Saturday, 9 November 2013

Is this thing on?

If you're a regular reader (hello, both of you!) you may have noticed a slight gap in the update schedule since, er, July.

One of the great pleasures of having only a small following is being able to sod off for a few months without upsetting anyone. I imagine if ChemBark or Derek Lowe did so, their inboxes would very quickly fill with confused messages checking if they were still breathing...

Since I updated earlier tonight with a short rant I thought I'd take the chance explain my absence and to engage in some shameless self-promotion.

A big part of this is that I've simply been online less, and therefore had much less involvement in the chemistry community - and less to say. Maybe that's a blessing. What writing I have done has been published elsewhere: I wrote a blogroll and book review for Nature Chemistry, and some news articles for The Conversation. All of which is pretty exciting: it's great to get feedback on my writing from professional editors and I feel that I've learned a lot.

Final comment: there won't be a #chemclub review for November, as unfortunately the contributor this month had to cancel due to more important commitments. Regular updates resume next month.

Unlikely results?

The Economist's "daily chart" from October 21st came with a striking headline: "Why most published scientific research is probably false".

The accompanying video explains that under certain assumptions, we are drawn inexorably to the conclusion that most scientific results are false. I won't outline their logic here: the video is only a minute and a half long, so go ahead and watch it. I'll wait.

Did it annoy you as much as it annoyed me? The claim - "most published scientific research is probably false" - rests on three assumptions:
  • Most science uses a statistical significance cut-off of p=0.05, and effectively no 'insignificant' results are published.
  • 10% of hypotheses tested will turn out to be correct.
  • The false negative rate is high - possibly as high as 40%.
This is nonsense.

Maybe I'm a little late to the game in criticising this, as it's a few weeks old now, but it's going around my Facebook feed this weekend without any criticism so I thought I'd comment. The tl;dr of this post is "your headline is bad and you should feel bad".

Firstly: entire disciplines that are firmly "science" rarely if ever use statistical significance as a criterion for publication, or have far more demanding requirements. Most organic chemistry falls into the former group, and particle physics into the latter. Are these disciplines negligible to science as a whole?

Secondly, the ratios of true to false positives and negatives in this thought experiment is heavily dependent upon the starting assumption that 10% of hypotheses tested are true. I don't particularly object to the number they chose - but it is arbitrary and unjustified. If we're a little more generous to scientists then the problem diminishes (and vice-versa). If you think scientists are pretty good at generating likely hypotheses then when you crunch the numbers in this fashion, you'll get a result that can inspire confidence. If you think we're generally on fishing expeditions, testing a dozen hypotheses for every validation, then you'll end up with a pessimistic view of the literature.

Finally, at the end of the video the author ramps up the estimate of the false negative rate to utterly overwhelm the number of true positives and draw the conclusion in the headline. What's the basis for the estimation of false negative rates here?

The whole thing smacks of an idealistic view of science as a discipline with unified practices, which operates according to a classic philosopher's view of the scientific method with a healthy dose of publication bias: all scientists operate by testing well-defined hypotheses one by one and analysing the results through statistics, publishing only those with p<0.05.

This kind of thought experiment can usefully explain the crisis of reproducibility in certain disciplines which do rely heavily upon p values for publication, such as some biomedical sciences. But the unqualified extrapolation of this to Science™ is absurd. The Economist can do better.

(This Saturday night rant brought to you by an overabundance of caffeine.)

Sunday, 13 October 2013

#chemclub Reviews: Cucurbiturils

This month's review is by Chad Jones, who is finishing up his PhD studying ion-neutral gas phase interactions. He blogs and podcasts at The Collapsed Wavefunction.

The most amazing type of chemistry is supramolecular chemistry. Now, you may or may not agree with that statement, but it is my goal in this review to at least convince you that supramolecular chemistry is among the most amazing. 

The Dearden group at Brigham Young University (my group) studies cucurbiturils, macrocycles so named for their distinct resemblance to a pumpkin (who are members of the plant family cucurbitaceae).

See? Pumpkins!
This review is by no means a complete look at the chemistry of cucurbiturils. There's no way I could do that - I'd like to graduate and I don't see this review making a big dent in that project. If you'd like more info I'll point you to any number of review articles4, patents5-7special journal issues, and special interest conferences

History and Overview

Robert Behrend initially synthesized cucurbiturils in 1905 by the condensation of glycoluril with formaldehyde.1 Behrend likely synthesised a range of cucurbituril derivatives including CB[5], CB[6], and CB[7]. Although he found that the newly synthesized substance had a high affinity for alkali metals and organic dyes,2 the structure of this interesting compound was not discovered for another 75 years. In 1981 Mock et al. used x-ray crystallography to show that cucurbit[n]urils, CB[n]s, are macrocyclic molecules with n repeating glycoluril monomers.3 Mock’s initial studies of cucurbiturils focused on CB[6]. Since then a number of homologues (CB[5]-CB[8], CB[10], and CB[5]@CB[10]) have been synthesized.4 A wide variety of derivatives have also been synthesized, each with a specific solubility, binding stoichiometry, and guest affinity. 

