In response to valuable comments from Prof. Andrei Kutateladze (University of Denver), we have rechecked the structures of the products obtained by the reactions of 10a−10c with HF/pyridine and found that they were incorrectly assigned. The products were not 11a−11c but (Formula Presented)simply the TBS- and MOM-deprotected products 10a′−10c′. (See the Supporting Information.) Compounds 5a−5c used in the reported biological experiments had been synthesized via the LiOH-mediated domino reaction of 12a−12c to give 13a−13c, removal of the MOM group of 13a−13c, and oxidation of the resulting secondary alcohol. They would not have been available via the HF/ pyridine-mediated reaction of 10a−10c, as we had suggested. The yields of the sequences have been revised (Scheme 1). We reconfirmed that the key HF/pyridine-mediated domino reaction of 10d does indeed proceed to afford 11d in good yield, as we had reported. (See Scheme 2 of the original manuscript.) Taking account of the results shown in Scheme 2C, the applicability of the new HF/pyridine-mediated domino process appears to be limited to substrates with an sp3 substituent at the C17 position. Fortunately, this issue has no impact on the successful synthesis and biological evaluation of the designed molecules 5a−5d or on the discovery of the potent NF-κB inhibitor 5d, which were the key features of the published paper.
All Science Journal Classification (ASJC) codes
- Physical and Theoretical Chemistry
- Organic Chemistry