Picture it…


Leave a comment

Common Laburnum, Laburnum anagyroides

What does the word “chemical” conjure in your mind? For many people chemical means a dangerous and even forbidden substance. Likewise, laburnum is a plant that has a bad image, reported as poisonous and capable of killing children. Yet our world is full of chemicals, many of which we enjoy (caffeine, sucrose and alcohol, for instance) and even the more reactive chemicals can be incredibly useful, if used properly. Here we will explore how the “deadly” chemicals in laburnum might be used to help people stop smoking.

Laburnum headline image

Laburnum leaves and seed pods, open to reveal the seeds within, resting on sample trays, with sample vials.

1) In the above picture we see sample vials, some with their lids on and some with lids off. Notice the lids have white membranes in the top; this allows the sample to be taken up with a syringe and needle, without having to open the sample vial. The leaves and seed pods of laburnum rest on blue and yellow sample vial trays. Trays like this are very helpful for lining up numerous samples, when monitoring a reaction or testing a process, for example. The molecular structure of cytisine, the chemical responsible for laburnums toxicity, is nestled amongst the laburnum leaves.

Modern society is now more aware of the effects of pollution on our health and the environment than ever before. We like to buy organic foods and we listen to scientists and politicians argue about green and renewable energies. This has perhaps led to the belief that the chemistry laboratory is a dangerous place and that chemicals are bad. Whilst both laboratories and chemicals can be dangerous, if not treated with respect, the laboratory is often the place where new, “green” and “organic” products are developed.

In a similar way, laburnum has suffered a reputation as a poisonous plant. Common laburnum is perhaps not quite so common anymore. This beautiful tree, used by property developers to conjure the image of a nice place to live (fancy a house on “Laburnum Avenue”?), has been brought into such disrepute that many gardeners will not allow it to grow. The Poison Garden highlights an article that may be the cause of this bad image (thepoisongarden.co.uk, Pontifications on Poison, Saturday 17th March 2012). In 1974, “Look and Learn” published an article entitled “Beautiful but deadly – poisonous plants in gardens and allotments” which claimed that the laburnum was responsible for more plant poisonings in the UK than any other plant and that nobody should ever touch the seeds or the bark of the trees (Look and Learn, Issue 663, 28th September 1974, published digitally on Wednesday 14th March 2012).

All parts of the laburnum are known to be poisonous, but it is the seed pods which are most often reported as causing problems, as children mistake them for pea pods. Whilst there are cases of people, both children and adults, suffering nausea, vomiting, sleepiness and convulsions after ingesting laburnum, there are very few cases of fatalities related to the tree. Interestingly, a report by the British Medical Journal called “Accidental poisoning deaths in British children, 1958 – 77” deals with 598 deaths, only 3 of which were attributed to plants, and none of which are associated with laburnum (N. C. Fraser, Br. Med. J., 1980, 280 (6231), 1595 – 1598). Contrast this with a report, printed in The Lancet in 1979, after the “Look and Learn” article, entitled “Have you Eaten Laburnum?” (R. M. Forrester, The Lancet, 1979, 313 (8125), 1073). This later item suggested that there were 3,000 hospital admissions for laburnum poisoning each year, none of which resulted in fatalities and the author suggested that many hospitalisations were unnecessary. These contrasting articles, highlighted by the Poison Garden blog, show just how dramatically our attitudes to “chemicals” can change, even in such a short period of time.

2) It is the chemical (–)-cytisine, present in all parts of the laburnum tree, which is responsible for causing sickness. However, these “side-effects” have also been exploited in beneficial ways. Thousands of years ago, American Indians consumed the seeds of laburnum for their purgative effects, and during the Second World War, laburnum leaves were used as a tobacco substitute. More recently, (–)-cytisine has been marketed as a smoking cessation aid, Tabex, but is not available in the UK.

Structures of cytisine and nicotine

Structures of cytisine and nicotine

Chemists use molecular structures as maps, to help them understand how a chemical might behave, in terms of its chemical and physical properties. If we look at the structures of S-nicotine, responsible for the addictive nature of tobacco, and (–)-cytisine, it is perhaps difficult to see why laburnum leaves might act as a tobacco substitute. At first glance, there doesn’t appear to be much similarity. Both S-nicotine and (–)-cytisine have ring structures and both have some atoms which organic chemists would call “heteroatoms” in the rings, this means atoms which are not carbon or hydrogen. In the case of S-nicotine, two nitrogen atoms (denoted as N in the structures) are the heteroatoms, whilst in (–)-cytisine, there are three heteroatoms, two nitrogens and one oxygen (denoted, respectively, as N and O in the structures).

Heteroatoms are important when looking at molecular structures because they behave in a different way to carbon. For example, nitrogen and oxygen are described as more electronegative and have more electrons and more associated negativity, than carbon atoms. This means that carbon-nitrogen, or carbon-oxygen bonds are polarised; there is more electron density, and more negative charge, at the heteroatom end of the bond than the other. So we can see that both S-nicotine and (–)-cytisine have regions of polarity, centred about the heteroatoms. However, this alone doesn’t really explain why S-nicotine and (–)-cytisine cause similar interactions within our bodies.