Often, very small changes in the structure leads to wildly different binding preference. As an example of this, look at cucurbit[5]uril (CB[5]) compared to decamethylcucurbit[5]uril (MC[5]). The only difference is a permethylated equator. 

Now, a passive look at these molecules says that their binding preferences should be about the same. After all most of the binding takes place in the cavity, a location that is far removed from the equatorial methyl groups. In both cases the portal diameter is about 2.4 Å. However, in the Dearden group our experience that these two cages have very different binding preferences. We have found that MC[5] is a much more rigid cage, which in turn affects the interactions inside the cavity. To me this subtle, unexpected change, is a perfect example of why cucurbiturils and their derivatives are such amazing molecules.

Cucurbituril formation and derivatives

For those interested in synthesis (which isn't really my thing) I'll quickly go over a few interesting derivatives of cucurbiturils. The original synthesis, which was later described by Mock et al., involved a two-step reaction. The condensation of glycoluril and excess formaldehyde in HCl was followed by a H2SO4 rinse. A high-yield (~82%), one-step synthesis was achieved by Buschmann by condensation in H2SO4.8 The mechanism of formation of cucurbiturils, however, was not understood until more recently.

The Isaacs group began studying the mechanism of cucurbituril formation by synthesizing a methylene bridged glycoluril dimer. To reduce reactivity, o-xylene units were attached.9,10 The result was two isomers with a pH dependent equilibrium.11 These isomers act as a type of “molecular clip” as the C-shape is able to partially encapsulate a guest, while the S-shaped isomer cannot.12 Understanding the energetics of these isomers was the first step to understanding cucurbituril formation, and led to the discovery of several cucurbituril derivatives.

Some of these derivatives include: i-CB[6] and i-CB[7], which include a single inverted glycoluril ring13; Bis-nor-seco-CB[10], a figure-eight style cucurbituril with single CH2 bridges instead of the paired bridge as seen in cucurbituril14; Nor-seco-CB[6],  a cucurbit[6]uril compound; Nor-seco-CB[6]  with a single attached phenyl group15; hemicucurbit[n]uril, compounds which can be visualized as cucurbiturils cut at the equator16; Bambus[6]uril, a combination of hemicucurbiturils and cucurbiturils17; decamethylcucurbit[5]uril, where methyl groups replace hydrogen atoms on the equator18; and CB*[5], where cyclohexane rings replace the hydrogen atoms on the equator18

Binding Properties

Cucurbiturils are highly symmetric, having two carbonyl portals with a high binding affinity for cationic guests. This binding is due to a combination of ion-dipole interaction involving the carbonyl groups and hydrophobic forces of the inner cavity. The latter is less well-understood, in part because of the lack of studies on neutral compound binding.

Image credit: see ref 24
The electrostatic potential map of CB[8] makes the cation receptor properties of cucurbiturils easy to understand. The electron dense carbonyl groups become efficient binding sites for cations. Also, the inner cavity is electron-deficient, which hints at the cavity acting as a hydrophobic region. Kaifer et al. demonstrated the importance of the inner cavity on binding preferences.19 Methylviologen binds as an inclusion complex, with a binding constant ~106 M-1 in an aqueous, salt-containing solution. This binding is due, in part, to the position of the charged nitrogens, which line up well with the carbonyl portals of cucurbiturils. Interestingly, loss of one of the two charges does not significantly affect the binding affinity. This hints that ion-dipole interactions are not the sole (and perhaps not even the primary) contributors to the binding motif. These results enforce the idea that hydrophobic interactions within the inner cavity play an important role in cucurbituril chemistry.

Understanding the unique binding of cucurbiturils requires an understanding of the microenvironment of the inner cavity.  Solvatochromic probes are useful tools to study such microenvironments. These fluorescent molecules (xanthene, coumarine, oxazine, and cyanine to name a few) have characteristic shifts in their fluorescence that are dependent on the polarity of the solvent environment.20 Nau21 and Wagner,22 using solvatochromic probes, have shown that the inner cavity of cucurbiturils is a very non-polar microenvironment. Further studies showed an environment most similar to n-octanol.23

Beyond the polarity of the inner cavity, the polarizability has also been studied using solvatochromic probes, such as 2,3-diazabicyclo[2.2.2]oct-2-ene. The inverse oscillator strength of the near-UV absorption band of DBO, a solvatochromatic dye, has a linear correlation to the polarizability of the solvent environment. Results showed that the polarizability of the inner cavity is extremely low (0.12), below that of even perfluorohexane.  As a comparison, most other macrocycles have a cavity whose polarizability is similar to that of alkanes or water. The interior of CB[7] to date has the lowest measured polarizability.

Image credit: see ref 24.

