From the results of previous experiments with the
From the results of previous experiments with the purvalanol B affinity matrix, it is not clear whether the selective capture of parasite CK1 enzymes is a reflection of their high abundance or rather their affinity for the ligand compared with other kinases present in the lysates. Based on comparisons with the TgCDPK1 , the presence of which was also monitored during column chromatographic fractionations by kinase assay and by Western blot (Fig. 5), TgCK1α is not an unusually abundant kinase in T. gondii tachyzoites. From kinase assays using selective peptides as substrates and from Western blot experiments with multiple enzyme-specific anti-peptide antisera, we conclude that TgCK1α is quantitatively less abundant than TgCDPK1 in tachyzoite lysates. Therefore, we believe that the IC50 values (∼100nM) obtained in dose response titrations of purvalanol B with recombinant or native TgCK1α do indeed reflect an affinity of the inhibitor for the enzyme and further validate the affinity ligand approach for identifying intracellular targets of protein kinase inhibitors. Among the CDK inhibitors tested, purvalanol B and aminopurvalanol show the highest degree of selectivity for the parasite TgCK1α enzyme compared with the TgCK1β isoform or a mammalian CK1 homologue, rat CK1δ (Table 2). Our data are consistent with earlier reports that document the lack of activity of these compounds against mammalian CK1 enzymes as well as against a variety of other non-CDK kinases tested , , , . With respect to the TgCK1α enzyme, selective inhibitors such as purvalanol B and aminopurvalanol share structural features (polar phenyl-ring substituents) that are absent from the relatively inactive purvalanol A (Fig. 7). Against mammalian CDK enzymes, all three compounds show similar levels of potency in the low nM range , . However, available data suggest that parasite CDK may not be a primary target of purvalanol compounds in apicomplexan parasites. Purvalanol A inhibits erythrocytic growth of P. falciparum (IC50 0.5μM) yet has poor activity against P. falciparum PK5/p25 in vitro (IC50 8μM) , . Furthermore, in Midostaurin with affinity chromatography experiments with mammalian cell extracts, the purvalanol B ligand failed to capture CDK enzymes from either T. gondii or P. falciparum. Compared with purvalanol A and aminopurvalanol, purvalanol B is relatively inactive against proliferating cancer cells ,  as well as against P. falciparum  and T. gondii in whole cell assays (Table 2). The lack of in vivo activity of this compound has been attributed to the carboxyl group substituent at R2, which may reduce cell permeability , . In contrast, aminopurvalanol (IC50 0.36μM) inhibits T. gondii growth with an efficacy approaching that of compound 1 (IC50 0.2μM), a coccidiostat and well-characterized inhibitor of PKG , . Considering the activity of aminopurvalanol against TgCK1α in vitro (IC50 40–60nM), it is tempting to propose TgCK1α as a significant intracellular target of this compound in T. gondii. Nevertheless, the possibility of other intracellular targets cannot be excluded. Aminopurvalanol, which has modest activity against erythrocytic P. falciparum (IC50 6μM, ), has been shown to inhibit Pfnek-1 (IC50 200nM ), a homologue of the Aspergillus nidulans NIMA cyclin-independent serine-threonine kinase that is required for entry into mitosis . A more definitive assessment of the relative selectivity of the purvalanol compounds toward cyclin dependent kinases (CDKs) is also needed. This should now be possible with the cloning and characterization of a CDC2 cyclin-dependent kinase from T. gondii, and from Eimeria tenella, as well as at least five distinct CDK enzymes from P. falciparum. Indirubin-3′-monoxime, the only other CDK inhibitor with sub-μM activity against the T. gondii parasite, showed activity against both TgCK1α (IC50 124nM) and TgCDPK1 (IC50 172nM). Inhibition of TgCDPK1 by a structurally unrelated compound KT5926 (IC50 100nM) that blocks parasite attachment to and invasion of host cells, has implicated this enzyme in these processes . Since indirubin-3′-monoxime also blocks parasite attachment (R. Donald, unpublished data), TgCDPK1 is a possible intracellular target for this kinase inhibitor as well. The activity of Compound 1 against TgCDPK1 (IC50 63nM) and TgCK1α (IC50 107nM) raises the possibility that inhibition of these enzymes might potentiate the efficacy attributable to inhibition of the primary PKG target (IC50 0.7nM) , . With the inevitability of drug resistance arising with the use of new therapies, the activity of compounds against multiple essential enzymes is potentially beneficial, assuming that the activity does not extend to important enzymes of the host.