Fortunately chemists have another trick up their sleeve and this is in the form of computational chemistry. Computational chemistry allows us to model a molecule in three dimensions and to take into account the energy that a molecule might have. For example, if we put regions of negative charge closer together, this would probably be unfavourable and would cause an increase in the energy of the molecule.

Computational models of (-)-cytisine and S-nicotine

Computational models of (-)-cytisine and S-nicotine

If we look at the computational models of (–)-cytisine (on the left in the above picture) and S-nicotine (on the right), we can see that the location of the heteroatoms, in a three-dimensional molecule can be quite similar. Note that these structures are not the lowest possible energy shapes of the molecules. However, they do show that, when acting upon sites in the body, both S-nicotine and (–)-cytisine can adopt a shape such that the regions of negative charge, the heteroatoms, denoted here in blue, for nitrogen and red for oxygen, are in similar places. See, for example, where the red oxygen in (–)-cytisine mimics the position of the blue nitrogen in S-nicotine. It is this ability of (–)-cytisine to adopt a similar shape to S-nicotine which allows it to act on the same receptors as nicotine. The ability of cytisine to interact with nicotine receptors in the body with higher activity, yet lower toxicity, has led to its use as a replacement for nicotine (take a look at P. Ruckatooa, C. Haseler, R. van Elk, A. B. Smit, T. Gallagher, T. K. Sixma, J. Biomol. Chem., 2012, 287, 23283 – 23293, for structure reactivity studies).

3) Scientists and medical practitioners have been fascinated by the properties of (–)-cytisine for over 100 years. Initial attempts at extracting the chemical efficiently proved frustrating, as a report from 1829 shows, it was reported as “attracting moisture from the atmosphere… insoluble in ether, soluble in water… it is not precipitated…” (A. T. Thomson, The London Medical and Physical Journal, 1829, 366 (62), 93 – 95, The Laburnum, Remarks on the Poisonous Properties of the Laburnum).


[3H]-Cytisine, the most commonly used labelled version of cytisine, which has enabled understanding of how nicotine receptors work.

Extraction from laburnum is now more efficient, producing up to 1.5 g of cytisine per 100 g of seeds (A. J. Dixon, M. J. McGrath, P. O’Brien, Org. Synth., 2006, 83, 141), and the seeds are relatively cheap, around 60£/kg (Vilmorin.com). Producing pure cytisine in an affordable and efficient manner is vital for the continuation of research in this area. Labelled derivatives of cytisine, specifically 3H-cytisine, have been used as radioligands and have contributed significantly to the understanding of nicotinic neurotransmission (J. Rouden, M.-C. Lasne, J. Blanchet, J. Baudoux, Chem. Rev., 2014, 114 (1), 712 – 778).

(–)-Cytisine has appeared as a lead candidate to enable the development of derivatives with specific central nervous system interactions which also have minimal side-effects. Thus developing methods to synthesise the tricyclic, fused-ring system has been the focus of much research (C. C. Boido, B. Tasso, V. Boido, F. Sparatore, Il Farmaco, 2003, 58 (3), 265 – 277). Multiple routes exist to the racemic (±)-cytisine and effective methods for resolving the racemic compound have been documented. For example, van Tamelen, used d-camphor-10-sulfonic acid, to give relatively good yields of enantiopure (–)-cytisine (E. E. van Tamelen, J. S Baran, J. Am. Chem. Soc., 1958, 80, 4659 – 4670).

However only one route has thus far been reported which allows direct synthesis of (–)-cytisine in 12 steps, with a 9% overall yield (B. Danieli, G. Lesma, D. Passarella, A. Sacchetti, A. Silvani, A. Virdis, Org. Lett., 2004, 6 (4), 493 – 496). The key step is the use of a ring-closing metathesis, to construct the N-heterocycle, utilising a CBz protecting group, to avoid N-chelation of the catalyst inhibiting reaction. The tricyclic compound, resulting from this ring-closing metathesis, would allow derivatives, based on the cytisine tricyclic motif to be developed. The synthesis relies on the readily-available compound, cis-piperidine-3,5-dimethanol, as the source of the chiral centres. Importantly, since both enantiomeric forms of cis-piperidine-3,5-dimethanol are available, this synthesis would allow both (+)- and (–)-cytisine to be formed.

Synthesis of (-)-cytisine

The synthesis of (-)-cytisine, in 12 steps with a 9% overall yield.

Closer to the Picture It home, Gallagher’s group, here at the University of Bristol have developed methods for the synthesis of (+)-cytisine, in 10-steps with 3.7 % overall yield (D. Gray, T. Gallagher, Angew. Chem. Int. Ed., 2006, 45, 2419 – 2423). This synthetic method relies on enzymatic resolution as key to selectively producing the desired enantiomer. Whilst it may not produce the desired cytisine, the flexibility engendered by incorporating key nitrogen-carbon bond and carbon-carbon bond forming steps between rings allowed the synthetic method to be expanded to the synthesis of other lupin alkaloids, such as (±)-anagyrine and (±)-thermopsine.

Synthesis of (+)-cytisine

The synthesis of (+)-cytisine, in 10 steps, with 3.7% overall yield.

Contributors: Jenny Slaughter (research, words, photos, images and schemes), Richard Stevenson (research), Aurelien Honraedt (ideas and editing) Natalie Fey (computational structures).

Our thanks also to Kevin Stuckey for supplying the laburnum cuttings.