An interesting consequence of such low polarizability is a decrease in the radiative decay rate, kr, of fluorescent states.21 This decay rate is determined by the Stickler-Berg equation:

Where Ff is the fluorescence quantum yield, tf is the fluorescence lifetime, and n is the refractive index of the environment. Some fluorescent dyes, when bound to cucurbiturils, display their longest recorded fluorescent lifetimes. This effect is seen even when the fluorescent dye is too large to fit completely inside the cucurbituril, as is the case with DBO.24 

What is even more surprising, given that complexation leads to decreased radiative decay, is that the quantum yield (defined as the ratio of the radiative decay rate to the sum of all decay rates including non-radiative decay) of a dye when complexed with CB[7] increases.21 The significance of this statement is that cucurbituril complexation both decreases the radiative decay rate and increases the fluorescence quantum yield, a scenario that the Stickler-Berg equation seems to prohibit. This paradoxical result seems to suggest that cucurbituril complexation protects the dye from non-radiative relaxation pathways.21

A microenvironment with such a low polarizability that is able to affect the encapsulated molecule to this degree is interesting, albeit insufficient, evidence to support a controversial postulate by Cram that the inner cavity of molecular containers represents a new phase of matter.25

And let's face it: Nothing sounds sexier than the phrase "new phase of matter"

Further evidence for this sexy idea can be found in a more recent paper by Martin Czar. Czar used acridine orange as a probe of the inner cavity characteristics of CB[7]. He studied the fluorescence of acridine orange - both bound to CB[7] and on its own - in solution and in the gas phase. It turns out that when acridine orange is bound inside CB[7] it fluoresces as if it were in the gas phase - even if it isn't! Now that's an oversimplification, and I don't have time to describe the subtle details, but I seriously suggest you read that paper.

In case you're just skimming this GO BACK! That last paragraph (more specifically the paper I linked to in it) is one of the coolest things I've seen. It's really what I would consider one of the best papers I've read this year.

Cucurbiturils as synthetic receptors

Cucurbiturils not only have high binding affinities, but the binding can also be very specific to functional groups present in the guest. Methyl viologen (V2+) is an excellent example of this behavior. The charged nitrogens are positioned such that each is allowed to interact with the carbonyl portals. The reduction of doubly protonated methyl viologen (V2+ + e- → V+) has a smaller affect on binding affinity than would be expected if only interaction with the cucurbituril carbonyl portals were responsible for binding.

Methyl viologen (V2+)

Expecting to see this same behavior in other guests, a collaboration between Gibson, Inoue, Isaacs, Kim, and Kaifer26 instead demonstrated the unique binding of cucurbit[6]uril to ferrocenyl (Fc) guests. While FcOH and FcMe+ were found to bind strongly to CB[7],  binding to the sterically similar FcCOO- was not seen. The lack of binding can be attributed to the negative charge. Binding of FcCOO- to cucurbit[7]uril would  place the carboxylate functional group close to the electron-dense carbonyls of the cucurbituril. While this behavior of electrostatic repulsion is not at all unexpected, it does show a stark difference between cucurbiturils, which are capable of strong binding affinities and unique functional group specificity, and cyclodextrins, which in this case lack functional group specificity, binding to FcCOO- as well as FcMe+ and FcOH. It is this specificity and ultrahigh binding affinity that opens the door to possible applications in biology as synthetic receptors.

The binding of cucurbiturils to amino acids, peptides, and proteins has been studied. CB[6], as a host molecule, is too small to encapsulate amino acids.  However, Buschmann et al. have shown using isothermal titration calorimietry (ITC) that not only do amino acids have a high binding affinity (Ka ~1 × 103 M-1), but that binding affinity has little variability as the size of the amino acid is increased.27 From this data, the Buschmann group determined that the amino acids have very little interaction with the inner cavity and instead form externally bound complexes. Tao and coworkers later showed that the cavity of CB[7] was large enough to encapsulate two equivalents of phenylalanine, leucine, or tyrosine inside the cavity.28

An interesting application to cucurbiturils as synthetic receptors was demonstrated by Urbach et al. Capitalizing on the strong binding affinity of CB[7] to phenylalanine, a direct assay for insulin is reported. A fluorescent dye is first complexed with CB[7], which acts as a fluorescent quencher. When aliquots of insulin are added an increase in fluorescence is seen as the phenylalanine residue binds within CB[7], displacing the fluorescent dye. The effect is quantitative, and is seen even in a mixture of common blood proteins.29

Image credit: see ref 29.
In this review I may not have convinced you that supramolecular chemistry is the most amazing type of chemistry, but hopefully I've at least shown you something cool. 


1.         Behrend, R.; Meyer, E.; Rusche, F., Justus Liebigs Ann. Chem. 1905,  (339), 1-37.
2.         Meyer, E., Inaugural-Dissertation, Heidelberg, Germany. 1904.
3.         Freeman, W. A.; Mock, W. L.; Shih, N. Y., J. Am. Chem. Soc. 1981, 103, 7367.
4.         Nau, W.; Scherman, O. A., Israel Journal of Chemistry 2011, 51 (5-6), 492-494.
5.         Kim, K.; Park, K. M.; Ko, Y.; Selvapalam, N.; Nagarajan, E. R. Stationary Phase and Column Using Cucurbituril Bonded Silica Gel, and Seperation Method of Taxol Using the Column. Aug. 23, 2011.
6.         Inoue, Y.; Rekharsky, M.; Kim, K.; Ko, Y.; Selvapalam, N. Method for Determination of Presence or Absence of Peptide Compound PYY. Aug. 9, 2011.
7.         Ciaramitaro, D. A.; Zentner, B. A.; Lieux, A. J.; Merrit, A. R. Thermosetting Coatings for Particulate Materials and Methods of Application. Sep. 6, 2011.
8.         Buschmann, H.-J.; Fink, H.; Schollmeyer, E. Preparation of cucurbituril. DE19603377A1, 1997.
9.         Witt, D.; Lagona, J.; Damkaci, F.; Fettinger, J.; Isaacs, G. L., Org. Lett. 2000, 2, 755-758.
10.       Wu, A.; Chakraborty, A.; Witt, D.; Lagona, J.; Damkaci, F.; Ofori, M. A.; Chiles, J. K.; Fettinger, J.; Isaacs, G. L., J. Org. Chem. 2002, 67, 5817-5830.
11.       Chakraborty, A.; Wu, A.; Witt, D.; Lagona, J.; Fettinger, J.; Isaacs, G. L., J. Am. Chem. Soc. 2002, 124, 8297-8306.
12.       Lagona, J.; Fettinger, J.; Isaacs, G. L., J. Org. Chem. 2005, 70, 10381-10392.
13.       Isaacs, G. L.; Park, S. K.; Liu, S.; Ko, Y.; Selvapalam, N.; Kim, H.; Zavalij, P.; Kim, G. H.; Lee, H. S.; Kim, K., J. Am. Chem. Soc. 2005127, 18000-18001.
14.       Huang, W. H.; Liu, S.; Zavalij, P.; Isaacs, G. L., J. Am. Chem. Soc. 2006, 128, 14744-14745.
15.       Huang, W. H.; Zavalij, P.; Isaacs, G. L., Org. Lett. 2008, 10, 2577-2580.
16.       Miyahara, Y.; Goto, K.; Oka, M.; Inazu, T., Angew. Chem. Int. Ed. 2004, 43, 5019-5022.
17.       Svec, J.; Necas, M.; Sindelar, V., Angew. Chem. Int. Ed. 2010, 49, 2378-2381.
18.       Day, A.; Arnold, A.; Blanch, R., Molecules 2003, 8, 74-84.
19.       Kaifer, A. E.; Li, W.; Yi, S., Israel Journal of Chemistry 2011, 51 (5-6), 496-505.
20.       Reichardt, C., Solvents and Solvent Effects in Organic Chemistry. Willey-VCH: Weinheim, 1988.
21.       Nau, W. M.; Mohanty, J., Int. J. Photoenergy 2005, 7, 133-141.
22.       Rankin, M. A.; Wagner, B. D., Supramol. Chem.  2004, 16 (7), 513-519.
23.       Mohanty, J.; Nau, W. M., Angew. Chem. Int. Ed. 2005, 44 (24), 3750-3754.
24.       Nau, W. M.; Florea, M.; Assaf, K. I., Israel Journal of Chemistry 2011, 51 (5-6), 559-577.
25.       Cram, D. J., Nature 1992, 356, 29-36.
26.       Rekharsky, M. V.; Mori, T.; Yang, C.; Ko, Y. H.; Selvapalam, N.; Kim, H.; Sobransingh, D.; Kaifer, A. E.; Liu, S.; Isaacs, L.; Chen, W.; Moghaddam, S.; Gilson, M. K.; Kim, K.; Inoue, Y., PNAS 2007, 104 (52), 20737-20742.
27.       Buschmann, H. J.; Schollmeyer, E.; Mutihac, L., Thermochim. Acta 2003, 399, 203-208.
28.       Yi, J. M.; Zhang, Y.; Cong, H.; Xue, S.; Tao, Z., J. Mol. Struct. 2009, 933, 112-117.
29.       Urbach, A. R.; Ramalingam, V., Israel Journal of Chemistry 2011, 51 (5-6), 664-678.
30.       Isaacs, L. Israel Journal of Chemistry 2011, 51 (5-6), 578-591.
31.       Stancl, M.; Svec, J.; Sindelar, V., Israel Journal of Chemistry 2011, 51 (5-6), 592-599.

Sunday, 8 September 2013

#chemclub Reviews: C-H Activation

This month's review is by Kat, aka the Grumpy Chemist, a PhD student working on Pd-catalysed reactions. You can find her on Twitter as @Chemistry_Kat.

Biaryls – what are they good for anyway?
Biaryls are those nice, flat little structures that are very common in natural products, pharmaceuticals, agrochemicals or even functional materials and make one forget that there’s such thing as stereoisomers.

Here is where I usually pull out the statistics: according to Novartis, one of their best-selling drugs ($4 billion in 2012!) is Valsartan, used to treat high blood pressure. Valsartan contains a biaryl group, and it is only one of many. Last month, Jess’article on fluorinated drugs featured Lipitor, a drug with several interconnected arenes.

