all of them? Nope! The approval is only for adult patients, which might be around 80% of patients, and only in cases with severe hemophilia B, maybe around 60% of the patients. So 6K times 80% times 60% – some of you might do it in their head –is 2’880. But why is the number I have here different? Well, I mentioned some patients develop antibodies or inhibitors to factor IX replacement therapy. Because Hemgenix does not have data in these patients, we will exclude them too. This reduces the eligible pool by 2%.
Gene THerapies Are Unique
Now just multiply this number with price? Think again! Will all patients use Hemgenix, immediately? Clearly not – most of them will never receive it due to limited coverage due to the high cost, and the uptake will only gradually increase with time.
It’s probably fair to say that not much more than 10% of patients will ever get Hemgenix, and we also need to assume a build over time – let’s say a one or two percentage points every year. Hemgenix is a one-time treatment. It practically cures patients or it “does not work”. Therefore, more patients being treated reduces the addressable pool.
But hey, this therapy will also not be the only one. There is another hemophilia B gene therapy (Pfizer/Spark). Even if we assume that the first-to-market therapy will dominate, the second product will also have some penetration. Obviously, this all depends on its relative efficacy, safety and importantly price. Here, I simply assumed that the new product would have 50% of Hemgenix’ uptake, delayed by one year.
This means the blue line gives us the actual pool we can treat with each product in every year, factoring in the total number of patients previously treated with Hemgenix and the Pfizer/Spark product. I hope you could follow.
Now we can take the number of new patients treated with Hemgenix, calculated by multiplying the phased penetration with the untreated addressable pool, and multiply it with the price per patient. There are also some less important pricing assumptions.
How Profitable is The World’s most expensive drug?
Do you notice anything about the result? The number of Hemgenix-treated patients actually peaks after just a few years, maybe even staying below 200 patients per year. This translates into peak sales of $630M.
Definitely nothing to sneeze at, but we also have to be real here – many mega-blockbusters with lower pricing are much more lucrative for companies due to steady revenues and higher number of patients. Throw the quarter million-per unit production cost on top, program development cost, and all your other functional and SG&A cost … Just because something is the world’s most expensive drug, doesn’t mean it’s highly profitable or overly “capitalistic”.
And let’s once again re-iterate the topic of low probability of success. With bad luck, developers can easily risk all profits of one drug while financing others. Even the simplest medicines can simply fail to demonstrate efficacy – but the development of innovative therapies like gene or cell therapies is another beast. Very recently, the FDA put a clinical hold on Arcellx’s Phase 2 investigating a CAR-T cell therapy following the death of a patient. Just half a year prior, Gilead invested over $300M to secure development and commercialization rights. This shows how fast the profits gained on some products can disappear.
Thank you for sticking through to the end! As always, I hope you’ve learned something!
Key references
Lonafarnib: First Approval | Drugs. 2021; 81(2): 283
A Novel Iodide-Catalyzed Reduction of Nitroarenes and Aryl Ketones with H3PO2 or H3PO3: Its Application to the Synthesis of a Potential Anticancer Agent | Org. Lett. 2011; 13 (19): 5220
Impact of farnesylation inhibitors on survival in Hutchinson-Gilford progeria syndrome | Circulation 2014;130(1): 27
Gene Therapy with Etranacogene Dezaparvovec for Hemophilia B | N Engl J Med 2023; 388(8): 706
Etranacogene dezaparvovec for hemophilia B gene therapy | Therapeutic Advances in Rare Disease. 2021;2. doi: 10.1177/26330040211058896
Structure-activity relationship study to improve cytotoxicity and selectivity of lonafarnib against breast cancer cells | Arch Pharm (Weinheim) 2023; 356(4): e2200263
Deloitte | Measuring return from pharmaceutical innovation 2022 – IQVIA | Global Trends in R&D 2021
