My first reading of this paper left the impression that it was an excellent illustration of the power of NMR chemical shift predictions being used to confirm or question the validity of a proposed structure. In this case the reported data for the alkaloid lamellarin did not appear to be consistent with those calculated for the proposed structure. This then led to the use of the Structure Elucidator package to generate a suite of structures more consistent with the reported data. The best candidate turns out to be lamellarin F, first reported in 1988 (J. Org. Chem. 53, 1988, 4570-74). Surprisingly, this fact is not mentioned in this paper. If it had been recognized, then a comparison of the NMR data published for lamellarin F with that for lamellarin would have shown a high degree of consistency, after allowance was made for the different solvents used in obtaining the spectra (DMSO vs CDCl3). Only two values appeared to be markedly different, those for C6 and C6a. Interestingly, from Fig 2 it can be seen that the calculated value for C6a is nearer to that reported for lamellarin F than that claimed for lamellarin gamma. The situation with data for C6 is more problematic. Another anomaly is the reported mp of 273-6 for lamellarin F, but 163-5 for lamellarin gamma. However, lamellarin gamma is reported as a white solid, with no indication of any solvent being used for recrystallisation, whereas lamellarin F was crystallized from DMSO.
The structure for lamellarin gamma should never have been accepted for publication, as it is quite clear (as pointed out in this paper) that there was a lack of identity in chemical shift data for C12 and C16, which should have been expected for the local symmetry in the structure proposed. Interestingly, the authors of the lamellarin gamma paper pointed out the similarity of the UV data for their compound with those for lamellarin F, but made no reference to the similarities in chemical shift data.
In the presentation of this paper, the compound should be referred to as lamellarin gamma (greek sign), not lamellarin g, in order to make comparison with the original more obvious.
PS. This feedback box should support the input of Greek Characters
There are a couple of characters in the comments that are missing. This is because the comments box does not support Greek Characters. We should consider an HTML editor for reviewers.
I think John is looking at the article in Firefox.In Firefox the gamma in the article shows as a g but it looks fine in Internet Explorer. Browser dependent font display unfortunately.