The long road from cross coupling to C-H activation
Everybody loves palladium-catalysed cross couplings. And they have good reason to do so: it’s a straight-forward and versatile class of reactions. Cross-couplings can connect sp to sp2 carbons (e.g. Sonogoshira coupling), sp3 to sp2 carbons (e.g. Fukuyama coupling), or my personal favourite, sp2 to sp2 carbons (e.g. Suzuki, Stille, Negishi couplings).

Since their discovery in the 1970s people constantly refined and improved upon the cross coupling reaction conditions. In 2010 Heck, Negishi and Suzuki, three of the “big players” in the field of palladium-catalysed cross-coupling, received theNobel Prize for their discoveries made almost 40 years earlier.

Traditionally, cross couplings need pre-functionalisation of both coupling partners, usually one as a (pseudo)halide, one as a organometal (or metalloid). The Suzuki coupling, one of the most common cross-couplings, uses arylboronic acids for coupling with arylhalides or even a pseudohalides like aryltriflates. Making the functionalised arenes, such as the arylboronic acid and the arylhalide needed for a Suzuki coupling, of course means additional synthetic steps. Placing and then, in the process of cross-coupling, removing the functionalities (boronic acid & halide) also generates usually solid, sometimes toxic, waste products.

Click figures for a larger version!
With cross-couplings like these well developed the next logical step, realised mainly within the last two decades, was to replace one of the coupling partners with a non-pre-functionalised arene. This is exactly what C-H activation is all about. Most people chose to replace the organometallic, as it is generally the more difficult to make and less stable coupling partner, rather than the (pseudo)halide.1

Taking it one step further one could imagine coupling two arenes but not pre-functionalising any of them. This is called oxidative coupling. It is currently the most difficult and still least developed method because of the stability and ubiquity of C-H bonds.

The trouble with regioselectivity
The problem with C-H activation is that most arenes have so darn many C-H bonds that can potentially react with palladium. The reactivity of arenes in C-H activations is extremely dependant on their substitutents. Despite being a major research focus, we’re not at the point yet where we can take any given arene and react it specifically at a desired C-H bond. We have mostly got the reactivity figured out: there are tons of methodologies of Pd-catalysed C-H activation out there that can bring a certain substrate type to reaction, but oftentimes the regioselectivity in arenes with more than one C-H bond leaves a lot to be desired. Something like toluene, that has no particular electronic properties making it either very electron-rich or electron-poor, usually gives mixtures of regioisomers in C-H arylation (if it reacts at all).

Solution 1: A directing group
A common way around this problem is to use a directing group. These are able to coordinate and direct the Pd catalyst to a certain position on the ring to perform the C-H activation regioselectively, and usually come in form of a heteroatom. Most, like phenols, amides, ketones, aldehydes, imines, heterocycles like pyridines or quinolones and many others, direct the reaction ortho to the directing group. Meta and para C-H activations are a little more unusual.

An example for a meta-selective C-H activation uses a specially designed “template”-type directing group for meta-olefination. This directing group can not only override effects from other potential directing groups groups, it also can be removed fairly easily.2

Carboxylic acids are excellent directing groups for palladium. Coincidentally (and conveniently), they can also be removed through decarboxylation. Combine the two and you get the cunning way to meta-selective C-H activation by arylating an ortho-substituted benzoic acid and then in-situ removing the carboxylic acid group: voilà, meta-subsitution.3

Solution 2: The electronics of the substrate
Heteroarenes like thiophene, furan and many others are so electron-rich that they can react with palladium(II) in an SEAr type reaction, usually with great regiocontrol for the most nucleophilic position. A lot of them react preferentially next to the heteroatom. Indoles are a little special in that respect: depending on the group on the nitrogen, the electronic properties can be tweaked towards favouring either C2 or C3 arylation.4

On the other end of the spectrum there are the very electron-deficient arenes like pentafluorobenzene. Even though there are exceptions, a rule of thumb is: the lower the pKa of the aromatic proton, the better its reactivity. Being thought to be quite unreactive for a long time, polyfluoroarenes and other electron-poor arenes can indeed react beautifully in C-H activation reactions, even at room temperature, through a mechanism coined CMD (concerted-metallation-deprotonation). Initially only suggested by a handful of groups, it is now an accepted mechanism that has been studied intensively in the last couple of years.5

And then there are those arenes which don’t have strong intrinsic electronic features, benzene being the godfather of all of them. One thing the currently available arylation procedures all have in common is that they use the “unactivated arene” in a huge excess, usually as the solvent. It’s a good thing that benzene, toluene and others are fairly cheap and readily available.6

To be continued
Now you know that palladium is pretty good at the whole C-C bond formation thing. People have thought of are some ingenious and elegant solutions to the problems of reactivity and selectivity. There’s no “the one (palladium) ring” for all possible substrates out there yet, but judging by the substantial number of new Pd-catalysed syntheses that are published every week, people work hard to find it. I’m certain that people will come up with more ingenious directing group designs that can be easily attached and detached on a very wide range of substrates for convenient C-H activation. Another thing I hope will get developed further in the future is oxidative coupling. Wouldn’t it be great to take any two arenes and just “snip off” the C-H bonds and make them into a C-C bond instead? Just like crafting, only with a transition metal as the scissors and glue and atoms as the paper.

