Evidence for two nonidentical drug-interaction sites in the human P-glycoprotein

Human P-glycoprotein (Pgp) confers multidrug resistance to cancer cells by ATP-dependent extrusion of a great many structurally dissimilar hydrophobic compounds. The manner in which Pgp recognizes these different substrates is unknown. The protein shows internal homology between its N- and C-terminal halves, each comprised of six putative transmembrane helices and a consensus ATP binding/utilization site. Photoactive derivatives of certain Pgp substrates specifically label two regions, one on each half of the protein. In this study, using [125I]iodoarylazidoprazosin ([125I]IAAP), a photoactive analog of prazosin, we have demonstrated the presence of two nonidentical drug-interaction sites within Pgp. Taking advantage of a highly susceptible trypsin cleavage site in the linker region of Pgp, we characterized the [125I]IAAP binding to the N- and C-terminal halves. cis(Z)-Flupentixol, a modulator of Pgp function, preferentially increased the affinity of [125I]IAAP for the C-terminal half of the protein (C-site) by reducing the Kd from 20 to 6 nM without changing the labeling or affinity (Kd = 42–46 nM) of the N-terminal half (N-site). Also, the concentration of vinblastine (Pgp substrate) and cyclosporin A (Pgp modulator) required for 50% inhibition of [125I]IAAP binding to the C-site was increased 5- to 6-fold by cis(Z)-flupentixol without any effect on the N-site. In addition, [125I]IAAP binding to the N-site was less susceptible than to C-site to inhibition by vanadate which blocks ATP hydrolysis and drug transport. These data demonstrate the presence of at least two nonidentical substrate interaction sites in Pgp.

Structure-activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming.

We present a general approach to forming structure-activity relationships (SARs). This approach is based on representing chemical structure by atoms and their bond connectivities in combination with the inductive logic programming (ILP) algorithm PROGOL. Existing SAR methods describe chemical structure by using attributes which are general properties of an object. It is not possible to map chemical structure directly to attribute-based descriptions, as such descriptions have no internal organization. A more natural and general way to describe chemical structure is to use a relational description, where the internal construction of the description maps that of the object described. Our atom and bond connectivities representation is a relational description. ILP algorithms can form SARs with relational descriptions. We have tested the relational approach by investigating the SARs of 230 aromatic and heteroaromatic nitro compounds. These compounds had been split previously into two subsets, 188 compounds that were amenable to regression and 42 that were not. For the 188 compounds, a SAR was found that was as accurate as the best statistical or neural network-generated SARs. The PROGOL SAR has the advantages that it did not need the use of any indicator variables handcrafted by an expert, and the generated rules were easily comprehensible. For the 42 compounds, PROGOL formed a SAR that was significantly (P < 0.025) more accurate than linear regression, quadratic regression, and back-propagation. This SAR is based on an automatically generated structural alert for mutagenicity.