Total Synthesis of a Reported Fluorometabolite from Streptomyces sp. TC1 Indicates an Incorrect Assignment. The Isolated Compound Did Not Contain Fluorine

Abstract

3,5-Di-tert-butyl-4-fluorophenylpropionic acid (1) was recently reported as a natural product from Streptomyces sp. TC1. This was a notable disclosure because fluorinated natural products are exceedingly rare, and in this case it suggested that the bacterium had the capacity to mediate an enzymatic aryl fluorination reaction. However, a synthesis of the putative metabolite 1 demonstrates that the spectroscopic data are inconsistent with the proposed structure. There is no evidence that the isolated compound contained a fluorine atom.

The value of universally available raw NMR data for transparency, reproducibility, and integrity in natural product research

“…TC1,289 which suggested the existence of an enzyme capable of mediating an aryl fluorination reaction. This report attracted the attention of two different groups, who via synthesis of the putative fluorinated natural product, and based on the analysis of 1 H, 19 F and 13 C NMR spectra, both demonstrated the absence of fluorine in the secondary metabolite 290,291. While those follow up studies essentially refute the initial study, perhaps a more thorough analysis of the NMR data at peer review, including examination of raw NMR data, would have prevented the need for such research.…”

Identification of a fluorometabolite from Streptomyces sp. MA37: (2R3S4S)-5-fluoro-2,3,4-trihydroxypentanoic acid

“…This is the rst secure identication of a new uorinated natural product since 1998. [26][27][28][29] Experimental section 19…”

Structure of a Putative Fluorinated Natural Product from Streptomyces sp. TC1

“…This intriguingly suggested the existence of an unprecedented biocatalytic system that achieves aromatic fluorination, a finding challenged by O'Hagan and co-workers, who recently synthesized 1 and showed that the spectroscopic data of natural and synthetic 1 did not match. 14 The actual structure of the initially isolated metabolite, however, was not further investigated in this work. In parallel to O'Hagan's group, we set out to elucidate the true identity of the original compound, triggered by major inconsistencies in the spectroscopic data of Marimuthu et al 13 Excited by the yet unprecedented structure of 1 ( Figure 1A) presented in the work of Marimuthu et al, 13 we had a close look at the reported analytical data of this secondary metabolite.…”

“…Equally disturbing was the provided 19 F NMR spectrum of 1, with an extremely broad signal (approximately 40 ppm), which is not consistent with the common sharp signals of low molecular weight organofluorines. 14 The summation of these inconsistencies led us to have a careful look at the original 1 H NMR spectrum (Figure 2). At first sight the NMR data reported in Table 1 of the original manuscript seemed to match the provided spectrum (not the elucidated structure), with the two aromatic protons a1 and a2, the two signals b1 and c1 representing the methylene units, and two signals for the methyl groups at two tert-butyl functionalities.…”