1 A few reviews for further reading: Chem. Rev. 2007, 107, 174; Chem. Soc. Rev. 2011, 40, 4740; Acc. Chem. Res. 2009, 42, 1074; Chem. Soc. Rev. 2011, 40, 1885
2 Nature 2012, 468, 518
3 Angew. Chem. Int. Ed. 2011, 50, 9429
4 Chem. Rev. 2011, 111, PR215
5 Org. Lett. 2010, 12, 2116; J. Org. Chem. 2012, 77, 658

6 J. Am. Chem. Soc. 2006, 128, 16496

Friday, 2 August 2013

#chemclub Reviews: Fluorinated drugs

This month's review is by JessTheChemist, who did her PhD working with fluorine chemistry, and is now a postdoc researching chiral amines. She blogs at The Organic Solution.

Fluorine, why do I love thee so?

Although I am no longer a fluorine chemist, I thought that I would take you on a journey into what was the background of my PhD research all those years ago. Interest in this small but potent atom has boomed over the last decade. Since there is so much literature on fluorine chemistry, I am going to show you that the addition of fluorine to a molecule can have wide ranging benefits, particularly in the pharmaceutical industry.


Organo-fluorine compounds are pretty rare as natural products, although some do exist, for example, in 1943 fluoroacetate was isolated1. It may come as a bit of a shock that at least 20–25 % of pharmaceuticals contain at least one fluorine atom. These fluorine-containing drugs can be used in the treatment of cardiovascular diseases, obesity, and bacterial and fungal infections. Approximately 15 % of all drugs launched worldwide over the past 50 years are fluorinated2. One of the first successful fluorinated drugs was 5-fluorouracil which I previously wrote about in my #toxiccarnival post from 2012.  
Fluorinated organic molecules are known to perform a wide range of biological functions3. Cipro (Bayer AG) is an antibacterial, which is the fifth most prescribed antibacterial in the USA. Alternatively, Lipitor (Pfizer), which is a member of the statin drug class used for lowering blood cholesterol, has been the world’s bestselling drug for nearly a decade and has made a staggering $125 billion over 14 years.

There are a variety of reasons are behind the importance of the fluorine atom in drugs. The addition of fluorine to the 4-position of an aromatic ring is thought to prevent oxidation by cytochrome P450 due to the C-F bond being so strong4. If you want to learn more about properties of the C-F bond, then read this excellent O’Hagan review5. Substitution of hydrogen by fluorine, however, can also change the conformational preferences of a small molecule because of size and stereo-electronic effects as well as the pKa and lipophilicity of the molecule. These factors are discussed in more detail below.


The introduction of fluorine in to a molecule can have a significant effect on the acidity or basicity of a molecule due to its large inductive electron-withdrawing effect (electronegativity of F is 3.98). The effect is predictable as pKa generally decreases on increasing fluorination. For example, the pKa of acetic acid is 4.76 while the pKa of trifluoroacetic acid is 0.23.

The pKa of a drug can also have an impact on its bioavailability. For example, in a series of 3-piperidinylindole antipsychotic drugs, it was found that fluorination decreased the basicity of the amine, thereby improving bioavailability6. Changes in pKa can have effects on a variety of other parameters including physio-chemical properties (solubility), binding affinities (potency and selectivity), absorption, distribution, metabolism and safety issues2.

In HIV drug therapies, combinational therapies involving Efavirenz (Bristol-Myers Squibb) have been found to be the most active against the retrovirus. Structure–activity relationship studies have shown that the presence of the trifluoromethyl group improved drug potency by lowering the pKa of the cyclic carbamate, which makes a key hydrogen bonding interaction with the protein7.


For a drug molecule to pass through a cell membrane its lipophilicity (greasiness) must be such that it can pass into the lipid core but not become trapped within it. Increased lipophilicity often leads to increase in blood-brain barrier permeability. In general, exchange of a hydrogen atom by a fluorine atom leads to a more lipophilic molecule, however although there are exceptions, especially in aliphatic chains due to the strong electron withdrawing nature of fluorine8. Aromatic fluorination increases lipophilicity due to the good overlap between the fluorine 2s or 2p orbitals with the corresponding orbitals on carbon. This makes the C–F bond non-polarisable and, therefore, contributes to increased lipophilicity of the whole molecule9.

Steric Effects

Substitution of a hydrogen or hydroxyl group, for example, for a fluorine atom, in biologically active molecules is thought to be tolerated because the van der Waals radius of fluorine (1.47 Å) is in-between that of oxygen (1.57 Å) and hydrogen (1.20 Å). This means that fluorine substitution only changes the steric demand at receptor sites slightly. This can be seen by the use of the fluorovinyl group (C=CHF) as a replacement for the peptide amide bond10 for example in the antibacterial agent, REF883911.

Similarly, the introduction of a trifluoromethyl group (CF3) within a molecule can change the steric bulk significantly as its van der Waals volume is estimated to be similar to that of an iso-propyl group, with both groups having an effective van der Waals radius of 2.20 Å9. These steric variations combined with the high electronegativity of the fluorine atom, can lead to changes in preferred molecular conformation upon fluorine substitution which can be seen in Prozac, the famous antidepressant drug.

