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: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
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
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
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
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

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

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
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

« Hexacyclinol: A Forensic Case | Main | Genentech's Culture: At Risk or Not? »

February 20, 2009

Hexacyclinol - Another Request

Email This Entry

Posted by Derek

I'm taking the day off from cranking out the medicines of tomorrow (OK, the day after tomorrow), so there will be no post today.

I did want to add something about yesterday's post on the La Clair/hexacyclinol controversy. I'd like to ask that people not fill up the comments with ad hominem remarks or potentially libelous statements about La Clair himself. I don't mind saying that the evidence so far makes it very hard for me to believe his original paper, and I also have to say that I haven't seen any convincing explanations for all the discrepancies that have turned up. And I think that those opinions are shared by many people who've followed the story.

But let's keep it on a scientific plane, if possible. Opinions on NMR spectra and the like are one thing, but personal insults are another, and those we don't need. I try not to have to go in and hose out the comments sections around here.

Comments (15) + TrackBacks (0) | Category: Blog Housekeeping | Chemical News | The Scientific Literature


1. Hap on February 20, 2009 1:19 PM writes...

Sorry if I said anything bad (well, the goalposts comment was catty). I don't wish Dr. La Clair any ill, but it would be nice to have a complete picture of the whole hexacyclinol shebang.

Permalink to Comment

2. HOMO-LUMO on February 20, 2009 1:43 PM writes...

To me the point is how an article without any supporting information and with notorious mistakes and irregularities got in Angewandte Chemie.Usually referees ask for lots of stuff before accepting a good paper with isolated and punctual mistakes. Another issue is why some people does not need to comply with rules applicable to almost everybody.

My earlier point is that without controversy the system would not be corrected.

Permalink to Comment

3. Anonymous on February 22, 2009 9:16 AM writes...

HOMO-LUMO: I hope you are kidding. ACIE, along with JACS and other journals, are full of papers with incomplete data for key compounds, including target molecules of natural product syntheses. After years in the field, it is clear to me that there is no real honor for being in one of these "high impact" journals. It rather implies that you know someone on the editorial board or have suggested a very, very friendly (read: they are your friends) set of reviewers. Sadly, the literature has degenerated to this state, with the funding agencies are blindly following the same folly.

Permalink to Comment

4. HOMO-LUMO on February 22, 2009 1:45 PM writes...

Exactly, that´s why controversial cases are good to show how the system is far from perfect and needs fixing.

Always there is an honour in been in that high impact journals if your science is relevant, influential and reproducible.

Permalink to Comment

5. Mad Chemist Chick on February 22, 2009 1:46 PM writes...

I find it very frustrating that you can pull 10 references for procedures and you are lucky if 2 work as described.

For the older literature, this doesn't seem to be a problem. Give me a procedure from a 1940s JACS any day over a 2008 JACS.

Permalink to Comment

6. Anonymous on February 22, 2009 2:13 PM writes...

I will say that J.Org.Chem., and Org.Lett, are now very fastidious about requiring complete compound characterizations in the Supp Info. This makes me more confident that the syntheses described will be succesful. You may not get the same yield, perhaps, but at least you will get product.

Permalink to Comment

7. HOMO-LUMO on February 22, 2009 4:45 PM writes...

Tetrahedron till the 90s is also quite good. In my little experience 8-9 out of ten preps works.

Chemistry before the 70s was a craftmen job, without the comprehensive analysis techniques exiting nowadays is amazing how so many things were discovered and developed. I guess that is one reason why reproducibility was at that time at the top of the academics priorities when publishing a paper.

Permalink to Comment

8. Bill on February 22, 2009 10:24 PM writes...

HOMO-LUMO has hit upon a key point. In the past, most of the top organic chemists were focused on structure elucidation of natural products. This required a wide variety of highly developed lab skills with success still not guaranteed. A good
example was the German chemist (Loucks?) who spent
over 40 years working on the structure of strychnine, only to be 'scooped' by R. B. Woodward. Now modern NMR methods using state-of-the-art equipment could determine the structure of strychnine in, at most a weekend, using no more than 1 mg of sample. In fact, as an NMR spectroscopist, I have been accused by senior organic friends as being one of the ones who
ruined the natural product area by making it too easy. While there is an element of truth in this, nevertheless, one type of skill set has been replaced by another. Modern spectroscopic methods, incorrectly applied, can still give wrong structures. I believe that hexacyclinol is an example where inadequately resolved 2D spectra gave ambiguous results which the original authors interpreted incorrectly, yielding the wrong structure. This is not uncommon, as evidenced by a 2005 article by K.. Nicolau which listed well over 100 structures where spectroscopic methods gave wrong stuctures, mostly I suspect because
authors convinced themselves that their spectra fit their prefered structure, rather than letting
the data lead them to the structure. Overall, I believe that authors, reviewers and editors are all too willing to accept limited data as proving structures, which may at least in part for literature synthetic methods which don't work in
other hands. Certainly. there is a pressing need for editors to insist on detailed procedures and
actual spectra (if only as supplemental material) so that literature reports are reliable and can be reproduced.

Permalink to Comment

9. JAB on February 23, 2009 9:40 AM writes...

Another factor is that for natural product structure elucidation by spectroscopy, referees do not get the level of experimental detail required to be sure that the structure is correct, nor do they have the time to work all the way through the structure proof. A referee can only say that the structure proposed is consistent with the data (or not) and that the elucidation has been documented properly. This applies to truly new structures - if the compound class is well known it's not much of an issue. But if you propose a new carbon skeleton you need to bring more to the table, and some chemical transformations are always a good thing. I don't really consider a novel structure secure until there's a good x-ray structure completed.

