Corante

About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: derekb.lowe@gmail.com Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Emolecules
ChemSpider
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
PubChem
Not Voodoo
DailyMed
Druglib
Clinicaltrials.gov

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
Kilomentor
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
ChemBark
Realizations in Biostatistics
Chemjobber
Pharmalot
ChemSpider Blog
Pharmagossip
Med-Chemist
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
SimBioSys
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Business|Bytes|Genes|Molecules
Eye on FDA
Chemical Forums
Depth-First
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa


Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
FuturePundit
Aetiology
Gene Expression (I)
Gene Expression (II)
Sciencebase
Pharyngula
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net


Medical Blogs
DB's Medical Rants
Science-Based Medicine
GruntDoc
Respectful Insolence
Diabetes Mine


Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem


Politics / Current Events
Virginia Postrel
Instapundit
Belmont Club
Mickey Kaus


Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Chemistry Jobs Roundtable: What About Tenure? | Main | Science Jobs Roundtable - Day Four »

December 15, 2010

What A Paper Doesn't Have In It

Email This Entry

Posted by Derek

So, should I apologize for the tone of yesterday's post? I'm not ready to yet, and I'll explain why.

Here's a general rule: if you find yourself having to make excuses for a scientific paper, for the key experiments that weren't done and the rationales that don't appear, then something is wrong. A paper shouldn't make you have to assume that the authors will get around to running Experiment X and Comparison Y and Test Z, and put that in their next manuscript. (If this makes you think of the recent arsenic bacteria controversy, you're right on target). Anything like that which immediately comes to mind should be in the first manuscript. If the authors haven't put it in there, then it's the job of the referees to tell them to go back and do it. And whoever refereed this paper did an incompetent job.

Now, a quick technical note. It's true that medium-sized rings can form isolable atropisomers in some cases. (An atropisomer, for those who aren't chemists - or chirality geeks - is a compound that can exist as two mirror-image forms just through some sort of constrained movement of its parts. A putatively flat compound that comes around and wraps over itself like a screw thread is one example, as is one with a bond that should be freely rotating but it blocked by large side chains from doing so).

Problem is, this compound sure doesn't look like one of those cases. It has no bulky groups that have trouble getting out of each other's way, and it has no helical chirality. The only thing it has, to my eyes, is a nitrogen that could only lead to new isomers if its barrier to inversion were really, really high - and it shouldn't be. For nevirapine, there really should be no way to isolate such an isomer at room temperature. If anyone can provide evidence for isolable atropisomers in a system as small and lightly substituted as this one, I'll certainly consider eating my words - but not until then.

But that brings up a larger point. This is actually one of the things that makes me think this paper is bogus: the presence of such a compound really would be the big selling point of the manuscript, if the authors had thought about it. Instead, their main focus is on how the structure turns out to be a natural product (which I have trouble believing, too),and not on the bizarre nature of it being chiral. The chirality, frankly, seems to be an afterthought, the way things are written. The word "atropisomer" does not make an appearance. References to the other dibenzo ring systems that have shown this interesting phenomenon (which were worth papers all their own) are not cited. There is no mention of a nitrogen inversion. (If that were the explanation, you'd also expect that heating up the sample would eventually start flipping the molecules past that barrier and removing the optical activity - but there's no mention of any such thing). You'd never know that there was something interesting going on, because the authors give us no reason to believe that they knew that, either.

It would also be quite interesting, if you could have such a thing as an optical isomer of nevirapine, to see what its activity would be on its enzyme target, reverse transcriptase. And if nevirapine could exist as enantiomers, how about running the synthetic material down some chiral columns to see if you could resolve it? Then show that your new optically active stuff is only one of those peaks; that would be pretty convincing. Not done, not done, not even mentioned, etc.

Here's another consideration: as mentioned yesterday, I don't think that this structure has been given very good characterization (the X-ray data seem insufficient to talk about chirality). One of the comments to yesterday's post wondered why the authors didn't show HPLC traces with and without a spike of the authentic drug material. That's an excellent idea, and it's something that would be worth showing in an NMR spectrum, too. Claiming that you found nevirapine in a plant is quite weird - you'd want to really hammer down the fact that everything is identical. But these experiments haven't been done, either.

Allow me to mention one more oddity. The authors actually make reference to "optically active nevirapine" (their footnote #10), but the reference they cite (the original paper from Boehringer Ingleheim) does not, as far as I can see, mention any such thing. And that's because no one has ever mentioned any such thing, and that's because I don't think it exists, outside of (just maybe) a low-temperature NMR experiment.

