Science1 publisher2 min readPublished
Vienna chemists make strained amine rings without rigorously excluding air or water
The strained three-carbon ring has long been a plausible stand-in for a gem-dimethylamine. The obstacle was installing it on a molecule that already has chemistry on it, and a Vienna group now reports a mild way in.
The Scientist · Science desk
What happened
- Nuno Maulide's group at the University of Vienna reported a hydroaminoalkylation that builds aminocyclopropanes under mild conditions from an easily accessible, bench-stable hemiaminal.
- For many substrates, dissolving that hemiaminal in the right solvent starts the reaction at room temperature, with no need to rigorously exclude air or moisture.
- Products included aminocyclopropanes carrying unprotected alcohols, carbonyl groups and phthalimides, functional groups that give established routes significant trouble.
- A one-pot, three-component version brings the components together in a single process from cheap commercial starting materials rather than making the hemiaminal first.
- In the proof of concept, swapping phentermine's gem-dimethylamine for a cyclopropylamine kept activity at two transporters while suppressing dopamine release through the dopamine transporter.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability Installing the ring late, on a scaffold that already carries reactive groups, becomes a one-flask option rather than a protecting-group campaign, which is the case medicinal chemists actually face.
- constraint The property that usually motivates this swap, metabolic stability, is not the property the reported experiment measured, so a team hoping to fix a clearance problem still owes its own assays.
- decision With gram quantities available in a single run, the question in front of a project shifts from whether the route can be developed to which lead is worth spending the swap on.
- precedent A dependence-risk read built on a correlated pharmacological signature will invite behavioural studies on cyclopropyl analogues of other amine stimulants, and the chemistry cannot answer those.
Established routes to aminocyclopropanes lean on highly reactive chemicals, metal reagents, multi-step sequences, or starting materials that are a project in themselves, and they generally want glassware kept free of air and water [5]. Conditions like that choose the substrate for you. They limit which other groups can be present while the ring is being built, which is the wrong constraint for medicinal chemistry, where the molecule you want to modify is already complicated [6].
The work, published in Angewandte Chemie International Edition [12], moves the difficulty out of the conditions and into the reagent. Three classes of group that the account identifies as trouble for established methods came through intact [19], and the reaction also ran on a structurally complex molecule derived from estrone [10], which is the closer test of a late-stage method. The chemistry works with alkynes as well [13].
One run went past a gram and still returned 84% [11], which rules out the milligram-only failure mode that kills a lot of promising methodology. How yields sit across the rest of the substrates is not something this account of the paper gives. The three-component variant is worth a step count rather than an adjective: it removes one isolated intermediate, the hemiaminal, from the sequence [20].
The test compound was a close relative of phentermine, an amphetamine-like stimulant known to act on neurotransmitter transporters [15]. Suppressed dopamine release through the dopamine transporter is described in the account as a property that has been linked to reduced risk of dependence [17], and a link of that kind is a correlation drawn in the literature rather than a behavioural measurement on this molecule. No behavioural experiment is reported.
Metabolic stability is one of the things bioisosteric replacement is supposed to move [4], and it is the usual reason a project reaches for this particular swap, which is the property that actually changed hands here. What the experiment measured was transporter pharmacology. Nuno Maulide's framing is that replacing a structural element with an aminocyclopropane can substantially alter biological behaviour without eliminating the desired activity [18], and the phentermine analogue supports exactly that, in one molecule, at two transporters. Whether it generalises is now a cheap question to ask, which is the real contribution.
What to watch
- The full paper's scope table: whether yields across substrates resemble the 84% seen in the gram-scale run.
- Any microsome or clearance data on the cyclopropylamine phentermine analogue, since metabolic stability is the swap's usual selling point.
- Self-administration or other behavioural work testing whether suppressed dopamine release in this analogue tracks lower dependence liability.