Total synthesis of decarboxyaltenusin

The total synthesis of decarboxyaltenusin (5’-methoxy-6-methyl-[1,1’-biphenyl]-3,3’,4-triol), a toxin produced by various mold fungi, has been achieved in seven steps in a yield of 31% starting from 4-methylcatechol and 1-bromo-3,5-dimethoxybenzene, where the longest linear sequence consists of five steps. The key reaction was a palladium-catalyzed Suzuki coupling of an aromatic boronate with a brominated resorcin derivative.


Results and Discussion
In a retrosynthetic analysis, we envisioned a Suzuki coupling of two suitably substituted arenes. Silyl protecting groups like the  tert-butyldimethylsilyl group (TBS) were considered appropriate for all projected reaction steps. The boronate moiety 6a was prepared starting with 4-methylcatechol (2), which was initially protected with tert-butyldimethylsilyl chloride in the presence of 4-(dimethylamino)pyridine (DMAP) and imidazole (Scheme 2) according to a published procedure [20]. The thus obtained bis(silylether) 3 was then brominated with N-bromosuccinimide (NBS), where the utilization of acetonitrile as solvent [21] instead of carbon tetrachloride [22] furnished a close to quantitative yield of bromide 5a, though a prolonged reaction time of 72 h had to be accepted. The subsequent formation of boronate 6a was accomplished through a metal-halogen interchange with butyllithium and trapping with 2-isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane [23].
The electrophilic compound suitable for the projected cross coupling was obtained by mono-demethylation of commercially available 1-bromo-3,5-dimethoxybenzene (7) with boron tribromide (Scheme 3). A satisfactory yield of phenol 8 was observed with 0.9 equivalents of the Lewis acid, while the utilization of 1.5 equivalents led to a significant overreaction with the predominant formation of 5-bromobenzene-1,3-diol (59%) together with a smaller amount of the required product 8 (30%). The subsequent protection with the TBS group yielded the known aryl bromide 9a [24] with a 73% yield. The Suzuki coupling of boronate 6a and aryl bromide 9a using palladium acetate and cesium carbonate in the presence of the ligand SPhos [25] yielded biaryl 10a with virtually quantitative yield (98%, Scheme 4). Unfortunately, the removal of nonidentified byproducts and of 9a (which had been used in a 1.2-fold excess) turned out to be very laborious. Moreover, the subsequent deprotection to the natural product 1 could not be achieved sufficiently: After treatment of 10a with tetrabutylammonium fluoride (TBAF), the signals of 1 could be detected in a 1 H NMR spectrum of the crude product, but purification and isolation of the product by column chromatography was not possible -neither with conventional nor with reversed phase methods. It turned out that the high polarity of the triol complicated its separation from other polar side products.
To circumvent this problem, we decided to use a different protection group strategy and to employ hydrogenolytically cleavable benzyl groups. The synthesis of the benzyl-protected boronate was here achieved with a modified strategy including bromination [21] of 4-methylcatechol (2) to the known bromide 4 [26] and subsequent benzyl protection to the bis(benzyl ether) 5b using standard conditions (Scheme 2) [27]. The preparation of boronate 6b applying the conditions used for the silylated substrate 6a (vide supra) led to a mediocre 44% yield, but   [7]; b data published by Wang et al. [2]; c the data are given in ascending order. Assignments in the original papers are in agreement with those given for the synthesized product (except footnote e); d data published by Xiao et al. [7]; e the assignment in the original paper is: 143.5: C-4, 144.5: C-3, and 144.8: C-1'; f a superscript letter is included after these numbers in the original paper, but a corresponding footnote is missing. It can be assumed that the assignment of these signals had been considered questionable. Decarboxyaltenusin (1) was screened for toxicity towards human HeLa cells but proved nontoxic at biologically relevant concentrations and showed an LD 50 value of above 50 μM. This screening was performed by measuring the cell viability using an MTT assay, where the viability is assessed based upon the reduction of the yellow tetrazolium MTT [3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide] by metabolically active and hence viable cells. The resulting intracellular purple formazan was quantified spectrophotometrically. The cell viability was calculated based on an untreated control. The cell viability cut-off was <70%. At a 0.5 μM concentration of decarboxyaltenusin (1), it was 103% ± 1%, at 5 μM it was 94% ± 1.4%, and at 50 μM 92% ± 1%.

Conclusion
The total synthesis of decarboxyaltenusin (1) was achieved in seven steps in a yield of 31% starting from 4-methylcatechol (2) and 1-bromo-3,5-dimethoxybenzene (7), where the longest linear sequence has consisted of five steps. Decarboxyaltenusin turned out to be nontoxic towards human HeLa cells.

Supporting Information
Supporting Information File 1 Experimental procedures and NMR spectra of all new compounds and of decarboxyaltenusin (1).