No, I'm still not buying this. I'm upset with the authors for having proposed such a thing with such thin evidence, but I'm really more upset with the editors of what is supposed to be a reputable journal for publishing it.

Comments (27) + TrackBacks (0) | Category: The Scientific Literature


COMMENTS

1. anana-mouse on December 16, 2010 10:36 AM writes...

Please, let this rest. I can't tell you how many papers of dubious quality comes from thence. I could quote you many of this ilk on how the clay montmorilonte "catayses" every reaction one could concieve.

Permalink to Comment

2. simon on December 16, 2010 10:36 AM writes...

One of these guys have to explain why this paper went through.

Editors
Professor L. Ghosez
Professor G.-Q. Lin
Professor S.F. Martin
Professor W.B. Motherwell
Professor G.P. Pandey
Professor B. M. Stoltz
Professor R.J.K. Taylor
Professor K. Tomioka

Permalink to Comment

3. excimer on December 16, 2010 11:05 AM writes...

A much better analysis of the crystal structure of nevirapine can be found here. The molecule itself crystallizes in the centrosymmetric space group P21/c, meaning that if the molecule is chiral, the asymmetric unit must exist as a racemate, which it does (PABHIJ01 on the CSD if you have access). So, yes, the molecule is chiral but exists as a racemate in the solid state.

Permalink to Comment

4. Anonymous on December 16, 2010 11:10 AM writes...

Hi Derek.

you say the following about atropisomers. "Now, a quick technical note. It's true that medium-sized rings can form isolable atropisomers in some cases. (An atropisomer, for those who aren't chemists - or chirality geeks - is a compound that can exist as two mirror-image forms just through some sort of constrained movement of its parts. A putatively flat compound that comes around and wraps over itself like a screw thread is one example, as is one with a bond that should be freely rotating but it blocked by large side chains from doing so)."

All this is true. But unless a compound has a chiral center, a compound can form both opposite (enantiomeric) atropisomers just as easily and thus no single enantiomeric form would be enriched over the other. The mixture would therefore still be racemic and still not show optical activity. (I should note that it is in principle possible to form an enantiomers form of an atropisomer and then to separate them if the energy barrier would be high enough. Heating such a sample should make the optical activity disappear quickly.

My 2 cents worth.

Permalink to Comment

5. p on December 16, 2010 11:35 AM writes...

anon, it's true that in the lab it would be hard to make. But if we accept that atropisomers are possible for the compound AND that it is a biosynthesized natural product, it doesn't seem a big jump to suggest the enzymes that make the molecule preferentially make one enantiomer.

It should have a higher than average barrier to inversion as that barrier is anti-aromatic. Whether that is high enough or not, I have no idea.

I agree with Derek that is annoying when the first experiments you think of - and very easy ones to do as well - aren't done. It isn't as if anyone is suggesting a 9 month set of experiments.

Permalink to Comment

6. Phil on December 16, 2010 11:45 AM writes...

@ simon

No way Stoltz was the editor responsible.

@ Derek

Don't apologize. Your skepticism is completely justified. I don't understand how any rational scientist could think the burden of proof is on ANYONE but the authors. As the reader, you should be convinced by the paper. You shouldn't have to convince yourself.

Permalink to Comment

7. Anonymous on December 16, 2010 12:04 PM writes...

Maybe I'm tired (or stupid, or both), but wouldn't it have optical activity if one of the conformations of the 7 membered ring was favored over the other? It doesn't seem likely for that structure but...

Permalink to Comment

8. Hap on December 16, 2010 1:03 PM writes...

No, because there's no reason for one enantiomer to be favored in the absence of some sort of chiral inducer. Enantiomeric conformations should be of equal energy, and not distinguishable in the absence of a chiral inducer. If enantiomers are possible, diastereotopism should show up in the NMR if they interconvert slowly enough, which would also be a requirement to see optical activity. If they can't be distinguished by NMR, then they probably can't be isolated either.

It shouldn't be the reader's job to substantiate the authors' claims - that's the authors' job. If they haven't substantiated themselves, well, they earned contempt for themselves. Also, while some people may want to read the "Journal of 'Look, I Just Thought Up Some Cool Sh%t Up While I Was Smoking My Bong'", I don't think they'd get enough of them to justify their subscription fees.

Permalink to Comment

9. strayxray on December 16, 2010 1:45 PM writes...

As someone mentioned yesterday, this molecule crystallized in space group P-1.

That means the crystal is made up of the structure drawn in their paper AND it's mirror image, since the "-1" means there is an inversion center.

http://materials.cmu.edu/degraef/pg/pg_bar1.gif

They crystallized both forms of the molecule (with diethylether in the lattice as well). How is the solution optically active?

Permalink to Comment

10. Jose on December 16, 2010 3:30 PM writes...

Damn, hard to believe that piece o'fluff got into Tetrahedron. "Tetrahedron publishes experimental and theoretical research results of outstanding significance and timeliness in the field".

RBW must be spinning in his grave....

Permalink to Comment

11. Hap on December 16, 2010 3:37 PM writes...

I keep thinking of atropisomeric amide isomers when the isomerism (?) is caused by atropisomers at N. I still don't get it, but I guess I'm not alone. Hurray.

Permalink to Comment

12. Phil on December 16, 2010 4:12 PM writes...

@3 and 9:

I do believe that in the solid state, this molecule will adopt both chiral conformers, but the crux of most of the objections is that this molecule would not be conformationally stable in solution at room temperature. Such molecules are usually not considered enantiomeric.

Permalink to Comment

13. Iridum on December 16, 2010 4:47 PM writes...

In fact tertiary amines (with R1, R2 and R3 being different from each other and R4 being the lone pair) are not considered chiral compounds.

The two "enantiomers" are not stable.

Permalink to Comment

14. Jose on December 16, 2010 5:32 PM writes...

There are examples (Clayden has done a lot of work)

http://www.sas.upenn.edu/~pwalsh/Papers/chirality-15-615.pdf

But I don't see enough evidence to support it here, and the authors sure as hell don't reference anything relevant.

Permalink to Comment

15. Tex on December 16, 2010 9:01 PM writes...

@13 Iridium

Actually, the amines you describe are chiral - there are non-superimposable mirror images. The issue pertains to the barriers for inversion, which if sufficiently high results in resolvable enantiomers. As you indicate, for simple amines this is not feasible as the inversion barrier is too low, on the other hand there are many examples of compounds in which the only chiral center is nitrogen and the two enantiomers are resolvable. In many cases, this involves the nitrogen being in a small ring for which pyramidal inversion is very difficult (highly strained transition state), but there are cases known where the nitrogen atom is not constrained in a ring, but with suitable choices of nitrogen substituents undergoes very slow pyramidal inversion, and thus are resolvable (at least partially so)

Permalink to Comment

16. Fbomber on December 16, 2010 9:29 PM writes...

Sigh......
Hap and Jose are on to something, but the stable clayden and (what I call) curran amides are all tertiary.
The key to atropisomerism is time, which of course is related to energy of inter conversion. Oki defined it almost a century ago as those chiral conformers whose half-life to interconversion was at least 1000s at rt. This definition is accepted by those in the field, such as clayden (who b the way has a nice paper in angew on the topic as it pertains to drug discovery). The barrier is too low for nevirapine and so the half life way too short for it tohave been made chiral biosynthetically (lol) and then isolated and characterized.
The bottom line is that this paper is shabby at best- peer review has let us down on this one.

Permalink to Comment

17. Fbomber on December 16, 2010 9:37 PM writes...

Oppsy. It was in the 80s for Oki. My bad. Peer review would havecaught that.....

Permalink to Comment

18. Paul on December 17, 2010 2:51 AM writes...

Pardon my ignorance, but what does it mean to be "a patent busting exercise"? How would a paper like this bust a patent, and what patent is it busting?

The concept isn't quite clear from the Wikipedia article on the EFF initiative, and I'd love for a perspective specific to chemistry.

Permalink to Comment

19. iridium on December 17, 2010 3:35 AM writes...

@ 14Tex

I was just trying to make a simple example to say that the energy for inversion is the key factor.
But probably I simplified too much....and of course you are right.

Permalink to Comment

20. pandora on December 17, 2010 9:59 AM writes...

The rotation barrier of the bond linked with cyclopropyl is about 10 kcal/mol, so it is possible to be optical active in rt. If you change the cyclopropyl to a even bigger group, it sure will be optical active.

Permalink to Comment

21. Hap on December 17, 2010 10:00 AM writes...

This post is likely an example of a clear patent-busting exercise - in that case, I think Dr. Reddy's was trying to claim that the method used to make the compound was obvious and therefore not patentable - hence busting the patent. (The counters were that the obvious method fails, for a variety of reasons.)

I don't think the authors are trying to bust the patent in that sense, but they are probably trying to claim another source for the compound, perhaps so they can sell it? Other than that, I don't know.

Permalink to Comment

22. Chemist turned banker on December 17, 2010 10:19 AM writes...

@18 Paul

Patent busting is an attempt to invalidate a patent, in this case by arguing that nevirapine, as a "natural product" is unpatentable subject matter. You can patent natural products if you draft your claims suitably (most often by imposing purity constraints or demonstrating unexpected properties), but you can't claim the molecule per se. In this case, BI's '972 patent claims "1. 11-Cyclopropyl-5,11-dihydro-4-methyl-6H-dipyrido[3,2-b:2',3'-e][1,4]diazepin-6-one, or a pharmaceutically acceptable salt thereof. ", so could fall as invalid under Section 101 of the Patent Act.

Of course, that presupposes that this paper is correct, and I am with the majority on that point...

Permalink to Comment

23. wwjd on December 17, 2010 10:30 AM writes...

If it is a natural product, could you isolate it and sell it as a nutritional supplement? (I understand that it is not really natural, but the authors are claiming it to be). Maybe saying this supplement helps give you energy and fights general malaise (espicially for those infected with a particular virus). These statements have not been evaluated by the FDA nor is this product meant to treat any disease.

Permalink to Comment

24. anchor on December 17, 2010 3:23 PM writes...

#23 a very valid point and knowing that nutrition industry is not well regulated.

Permalink to Comment

25. chiral on December 17, 2010 8:41 PM writes...

Derek,

Nice post on this one. There are a number of interesting things associated with this paper, as indicated by the many posted comments.

I read the paper, it seems to be a reasonable research with detailed experimental procedures and various characterization data, and notably a X-ray structure. I would think this 3-page paper would be more appropriate for Tetrahedron Letter instead of Tetrahedron, since it is really a preliminary report on an interesting "discovery", without extensive direct comparison to the synthetic molecule and evidence and discussion on its chirality and optical activity.

If the experimental and claims are true, it would be a sensational finding.

At this point, there are two obvious things that bothers me. (1) why did an Et2O molecule show up in the X-ray structure, when it was crystalized in EtOAc; (2) in ref 10, the NMR spectra of the synthetic molecule were determined in DMSO-d6, but why did the authors avoid a direct comparison by using CDCl3? These things at least demonstrate the apparent sloppiness in their research.

But I would still hope this is a true claim, based on their tedious effort in getting this result. If so, it would make more sense that this is from someone who should be competent enough to conduct this type of research. The lead author seems to have published >40 papers in the past years in some good journals on natural products, X-ray structures, HPLC methods, medicinal chemistry and synthetic methodology. If this is the same person, this author seems to have research experience in the US, and worked with Dr./Prof. Rao who is a legendary accomplished personality originally from India.

Therefore, I really hope this interesting report, despite of sloppiness, was based a solid science, which could have been published in Nature, if solid evidences are provided. Otherwise, it would not make any sense for them to do so.

Permalink to Comment

26. Hap on December 22, 2010 10:20 AM writes...

The NMR of nevirapine (J. Comp.-Aided Mol. Design, 2001, pp. 997-1004) shows three 1H signals for the cyclopropane (I assumed, perhaps wrongly, that they compared to the real spectrum). That isn't consistent with the nitrogen configuration being stable - otherwise, all five Hs should be different (the ones on the concave and convex sides of the tricyclic system sould differ, and their assignments indicate otherwise). I'd have to have some explanation why the original spectrum didn't have 5Hs while the current authors' spectrum did before I'd buy the current paper.

Permalink to Comment

27. RC Mishra on February 13, 2011 10:29 PM writes...

I got to know about this through one of my colleague here at UGA, when we found the original article, we saw that its from the same institute where we have done our PhD.
We are contemplating the points to discuss with the scientist involved in structure determination.
Let's see what comes out.

Permalink to Comment

POST A COMMENT




Remember Me?



EMAIL THIS ENTRY TO A FRIEND

Email this entry to:

Your email address:

Message (optional):




RELATED ENTRIES
Gitcher SF5 Groups Right Here
Changing A Broken Science System
One and Done
The Latest Protein-Protein Compounds
Professor Fukuyama's Solvent Peaks
Novartis Gets Out of RNAi
Total Synthesis in Flow
Sweet Reason Lands On Its Face