Permalink to Comment

10. Retread on February 23, 2009 9:44 AM writes...

To Bill #8

"I have been accused by senior organic friends as being one of the ones who ruined the natural product area by making it too easy."

Well, I don't know if you ruined it, but the logical determination of structure based on chemical reactions was a beautiful thing. In the fall of 1960, we had a series of lectures on the chemcial determination of the structure of morphine by David Ginsburg. My lecture notes cover 46 pages. Thanks be to God, I wrote them up afterward and saved them. An exquisite chain of logic and chemistry.

A similar type of deductive logic was employed in the detailed neurologic examination of the 60s and 70s which could localize the site (or sites) of pathology causing the problem. The CAT scan and later the MRI made this nearly irrelevant. No one wants to go back to that era. Not the patient and not the doctor.


Permalink to Comment

11. Hap on February 23, 2009 11:18 AM writes...

1) We are capable of isolating lots of things in very small quantities which are likely to be hard to characterize chemically (even the spectroscopic characterization can take up most of the sample). Actinide chemists do some chemistry on really small scale, but I don't know if the techniques are applicable or available to natural products chemists.

2) It doesn't require lots of NMR to mess up a natural product structure - the structure of diazonamide A was incorrect with a crystal structure to support it.

3) There was probably a lot of good careful chemical determination of natural product structure before NMR, but we might be comparing the determinations now with the older determinations we can remember, which are likely to be remembered because they were good. That doesn't say anything about the crappy ones (or, to someone who hasn't read the older literature as much, whether there were crappy ones).

Permalink to Comment

12. Bill on February 23, 2009 2:06 PM writes...

Hap, I still remember about 45 years ago, hearing an R. B. Woodward synthesis seminar which was a beautiful examnple of scientific logic and rigor. Unfortunately, as you suggest, few others were in his class. My concern at present is that many organic chemists have become complacent in assuming that modern spectroscopic methods have made structure elucidation routine and relatively foolproof.
This may have let to the habit of authors of not fully documenting their methods and structure proofs and referees and editors tolerating this.
Unfortunately, these methods are still error prone, particularly in non-expert hands, and even highly experienced users of NMR often still have problems with stereochemistry (and occasionally skeletal structures). Also, as pointed out, x-ray crystallographic data can also be misinterpreted. These problems are now generally recognized in the natural product community, but not nearly as much by synthetic chemists and I wonder how many errors there are in the synthetic literature because of this.

Permalink to Comment

13. Retread on February 23, 2009 4:52 PM writes...


The Woodward evening seminars certainly were a thing unto themselves. The same notebook with the Ginsburg lectures also has my notes on two of them 20 Oct '60 M. Rosenblum on The Von Richter Rearrangement and 8 June '61 Stevenson on triterpene rearrangements. I wish I'd taken more of them down, but more often than not there was no speaker so Woodward would start things off by writing a problem on the board, followed by a serious game of king of the mountain by those in the audience. Anyone could go up to the board with a solution. It was an occasion for some serious beer if you got one right. Ken Houk solved a few as an undergraduate as I recall.

Permalink to Comment

14. NP on February 24, 2009 5:25 AM writes...

I agree with you completely. I once spent a week pouring over a couple hundred reports of NMR data on a relative common synthetic intermediate and was amazed at the discrepancies in the spectroscopic data. Proton chemical shifts were often reported as large ranges (i.e. 2.0-1.9) for a single resonance while 3JCH values were rarely reported even between chiral centers; instead the ubiquitous “multiplet� was used. People forget that the couplings are actually more important than chemical shifts. The latter can vary greatly, but if you can’t explain the couplings the structure has to be wrong. Another example of this attitude is the practice of simply listing chemical shift for proton and carbons without assigning those signals to specific atoms. Just because you have the correct number of carbons and hydrogen doesn’t mean the structure is correct. As NMR techniques become more common, the rigour with which they are applied has diminished.
For a referee, it is hard to disprove a proposed NP structure given the data commonly provided. Halipeptin A was one example where the referees on the 2001 isolation paper proposed what actually turned out to be the correct structure established in the 2002 manuscript. In general though, you have to assume the tabulated spectroscopic data are accurately. Copies of the COSY, HMBC, and HSQC data in the supporting materials are more or less useless, as the resolution is completely inadequate unless extensive expansions are included. Perhaps the solution for natural products structures is to require the raw data (Bruker, Varian, Jeol NMR files) to be provided to the reviewers if not posted on-line at the journal sites. At least the accuracy of the tabulated data could then be checked.

Permalink to Comment

15. Bill on February 24, 2009 2:19 PM writes...

To NP: This would certainly help, if for no other reason that it might allow a reviewer or reader to
reprocess the data better and actually get more reliable data than the original authors. However, the main problem that I have seen, as a reviewer and former journal editor, is that many authors either use default parameters provided by the manufacturer or 'recipes' from a well known book (currently with '200 and more' and more recipes) in acquiring 2D NMR data. The problem is that these seem designed to get 'quick and dirty' spectra for simple test molecules but yield poorly resolved and ambiguous data for many real life problems, particularly where spectral crowding is a problem. I strongly suspect that this was a major factor in the original incorrect structure for hexacyclinol. I am reminded of the old English proverb 'penny wise, pound foolish' since, in attempting to save time, they actually waste time by obtaining inadequate data.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry