Seeman surveys scientists' esteem for structural determination of strychine: Robinson or Woodward?

 Citation:

 Seeman, J. I.; House, M. C. "For Its Size, the Most Complex Natural Product Known." Who Deserves Credit for Determining the Structure of Strychnine? ACS Cent. Sci. 2022, 8, 672-681. 

Before reading the rest of this post, read the following question:

Professor A published the correct structure for a natural product. 

Six months later, Professor A characterized his correct structure as “inadequate” and “questionable” and stated “feasible alternatives of a more complicated character have been devised.” He then published an alternative (and wrong) structure which he characterized as “the best working hypothesis at present available.” 

Shortly thereafter, Professor A changed his mind again, writing “we revert, as the best hypothesis to guide future work, to an earlier [and correct] structure... Several slight modifications of this structure are feasible, and these have special advantages and disadvantages, which must be discussed at a later date, especially since it is probable that crucial experimental tests can be devised.” 

Two months later, Professor B published the correct structure. Professor B was unambiguous in his conclusion, saying, “The new evidence completes the inferential proof” for what is the correct structure. Following Professor B’s suggestion, the relevant scientific community, including Professor A, considered the structure determination solved. 

Who should be recognized as the discoverer of the correct structure? 

Professor A, Professor B, or both? Settle on an answer now!

Summary Figure: 

 

Background

Back in ye olden days, when analytical techniques like NMR, mass spectrometry, and X-ray diffraction didn't exist, figuring out the structure of an unknown compound was immensely difficult. The 1927 and 1928 Nobel prizes were awarded, back to back, to Heinrich Wieland and Adolf Windaus for their work identifying the structures of bile acids and steroids, respectively. 

 Without many ways of "looking" at the structure of a substance, one of the key methods to figure out the structure was by chemical degradation. This meant reacting the compound in a known way (the example Seeman uses in another work is methylation with methyl iodide). Then, if the product was different from the starting material, there was a functional group that was able to react via that degradation method. For example, alkaloids with a tertiary amine (like strychnine) are able to be methylated at the nitrogen. This new compound could be compared via elemental analysis, melting point, or specific functional group tests (or IR and UV/Vis, once those were invented)

 

For more information on this method, I recommend Seeman's other article "On the Relationship between Classical Structural Determination and Retrosynthetic Analysis/Total Synthesis" (https://onlinelibrary.wiley.com/doi/abs/10.1002/ijch.201700079 for a brief summary or Roald Hoffmann's book "Classical Methods in Structure Elucidation in Natural Products" (https://onlinelibrary.wiley.com/doi/book/10.1002/9783906390819) for a longer historical examination with a lot of examples. 

In contrast, the other non-spectroscopic way to determine the structure of a compound is via synthesis: build the molecule up from smaller fragments via known chemical reactions, then compare the physical properties of the product to the target. This is effectively the reverse of chemical degradation. The major advantage of total synthesis is that it is much less ambiguous: if you know the starting material you put in the reaction, and the expected reaction outcome, and the product's physical properties precisely match the target, you can be as close to certain that you have made the target compound. To this day, total synthesis continues to be used as a means of identifying the structure of compounds that may have been misassigned.

(For examples, see: https://macmillan.princeton.edu/wp-content/uploads/structure-misassignment-gp-meeting-JML.pdf) 

So back in the ancient times, structural determination was the state of the art for identifying the structure of a natural product. Total synthesis relies on at least some prerequisite understanding of the structure to know what you're trying to make.

In the first half (but second quarter) of the 20th century, strychnine was an interesting target for chemists due to its complexity, representing the entire alkaloid class. Strychnine was somewhat famous, featured as a poison in mystery novels and occasionally touted as a potential medicine. 

Robert Robinson was one of the preeminent organic chemists of the time (the Robinson annulation is named after him), and would win the Nobel prize (solo) in 1947 for his work on the structure of alkaloids. A large part of this included attempts to determine the structure of strychnine via chemical degradation, which he had consistently published in 40+ papers since 1910. For more information, see another one of Seeman's articles, "From Decades to Minutes: Steps Toward the Structure of Strychnine 1910–1948 and the Application of Todays Technology" (https://onlinelibrary.wiley.com/doi/epdf/10.1002/anie.201916566) 

Robinson, as Seeman repeatedly points out, was a proponent of publishing hypotheses to engage in discussion with the broader chemical community. Over this time period, he published nearly 20 plausible structures of strychine, each supported by experimental evidence but not necessarily definitive. For example, in February 1946 he published a hypothesized structure of strychnine but included the statement "The available evidence bearing on the problem is insufficient and a decision cannot yet be made with confidence." Two weeks later, in another publication on the subject he stated, "we now believe that the [proposed] structure [I] interprets the whole behavior of strychnine better than any other." 

In hindsight, these statements were in fact describing the true structure of strychnine- Robinson had in fact proposed the correct structure. However, the evidence for the structure remained as suspect and circumstantial as ever, and could not be satisfactorily or definitively proven.  

Woodward, meanwhile, published a single (incredibly short) paper providing the structure of strychnine,  stating the famous phrase "the new evidence completes the inferential proof for a particular expression of strychnine." Woodward would not publish the complete synthesis of strychnine for several years, but this paper (and a subsequent follow up in 1948) concluded the bulk of the debate on the structure of strychine. 

Hilariously, the structure of strychnine would be solved by XRD in 1950. 

What was the question again?

There is no doubting that the effort that went into solving the structure of strychnine was not wasted. The purpose was not necessarily to unambigiously determine the structure for a utilitarian purpose. The efforts of Robinson and Woodward pushed the field of organic chemistry forward, providing new technologies that continue to find use in modern methods. Robinson's efforts constituted some of the most advanced use of degradation for structural elucidation, and his hypotheses about connecting the structure of alkaloids to their biosynthesis were ahead of their time. Woodward was the greatest total synthesis chemist of his era, and if not for his early death he likely would have received a second Nobel prize. 

But who deserves the credit for being first? 

In my opinion, the reason the debate is interesting is because it represents a two different schools of thought around publication. 

The first school of thought is that publication is designed to be a formal means of communication with the scientific community. When multiple research groups are working on similar (or even the same) problem, some degree of collaboration or communication is likely to help push projects forward. While it is possible to directly communicate with each other (via letters, conferences, in person meetings, seminars, and nowadays email), this does require a personal connection that may not be present. That favors a small clique of researchers who studied with each other, freezing out other researchers. Furthermore, this sort of personal correspondence can lead to fraud, as is famously the case with Jan Hendrik Schon's unpublished fradulent recipe for sputtering aluminum oxide, that was unable to be reproduced.  Therefore, it would be better to publish your hypotheses often, even if they are not certain, because it provides regular updates to the rest of the field about the current state of the art. This also provides an outlet for negative data, as it furthers ongoing research without needing to state a more formal proof. 

The second school of thought is that publication is designed to be a permanent record of results from the scientific community. Publications should represent well-executed, definitive results that other scientists can trust and then build upon. When scientists communicate, their records should be unambiguous and complete, allowing each paper to stand on its own merits. This prevents overloading the publication record with large quantities of dubious-quality work. In my opinion, this second school of thought has come to dominate modern scientific discourse. Each paper must tell a complete story, with rigorous, high-quality work that provides persuasive evidence that its goals have been achieved. 

So what do most people think?

Seeman and House sent this question to a large number of faculty (both inside and outside chemistry departments) to see how people involved in cutting-edge research view credit distribution. Overall, about 60% of respondents said both professors deserved some credit in the structural elucidation of strychnine. However, the authors of the survey noted this could also represent the path of least resistance- it is obvious that both have contributed in some way to the discovery of the correct structure, so answering "both" avoids making a difficult decision. Within non-chemist respondents, about 34% of respondents credited Prof. A (Robinson), while 10% credited Prof. B (Woodward) [the rest answered both]. Within chemist respondents, this was flipped: about 21% credited Prof. B (Woodward) while 14% credited Prof. A (Robinson) [again, the rest answered both]. The authors hypothesize that either (A) chemists distribute credit differently than other disciplines, with a chemists potentially placing higher value on not changing opinion, or (B) chemists understand the potential tools Prof. A and B were using better, and therefore judge more harshly due to their own experiences. 

I asked this question to two groups of chemists, each with ~10 people responding. In the first group, 7 people responded "both", 3 people responded "B" (Woodward) and no one responded "A" (Robinson). In the second group, 8 people responded "both", 2 people responded "B", and 2 people responded "A". I find it very amusing that this statistically insignificant sample size followed the study results fairly closely. 

The two authors of the survey gave their thoughts at the end of the paper, both of whom assign Woodward with the bulk of the credit.  

I personally credit Woodward with the first unambiguous structural determination of strychnine. I believe that Robinson's efforts were valuable towards the development of synthetic organic chemistry. However, the ultimate goal of any project is to come to a conclusive end. Woodward ended the debate over the structure of strychnine. 

Who do you think deserves the credit for discovering the structure of strychnine?