Prozac acts by selectively inhibiting the reuptake of serotonin, allowing the neurotransmitter to activate its specific receptor. It has been shown that the addition of a trifluoromethyl group in the 4-position of the phenolic ring increases the potency for inhibiting 5-HT uptake 6-fold, compared to the non-fluorinated equivalent.  It is believed that the size of the trifluoromethyl group at this position allows the phenoxy- ring to adopt a conformation which favours binding to the serotonin transporter6.

Electrostatic Interactions

It has been shown that the C–F bond dipole adopts a Bürgi-Dunitz* type trajectory to amide carbonyl groups on the peptide backbone. This electrostatic effect is weak relative to other protein–ligand binding interactions but the C–F bond dipole contributes to optimising how drugs orientate at the binding site of their target enzyme/protein5.


Isn’t fluorine great? As you can see, the addition of a fluorine atom can change the properties of a molecule in many different ways. I am sure there are many other properties and benefits of fluorine that scientists are yet to find. It is certainly an exciting time to be a fluorine chemist. There are some very exciting chemists working in fluorine chemistry and, as such, I very much look forward to seeing how the field improves of the next few years, particularly in relation to selective fluorination of aromatics for use in positron emission tomography (PET) imaging.

*I met Dunitz once at ETH, what a legend!

Friday, 12 July 2013

Punching Up: chemophobia and DHMO

Over at Pharyngula, Chris Clarke has some pointed words for douchebags chemists about a popular satire of chemophobia: the dangerous and ubiquitous chemical DHMO, or dihydrogen monoxide.

It's a pretty popular hashtag on Twitter and I'm not exempt from joining in the joke, so Chris' criticism includes me. In brief, his point is that this joke unfairly mocks the uneducated:
The dihydrogen monoxide joke punches down, in other words. It mocks people for not having had access to a good education. And the fact that many of its practitioners use it in order to belittle utterly valid environmental concerns, in the style of (for instance) Penn Jillette, makes it all the worse — even if those concerns aren’t always expressed in phraseology a chemist would find beyond reproach, or with math that necessarily works out on close examination.
(Emphasis in original)

I largely agree with Chris. As I've noted before, when responding to chemophobia we have to avoid the temptation to mock others' ignorance or to be condescending. It's unpleasant and counter-productive, and risks alienating people we ought to be communicating our work to and helping to educate. It gives us a bad name, makes us look arrogant and out of touch with the public, and is dismissive of legitimate or at least understandable concerns. None of what follows is intended to defend making jokes at the expense of the uneducated, or knee-jerk tribalism that 'belittles utterly valid concerns'.

Despite these important criticisms, I do think the DHMO trope has some merit. Conveniently, there's even a recent example to illustrate this for me. Buzzfeed recently ran a list of 8 food additives that are supposedly banned everywhere but the US. It's a classic of the chemophobia genre: scare-mongering, entirely (and probably deliberately) ignorant, and widely-read (over 5.2 million views and over 500,000 likes/tweets/shares at the time of writing). It's also a great example of churnalism and seems to have been lifted directly from a health food book.

The fears the article tries to whip up are pretty much baseless: Derek Lowe has pulled apart this article at length and he illustrates that the average high school student has a better grasp of chemistry than the author of this rubbish. Mark Lorch responded differently, with a satirical piece in the Guardian listing six chemicals that ought to be banned... including DHMO.

Mark's response illustrates the correct (in my view) application of the DHMO trope. He takes the same approach to everyday chemicals as the BuzzFeed writer: cherry-pick alarming-sounding properties, overextrapolate from extreme conditions to everyday conditions, and generally try to whip up as much unfounded fear as possible whilst ignoring logic at every turn. In doing so, he exposes these flaws in the original piece and hopefully provokes some thought amongst those who might otherwise have bought it. 

It's not an approach that's to everyone's taste, but I think it's in the tradition of great satire: it punches up and is instructive rather than cruel. The authors of the BuzzFeed article and the book it's based on are not uneducated and have a huge platform: hardly the same demographic Chris rightly defends. Chris' comments are a welcome reminder that satire has to be deployed deliberately and at the right targets.

A final thought: how widespread do you think the ignorance Chris describes (and the BuzzFeed article preys upon) is?

Less than 10% of those who read the BuzzFeed article 'liked' it on Facebook, assuming everyone who 'liked' it read the damn thing first. Only 2-3% shared it on Twitter or Facebook. This isn't an unusually high attrition rate for articles like this, but I have to wonder how many people who didn't share it either knew outright that it was crap, or quickly googled it and found out.  

Edit: I spelled Chris' surname incorrectly and have now rectified it. Sorry about that.

At SciAm Blogs, Janet Stemwedel has responded to Chris' post. Her response is considerably more interesting than mine and worth your time.

