Development of novel analogs of the TRi-1 and TRi-2 selenoprotein thioredoxin reductase inhibitors with initial assessment of their cytotoxicity profiles.

Abstract

The human selenoprotein thioredoxin reductases are encoded and expressed as separate cytosolic (TXNRD1) and mitochondrial (TXNRD2) isoforms, with both having been suggested as potential anticancer drug targets. The TRi-1 compound was recently shown to preferentially inhibit TXNRD1 in cells, while the TRi-2 compound seems to target both isoforms in a cellular context, albeit to different extent. Attempting to evaluate whether TXNRD1 or TXNRD2 inhibition most closely correlates with cytotoxicity we here synthesized several analogs of both TRi-1 and TRi-2, including triphenyl phosphonium derivatives designed to accumulate in mitochondria. We evaluated 11 compounds in comparison with TRi-1 and TRi-2 with regards to inhibition of TXNRD1 and TXNRD2 in pure enzyme assays, and their cytotoxicity profiles towards human lung adenocarcinoma A549 cells, with constitutively high NRF2 activity and thus potent antioxidant defense. Human squamous cell carcinoma FaDu cells with lower NRF2 and lower TXNRD1 activity were much more sensitive to the compounds. The results strengthen the notion that compound-derived inhibition of either TXNRD1 or TXNRD2 can yield cytotoxicity in human cancer cells. Comparing two pairs of matched inhibitor scaffolds for the effects of adding a triphenyl phosphonium group, we found that this moiety ensured minimal inhibition of cellular cytosolic TXNRD1 activity, as assessed using the specific RX1 activity probe, while maintaining cytotoxicity, which was thus likely involving targeting of TXNRD2 in the mitochondria. Our results represent a blueprint for initial evaluations of novel small molecule inhibitors of TXNRD1 and TXNRD2, correlating such inhibition of pure enzymes as well as in cells in relation to their cytotoxicity.