A series of indeno[1,2-b]indole-9,10-dione derivatives were synthesized as human casein kinase II (CK2) inhibitors. The most potent inhibitors contained a N(5)-isopropyl substituent on the C-ring. The same series of compounds was found to also inhibit the breast cancer resistance protein ABCG2 but with totally different structure-activity relationships: a N(5)-phenethyl substituent was critical, and additional hydrophobic substituents at position 7 or 8 of the D-ring or a methoxy at phenethyl position ortho or meta also contributed to inhibition. The best ABCG2 inhibitors, such as 4c, 4h, 4i, 4j, and 4k, behaved as very weak inhibitors of CK2, whereas the most potent CK2 inhibitors, such as 4a, 4p, and 4e, displayed limited interaction with ABCG2. It was therefore possible to convert, through suitable substitutions of the indeno[1,2-b]indole-9,10-dione scaffold, potent CK2 inhibitors into selective ABCG2 inhibitors and vice versa. In addition, some of the best ABCG2 inhibitors, which displayed a very low cytotoxicity, thus giving a high therapeutic ratio, and appeared not to be transported, constitute promising candidates for further investigations.

CK2α and CK2α′ are the two isoforms of the catalytic subunit of human protein kinase CK2, an important target for cancer therapy. They have similar, albeit not identical functional and structural properties, and were occasionally reported to be inhibited with distinct efficacies by certain ATP-competitive ligands. Here, we present THN27, an indeno[1,2-b]indole derivative, as a further inhibitor with basal isoform selectivity. The selectivity disappears when measured using CK2α/CK2α′ complexes with CK2β, the regulatory CK2 subunit. Co-crystal structures of THN27 with CK2α and CK2α′ reveal that subtle differences in the conformational variability of the interdomain hinge region are correlated with the observed effect. In the case of CK2α′, a crystallographically problematic protein so far, this comparative structural analysis required the development of an experimental strategy that finally enables atomic resolution structure determinations with ab initio phasing of potentially any ATP-competitive CK2 inhibitor and possibly many non-ATP-competitive ligands as well bound to CK2α′.