Wednesday, 10 July 2013

#chemclub Reviews: The Pummerer Reaction

This month’s review looks at the Pummerer rearrangement. I first met this reaction as an undergraduate and liked it for a few reasons: for one thing, it worked well! It has a simple mechanism that can lead to diverse behaviour and is always nice in group problem sessions. This brief discussion will cover the basic details of the reaction and a few interesting variants.

The Pummerer rearrangement was discovered in the early 20th century, although probably not by Rudolph Pummerer (a recurring theme with named reactions). Closely-related reactions were reported by Fromm & Achert and by Smythe before Pummerer’s first papers on the subject were published in 1909 and 1910, but it languished in obscurity until the 60s. It was only then that Pummerer’s name was attached to the reaction (for a nice discussion of the history of this chemistry see the review by Feldman). Pummerer's career included a great deal of sulfur chemistry but this class of reactions was not a great focus of his work; perhaps more important than his chemical legacy is his involvement in the relaunch of Angew. Chem. in 1947.

Sunday, 30 June 2013

#chemclub Roundup 12

Once again, here's a roundup of the best of the #chemclub Twitter feed from the last two weeks. (Confused? Read this.) The second review will go online early in this week; this unfortunately coincides with the death of Google Reader tomorrow. If somehow you both use this reader and live under a rock, you may need a replacement; personally I'm using Feedly, which now has a pretty green button on this page.

Tuesday, 18 June 2013

Acc. Chem. Fail.

Last week there was an online campaign to create pressure to publish negative results. The potential benefits are obvious, and are nicely summed up in the cartoon by Nik Papageorgiou: if we had a database (like Reaxys) of "stuff that doesn't work", we could all save time, effort, and money going down futile routes. Much of it has been coming from biological quarters, but it's something that chemists have proposed too.

(Quick disclaimer: I've not really been part of this conversation, so if my comments have been rebutted elsewhere please do correct me.)

I'm a little skeptical of whether it's really as simple as "publish your negative results!", at least when it comes to chemistry*. It's not enough to say "we tried these conditions and it didn't work". This isn't going to be too helpful; there are a million reasons why a particular reaction might not work in your hands (look at BlogSyn for a detailed example of this). For such a resource to be useful it has to be thorough: you have to try to pin down why your reaction doesn't work, and that's not a trivial matter.

The journal behind last week's campaign even say as much:
"For negative and null results, it is especially important to ensure that the outcome is a genuine finding generated by a well executed experiment, and not simply the result of poorly conducted work.  We have been talking to our Editorial Board about how to try to avoid the publication of the latter type of result and will be addressing this topic and asking for your input in a further post in the next few days."
I don't know how it is in the biomedical sciences, but in chemistry I'm not sure it's going to be clear up front why a reaction doesn't work. Ensuring that it is a genuinely negative result will take time, and is likely to be of limited interest to the wider community; understanding why the reaction doesn't work is yet more work, but will be much more useful.

Who is going to take the time and effort to really thoroughly study a failed reaction and figure out why it works except a methodology group that is already studying that chemistry in depth?

To illustrate my point I'll use an example from my own area, self-replicating molecules. In 2008, Vidonne and Philp reported an attempt to make a self-replicating rotaxane. I'm going to stop here to express my sincere admiration at the scale of a project like this. I'm not aware of any other attempts to achieve something like this; it blows my mind a little bit.

From Tetrahedron 2008, 64, 8464–8475.
The paper is a thorough study of the system and runs to 12 pages. They report careful planning and a detailed kinetic model; the synthesis and analysis of the system; and experiments to figure out exactly what is occurring in this complex system and why it didn't behave as expected. Judging by the abstract, this represents a huge portion of Vidonne's thesis.

This is the kind of detailed work needed to make negative results worthwhile - both to publishers and to other researchers. Anyone can break a reaction, but it takes time and attention to detail to turn that into useful knowledge.

That said, there are lesser steps that can be taken to get useful negative results into the literature without expending so much effort. For example, methodology papers could include tables of substrates or conditions that didn't work in their SI; synthesis papers could (and often do) discuss methods that failed for them.

An interesting alternative is the robustness screen recently proposed by Collins and Glorius. They describe a standardised 'kit' that may allow chemists to quickly get an idea of whether a particular set of reaction conditions is likely to tolerate functional groups and so on. One strength of this idea is that it would require chemists to report negative results: "our conditions tolerate A, B, and C, but are shut down by X, Y, and Z".

To sum up: it's easy to say "publish your negative results!", but in chemistry at least it's not clear that it's that straightforward. To be worth publishing, or worth anything, you have to have an idea why the results are negative, or negative results need to routinely reported alongside positive results.

What do you think negative results could contribute to chemistry? What information would you need for a negative result to be useful to you?

* to clarify: none of my comments are meant to generalise to all of science, or beyond organic chemistry, really. In other fields this may well be more straightforward.
** thanks a big huggy bunch to @PeONor and @craigdc1983 for having a look over this post before it went up.

Sunday, 16 June 2013

#chemclub Roundup 11

Here's the usual roundup of papers from the #chemclub Twitter feed and various chemistry blogs in the last fortnight. If you're not sure what #chemclub is, click here.