Reply to 'Corrected structure of mirin, a small-molecule inhibitor of the Mre11–Rad50–Nbs1 complex'

Shaun Peterson et al. reply:

We recently identified mirin from a forward genetic screen of a 10,000-compound library from Chembridge Corporation¹. The structure of mirin was assigned by the Chembridge database to be 6-(4-hydroxyphenyl)-2-thioxo-2,3-dihydro-4(1H)-pyrimidinone (Chembridge number 6451023, compound 1; note that for clarity we use the same compound numbering scheme as Garner, Pletnev and Eastman). Following the original screen, we obtained more compound from Chembridge to confirm the original finding and performed additional characterization of the inhibitor. During the course of our studies, Chembridge ran out of compound number 6451023. We identified another source for 6-(4-hydroxyphenyl)-2-thioxo-2,3-dihydro-4(1H)-pyrimidinone (Scientific Exchange, Inc.) and completed our studies with this compound. Notably, the compounds from Chembridge and Scientific Exchange had the same orange-red color and displayed similar activity in our assays. We did not realize until later that both companies had obtained this compound from the same source.

Using the structure assigned by Chembridge, we synthesized compound 1 at the Memorial Sloan-Kettering Cancer Center Organic Synthesis facility and obtained a white product that was inactive in a cellular assay. At approximately the same time, we were contacted by Garner, Pletnev and Eastman, who had independently synthesized compound 1 and found it to be inactive. They went further and determined the structure of the active compound. We obtained a sample from Garner, Pletnev and Eastman of the active compound, Z-5-(4-hydroxybenzylidene)-2-imino-1,3-thiazolidin-4-one (compound 3), and independently synthesized Z-5-(4-hydroxybenzylidene)-2-thioxo-1,3-imidazolidin-4-one (compound 2) described in their correspondence using selective methodology² that avoids concomitant formation of compounds 2 and 3 (Scheme 1).

We envisioned that synthesis of compounds 2 and 3 could readily be achieved starting from 2-thioxoimidazolidin-4-one (Scheme 1a) and 2-iminothiazolidin-4-one (Scheme 1b), respectively. Given the expected reactivity of the methylene center, we were confident about the regiochemical outcome of the condensation of these two starting materials with 4-hydroxybenzaldehyde. The products of these two transformations differed in proton NMR and in their color and biological activity. Indeed, upon condensation with 4-hydroxybenzaldehyde in the presence of sodium acetate in refluxing acetic acid, 2-thioxoimidazolidin-4-one and 2-iminothiazolidin-4-one gave a clay-yellow compound 2 and an orange-red compound 3, respectively.

We then compared the activity of the following compounds in cell-free extracts: (i) original compound from Scientific Exchange (compound 3), (ii) compound 3 from Garner, Pletnev and Eastman and (iii) compounds 1, 2 and 3 from our synthesis. We found that compound 3 independently synthesized by Garner, Pletnev and Eastman and by us, and the original compound from Scientific Exchange, all had a similar activity in cell-free extracts. In contrast, compounds 1 and 2 were inactive in the assay.

We therefore fully agree with the conclusion of Garner, Pletnev and Eastman that Z-5-(4-hydroxybenzylidene)-2-imino-1,3-thiazolidin-4-one (compound 3) is the correct structure of mirin, and we would like to thank them for identifying the correct structure.

(a) Synthesis of Z-5-(4-hydroxybenzylidene)-2-thioxo-1,3-imidazolidin-4-one (2). (b) Synthesis of Z-5-(4-hydroxybenzylidene)-2-imino-1,3-thiazolidin-4-one, correctly assigned as mirin (3).