Combinatorial Pharmacophore Modeling of Organic Cation Transporter 2 (OCT2) Inhibitors: Insights into Multiple Inhibitory Mechanisms

Organic
cation transporter 2 (OCT2) is responsible for the entry
step of many drugs in renal elimination, of which the changing activity
may cause unwanted drug–drug interactions (DDIs). To develop
drugs with favorable safety profile and provide instruction for rational
clinical drug administration, it is of great interest to investigate
the multiple mechanisms of OCT2 inhibition. In this study, we designed
a combinatorial scheme to screen the optimum combination of pharmacophores
from a pool of hypotheses established based on 162 OCT2 inhibitors.
Among them, one single pharmacophore hypothesis represents a potential
binding mode that may account for one unique inhibitory mechanism,
and the obtained pharmacophore combination describes the multimechanisms
of OCT2 inhibition. The final model consists of four individual pharmacophores,
i.e., DHPR18, APR2, PRR5 and HHR4. Given a query ligand, it is considered
as an inhibitor if it matches at least one of the hypotheses, or a
noninhibitor if it fails to match any of four hypotheses. Our combinatorial
pharmacophore model performs reasonably well to discriminate inhibitors
and noninhibitors, yielding an overall accuracy around 0.70 for a
test set containing 81 OCT2 inhibitors and 218 noninhibitors. Intriguingly,
we found that the number of matched hypotheses was positively correlated
with inhibition rate, which coincides with the pharmacophore modeling
result of P-gp substrate binding. Further analysis suggested that
the hypothesis PRR5 was responsible for competitive inhibition of
OCT2, and other hypotheses were important for interaction between
the inhibitor and OCT2. In light of the results, a hypothetical model
for inhibiting transporting mediated by OCT2 was proposed.