A re-examination of the difluoromethylenesulfonic acid group as a phosphotyrosine mimic for PTP1B inhibition Original Research Article

Protein tyrosine phosphatase 1B (PTP1B) is involved in the down-regulation of insulin signaling and is a well-validated therapeutic target for the treatment of diabetes and obesity. Key to the design of potent inhibitors of PTP1B is a moiety that effectively mimics the phosphate group of the natural phosphotyrosine substrate. Difluoromethylsulfonomethylphenylalanine (F2Smp) is one of the best monoanionic pTyr mimics reported to date. However, the difluoromethylenesulfonic acid (DFMS) group as a phosphate mimic has not been carefully evaluated in the context of a non-peptidyl platform. Here we present a careful examination of the DFMS group as a phosphate mimic. This was achieved by first constructing an analog of a previously reported high affinity, non-peptidyl PTP1B inhibitor (compound 2, IC50 = 8 nM) in which a difluoromethylenephosphonic acid group is replaced with the DFMS moiety (compound 6). We also report the synthesis of its non-fluorinated methylenesulfonic analog (compound 7), as well as two other derivatives in which a distal sulfonamide moiety is replaced with a difluoromethylenesulfonamide group (compounds 8 and 9). Compounds 2 and 6–9 were examined as PTP1B inhibitors. Replacing the distal sulfonamide moiety with a difluoromethylenesulfonamide group had only a modest effect on inhibitor potency. However, compound 6 was approximately a 1000-fold poorer inhibitor than compound 2. Most significantly, inhibition studies with compound 7 and a peptide bearing sulfonomethylphenylalanine revealed that the fluorines have little effect on the potency of the DFMS-bearing inhibitors. This is in contrast to a previous assumption that the fluorines in DFMS-bearing inhibitors contributed significantly to their potency. This may in part explain the large difference in potency between the DFMS and DFMP-bearing compounds. These results also demonstrate that sulfonomethylphenylalanine, a pTyr mimic that is readily constructed, is a relatively good pTyr mimic in comparison to most others that have been reported when examined in the context of the DADE-X-LNH2 peptide platform.